Luminous Landscape Forum

Site & Board Matters => About This Site => Topic started by: JohnBrew on September 26, 2009, 09:16:32 am

Title: Rant 23
Post by: JohnBrew on September 26, 2009, 09:16:32 am
Michael - hoisted on his own petard.
Title: Rant 23
Post by: michael on September 26, 2009, 09:58:36 am
Yes – and sometimes it's necessary.

Michael
Title: Rant 23
Post by: EsbenHR on September 26, 2009, 11:04:09 am
The way I see it, the quality of "100% crops" is the only pressure the manufacturers gets to keep the pixels at a reasonable size. The only ones who could possibly put a 1um^2 pixel on a fullframe DSLR to good use are the people in the marketing department.

To keep them at bay we need to look at the pixels and tell them "your 864 megapixels stinks worse than my breakfeast".

If they can not make their pixels have a high quality, stop making them. Bin them at high ISO or something, but do not pretend that the engineers won the battle with the PR department.
Title: Rant 23
Post by: ErikKaffehr on September 26, 2009, 11:27:44 am
Well,

We can post bitmaps on the net, prints are a bit hard. It's also very difficult to compare images, even prints, because much depends on viewing distance, contrast and other factors. My suggestion is:

Shooting under controlled conditions and making reproducible images is helpful.
Best way is to release raw files, that way anyone can print the images to taste and draw conclusions.

Best regards
Erik



Quote from: EsbenHR
The way I see it, the quality of "100% crops" is the only pressure the manufacturers gets to keep the pixels at a reasonable size. The only ones who could possibly put a 1um^2 pixel on a fullframe DSLR to good use are the people in the marketing department.

To keep them at bay we need to look at the pixels and tell them "your 864 megapixels stinks worse than my breakfeast".

If they can not make their pixels have a high quality, stop making them. Bin them at high ISO or something, but do not pretend that the engineers won the battle with the PR department.
Title: Rant 23
Post by: timparkin on September 26, 2009, 01:35:27 pm
Quote from: michael
Yes – and sometimes it's necessary.

Michael

But! But! -- If I get you right then it depends on what printer, what viewing conditions, what paper, what processing you apply to the image, etc, etc..

I agree that 100% doesn't tell the whole story but it *informs* more than anything else as long as you know what you are looking at ..

I *know* from looking at a 24TSEmk2 image from a 5dmk2 at 1005 that it will look sharp in the corners (sharper than a 17-40). Would I be able to get this information from a 10x8 print on a dye sub printer? No! Will I get this information from a 40x60" print from a 5Dmk2? probably..  

What you definitely have with a 100% view is *all of the information that you have* - it's up to you to interpret what this means. In a print, you have *one interpretation* of the results. If this is your interpretation then it is significant but if it isn't your interpretation then it's fairly useless...

Anyway - just defending the *pointless* practise of using 100% images as part of purchase consideration (for instance)

Tim

http://www.timparkin.co.uk (http://www.timparkin.co.uk)
Title: Rant 23
Post by: michael on September 26, 2009, 01:51:53 pm
No, the point isn't to turn the print into a means of technical measurement. Anything but.

What I was trying to get at is that it's what ends up on the final print (however it's made) that is the ultimate arbiter. If I can see the difference in the final work of art, then the difference is meaningful. If I can't, then it may be there but of only academic interest.

Michael
Title: Rant 23
Post by: ErikKaffehr on September 26, 2009, 02:15:53 pm
Michael,

Do you have any way to describe what you can see in a print? I see only two solutions:

1) Distribute the "RAW"
2) Make "objective" measurements

Best regards
Erik

Quote from: michael
No, the point isn't to turn the print into a means of technical measurement. Anything but.

What I was trying to get at is that it's what ends up on the final print (however it's made) that is the ultimate arbiter. If I can see the difference in the final work of art, then the difference is meaningful. If I can't, then it may be there but of only academic interest.

Michael
Title: Rant 23
Post by: John Camp on September 26, 2009, 02:37:38 pm
Another consideration:

More and more photographs -- even good photographs -- are "monitor only." Advertising and news people are jamming more and more photographs onto the net...one reason that newspapers are failing. "Wall art" photographs are exceptions, and a tiny percentage of the photography business. That means that we are developing a new monitor-based aesthetic, and somebody, somewhere, should really be working hard on a system to maximize the quality of monitor-based photographs. I think we are not far from the day when most monitors will look like the current Mac cinema displays, and that putting up an 18x24 inch photo will be perfectly feasible. While you can't tell what you need to know about a print photo from looking at a monitor, the reverse may also be true -- that the camera and technique qualities that make a good print may not make a good monitor photo.

More and more top-end fine-art photographers seem to be displaying their photos in back-lit boxes, almost as if they are driving toward monitor displays. The whole aesthetic of internally-lit photos is shifted when the skies can actually be bright, as opposed to light-colored.

I know this isn't what MR is doing, but more and more people are...  

Title: Rant 23
Post by: ErikKaffehr on September 26, 2009, 02:58:19 pm
Hi,

Monitors are like 2 Mega Pixel devices. So if we shoot for monitors everything will do..., The person behind the lens still matters!

Best regards
Erik


Quote from: John Camp
Another consideration:

More and more photographs -- even good photographs -- are "monitor only." Advertising and news people are jamming more and more photographs onto the net...one reason that newspapers are failing. "Wall art" photographs are exceptions, and a tiny percentage of the photography business. That means that we are developing a new monitor-based aesthetic, and somebody, somewhere, should really be working hard on a system to maximize the quality of monitor-based photographs. I think we are not far from the day when most monitors will look like the current Mac cinema displays, and that putting up an 18x24 inch photo will be perfectly feasible. While you can't tell what you need to know about a print photo from looking at a monitor, the reverse may also be true -- that the camera and technique qualities that make a good print may not make a good monitor photo.

More and more top-end fine-art photographers seem to be displaying their photos in back-lit boxes, almost as if they are driving toward monitor displays. The whole aesthetic of internally-lit photos is shifted when the skies can actually be bright, as opposed to light-colored.

I know this isn't what MR is doing, but more and more people are...
Title: Rant 23
Post by: Schewe on September 26, 2009, 03:02:19 pm
Quote from: John Camp
Another consideration:

Well, lets see, a P65+ back will give you 60.4 MP and 8976 x 6724 pixels. Currently, my 30" display can show 2560 x 1600 pixels. That means to show my full frame crop on the display I would have to downsample it by 3.5x.

That's the fallacy of that argument...even a 6.3 MP at 3072 x 2048 would need downsampling to display on a 30" display.

Ok, my iPhone at 1.9 MP and 1600 x 1200 pixels would be insufficient. Shucks...I guess I won't be able to get by using my iPhone...

Again, Michael's point is that evaluating an image at 1:1 tells you nothing about what the image will look like printed nor downsampled for display/web use. Image sharpness and noise get substantially impacted by any size interpolation–fact is, downsampling is a great way of reducing noise.

If you have an LCD the adds are you'll be seeing close to 100 PPI. If you print out at 300 PPI, what you are seeing on the display is 3X the size when you look at the image at 100%. If you reduce the image even to 50% the image will be too large on the display and you'll be looking at the image with a low resolution. The only way to truly evaluate an image for print is to print it on the printer and medium you'll be using (with the proper image sharpening of course!)

And, until we get displays in the range of 180-220 PPI, the computer display will simply be a very poor gauge for the detail you'll be able to expect in print. Smaller displays like the iPhone actually do have 180 PPI which makes them look really good. heck, color management is far easier and more advanced than image detail...
Title: Rant 23
Post by: John Camp on September 26, 2009, 03:14:26 pm
Quote from: Schewe
Well, lets see, a P65+ back will give you 60.4 MP and 8976 x 6724 pixels. Currently, my 30" display can show 2560 x 1600 pixels. That means to show my full frame crop on the display I would have to downsample it by 3.5x.

That's the fallacy of that argument...even a 6.3 MP at 3072 x 2048 would need downsampling to display on a 30" display.

YOUR error is to think I'm making an argument. I'm not. I'm describing a situation. A newspaper provides professional journalists with professional editors making carefully judged selections of the important news of the day. They're failing. In the meantime, some blogger who's pulling his opinions out of a place where the sun don't shine, becomes famous. That's a situation, not an argument.

A cell phone won't make the best possible video display shots because the lenses and responses are crappy.
You talk about your P65+, I'd like to see you take your P65+ onto a photo platform in the endzone of a football stadium and zoom from a full view to the player's face down on the field...you can't, because you don't have the $70,000 lens needed to do that.

The point being...maybe you've got the wrong equipment for the developing aesthetic. Maybe you've got the 21st century equivalent of the best glass-plate negative camera after the arrival of film. Maybe what we need are highly refined 4mp cameras that will shoot both video clips and stills with very strong tracking zoom lenses, at ISO104,000, because that's what monitors need. And you don't have one...but you may get one.

JC

Title: Rant 23
Post by: Daniel Browning on September 26, 2009, 04:03:44 pm
I liked the rant, Michael. Thanks for posting it.

"I alternately chuckle and get steamed up when I read someone on a web forum either condemn or praise a camera or lens based on a web images. This is utter nonsense."

Agreed; there are very few lens/camera issues that can be evaluated from such small images: bokeh, distortion, flare, perhaps a few others.
Title: Rant 23
Post by: pete_truman on September 26, 2009, 04:09:00 pm
A refreshing rant and one I entirely agree with. Thanks.

Amusing to see how this is now being debated... chuckle  
Title: Rant 23
Post by: Schewe on September 26, 2009, 04:52:25 pm
Quote from: John Camp
A cell phone won't make the best possible video display shots because the lenses and responses are crappy.
You talk about your P65+, I'd like to see you take your P65+ onto a photo platform in the endzone of a football stadium and zoom from a full view to the player's face down on the field...you can't, because you don't have the $70,000 lens needed to do that.

Oddly enough if all you need is a 4"x5" halftone output, the iPhone (if the original is well shot) can produce an image roughly equal to the 60 MP P65+ back. From the end zone, perhaps the optics of the Mamyia 6x4.5 system is limited...could shoot from the sidelines though.

I always get a kick out of photographers talking "video". Different beast, different skill set. You don't shoot video with the expectation of successfully pulling great still frames. Nor can a videographer really appreciate a still, unmoving frame. Two different art forms.
Title: Rant 23
Post by: laughingbear on September 26, 2009, 06:58:34 pm
Greetings from Ireland,

I experience a great deal of frustration in trying to create pictures for web, one reason is that I did not manage to get across the point of "This is how the print will look", at all, and frankly I gave up on it.

I always wondered about Alain's 600 pixel pics and was tempted to believe I need a much bigger size to represent my work, but as a matter of fact, in this case, bigger ain't better.

Someone who has printed himself a lot, well he probably has a better judgement of what this shiteRGB will look like on Hahnemuehle Bamboo ____fill in paper of your choice, but that is only because he has printed himself a lot, which is not true for the average client who will purchase a print of course.

As for stereotype 100% pixel show me your noodle attitudes all over internet foras, well, yeah, it is really amusing isn't it.

I guess the question remains, what can we do to showcase final print quality in better ways with the given technology? Anything at all?

Posting a A4 size demo of a possible 40"x60" print on the same paper probably has more of an impact than anything that even the very best 30" screens can reproduce.

As for pixelpeepers, I suggest a can of Hahnemuehle protective spray to be applied three times, let screen dry for 10 minutes and turn 90 degress before you apply the next layer, onto the 129.99 Dollar wallmart HD ready before judging final sharpening. ....ahem, no offence, just could not resist....
Title: Rant 23
Post by: Steven Draper on September 26, 2009, 07:35:55 pm
It is amazing how many articles appear on this website just as I start to consider the very subject.

Recently an Epson 7900 arrived in my Studio and I have worked very hard on my prints, pushing some 35 mm file (D2x and D700) to twice their previous size (24x16 being the largest I've gone so far.) The results have made me very happy and are a step or two up from the B9180 (for my eyes anyway)

But onto the topic. Last weekend I was part of a studio tour in Prince Edward County, Ontario with many people visiting my studio and viewing my prints. They were absolutely amazed when I showed them 100% crops on my screen and then the print, they could not believe how ugly 100% screen crops can look. Some folks did pixel peep the prints, (and gave good comments) but most just let themselves become immersed in the experience of viewing reasonable sized prints. No-one said that is a B9180 print and that one an Epson, although the size generally would give them away.

And for that very reason I am about to revamp my websites and remove my art most of my art images and produce just an online catalogue, leaving my website for my commercial work, much of which is destined for the web anyway.


Steven





Title: Rant 23
Post by: feppe on September 26, 2009, 08:08:39 pm
While I agree fully with Michael's point, it misses the much more important rant: what on earth is the point of 10+ megapixel point&shoot cameras, 20+ megapixel dSLRs and 50+ megapixel MFDBs when 99.999999% of all photography is only ever seen on the screen? I'm fully aware LL is in large part for and of fine art photographers, but out in the real world people don't do prints beyond a 4x5 (that's inches, not feet). And that's a dying art as well. Even those who hang pictures are moving towards digital picture frames. Printed magazines are going extinct, and Kindle 5, Sony Reader 7 or disposable digital paper will kill the last few which remain. A slow, painful uncalibrated 72 dpi sRGB screen at a time.

Sure, we'll have affordable and large 300dpi screens eventually. Then again, current trend seems to be pushing video and 3D stills rather than screen resolution, so the 5DII and P65 owners might be out of luck for quite a while.

Now, I have a 80cmx80cm MF shot I made on my wall and will have more after I move, and like a good print as much as most others here. I even go to exhibitions. But I and especially the average LL forumite is hardly the standard, even in the industry. Just how many photographers get their living from prints, and how many of those out of prints larger than a magazine spread? 20%? 10%? 1?

It's not only us, though. Canon, Nikon, Sony, Hasselblad, Phase, etc. need to wake up. I've ranted elsewhere  (http://luminous-landscape.com/forum/index.php?showtopic=37289)about the lack of innovation in the industry which I won't rehash here.

My point here is that what you can show on the web is all that matters in 99.99999% of the cases. If you can't produce work which looks stellar on a 72dpi screen, you're shit. Unfortunately the photographer has almost zero control of how their work is seen on an uncalibrated 5-year-old LCD through color mismanaged Internet Explorer. I presume that's why we have so many ghastly HDR, prickly oversharpened, surypy oversaturated, morbid draganized, overvibranciated and too much frigging local contrast enhancement*, making your average day even on most fine art sites like a nightmarish stroll through a kitschy version of Alice In Wonderland - but perhaps I give too much credit to the photographers thinking it's not their fault.

For photographers this means that we should urge hardware manufacturers and software developers to move to aRGB (or wider) color standard, make monitors and cameras factory-calibrated and "good enough" (no one calibrates outside this audience), create fully color-aware workflows which don't require an IT degree to implement, and write web browsers which consistently, correctly and by default use color profiles. I know these are mostly pie-in-the-sky dreams, but I'd much rather have companies spend my investments on those, rather than putting frigging microprisms or goddamn AA filters on my frigging cameras.

Ooh, it feels good, ranting.

* I swear if Lightroom made Clarity scale go to 11 people would still ask for 12. And yes, I know it goes to 100.
Title: Rant 23
Post by: AJSJones on September 26, 2009, 08:20:29 pm
"I would feel less riled up about this if it wasn't for the fact that I encounter people almost daily that come to my gallery and say, "I can't believe how different your images look in prints to how they look on your web site." It's almost as if these folks are reading from the same script. Talking to other photographers who both show their work online and then in large prints, I hear the same thing; how people can't believe the differences. In fact I know quite a few widely exhibited fine art photographers who refuse to show their images online because they feel that this misrepresents how their work should be seen."

Boom goes the petard indeed!

Surely these folks are expressing their surprise/frustration with small monitor images and their inability to represent the print, just like you and other respondents.  It's no different than trying to put together on on-line catalog for e.g., a painting exhibition, is it?  Why would one be surprised that the real thing is so different from the small thingy in the brochure?  So the reaction is to replace them with, what, names?  Is that an improvement?   "Hi I'm a photographer - here's my website: a text file of my titles and word descriptions of the images"  Of course the fine art photographer's work is meant to be seen as prints, but it doesn't exist if people don't know about it - and if it's not on the internet then a lot of people just won't know about it - whether we like that situation or not.  I don't understand the rile factor here:  If they are fine art photographers (or aspire to be) and have made large prints themselves, they'll already know the deal.  If they're not and they're unfamiliar with large prints, such as yours, they'll be pleasantly surprised and, thankfully, a bit more educated at the same time. They might also spread the word by mouth (or text message  )


Assessing the performance of a Leica from an 800 pixel image and concluding that "of course the latest FlipHD just blows it away - it goes up to 11" is just lack of education and surely just generates chuckles.  The huge teaching component of your website reveals your desire to educate, so I'm a touch surprised at the "steaming up" due to reactions and comments of the (as yet) uneducated.

I hope your rant provided some relief.
Title: Rant 23
Post by: David Sutton on September 26, 2009, 09:38:41 pm
Resolution, gamut, dynamic range. Pick any two.
For many of us that's the choice. And for web presentation it gets worse. All that has ever interested me is printing, and I have never gotten used to seeing what happens to an image when posted. Sometimes I sit here looking at images on screen and wonder what they really look like.
What is more interesting, however, is that there doesn't seem to be much discussion on the aesthetics of images seen with transmitted light. It is not as if it is anything new. After all, we have a millennium of stained glass. There is a rich tradition of story telling and pictorial styles here. What puzzles me is why I prefer not to view photos on a screen or digitally projected. Is it the refresh rate? Is it the limited viewing angle? I suspect that what is missing is the tactile pleasure of walking from one image to the next and back again. Being shown one image at a time on one or two screens is not the same. A slide show is venturing more into the realm of the aesthetics of cinema. The decision over what to look at and when is passed over to someone else. I want to get off my bum and move around. Switch the light on and off. Talk about it. Close and open the curtains. I enjoy the experience of using my whole body to view an exhibition or even a single photograph on a wall.
Also there is some educating to be done. Few people expect an on-line catalogue of paintings to look like the real thing. But many people have not experienced a real print. It's like taking someone who's only ever watched television to a wide-screen cinema.
My 2 cents worth, David
Title: Rant 23
Post by: DaveCurtis on September 26, 2009, 09:52:04 pm
Interesting rant and a rant I fully agree with.

This afternoon I've just sold several fine art prints printed on A2 Iford GFS printed with my 3800.

My client said "wow they don't look like that on the web. Look at all that detail ..."  

To me at least, the final print is what it is all about. As a medium small web sRGBs images  are completely different to large well printed print.




Title: Rant 23
Post by: Steven Draper on September 27, 2009, 12:31:50 am
Hi.

Years ago my father painted copies of Old Master images for people who did not want a cheap print of Constable's or Turners. I wondered why we had to go to the museums to see the actual pictures when they were in so many books, until you realise that much of the enjoyment of a real painting is actually being in its presence, not just looking at the elements of the picture. I also noted that the colours and detail were different without all those little printing dots!

Yet many people never saw the real painting yet enjoyed pictures in books or cheap prints from market stands. Yet do 'wet' artists get over fussed - most I know do not have web sites - so probably not.


Is the real issue not how other people enjoy our 'online'  photographic images i - and yes folks enjoy pictures they see on the web on really old monitors with really odd colours, but a fear that our images, often created using expensive equipment and slaved over for hours and hours will just be equalized with someone who has snapped something similar on a mobile phone an sent it straight to Flickr!

I think that people who are seriously interested in buying prints are also sophisticated enough to realize that the web is really just a catalogue of work, Yes they will probably enjoy getting to know a little bit about an artists style and flavour of work, and yes they may narrow a choice of prints down from a whole collection to two or three, but ultimately they are using the web as a tool, with the experience of a print in mind.

I'll also add that earlier this year I did get an exhibition based largely on a few emailed files and web site, although in fairness the curator had seen one file as a print at a show, therefore the ability to mentally visualize the others may well have been easier. And that is why people still come to galleries to see exhibitions - although the web may provide a check as to if the concept of what is being displayed is to taste - unless you have a big reputation!! So getting work in public display IS critical if you want to build a reputation - the more real images people have seen, the better they can anticipate what a print of a web image will look like.

For people who believe that the screen is the 'end product' then you can draw your own conclusions. However I can say - as chair of a successful artist run gallery - we've actually increased our sales this year over last - with photography doing very well. So the print is a long way from being dead. The biggest selling point of an image is the connection a viewer has with it, if they like it, it goes with the furniture and the price point is about right they will purchase. The exact details of the print, paper, whether the master file is sRGB or prophoto, do not matter, infact many do not want to know because they have no need to understand in order to enjoy the print. There probably comes a point for a few photographers when their name becomes the core selling feature, but that does not always mean great prints, although many will be, just that their name means something. However for the majority of middle ground photographer, the overall image NEEDS to be more creative and technically better than what the purchaser feels they could achieve with their own camera.... and that means pretty good now days!

S



















Title: Rant 23
Post by: Rob C on September 27, 2009, 04:03:33 am
Well, I have spent a lot of my time looking at fashion and beauty photographers' sites on the web and some have been amazingly impressive. Frankly, I doubt if any printing device can ever match the luminosity and sheer sense of life that springs out of the monitor when the original is wonderful. This is not sour grapes, either, because I have coaxed some (to me) quite impressive b/w from my B9180 with unexpected detail that I hadn't seen either. But there is something else, beyond detail - the same thing that the transparency used to have and print never did.

Perhaps there is a confusion here - is there the fear that the web image hasn't been prepared well enough for the medium? Is it the age-old defence system coming to the fore, saying that though the website looks good, my work is better, the shot the client used wasn't the best and so on?

That uncalibrated monitors don't help is beyond dispute, but are they all so bad, I wonder? My first monitor was a Mitsubishi Diamond Pro, I never calibrated it because I had neither the tools nor the desire nor understanding of digital problems at the time, but also because I was fascinated and highly impressed by the images that I did manage to raise on it: they looked stunning, straight out of the box (the monitor came out of the box, not the images). Of course I had no idea what the photographers saw in their studios, but it still looked great to me at home.

Perhaps we have to cut down on our expectations; if we sell prints via the web, then if the buyer likes what he sees there well enough to make an order, we should be happy that he will be even more happy when he gets the real print. If the theory holds good that is, that the print is going to be better, which is not always the case, even when you are trying to match colours yourself as you try for that magical, final piece of paper.

This theory would make perfect sense if control at the production end was also perfect, which for me it sure ain't, nor, I suspect, for many here, regardless of how expert they may wish to appear.

Rob C
Title: Rant 23
Post by: tho_mas on September 27, 2009, 10:32:36 am
Quote from: Schewe
Again, Michael's point is that evaluating an image at 1:1 tells you nothing about what the image will look like printed
As we are not able to judge about how an image will look like printed that also means by implication that softproof, curves etc. etc. are pointless ... so finally (pointed) a screen is not an helpful part of the equipment.
Of course the screen view never looks exactly as the print, especially regarding noise. But infact I feel that I can estimate quite well how a certain image displayed on screen will look like when printed (on different papers). That also means that I can judge about files of other cameras (within limitations of course).
Title: Rant 23
Post by: Quentin on September 27, 2009, 12:14:42 pm
While I agree with Michael's point for prints, if you shoot for stock, the pictures are judged and purchased based on the screen thumbnail or enlaged version.  I have many shots I'll never print and which are not that interesting in fine art terms, but they have to look noise free and sharp on screen because that is how buyers make their buying decisions - even if (eventually) the picture is destined for use in print in a book or magazine.  For these shots, appearance on screen is all that matters

Quentin
Title: Rant 23
Post by: Graeme Nattress on September 27, 2009, 04:18:17 pm
I can see how comparing prints is meaningful to the photographer taking the shots and doing the printing. They know the scene, the lighting conditions and have the final image that they wish to achieve in their heads. They can figure how this camera or that camera helped or hindered them on their journey from scene to print.

However, for judging dynamic range, say, it's hard enough when you were at the scene yourself. Eyes are not light meters. Even harder still for someone who was never their to infer what the lighting conditions were. That's where calibrated test charts come into their own. Even resolution and detail are tricky to judge on real world scenes.

Even with careful measuring, the test you do yourself is the one you should trust.

Graeme
Title: Rant 23
Post by: David Mantripp on September 27, 2009, 05:26:29 pm
Michael,

I think this sidewsipe:

“Now, if judging a camera or lens' performance by means an image on-screen is your idea of a good time – well, fine. But, except for the Flickr crowd that's not what fine photography is about – at least it isn't for me“

was pretty uncalled for. There is a scary amount of talent on show at Flickr, and frankly if it's good enough for Art Wolfe....   There are some very fine photographers out there who's work is actually very adapted to the online space, indeed in embraces the medium.

There is no single "Flickr crowd". Far from it. There are many. And frankly, the amount of time they spend "judging a camera or lens' performance by means an image on-screen" is approximately zero in my experience. Flickr ain't DPreview - or even The Luminous Landscape.

David


Title: Rant 23
Post by: Bill VN on September 27, 2009, 05:52:20 pm
Quote from: michael
No, the point isn't to turn the print into a means of technical measurement. Anything but.

What I was trying to get at is that it's what ends up on the final print (however it's made) that is the ultimate arbiter. If I can see the difference in the final work of art, then the difference is meaningful. If I can't, then it may be there but of only academic interest.

Michael

I am a little confused on the purpose of Mike's rant. I thought the essay's point was that you cannot evaluate the performance of a camera or lens from samples posted on the web, which is valid.

However, saying that you can evaluate a camera or lens from the output of an inkjet or laser printer is equally untrue. All laser and inkjet printers are halftone devices that print very fine multi color dots to produce an image. The resolution of a print is not directly dependent on the 1s and 0s a digital camera generates, but on the printer's halftone screening process, the inks and printheads used, the raster image processing (RIP) algorithms and the paper's ink dispersion rate.

On another note, if you are working in Photoshop, or any other pixel-based program, keep in mind that Apple displays can only accurately show a pixel image at a halving or doubling  of 100%--i.e. 100%, 50%, 25%, 200% 400%, etc. At other magnifications, which is usually the case if you type command-0, the screen image is dithered and inaccurate.
Title: Rant 23
Post by: Schewe on September 27, 2009, 06:57:48 pm
Quote from: Bill VN
On another note, if you are working in Photoshop, or any other pixel-based program, keep in mind that Apple displays can only accurately show a pixel image at a halving or doubling  of 100%--i.e. 100%, 50%, 25%, 200% 400%, etc. At other magnifications, which is usually the case if you type command-0, the screen image is dithered and inaccurate.


Actually, to be accurate, if you are using Photoshop CS3 and earlier, this is true. If you are using Photoshop CS4 then not true depending o n your GPU settings. If you are using Photoshop CS4 and you have GPU turned on, then resampling for screen display in Photoshop is being done using Bilinear. If you have GPU turned off then it's using Nearest Neighbor. So, one should be precise regarding the version you are using...
Title: Rant 23
Post by: michael on September 27, 2009, 08:01:55 pm
Quote from: Bill VN
I am a little confused on the purpose of Mike's rant. I thought the essay's point was that you cannot evaluate the performance of a camera or lens from samples posted on the web, which is valid.

However, saying that you can evaluate a camera or lens from the output of an inkjet or laser printer is equally untrue. All laser and inkjet printers are halftone devices that print very fine multi color dots to produce an image. The resolution of a print is not directly dependent on the 1s and 0s a digital camera generates, but on the printer's halftone screening process, the inks and printheads used, the raster image processing (RIP) algorithms and the paper's ink dispersion rate.

On another note, if you are working in Photoshop, or any other pixel-based program, keep in mind that Apple displays can only accurately show a pixel image at a halving or doubling  of 100%--i.e. 100%, 50%, 25%, 200% 400%, etc. At other magnifications, which is usually the case if you type command-0, the screen image is dithered and inaccurate.

David,

I meant no offense to Flickr or the people that use it. I was simply using it as a metaphor for web based photographic display.

Michael
Title: Rant 23
Post by: Tyler Mallory on September 27, 2009, 09:51:37 pm
Sir Edmund Hillary's comment about why he climbed Mount Everest "Because it's there." comes to mind. We look up close because we can. This was not such a thing with film because it was a pain in the butt to look at so many frames under such magnification.
I think what Michael's saying is that, just because we can, doesn't mean we must. And when we do, take what we're looking at with a grain of salt.
Title: Rant 23
Post by: Gordon Buck on September 27, 2009, 10:53:20 pm
I like prints.  Our house is full of (our own) large prints; in fact, we need a few more walls!

What if the cost and availability of large flat "monitors" changed favorably so that I could hang (framed?) monitors instead of framed prints?  Would the display and resolution need to be even more HD than current HD monitors?


Title: Rant 23
Post by: Ray on September 27, 2009, 11:55:37 pm
Quote from: gordonsbuck
I like prints.  Our house is full of (our own) large prints; in fact, we need a few more walls!

What if the cost and availability of large flat "monitors" changed favorably so that I could hang (framed?) monitors instead of framed prints?  Would the display and resolution need to be even more HD than current HD monitors?

It depends on the viewing distance. We should always bear in mind, if you walk up close to an impressionistic painting in an art gallery, it generally looks like crap. It's made to be viewed from a distance.

I like big photos on my wall so they can be appreciated from the average distance in the room that I find myself most of the time. It's why I bought an Epson 7600 some years ago.

My partner has a 50" Panasonic plasma TV which sits on a stand close to a 23"x35" print of mine hanging on the wall. Out of curiosity, I downsampled the same file from which the print was made, to around 2.5mb (converting also to sRGB) and displayed it on the 50" plasma through its built-in SD Card slot. I was amazed.

The image on the TV screen is about the same size as the print on the wall. From a comfortable viewing distance of 3-4 metres, there is no greater detail to be seen in the print, and the image on the TV monitor looks clearly more vibrant and generally preferrable, but not always. The print can be seen only through reflected light. As the nature of the light changes, so does the appearance of the print. There are changing subtleties as the ambient light changes, which result in changes in mood as reflected in the print.




Title: Rant 23
Post by: pegelli on September 28, 2009, 04:28:54 am
I agree 100% crops is nowhere near a guaranteed "end-all" to judge qualities of a lens or sensor, however I object to the addition of the italics words "(at least in part)" on the WHAT's NEW page. Some useful information or comparison can be had from these crops. Even Michael himself has posted 100% crops, for instance to point at noise characteristics. I believe with proper care it can still serve a useful purpose, even when posted on the web.

I still got good value out of this page (http://www.luminous-landscape.com/reviews/cameras/big-three-5.shtml), or should we conclude from this rant that this is a useless comparison, since the only thing that matters is how the images look in print  
Title: Rant 23
Post by: Bill VN on September 28, 2009, 01:30:17 pm
Quote from: michael
David,

I meant no offense to Flickr or the people that use it. I was simply using it as a metaphor for web based photographic display.

Michael

I am not David.
Title: Rant 23
Post by: JohnKoerner on September 28, 2009, 01:56:41 pm
With all due respect, there was a lot of simplistic thinking and overly-simplistic generalizations in said article, if not outright falshoods and self-mockery.

For starters, as has been suggested above, if the physical printed image is the end-all, be-all judge of the value of any given camera, then pray tell what is the point of Michael's own page-after-page, year-after-year articles and posts on the web ... all of which are replete with "digital, internet" images being used as 'proof' for his own voluminous takes and positions on said products? If this new article isn't self-mockery, then I don't see what is.

That being said, this entire tirade begs the question as to "on what" physical printers may a camera's quality be accurately judged? All of them? Or only a few? After all, if the printed image is the 'only' image on which a camera may be judged, then does this statement hold true for every photo printer available---or does this statement hold true for only just a hand-selected few? After all, can anyone seriously declare that the print quality produced on a cheesy multi-function devise, printed-out on cheesy paper, gives a better inclination of 'camera quality' than does the highest-resolution, color-calibrated monitor available today? I doubt that very much.

So aside from the problems with reality that the above scenarios represent, is another fact, which is simply that MORE images are seen, evaluated, and judged online, via the internet, than are seen, evaluated, and judged in person on fine art paper. Further, more images are seen, evaluated, and judged in simple magazines and books, too, than are done so on fine art paper. If anything, fine art photos are actually the LEAST-purchased, LEAST-used applications to which the majority of photos are in fact put and judged. There are more photos in books, in any one major city, than there are photos in all of the collections of the world put together.

Further, notwithstanding the additional simple-minded posit that 'all' Flicker contributors put out the same-quality images, captured on the same-quality cameras, there is the point to be reckoned with that the value of any image isn't in it's 'absolute pixel (or printed) detail and quality, but on the overall impact (or significance) that the image conveys. There are too many 'technically-great' mediocre images to mention, and too many fabulous images that have technical problems to resolve.

There are simply too many variables to consider, to be able to completely dismiss the value and power of the online internet image. Moreover, I would say the future suggests, if anything, that the physical medium of 'paper and ink' are in fact the "dying dinosaurs" on the decline, not the digital image. If anything, all evidence suggests it is the digital image that is ever-growing in importance overall, in most people's lives, than is the physical printed image. For example, my girlfriend and I are in a quandary as to which of our images we would like to display in our own home. Thus the question becomes 'which one' makes for the best presentation? Ray alluded to this also. My girlfriend and I decided that we would rather purchase a 40" x 30" LCD monitor, on which we could 'flash' a thousand of our very best photos ... each flashing every few seconds ... than we would to "print and display" any one of our images on paper. I mean, let's be real, which presentation would ultimately prove to be the greatest "display" in one's home?

Thus, in the end, I think the printed image will ultimately be the least-used, least-desired form of display for the majority of people in the not-too-distant future. This is not to say that the digital print has no value; it is simply to suggest that digitized, computer-monitor-viewed images are more ubiquitous and important to most people than what is being acknowledged in this article.


Jack

.
Title: Rant 23
Post by: Rob C on September 28, 2009, 02:10:41 pm
Old men obviously get confused, but I sort of miss where the OP made the statement that has carried the thread into the current mood of taking up positions.

I hadn't realised that, all along, it had been an either/or proposition.

Silly me.

Rob C
Title: Rant 23
Post by: michael on September 28, 2009, 04:14:48 pm
I'm not sure why the central thesis of my argument is proving so difficult for some people.

I am not saying that people shouldn't display their images on the web, on screens or in any other manner. I am simply pointing out that the full quality of a properly shot, edited and adjusted image from a high quality camera simply can not be properly seen electronically.

Point: The web is sRBG and images displayed are compressed JPGs, while a print can easily exceed Adobe RGB in gamut.

Point: A print from, say, an 18MP camera (certainly not the biggest) is some 5200 X 3500 pixels. A 30" display is just over 2 million pixels. Something has to give, no?

An 18MP camera file can produce a roughly 16X20 print at optimum printing resolution, allowing one to see everything that the camera and lens have to offer. To produce an image with comparable resolution on a screen would require  a 240-300 PPI screen, which do not exist. Yes, a 50" LCD or Plasma can show the image at 100%, but to view it properly one has to sit across the room which means that the resolution isn't properly visible the way it is on a print.

So my thesis isn't that prints are a superior means of showing ones images (though for me they are). I realize that many people never print. But, simply that with today's technology they can't show you everything that's in the file the way a print can. Or, if they can, they suffer from very high cost, the extremes of technology, or other impediments.

Michael
Title: Rant 23
Post by: ErikKaffehr on September 28, 2009, 04:31:03 pm
Hi!

Making RAW images available may be the best choice in my opinion, than you can download and print.

Regarding "scientific data" like Imatest and DxO-mark I'd suggest that they are probably correct, but not necessarily relevant. The reason they may be less relevant may be that we don't know how to interpret them.

My own testing seems to point in the direction that we need to go beyond A2 size to clearly see advantage for "full format" over APS-C, and this also applies to "medium format".

Subject may matter a lot, it's note really obvious which subjects are most demanding and in which respect they are demanding. The issue is quite complex. It's a good proposal that you should test with rental equipment, but that can be both impractical and expensive. Renting a Phase One equipment for two days here in Sweden is same money as a Canon 5DII, and there is no warranty that the equipment you rent is properly aligned.

There was some comparative shooting on different systems on Flickr, here:

http://www.flickr.com/photos/dos-chin/sets...6120567/detail/ (http://www.flickr.com/photos/dos-chin/sets/72157614936120567/detail/)

I downloaded the sample image for Nikon 3DX and Hasselblad HII-50 and printed both in A2-format trough my standard processing pipeline. In my view the Hassy image came out on tops when "pixel peeping" but the Nikon image was actually sharper in general.

I asked a friend of mine who has been working at one of Swedens top professional photo labs about his opinion and he was quite affirmative that the Nikon image was sharper. Looking on screen at actual pixels in the area of maximum sharpness there was no question that the Hassy image was sharper. My friend did not know which was Hassy or Nikon and I could not talk him into changing his mind that the Nikon image was the better one.

My explanation for this is:

- Subject was not really suitable for this test, much because lacking high frequency detail.
- Distance to subject was to short, longer distance would give smaller detail and larger depth of field
- Color and tonality was better for the Nikon image
- DOF was in favor of the Nikon image, but the lens was also stopped down beyond optimal aperture

The conclusion I draw from this:

- If sharpness is good enough for a certain size of print other factors will dominate.
- The benefit of larger formats is not obvious for sizes up to A2

Best regards
Erik





Quote from: Rob C
Old men obviously get confused, but I sort of miss where the OP made the statement that has carried the thread into the current mood of taking up positions.

I hadn't realised that, all along, it had been an either/or proposition.

Silly me.

Rob C
Title: Rant 23
Post by: Ray on September 28, 2009, 11:10:24 pm
I made a mistake in my previous post by writing that I downsized a large file I used for a 23"x35" print to about 2.5mb for display on a 50" plasma TV screen.

Of course I meant 2.5mp. One doesn't have to be precise because the TV will downsize a larger-than-necessary file to the exact size to fill display, although it helps to crop the image to the 16:9 aspect ratio.

Perhaps the time has come for Canon to go full circle and produce a D30 MkII using all the latest technology to provide those 3mp with the maximum DR and lowest noise possible, even better than a D3X image downsampled to 3mp. This is all one needs for a full HD display.
Title: Rant 23
Post by: pegelli on September 29, 2009, 03:14:56 am
Quote from: michael
I'm not sure why the central thesis of my argument is proving so difficult for some people.

I am not saying that people shouldn't display their images on the web, on screens or in any other manner. I am simply pointing out that the full quality of a properly shot, edited and adjusted image from a high quality camera simply can not be properly seen electronically.

Point: The web is sRBG and images displayed are compressed JPGs, while a print can easily exceed Adobe RGB in gamut.

Point: A print from, say, an 18MP camera (certainly not the biggest) is some 5200 X 3500 pixels. A 30" display is just over 2 million pixels. Something has to give, no?

Quote from "What's New":
We've all done it. There's a photograph or a 100% crop of an image online, taken with a camera or lens that we're interested in, and we judge its image quality capabilities (at least in part) on the basis of that screen image.

Wrong!


I think I'm getting the central thesis of your argument and even agree with it, however the words "(at least in part)" suggest you find absolutely no use of comparing 100% crops which I don't understand and is not supported by your points.

I'm not trying to argue, I'm just trying to understand the whole logic better.


Title: Rant 23
Post by: Rob C on September 29, 2009, 04:26:32 am
"I'm not sure why the central thesis of my argument is proving so difficult for some people.

I am not saying that people shouldn't display their images on the web, on screens or in any other manner. I am simply pointing out that the full quality of a properly shot, edited and adjusted image from a high quality camera simply can not be properly seen electronically."






Michael

Relax, the internet is the natural home of the blown hermeneutic. You fare neither better nor worse than anyone else who ventures to publish a thought!

Rob C

Title: Rant 23
Post by: JohnKoerner on September 29, 2009, 12:32:20 pm
Quote from: michael
I'm not sure why the central thesis of my argument is proving so difficult for some people.
I am not saying that people shouldn't display their images on the web, on screens or in any other manner. I am simply pointing out that the full quality of a properly shot, edited and adjusted image from a high quality camera simply can not be properly seen electronically.
Point: The web is sRBG and images displayed are compressed JPGs, while a print can easily exceed Adobe RGB in gamut.
Point: A print from, say, an 18MP camera (certainly not the biggest) is some 5200 X 3500 pixels. A 30" display is just over 2 million pixels. Something has to give, no?
An 18MP camera file can produce a roughly 16X20 print at optimum printing resolution, allowing one to see everything that the camera and lens have to offer. To produce an image with comparable resolution on a screen would require  a 240-300 PPI screen, which do not exist. Yes, a 50" LCD or Plasma can show the image at 100%, but to view it properly one has to sit across the room which means that the resolution isn't properly visible the way it is on a print.
So my thesis isn't that prints are a superior means of showing ones images (though for me they are). I realize that many people never print. But, simply that with today's technology they can't show you everything that's in the file the way a print can. Or, if they can, they suffer from very high cost, the extremes of technology, or other impediments.
Michael


Points made and points well taken.

Even amongst digital devices there is a tremendous difference in what one can see. For example, my finished images that I adjust on my color-calibrated NEC (in ProPhoto) don't appear anywhere near as colorful or vibrant when viewed online (converted to sRGB) via my web browser and cheap laptop. So I understand your point there, and I further understand that even the images which look so vibrant and rich in color on my calibrated NEC will look still more colorful and vibrant when actually printed well on archival paper from a quality printer. Again, points made and points well taken.

The trouble I have is, while your logic above is entirely sound, and the mood you convey above is entirely sober, such was not the frame of mind in which you began your original article. This is how you began your original article:

"I alternately chuckle and get steamed up when I read someone on a web forum either condemn or praise a camera or lens based on a web images. This is utter nonsense."

You then pretty much, and with a great deal of contempt, dismissed "the Flickr crowd" as a single living entity, from which you then went on to your central theme of fine art prints being the ultimate arbiter of equipment quality. Backpedaling a bit now does not change the mood of what you originally wrote. The paradox is, while the theme of your article may be true ultimately, the real truth is the vast majority of photographs do not get seen, judged, bought, or sold in a "fine art print" form. Moreover, there is an overall condescending theme to the original article towards any person whose judgments are being rendered through digital evaluation, while at the same time an automatic self-contradiction is raised by virtue of the existence of the "Luminious Landscape" very own digital presence and reviews.

Therefore, since this here is a forum to discuss your site and the topics of discussion, I would like to convey here that (as a reader looking to learn from your online articles, supported by your online digital images) when I read something from a man who is 'alternately chuckling, and then getting steamed up' over something he himself does, this doesn't conjure a feel of a level-headed thinker. In fact, when one is being tossed back-and-forth with such emotions as a writer, one can miss a lot of key points.

For, while you entitled your article, The Fallacy of Judging an Image Online, what seems to be two twin simultaneous fallacies of your own article is (1) to assume that the most important photography (to the most people) is fine art photography 'in print', and (2) that this position seems to make a mockery of all of your own online articles, all of which are supported with plethora of the very "digital web image evaluations" you dismiss as "utter nonsense."

Regarding (1), since many {if not the majority} of the viewing & buying decisions and judgments of photographs are in fact made based on the submission of electronic images, to dismiss the importance of how such digital images look (either online or on a monitor) is itself a real fallacy. In "The Photographer's Market" book, which lists just about every purveyor of photography known, almost every one of these purveyors ASKS FOR the submission of high res digital images on which to base their decisions on whether or not they will accept (buy) any photographer's work. Furthermore, regarding (2), I still don't understand how you can say this:

"But please, please, stop judging the technical quality of photographic equipment by looking at small web images. And, while 100% crops can be helpful in comparing certain technical aspects of image quality, this usually bears little real-world relationship to how a photograph will appear in a print. In my experience its rare that the pixel peeping that the online image analysis that many people love to do (doh! including myself) bears any real-world relationship to how an artist's image will appear in a final display print."
(Parenthsis added)

... and yet post as many photographs as you do digitally online as 'proof' of the positions you take regarding the quality of the lenses and software.

I don't think too many photographers judge photos based on "small" web images, but I do think 99% of judgments come from 100% crops. So it just seems pretty hard to reconcile the two positions, the fact you yourself have evaluated cameras and lenses (for years) on your website, always supported by large digital web images, and yet that you take the ultimate position you do above, that such judgments have little 'real-world' value. One is left scratching one's as to "why?" You even said, "In fact I know quite a few widely exhibited fine art photographers who refuse to show their images online because they feel that this misrepresents how their work should be seen." I would say this latter quote shows more integrity to the above belief system than does holding this belief and yet at the same time still posting articles with web images "as proof" for one's position on the quality of cameras and lenses.

In the end, the ultimate truth of your article is easily understood. The fine art print, in fine form, does show the most of what a camera and lens can do. It was not the truth of this that was the problem, it was the self-contradiction of utterly condemning the importance of any judgments being made of digital images at all, as nonsense, and squaring this with the plethora of your own online articles and judgments, all supported with digital web images, that raised an eyebrow for me. For while the very largest of "fine art prints," printed on the very finest printers, may in fact show the ultimate strengths and liabilities of any equipment, the truth is such end results are NOT how the majority of photos are in fact bought, sold, or even seen.

The majority of photographic images bought, sold, and judged in this day and age are in fact based on DIGITAL evaluations, long before they ever get to be printed, and so to dismiss the importance of digital evaluation (or how images appear digitally) is to dismiss the vast majority of real world photography ... and thus the vast majority of people ... as well as your own online, digital presence. The A1 fine art print may tell the ultimate tale, but ultimately it is the least-used medium that buyers of photography use to judge photography.

The reality is, almost no one has the luxury to be able to get a free camera and a free set of lenses, and then be at liberty to take as many photos as contents their heart, print-out their images on the finest paper with the finest printers ... and then do this with all of the name brands, makes, and models ... from which they may then make a purchase decision. This is not a reality for people. The purchase decisions most people make are based on online viewings and/or high-res digital viewings, and so such judgments are ultimately the most important.

Thus, if the point of your article is that the ultimate reality is to judge cameras and lenses based on fine-art prints, then I think all of the judgments and proclamations of your own articles online ought to be based only on a comparison of A1 and A2 prints, not on digital crops, so that the conclusions on your website product reviews are to be in harmony with your core beliefs.

Jack

.
Title: Rant 23
Post by: Alan Goldhammer on September 29, 2009, 01:07:19 pm
I'm hesitant about joining in the discussion because we didn't have a prayer of repentance yesterday for posting less than optimal images on the Internet.  So be it.  

I can only offer the following observations (which I think are true):
1.  most of those who view any of our images on the Internet do so on monitors that are much worse than ours;
2.  most of us (and those who we give or sell our prints to) have only limited wall space to display prints, whereas digital displays offer "almost" unlimited number of images;
3.  anyone (caveat: in my experience) who has seen a fine art print and the digital image, prefers the fine art print (though it doesn't guarantee a sale);
4.  there will be a continuing move towards more digital presentation of images (and probably Ansel Adams would not only approve, but be excited about the prospect were he to see all of the advances in the 25 years since his passing).

I still get excited seeing the print come off the Epson and seeing details in it that I've not seen on my monitor.  I would hope others feel the same way.
Title: Rant 23
Post by: Ben Rubinstein on September 29, 2009, 01:40:39 pm
Quote from: Alan Goldhammer
I still get excited seeing the print come off the Epson and seeing details in it that I've not seen on my monitor.  I would hope others feel the same way.

Scares me silly actually, how on earth are we supposed to be able to control our output if we don't have the slightest idea what it's going to look like until it's printed? I thought that WYSIWYG was supposed to be part of our modern workflow. Flatscreens that are far sharper, contrastier, brighter than the print that you are looking to achieve (yes with all the best profiling unless perhaps you spend more on your screen than on your camera), colours that you didn't know were there (ooh I'd really love that if I was trying to get skin tones right), etc, etc. Personally I'd prefer to edit within the constraints of WYSIWYG even if it means sRGB. At least I won't have any surprises.
Title: Rant 23
Post by: Alan Goldhammer on September 29, 2009, 02:17:20 pm
Quote from: pom
Scares me silly actually, how on earth are we supposed to be able to control our output if we don't have the slightest idea what it's going to look like until it's printed? I thought that WYSIWYG was supposed to be part of our modern workflow. Flatscreens that are far sharper, contrastier, brighter than the print that you are looking to achieve (yes with all the best profiling unless perhaps you spend more on your screen than on your camera), colours that you didn't know were there (ooh I'd really love that if I was trying to get skin tones right), etc, etc. Personally I'd prefer to edit within the constraints of WYSIWYG even if it means sRGB. At least I won't have any surprises.

Perhaps I wasn't clear in my post (or maybe I'm not clear in the following response).  I have a NEC profiled monitor and I get prints that are what I see on the screen, so I am controlling output.  However, a 13x19 print is a much different viewing experience than what is/was on the monitor.  It's a different visual experience, and yes because I can continue to look at the print, I do see details that were overlooked on the monitor (these have nothing to do with getting colors, tone, etc. correct for output).
Title: Rant 23
Post by: dwdallam on September 29, 2009, 03:11:09 pm
Quote from: michael
No, the point isn't to turn the print into a means of technical measurement. Anything but.

What I was trying to get at is that it's what ends up on the final print (however it's made) that is the ultimate arbiter. If I can see the difference in the final work of art, then the difference is meaningful. If I can't, then it may be there but of only academic interest.

Michael

And that really is the point. I mean photography is a visual medium. Anything outside the human eye's ability is really irrelevant in this respect. Also, for me, extra pixels aren't for creating better art (especially after the new IDS3 and D3X) but more for cropping and still having enough pixels to do as you wish with the image.  This is exactly why I will buy the next iteration of Canon 1DSMKIV, if the increase in pixels and noise warrant it, say 27-30MPs. I can't see needing more than that, at least for my needs.
Title: Rant 23
Post by: Ray on September 29, 2009, 09:06:27 pm
Quote from: Alan Goldhammer
Perhaps I wasn't clear in my post (or maybe I'm not clear in the following response).  I have a NEC profiled monitor and I get prints that are what I see on the screen, so I am controlling output.  However, a 13x19 print is a much different viewing experience than what is/was on the monitor.  It's a different visual experience, and yes because I can continue to look at the print, I do see details that were overlooked on the monitor (these have nothing to do with getting colors, tone, etc. correct for output).


Surely everyone understands that a print is a reflective medium whereas a monitor is transmissive. Turn the lights down and the shadows on the print will become impenetrable, whereas on the monitor they will remain just as detailed. Turn the lights off completely, in the evening, and you can't even see the print at all, wereas the image on the monitor will appear even more vibrant than it did with the lights on.

So of course a print is a different viewing experience. In a sense, the image on the monitor is a more reliable representation of the detail and tonality in the photo, as a result of the transmissive nature of the monitor which is far less affected by changing ambient light conditions.

I simply don't expect to see detail on any print that I make that I can't see on the monitor. In fact, the reverse is often the case, and that's because a 100% crop on my monitor usually represents a much larger print than I have the means of making. Shadow noise that is visible at 100% on the monitor, representative of a 6ftx4ft print or even larger, will be far less obvious on a 3ftx2ft print.

Of course, just as a 100% crop on the monitor should reveal more detail in the image than is likely to be visible on the print, the complete image at 5% or 12.5% on the monitor cannot display the detail and have the same over all impact of a 24"x36' print.

Perhaps my experience is different because I still use CRT monitors which generally have a better contrast ratio than LCD monitors. When I process an image for printing, using the 'proof color' facility in Photoshop, always ticking the 'paper-white' box, I sometimes have difficulty in successfully adjusting the image to match the vibrancy of the previously adjusted image for monitor display. Or, to put it another way, when I succeed in matching the vibrancy of the non-proof image, almost invariably I find there are large areas of out-of-gamut colors. Reducing the intensity of such areas of out-of-gamut color, either by reducing saturation or luminance (or sometimes increasing luminance) tends to dull the image.

However, I do appreciate that the average member of the public who is not a photography enthusiast is likely to use an uncalibrated monitor. In those situations, jpeg samples of images for print might give an inaccurate impression.
Title: Rant 23
Post by: BernardLanguillier on September 29, 2009, 09:42:19 pm
The real questions in the end are "what is a good photograph" and "what is a good print". I see a some confusion in this debate between photography and printing/marketing. IMHO Michael's article is about great prints, not about great photographs.

I would personnally argue that a photograph that doesn't look good on screen/on flickr, is probably not a good photograph, merely an average one.

However it is possible to do a great print of an average photograph, as well as a poor print of a great photograph. Both are not directly related.

Prints will reveal qualities in images that are hard to perceive on screen, and it is true that a high resolution large print will be a lot more impressive. Yet, all things being equal the image that looked good on Flick will look even better printed on an Epson 9900 and the best Baryta paper providing the print size matches the capture resolution.

Top end equipment or large stitches can extend the scope of the outputs that can be handled (size and colors) but will never be able to turn an average photograph into a great one. On the other hand, there is probably some truth to the belief that each image is best viewed at a given output size, some lending themselves more to very large sizes though.

I am sure that a lot of us will agree that there is far more photographic talent on Flickr than in the official circles of fine art printing, but these folks typically don't have access to the latest equipment and do typically not even attempt to print or to sell their images. It doesn't make their images less interesting nor talented. Their focus is great photography, not great prints.

This debate will not end until this distinction is explicitely done and agreed with.

One thing to keep in mind is that this distinction will reveal clear paradoxes since fine art printers often use online sites to market their prints to a large audience...

Cheers,
Bernard
Title: Rant 23
Post by: JeffKohn on September 29, 2009, 10:40:17 pm
Quote from: BernardLanguillier
IMHO Michael's article is about great prints, not about great photographs.
If that was the intent of the rant, I don't think it came through very well. Part of the problem is that he can't quite decide his premise: is it that we cannot predict print quality from on-screen images, or that "judging a camera or lens' performance" from on-screen images is a pointless exercise, that's only useful to pixel peepers and the Flickr crowd? He seems to be arguing both points at different times, just in the first few paragraphs.

Now, if you want to limit the argument to images that have been re-sized for viewing on the web, I would agree that those images are just about useless for judging the performance of today's cameras and lenses.

I also agree there are people on the net who will argue endlessly over minute differences that have little if any impact on final output, whether that output is a print or a web-sized online image. I can even understand someone getting tired of such arguments and feeling the need to rant a little. But to say that viewing full-resolution images on-screen is useless for judging camera and lens performance, well that's just absurd. In fact I would say that even if your intent is to judge eventual print quality, the premise is still not correct. It's true that a 100% on-screen image doesn't look exactly like the final print.  But that doesn't mean it won't tell you anything at all about the eventual print quality.

If two shots are taken of the same scene with two different lenses, and on-screen evaluation shows that one of them has obvious CA, light falloff, corner softness, barrel distortion, etc, it doesn't take a rocket scientist to figure out that the other image is going to make a better print. Maybe small differences that are detectable on-screen at 100% won't show up in prints until you hit a certain size, if at all. But an experienced print-maker will understand that, and will have an idea of what kind of differences are going to show up in the intended output. Comparing actual prints is the ideal, but that doesn't mean anything less is pointless.
Title: Rant 23
Post by: Jack Flesher on September 30, 2009, 05:22:48 pm
Quote from: michael
I'm not sure why the central thesis of my argument is proving so difficult for some people.

Michael,

Maybe because a majority of them have never bothered looking at a quality print in the first place?  

I don't know for sure, but have wondered the same thing.  Based on a few of the recent threads in your forums, it seems many of your readers don't own or have never even used the cameras they claim are inferior -- and worse, don't actually ever PRINT anything either!  Their opinions seem to based entirely on whichever sensor's raw data performed the best in a test some distant lab devised.  Regardless, they refuse to believe somebody that actually does own and use the gear in question and actually has made prints from it when they say, "IME x is better than y because of abc."  Of course that ultimately leads your test-heads to their "prove it" comments, demanding you show them the differences you see in your prints on the forum -------- VIA A WEB JPEG!  Yes indeed, the absurdity of their logic does make one wonder...

Cheers, and good luck!
Title: Rant 23
Post by: PierreVandevenne on September 30, 2009, 05:50:05 pm
In any case, the word "100% crops" appear 146  times on this site if one googles

100% crops site:www.luminous-landscape.com

and 44 times in reviews

100% crops site:www.luminous-landscape.com/reviews

They "tell the tale", "clearly show", etc...

They must have some use ;-)

AFAIC, using a printer recommended on this site, and applying the knowledge gained through its tutorials, the gamut issue is what I notice most, especially when contrast is low. The print is the result of some processing (sharpening, whatever...) and some dithering, and I don't feel it resolves better.  

Title: Rant 23
Post by: Rob C on October 01, 2009, 05:26:25 am
I came across this interesting statement earlier today, on this same topic of judgement and the rôle of the final print.

http://www.photo-i.co.uk/Reviews/interacti...V750/page_8.htm (http://www.photo-i.co.uk/Reviews/interactive/Epson%20V750/page_8.htm)

It's to be found just below a set of scans.

Rob C
Title: Rant 23
Post by: michael on October 01, 2009, 06:29:49 am
They must have some use ;-)

Well, of course they do. I never wrote otherwise. This completely misses the point of what I've written though.

A 100% view onscreen may tell you what you need to know about some aspect of image quality (except gamut, of course), but it tells you little about how the image from a particular lens or sensor will appear in a print.

At 100% a typical high res camera file will be somewhere between 65 and 100 inches across. Thats eight feet across!!

No one will look at a print that size from 24 inches away, which is how we normally view a screen.

Step back to the proper viewing distance for a 6 – 8 foot print and you stop seeing fine detail. The human eye simply can see it at that distance.

But, because a print has much higher resolution than, say, a 13X19" print which size is viewed at a distance of as close as 18 inches, ones ability to see fine detail and tonal transitions is greatly enhanced over a huge print (or screen) which just be seen at a greater distance.

In the real world a 100ppi screen of between 24" and 30" simply can not show you what an image looks like, except when one is looking at small section of it at 100%, and what does that have to do with anything other than technical analysis?

I'm amazed that this isn't intuitively obvious based on experience to anyone that is a working photographer.

The bottom line....

A proper size print shows you what the image looks like.

A screen shows you what a reduced resolution, reduced gamut and compressed version of the image looks like.

A 100% view of a section of an image onscreen shows you what you need to know to do a technical analysis of a sensor or lens, but may have little to do with how images may appear.


Michael
Title: Rant 23
Post by: Jack Flesher on October 01, 2009, 11:38:04 am
Quote from: michael
I'm amazed that this isn't intuitively obvious based on experience to anyone that is a working photographer.

My point above was that I think it IS apparent to anybody that is a working photographer who also prints.  

I think the problem is you (LL) have a small portion of your readership who fancy themselves "experts" because they understand the physics involved in the A/D imaging process, yet have never printed, and they are the ones who "don't get it."   This same group are frequent posters, posting voraciously as if their point of view is the holy grail of any and all aspects of digital imaging science and process...

Seriously, I wish you the best of luck convincing them --- I tried for the last time last week and have decided it's a total waste of my time and energy.  Too bad too, because they are generating a lot of bad information for your web readership...

Cheers,
Title: Rant 23
Post by: ErikKaffehr on October 01, 2009, 01:21:05 pm
Hi,

I mostly agree with Michael's view, also this time. Just want to point out two issues.

1) Screen has better dynamic range than print.
2) Looking at a large print at close range can be an "immersive" experiment. I shoot a lot of "panos" and they often represent a very wide angle. Viewing those prints at a close range makes you feel like a part of the picture. Hard to explain, I'd guess it has to do with peripheral vision and perspective.

Another thing I'd point out that I feel that projection in HD (1920x1080 pixels) is a great way to present pictures, but we need only two megapixels for that!

Best regards
Erik



Quote from: michael
They must have some use ;-)

Well, of course they do. I never wrote otherwise. This completely misses the point of what I've written though.

A 100% view onscreen may tell you what you need to know about some aspect of image quality (except gamut, of course), but it tells you little about how the image from a particular lens or sensor will appear in a print.

At 100% a typical high res camera file will be somewhere between 65 and 100 inches across. Thats eight feet across!!

No one will look at a print that size from 24 inches away, which is how we normally view a screen.

Step back to the proper viewing distance for a 6 – 8 foot print and you stop seeing fine detail. The human eye simply can see it at that distance.

But, because a print has much higher resolution than, say, a 13X19" print which size is viewed at a distance of as close as 18 inches, ones ability to see fine detail and tonal transitions is greatly enhanced over a huge print (or screen) which just be seen at a greater distance.

In the real world a 100ppi screen of between 24" and 30" simply can not show you what an image looks like, except when one is looking at small section of it at 100%, and what does that have to do with anything other than technical analysis?

I'm amazed that this isn't intuitively obvious based on experience to anyone that is a working photographer.

The bottom line....

A proper size print shows you what the image looks like.

A screen shows you what a reduced resolution, reduced gamut and compressed version of the image looks like.

A 100% view of a section of an image onscreen shows you what you need to know to do a technical analysis of a sensor or lens, but may have little to do with how images may appear.


Michael
Title: Rant 23
Post by: Ray on October 01, 2009, 08:12:38 pm
Well, I certainly don't understand what the fuss is about, outside of the fact that a rant is a rant. I'm not in complete agreement with Michael, but I do understand that print-making is an art in itself and that a print has a certain tangibility and offers an experience which is diferent from the image viewed only on the monitor.

Nevertheless, we go to a lot of trouble calibrating our monitors and creating profiles for our paper/ink combinations so that the print will match as closely as possible the color, hue and tonality of what we see on the monitor, and we do this so that we can process and adjust the image to taste before printing, and expect the result on print to be close to what we see on the monitor.

There's an option at the top of the Photoshop window to view any image at print size. It's not exact. However, if one wants a more exact size, one can adjust the percentage accordingly. Of course, if the print is large and the monitor is small, one may see only a crop of the print at one time. That's understood and the eye/brain makes allowances. If I view the full size print from the same distance I am from my monitor, then I understand that the detail I can expect to see on that portion of the print, as represented on the monitor, will be very close to the detail I see on the monitor. If I view the full image on monitor from a distance of twice its diagonal (for example) then I get an idea of the detail I can expect to see on the print, however large, viewed from a distance of twice the print's diagonal.

Of course, there's a variety of different paper types with surfaces ranging from dull matter to brilliant glossy. One can't expect the monitor to simulate such differences of surface texture, or indeed the metamerism that certain paper/ink combinations display. There may also be certain subtle differences in tonality due to certain hues being visible on the print which are not visible on the monitor, particularly if the color space of the image is ProPhoto RGB, but my own experience suggests that such differences are far less significant than the differences resulting from the use of either a poorly calibrated monitor and/or inadequate paper/ink profile.
Title: Rant 23
Post by: Jack Flesher on October 01, 2009, 08:59:30 pm
Quote from: Ray
There's an option at the top of the Photoshop window to view any image at print size. It's not exact. However, if one wants a more exact size, one can adjust the percentage accordingly.
Actually, if one wants those measurements to be exact, all they need do is set the PPI setting for their monitor appropriately in the CS preferences...
Title: Rant 23
Post by: Ray on October 01, 2009, 10:20:47 pm
Quote from: Jack Flesher
Actually, if one wants those measurements to be exact, all they need do is set the PPI setting for their monitor appropriately in the CS preferences...


Good point, Jack!

Well, there you have it. A good monitor, properly calibrated, is not a bad device for getting a fair idea of what your print will look like, using proof setup in PS of course, and a good paper/ink profile.
Title: Rant 23
Post by: David Mantripp on October 02, 2009, 05:47:41 pm
Quote from: Bill VN
I am not David.

You should count yourself lucky, 'cos if you were you'd be an intolerant bad tempered old git  :-)

I think Michael clarified what he meant ...  anyway Flickr is so vast I'm sure examples for any particular spin on its merits cound be discussed from here to eternity.
Title: Rant 23
Post by: Ray on October 02, 2009, 10:10:32 pm
I would also tend to disagree with the following comment from Michael in his article,

Quote
But what about looking at images at 100% on-screen? There one can see the differences – right? Well, yes, one can see the difference of how they look on-screen at 100% magnification. The fact that this has little to nothing to do with how an image will appear in a print seems to escape many people. This applies to judging high ISO noise, resolution and more.


.. a sentiment which is also echoed by the following comment from Jack Flesher.  
Quote
I think the problem is you (LL) have a small portion of your readership who fancy themselves "experts" because they understand the physics involved in the A/D imaging process, yet have never printed, and they are the ones who "don't get it." This same group are frequent posters, posting voraciously as if their point of view is the holy grail of any and all aspects of digital imaging science and process...

I would be very surprised if any from this small group of so-called experts obsessed with image quality at 100% on the monitor never print any of their images. I've made the point before, if you rely exclusively on the monitor or an HD display to view your photos, all you need is a 3mp camera which allows for a moderate amount of cropping till you reach the required 2mp to fill an HD screen.

Of course, there are advantages of a high-pixel-count camera. It allows more scope for cropping, and cropping effectively extends the reach of one's lens. Downsampling the full size image for display on monitor, HD TV, or 8x12" print also reduces noise, but surprisingly, if noise is a major problem in the full image at 100%, it can still be a problem at a much reduced size, although obviously not quite as great a problem. Downsampling reduces the appearance of noise, whether on print or monitor, but it's no magic bullet.

Here's an example of a shot taken at ISO 400 with my 20D a few years ago in Pompei. It's the sort of shot that should have been autobracketed for merging to HDR or for processing some other way, but I wasn't carrying a tripod; auto-alignment of images in those days, in PS, wasn't too good, and I was verey verey draaank at the taaame!!! (Just kidding   ).

With an exposure bias of -0.67EV, I got a compromise of totally blown sky and noisy shadows. (A clear example of the camera's DR not being adequate for the SBR of the scene). The noise in the shadows is still very noticeable at an image size of around 43"x24" (the size of my HDTV) viewed from a distance of about twice the diagonal.

Is Michael saying that such noise would not be apparent on a 43"x24" print viewed from the same distance? I'm pretty sure that it would be. I've attached the full image as processed for my HDTV. If your monitor is capable of 1920x1080, you will see the full image at 100%. If your monitor has lower resolution than full HD, you'll see only part of the image at 100%.

If you want to assess how this image would look on a 43"x24" print, follow Jack Flesher's recommendation and go to PS preferences, units and rulers; set the print resolution to 46 pixels/inch and your screen resolution to whatever it is (mine is currently 90 pixels/inch), click Print Size at the top of your PS window, step back from your monitor an appropriate distance that you imagine you would view either print or HDTV screen, and Voila!.. you've got a fairly good representation of what you could expect on either a 50" HDTV screen or 43"x24" print viewed from the same distance.

If you want to get an idea if the noise would still be visible on a postcard size print (it actually is still visible) then make the appropriate setting in PS preferences, or more quickly and easily just adjust the image size on screen to about 6"x4".

I should add that the following image is just slightly less than maximum quality jpeg compression of the original 6mb in order to conform with the maximum single upload size of 2mb on this site.

[attachment=16939:Pompei_laundry.jpg]
Title: Rant 23
Post by: pegelli on October 03, 2009, 08:49:40 am
Quote from: michael
They must have some use ;-)

Well, of course they do. I never wrote otherwise. This completely misses the point of what I've written though.

Michael, glad you explained, but in my opinion you did write they were of no use:

From what's new:
QUOTE: We've all done it. There's a photograph or a 100% crop of an image online, taken with a camera or lens that we're interested in, and we judge its image quality capabilities (at least in part) on the basis of that screen image.

Wrong!
UNQUOTE

Unless I'm really wrong interpreting the words (at least in part), which is possible since English is not my native language.
The way I understand that sentence is that even partly judging sensor and lens qualities from internet pics and 100% crops is wrong, or did you have a different thought in mind when you wrote that  
Title: Rant 23
Post by: Eric Myrvaagnes on October 03, 2009, 09:42:52 am
Quote from: pegelli
Michael, glad you explained, but in my opinion you did write they were of no use:

From what's new:
QUOTE: We've all done it. There's a photograph or a 100% crop of an image online, taken with a camera or lens that we're interested in, and we judge its image quality capabilities (at least in part) on the basis of that screen image.

Wrong!
UNQUOTE

Unless I'm really wrong interpreting the words (at least in part), which is possible since English is not my native language.
The way I understand that sentence is that even partly judging sensor and lens qualities from internet pics and 100% crops is wrong, or did you have a different thought in mind when you wrote that  
Pieter,

You are correctly parsing what Michael said in English (although I think the rest of the essay clarified his views). The problem is, in English we have a peculiar custom of insisting that the listener/reader should always understand what I meant regardless of what I actually said.

Title: Rant 23
Post by: Jack Flesher on October 03, 2009, 09:49:29 am
Quote from: Ray
Well, there you have it. A good monitor, properly calibrated, is not a bad device for getting a *fair idea* of what your print will look like, using proof setup in PS of course, and a good paper/ink profile.

As long as you leave the bolded modifier in I'd agree, however/but another thing is equally certain and I believe more to Michael's point:

When and where you can see differences between two images being proofed on your monitor you will almost certainly see those same differences in a good print, the converse is NOT necessarily true; you often detect differences in prints you cannot preview on your monitor!  The corollary point is that anybody who makes larger, wide-gamut prints with regularity knows this quite well, while folks who prefer theorizing about linear sensor response curves before the AD converter and post processing get applied, for whatever reason seem to refuse to acknowledge such differences can possibly even exist in the print side. You can lead a horse to water...  

Later,
Title: Rant 23
Post by: pegelli on October 03, 2009, 10:15:48 am
Quote from: EricM
Pieter,

You are correctly parsing what Michael said in English (although I think the rest of the essay clarified his views). The problem is, in English we have a peculiar custom of insisting that the listener/reader should always understand what I meant regardless of what I actually said.

Maybe that explains it, however a few points:

I think I've found another cause for the confusion. In the essay he says:

QUOTE: But what about looking at images at 100% on-screen? There one can see the differences – right? Well, yes, one can see the difference of how they look on-screen at 100% magnification. The fact that this has little to nothing to do with how an image will appear in a print seems to escape many people. This applies to judging high ISO noise, resolution and more. UNQUOTE

While the "What's new" text talks about judging lens and camera image qualities being judged from 100% crops.

So in my mind the confusion lies between the difference between how an image will actually look in print (which I agree you can only see in a print) and judging certain lens and camera image qualities, which in my mind you can still "in part" do from 100% crops.

Title: Rant 23
Post by: Ray on October 03, 2009, 10:24:24 am
Quote from: Jack Flesher
When and where you can see differences between two images being proofed on your monitor you will almost certainly see those same differences in a good print, the converse is NOT necessarily true; you often detect differences in prints you cannot preview on your monitor!

Jack,
You seem to be implying that two images that may look identical on the same monitor in the same color space, with same embedded profile and same rendering intent in proof setup, may print differently on the same printer using the same paper and ink. Is this true?

If it is true, then that is surely a printing problem which would require a test run before each print. I can appreciate that the latest Epson printers with improved color gamut may print subtle hues that are not visible on a monitor that may not be able to display even the full gamut of ARGB. However, in my experience such differences really are subtle and should not invalidate the comparison of image quality on the monitor.
Title: Rant 23
Post by: ErikKaffehr on October 03, 2009, 11:36:12 am
Hi,

Screens and printers are different. The color gamut volume of screens is actually larger than the gamut volume printers. Printers can reproduce some colors outside the screen RGBs but this does not apply to very bright or very dark colors. Especially LCD screens have large DR compared to print. Resolutionswise it's a different game. In my view an actual pixel view is much more demanding than print.

Also, ignorance of physics does not make for a better photographer but not necessarily a worse one, just keep in mind that we would not have the art if we did not have the physics.

Or to put it another way: Photography may be art but it's based on physics. We don't need to understand the physics in order to create art. On the other hand. art does not suffer from knowledge and actually understanding the process may even be helpful in improving the execution of the art.

Best regards
Erik

Quote from: Jack Flesher
As long as you leave the bolded modifier in I'd agree, however/but another thing is equally certain and I believe more to Michael's point:

When and where you can see differences between two images being proofed on your monitor you will almost certainly see those same differences in a good print, the converse is NOT necessarily true; you often detect differences in prints you cannot preview on your monitor!  The corollary point is that anybody who makes larger, wide-gamut prints with regularity knows this quite well, while folks who prefer theorizing about linear sensor response curves before the AD converter and post processing get applied, for whatever reason seem to refuse to acknowledge such differences can possibly even exist in the print side. You can lead a horse to water...  

Later,
Title: Rant 23
Post by: cmi on October 03, 2009, 12:33:21 pm
Quote from: ErikKaffehr
...and actually understanding the process may even be helpful in improving the execution of the art.

Absolutely Erik!

The artistic approach is about achieving practical results, it is not exact and purely result-centered. (One example, since the issue came up earlier: It is not important if sharpening increases resolution or edge contrast. You just adjust the slider and decide with your eyes.)

The engineering/scientific approach on the other hand is about scientific correctness, repeatability, exact comparisation. It is essential e.g. for Software and Hardware developers - applied science.

An artistic discussion in order to be useful should revolve strictly around the image and be held by real photographers. Technical explanations need to take the standpoint of the artist and come from someone with real working experience in the field. IF explanations come from a only theoretical side they must at least be understandable by an artistic mind, formulated towards that. So theoreticians and engineers should not post if they are not able to relate to the artistic standpoint and are not able to craft good and intuitive explanations. On the other hand, people who ARE able to give good technical explanation in the artistic context are always welcome, at least thats my experience.

Cheers,

Christian
Title: Rant 23
Post by: cmi on October 03, 2009, 12:55:20 pm
Oh and I forgot someting, what we have here at LL is some mixup of the both. The analytical side arguing why the artists are ignorant.  //edit:// And sometimes also the other way around //edit end// This is unfortunate. Obviously the artist can make his image without being a physicist. At the end I am not required to be an engineer to do photos. Quite the contrary. Also the artist is not obliged to take the position of a scientist into account. He will do his art without it. Intuition and analytic thinking, different beasts. Not do downtalk the importance of the technical happenings in the camera through. And no doubt the best people out there in the business of computer graphics utilize both their artistic and scientfic knowledge, thats out of dispute. But what we had happen here in the past where no enlightening technical discussions. Everything else but that. These where to large parts absurd, academic and childish disputes. But I dont want to expand on that.

At the end no question both sides scientific and artistic, have each their merit in their own respective context. But most importantly they should respect - and hopefully also value each other (through the latter being maybe a bit of a far call currently).

Christian
Title: Rant 23
Post by: bjanes on October 03, 2009, 01:19:07 pm
Quote from: Jack Flesher
When and where you can see differences between two images being proofed on your monitor you will almost certainly see those same differences in a good print, the converse is NOT necessarily true; you often detect differences in prints you cannot preview on your monitor!  The corollary point is that anybody who makes larger, wide-gamut prints with regularity knows this quite well, while folks who prefer theorizing about linear sensor response curves before the AD converter and post processing get applied, for whatever reason seem to refuse to acknowledge such differences can possibly even exist in the print side. You can lead a horse to water...  

Later,

Wide gamut is a misnomer for prints. As previously pointed out, a decent monitor has a much wider dynamic range than a reflection print. Likewise, the color gamut of prints is restricted. Bruce Fraser admonished that you should temporarily look away from the monitor when displaying a soft proof lest you be shocked by the reduced gamut of the soft proof.

When differences are apparent on the print that are not apparent on the screen, it is likely that the tonal range or color gamut has been mapped to fit into the limited gamut of the print. Such rendering is not usually done for screen output.

What digital displays lack is resolution. It is not possible to view a 60 megapixel image all at once on screen. However, the field of view for sharp human vision is quite limited, and one can not view a large print all at once, but must move the eyeballs so as to view small portions at a time. However, if one views the screen at a suitable distance so that the angular resolution in pixels per degree rather in pixels per inch is the same on both media, then a better comparison can be made. However, panning of the screen is not as easy or natural than merely moving one's glance around.

Has Mr. Flesher ever compared a high resolution large format transparency to his prints?
Title: Rant 23
Post by: soslund on October 03, 2009, 03:38:47 pm
Comedic interruption.....

I was reading this "rant" in between seeing patients in the Emergency Department where I work, when a disgruntled patient heaved his "petard" (aka "pee bottle" aka urinal) --yes, it was full of the warm yellow liquid--at no one in particular.  Thank goodness urine is sterile (mostly!!).  

Scott
Title: Rant 23
Post by: Ray on October 03, 2009, 08:39:15 pm
Quote from: bjanes
Wide gamut is a misnomer for prints. As previously pointed out, a decent monitor has a much wider dynamic range than a reflection print. Likewise, the color gamut of prints is restricted. Bruce Fraser admonished that you should temporarily look away from the monitor when displaying a soft proof lest you be shocked by the reduced gamut of the soft proof.


Well, I can certainly affirm that I sometimes struggle to get an image, as viewed on the monitor after ticking the 'proof colors' box, to match the vibrancy of the image without proof colors enabled. It's always possible after much adjustment of contrast, saturation and brightness to get reasonably close. However, after making such fairly radical adjustments to get the proof image looking close to the way it previously looked before enabling proof colors, one always finds that the adjusted image then looks even more vibrant, and sometimes preferrable, when one unticks the 'proof colors' box and toggles between the two modes.

So I would agree that the transmissive nature of the monitor gives it an advantage which the reflective nature of the print cannot match, but the point Jack appears to be making is slightly different. He claims that, if two images of the same subject taken with different cameras, viewed side by side on the monitor, at print size, look identical, it does not necessarily follow that they will look identical when both are printed on the same printer using the same paper, ink and profile.

Even my old Epson 7600 claims to be able to print a few shades and hues which are contained within the ProPhoto RGB gamut and which cannot be displayed on the average monitor, or perhaps even any monitor. Epson have now apparently increased that gamut capability with there latest printers such as the 7900.

In other words, if the two images have an embedded ProPhoto RGB profile and are adjusted on the monitor in the ProPhoto RGB working space so that they both look as identical as possible, they may not actually be identical because there are subtle hues which the monitor cannot display but which a good printer, such as the Epson 7900, can reproduce.

I've never actually witnessed such an effect myself. It would be very troublesome and time-consuming to try and isolate such an effect. The mere fact that a particular print does not appear to quite match the appearance of the image on the monitor does not demonstrate the point Jack is making. One needs to compare print with print of apparently identical images taken with different cameras.

Two candidates for such a comparison would be the Canon 5D and the Nikon D700, both of which I own. They both have a very similar pixel pitch and pixel count so it should not be difficult, after ajustment of temperature, tint and levels etc, to get two images of the same subject looking very similar on the monitor.

However, if I were to go to the trouble of making such a comparison and were to discover that there was no significant difference between the two printed images, over and above any small differences apparent on the monitor, it would not disprove Jack's point. It's really up to Jack to demonstrate his point, but I don't see how he can if the reasons for such differences are due to the print containing hues that are not visible on the monitor.
Title: Rant 23
Post by: Jack Flesher on October 03, 2009, 08:39:31 pm
Quote from: Ray
Jack,
You seem to be implying that two images that may look identical on the same monitor in the same color space, with same embedded profile and same rendering intent in proof setup, may print differently on the same printer using the same paper and ink. Is this true?

No Ray, just like Michael in his rant I am talking about similar images from two different camera systems looking similar in certain areas on the monitor, but those areas printing out differently in a properly color-managed workflow.  

So now do you see that Michael is right?  Some folks simply refuse to get it, instead trying to twist the topic to some other discussion.
Title: Rant 23
Post by: Ray on October 03, 2009, 08:47:58 pm
Quote from: Jack Flesher
No Ray, just like Michael in his rant I am talking about similar images from two different camera systems looking similar in certain areas on the monitor, but those areas printing out differently in a properly color-managed workflow.  

So now do you see that Michael is right?  Some folks simply refuse to get it, instead trying to twist the topic to some other discussion.

A very subtle effect, I bet. Remember Michael's comparison of the P45+ and Canon G10 on A3+ prints? No-one could tell the difference, except for the fact the P45+ print had a slightly shallower DoF which sort of gave the game away.
Title: Rant 23
Post by: JohnKoerner on October 03, 2009, 09:42:57 pm
Quote from: michael
A 100% view onscreen may tell you what you need to know about some aspect of image quality (except gamut, of course), but it tells you little about how the image from a particular lens or sensor will appear in a print.

I think your rant is aptly named "rant," because I think there is an enormous amount of exaggeration of what is essentially a triviality.

If the images we see onscreen really told us "little" about how they would appear in print, then what would be the point of having color-calibrated monitors, and then using our monitors to select from our entire montage of images WHICH photos we're going to print? There would be none.

If we took your logic of the unrelatedness of onscreen-image-to-print to the extreme, then we I guess we shouldn't use or depend our monitors to judge our photos at all; I suppose we should just print every image out that we ever take, and then base our analyses on how each image looks in print. This of course is ludicrous.

The truth is, the view that we see onscreen (on a properly-calibrated monitor) tells us ALOT about how our image is going to look on print (not "a little"), and this is why 100% of all photographers use their computer screens for this purpose, including yourself.

What really deserves the monker of "little" in this discussion is the amount of variance there actually is post-print. Sure, there may be tiny nuances in the final printed images, not seen onscreen, but there will be far more similarities than differences. Maybe these subtleties in the final print do warrant a subject for discussion ... but they are hardly worth a rant and litany against all onscreen evaluations. In the end, the biggest fallacy of all is to say that the images we see onscreen "tell us little" about how they will look in print; this is nothing more than an over-exaggeration. If anything, many of the errors seen blown-up onscreen don't show up in the final print.

Jack
Title: Rant 23
Post by: ErikKaffehr on October 04, 2009, 01:28:18 am
Hi,

Ray, your reasoning seems absolutely correct to me. One way to achieve that could be to take an image ProPhoto RGB image convert it to sRGB and than reconvert it to ProPhoto RGB. That would give us two images one with ProPhoto Gamut and another one beeing in ProPhoto RGB but with sRGB contents. Perhaps they would be identical on an sRGB screen and they may print different.

Best regards
Erik


Quote from: Ray
...

Even my old Epson 7600 claims to be able to print a few shades and hues which are contained within the ProPhoto RGB gamut and which cannot be displayed on the average monitor, or perhaps even any monitor. Epson have now apparently increased that gamut capability with there latest printers such as the 7900.

In other words, if the two images have an embedded ProPhoto RGB profile and are adjusted on the monitor in the ProPhoto RGB working space so that they both look as identical as possible, they may not actually be identical because there are subtle hues which the monitor cannot display but which a good printer, such as the Epson 7900, can reproduce.
Title: Rant 23
Post by: ErikKaffehr on October 04, 2009, 01:45:02 am
Hi,

There are many things to discuss. In a sense screens and prints are very different:

- Screens are high contrast while prints are low contrast
- Color spaces are vastly different
- Screen resolution is discrete there is no smaller detail than the pixels themselves

- If we look at low resolution images we don't see detail, there is no way to say if a Canon L lens is sharper than a "cheapo" zoom in 600x800 images.
- If we look at actual pixels we enlarge the smallest picture elements extremely, we would never observe that level of detail in prints because our eyes are not that very sensitive to detail. Our visual impression is dominated by feature sizes of perhaps 0.3 mm. For this reason minor differences visible on screen may just disappear in print, although they may show up when the print is inspected with a magnifier.
- The actual pixel view exaggerates the differences

Summing up:

- Small images contain very little information about pixel level image quality.
- Actual pixel view shows differences that cannot be see in print.
- IMHO an image that is sharper at actual pixel will never be worse in print than an image that is less sharp, if other parameters are constant, that is.

Best regards
Erik

Quote from: JohnKoerner
I think your rant is aptly named "rant," because I think there is an enormous amount of exaggeration of what is essentially a triviality.

If the images we see onscreen really told us "little" about how they would appear in print, then what would be the point of having color-calibrated monitors, and then using our monitors to select from our entire montage of images WHICH photos we're going to print? There would be none.

If we took your logic of the unrelatedness of onscreen-image-to-print to the extreme, then we I guess we shouldn't use or depend our monitors to judge our photos at all; I suppose we should just print every image out that we ever take, and then base our analyses on how each image looks in print. This of course is ludicrous.

The truth is, the view that we see onscreen (on a properly-calibrated monitor) tells us ALOT about how our image is going to look on print (not "a little"), and this is why 100% of all photographers use their computer screens for this purpose, including yourself.

What really deserves the monker of "little" in this discussion is the amount of variance there actually is post-print. Sure, there may be tiny nuances in the final printed images, not seen onscreen, but there will be far more similarities than differences. Maybe these subtleties in the final print do warrant a subject for discussion ... but they are hardly worth a rant and litany against all onscreen evaluations. In the end, the biggest fallacy of all is to say that the images we see onscreen "tell us little" about how they will look in print; this is nothing more than an over-exaggeration. If anything, many of the errors seen blown-up onscreen don't show up in the final print.

Jack
Title: Rant 23
Post by: Ray on October 04, 2009, 01:22:45 pm
Quote from: ErikKaffehr
Ray, your reasoning seems absolutely correct to me. One way to achieve that could be to take an image ProPhoto RGB image convert it to sRGB and than reconvert it to ProPhoto RGB. That would give us two images one with ProPhoto Gamut and another one beeing in ProPhoto RGB but with sRGB contents. Perhaps they would be identical on an sRGB screen and they may print different.


Erik,
Good idea! What I've done is take the saturated swatches in Photoshop to create an image of colored dots in a new document with embedded ProPhoto profile. I then duplicated the image and converted the duplicate to sRGB. I then applied the Bill Atkinson profile for Epson Enhanced Matte in proof setup, ticked gamut warning and brought all colors within gamut by selecting each color with the magic wand and reducing the saturation of that color only, by an amount just sufficient, but no more than sufficient, to bring the color back within gamut.

I then reconverted the sRGB image to ProPhoto and printed them together, side by side on the same sheet of paper, using absolute colorimetric rendering intent.

What I've found is that the saturated CMYK colors (red, yellow and green in the lower portion of the image below) are very similar whether sRGB or ProPhoto. The RGB colors (red, yellow and green in the upper portion of the image) differ the most. ProPhoto RGB-yellow looks clearly a more solid and saturated yellow than sRGB. ProPhoto RGB-red seems to have less yellow and the green looks more blue.

These differences are equally visible on both my monitor and the print. I haven't come across any significant color differences so far that appear on print but are not apparent on my monitor. Is there a flaw in my methodology, or have I just got a good monitor well calibrated? It's an old-fashioned Sony Trinitron.

Do you see the differences that I see in the image below?

[attachment=16953:sRGB___P...to_space.jpg]

Edit: Forgot to mention. The above image has an embedded ProPhoto profile. You'll need to open it in Photoshop to see what I'm talking about.
Title: Rant 23
Post by: ErikKaffehr on October 04, 2009, 03:31:28 pm
Ray,

I see a difference on the colors in the top and the middle on both of my color calibrated monitors, the first monitor is sRGB and the other one essentially AdobeRGB, I have no comments on your methodology and did not print.


Best regards
Erik

Quote from: Ray
Erik,
Good idea! What I've done is take the saturated swatches in Photoshop to create an image of colored dots in a new document with embedded ProPhoto profile. I then duplicated the image and converted the duplicate to sRGB. I then applied the Bill Atkinson profile for Epson Enhanced Matte in proof setup, ticked gamut warning and brought all colors within gamut by selecting each color with the magic wand and reducing the saturation of that color only, by an amount just sufficient, but no more than sufficient, to bring the color back within gamut.

I then reconverted the sRGB image to ProPhoto and printed them together, side by side on the same sheet of paper, using absolute colorimetric rendering intent.

What I've found is that the saturated CMYK colors (red, yellow and green in the lower portion of the image below) are very similar whether sRGB or ProPhoto. The RGB colors (red, yellow and green in the upper portion of the image) differ the most. ProPhoto RGB-yellow looks clearly a more solid and saturated yellow than sRGB. ProPhoto RGB-red seems to have less yellow and the green looks more blue.

These differences are equally visible on both my monitor and the print. I haven't come across any significant color differences so far that appear on print but are not apparent on my monitor. Is there a flaw in my methodology, or have I just got a good monitor well calibrated? It's an old-fashioned Sony Trinitron.

Do you see the differences that I see in the image below?

[attachment=16953:sRGB___P...to_space.jpg]

Edit: Forgot to mention. The above image has an embedded ProPhoto profile. You'll need to open it in Photoshop to see what I'm talking about.
Title: Rant 23
Post by: MarkKay on October 04, 2009, 05:12:45 pm
I think Michael makes some valid points in the Rant article. While monitors can help in certain comparisons, I agree with what Jack Flesher states in an earlier post.   However,  I am wondering whether or not there are hands on  reviews comparing monitors (after careful and accurate calibration),  which give the best results in terms of being able to display  some of the more subtle differences (e.g. widest gamut, tones, detail etc) discussed in this thread.  I suspect no single monitor is best for all. Perhaps I have just not seen such reviews.
Title: Rant 23
Post by: Jack Flesher on October 04, 2009, 05:39:11 pm
If you want to see small gamut shortfalls in print, an easy way to do it is this:

1)  download this printer evaluation image (it's a 40MB, Pro RGB tiff): http://www.outbackprint.com/printinginsigh...i048/essay.html (http://www.outbackprint.com/printinginsights/pi048/essay.html)

2) It is in Prophoto RGB and already sized to fit on a letter-size sheet,  so save a native copy somewhere safe, then print it out properly using whatever your favorite paper and profile combination is -- I recommend Epson's Premium (or Canon's or HP's) Luster as it has a larger gamut than most matte papers and is readily available. Go ahead and use Epson's (or HP's or Canon's) canned profile for your printer if you don't have a custom one.

3)  Now convert the test image to sRGB and print the sRGB version on the same paper using the proper paper profile -- the only difference is this time you'll be printing from sRGB converted to your paper profile, not Prophoto RGB to your paper profile as above.  

4) Now compare the prints side-by-side.  You should easily be able to see the poor rendering of the sRGB print on the greens and blues of the test patch squares just under the grayscale image in the test image.  If you can't see any differences there, you are not printing properly or your printer/paper combo has an inordinately small gamut to begin with...

Cheers,
Title: Rant 23
Post by: ErikKaffehr on October 04, 2009, 11:46:51 pm
Hi,

Interestingly there are a lot of scientists who also are excellent photographers.

Best regards
Erik


Quote from: Christian Miersch
Oh and I forgot someting, what we have here at LL is some mixup of the both. The analytical side arguing why the artists are ignorant.  //edit:// And sometimes also the other way around //edit end// This is unfortunate. Obviously the artist can make his image without being a physicist. At the end I am not required to be an engineer to do photos. Quite the contrary. Also the artist is not obliged to take the position of a scientist into account. He will do his art without it. Intuition and analytic thinking, different beasts. Not do downtalk the importance of the technical happenings in the camera through. And no doubt the best people out there in the business of computer graphics utilize both their artistic and scientfic knowledge, thats out of dispute. But what we had happen here in the past where no enlightening technical discussions. Everything else but that. These where to large parts absurd, academic and childish disputes. But I dont want to expand on that.

At the end no question both sides scientific and artistic, have each their merit in their own respective context. But most importantly they should respect - and hopefully also value each other (through the latter being maybe a bit of a far call currently).

Christian
Title: Rant 23
Post by: Ray on October 05, 2009, 12:11:10 am
Quote from: Jack Flesher
If you want to see small gamut shortfalls in print, an easy way to do it is this:

1)  download this printer evaluation image (it's a 40MB, Pro RGB tiff): http://www.outbackprint.com/printinginsigh...i048/essay.html (http://www.outbackprint.com/printinginsights/pi048/essay.html)

2) It is in Prophoto RGB and already sized to fit on a letter-size sheet,  so save a native copy somewhere safe, then print it out properly using whatever your favorite paper and profile combination is -- I recommend Epson's Premium (or Canon's or HP's) Luster as it has a larger gamut than most matte papers and is readily available. Go ahead and use Epson's (or HP's or Canon's) canned profile for your printer if you don't have a custom one.

3)  Now convert the test image to sRGB and print the sRGB version on the same paper using the proper paper profile -- the only difference is this time you'll be printing from sRGB converted to your paper profile, not Prophoto RGB to your paper profile as above.  

4) Now compare the prints side-by-side.  You should easily be able to see the poor rendering of the sRGB print on the greens and blues of the test patch squares just under the grayscale image in the test image.  If you can't see any differences there, you are not printing properly or your printer/paper combo has an inordinately small gamut to begin with...

Cheers,


Thanks for the link to that test image, Jack. (You're not such a bad bloke after all   .)

I've downloaded the image, printed the ProPhoto version next to the sRGB conversion (which I reconverted to the Prophoto profile for ease of printing since I'm on a 24" roll).  I have to say that your test image merely confirms what I either knew already or suspected might be the case.

I know that Premium Lustre has a slightly wider gamut than Enhanced Matte. I have both profiles loaded in my system, but unfortunately for me, changing matte black ink on my Epson 7600 is too much of a hassle. I'll leave that chore till my matte black cartridge is empty, or my Enhanced Matte roll is finished (I use the South African method). I'd like to upgrade to an Epson 7900 but apparently my old 7600 has no resale value, and it's currently performing flawlessly (touch wood).

Some comments. First, I don't need to be convinced that ProPhoto RGB has a wider gamut than sRGB and ARGB and that Epson printers are capable of printing certain shades that are outside the gamut of both sRGB and ARGB.

The issue, in relation to Michael's rant, is whether such differences in shades of color, reproducible on the print with embedded ProPhoto profile, are significantly different to what one might see on the monitor. My view is, they are not, based on the performance of my own equipment and profiles (courtesy of Bill Atkinson - thanks! Bill).

Examining the differences between the two images on Epson Enhanced Matte, I see there are very few areas in the real-world sections of that test image where the Prophoto print is different. The differences are mostly apparent in the colored squares where a couple of shades of green and a couple of shades of cyan have merged into one shade in the sRGB image.

I've long been of the opinion that a major difference between sRGB and ProPhoto RGB is in the rendition of yellows. If we examine the autumn scene of the woods in your test image (at 100%), there's no doubt in my mind that, on both print and monitor, those yellow, fading leaves are more golden in the ProPhoto image. That difference, oddly enough, is not reflected as significantly in the yellow squares.

I should stress also that such differences are more apparent on the monitor when proof setup is enabled, with either Enhanced Matte profile or Premium Lustre profile selected, in my situation.

The following crop of the colored squares from your test image shows the merging of the two shades of green and (separately) the two shades of cyan. The differences in the autumn woods' scene are only apparent in relation to my profiles for the Epson 7600. I tried to upload the two images of the autumn leaves with embedded Premium Lustre profile, to demonstrate the more golden effect of ProPhoto RGB, but failed. I presume the web cannot handle such profiles.

The image below has an embedded ProPhoto RGB profile. It's only meaningful in a program like Photoshop.

[attachment=16962:Colored_Squares.jpg]

Edit: Actually, the differences are so great that on my monitor those merged shades in sRGB, as opposed to the distinct shades in ProPhoto, are obvious without even opening the image in PS.
Title: Rant 23
Post by: ErikKaffehr on October 05, 2009, 12:33:32 am
Hi,

I have also printed the images, and noted the differences in the green patches after reading Jack Flesher's advice on reading the test images. After that I went back to Lightroom  and looked at the patches in comparison mode. On my iMac screen (essentially sRGB)  the patches were essentially the same. On my right screen (which essentially Adobe) I could see the same differences as in print.

This was interesting, thank you all!

Best regards
Erik


Quote from: Ray
Thanks for the link to that test image, Jack. (You're not such a bad bloke after all   .)

I've downloaded the image, printed the ProPhoto version next to the sRGB conversion (which I reconverted to the Prophoto profile for ease of printing since I'm on a 24" roll).  I have to say that your test image merely confirms what I either knew already or suspected might be the case.

I know that Premium Lustre has a slightly wider gamut than Enhanced Matte. I have both profiles loaded in my system, but unfortunately for me, changing matte black ink on my Epson 7600 is too much of a hassle. I'll leave that chore till my matte black cartridge is empty, or my Enhanced Matte roll is finished (I use the South African method). I'd like to upgrade to an Epson 7900 but apparently my old 7600 has no resale value, and it's currently performing flawlessly (touch wood).

Some comments. First, I don't need to be convinced that ProPhoto RGB has a wider gamut than sRGB and ARGB and that Epson printers are capable of printing certain shades that are outside the gamut of both sRGB and ARGB.

The issue, in relation to Michael's rant, is whether such differences in shades of color, reproducible on the print with embedded ProPhoto profile, are significantly different to what one might see on the monitor. My view is, they are not, based on the performance of my own equipment and profiles (courtesy of Bill Atkinson - thanks! Bill).

Examining the differences between the two images on Epson Enhanced Matte, I see there are very few areas in the real-world sections of that test image where the Prophoto print is different. The differences are mostly apparent in the colored squares where a couple of shades of green and a couple of shades of cyan have merged into one shade in the sRGB image.

I've long been of the opinion that a major difference between sRGB and ProPhoto RGB is in the rendition of yellows. If we examine the autumn scene of the woods in your test image (at 100%), there's no doubt in my mind that, on both print and monitor, those yellow, fading leaves are more golden in the ProPhoto image. That difference, oddly enough, is not reflected as significantly in the yellow squares.

I should stress also that such differences are more apparent on the monitor when proof setup is enabled, with either Enhanced Matte profile or Premium Lustre profile selected, in my situation.

The following crop of the colored squares from your test image shows the merging of the two shades of green and (separately) the two shades of cyan. The differences in the autumn woods' scene are only apparent in relation to my profiles for the Epson 7600. I tried to upload the two images of the autumn leaves with embedded Premium Lustre profile, to demonstrate the more golden effect of ProPhoto RGB, but failed. I presume the web cannot handle such profiles.

The image below has an embedded ProPhoto RGB profile. It's only meaningful in a program like Photoshop.

[attachment=16962:Colored_Squares.jpg]
Title: Rant 23
Post by: Jack Flesher on October 05, 2009, 09:20:31 am
Quote from: Ray
Thanks for the link to that test image, Jack. (You're not such a bad bloke after all   .)

SNIP

I know that Premium Lustre has a slightly wider gamut than Enhanced Matte. I have both profiles loaded in my system, but unfortunately for me, changing matte black ink on my Epson 7600 is too much of a hassle.
SNIP
Examining the differences between the two images on Epson Enhanced Matte, I see there are very few areas in the real-world sections of that test image where the Prophoto print is different. The differences are mostly apparent in the colored squares where a couple of shades of green and a couple of shades of cyan have merged into one shade in the sRGB image.

Ray,

You are welcome .

FWIW, I have found that the "slightly wider" gamut you mention between x600 UC MK on art paper and x600 UC PK on Luster is in actuality a pretty significant difference visually, especially on vegetation; EPPLuster just holds ink really well compared to most any MK alternative -- and more so on x800 and x900 printers.

PS: Another thing you'll immediately note with the MK art papers versus Luster, is comparing the overall print DR separation by using the light gray on white patches and and the dark gray on black patches.  With a good Luster profile you can see differences far closer to the edges of each transition zone than you will with any art paper -- and actually if you want to be really geek about analyzing it, you can continue to measure differences on a few of the edge patches with your spectro you cannot see visually.  FWIW...

Cheers,
Title: Rant 23
Post by: madmanchan on October 05, 2009, 09:33:54 am
Along the lines of Jack's post, I have some gamut plot examples here (two papers, plus an old ~sRGB display):

http://people.csail.mit.edu/ericchan/dp/Ep...800/gamuts.html (http://people.csail.mit.edu/ericchan/dp/Epson3800/gamuts.html)

the take-home point being that sometimes there are two different colors that can be seen on the display, which may not be seen in print, and similarly sometimes there are two different colors that appear in print, but cannot be distinguished on the display.
Title: Rant 23
Post by: Ray on October 05, 2009, 01:18:37 pm
Quote from: Jack Flesher
FWIW, I have found that the "slightly wider" gamut you mention between x600 UC MK on art paper and x600 UC PK on Luster is in actuality a pretty significant difference visually, especially on vegetation; EPPLuster just holds ink really well compared to most any MK alternative -- and more so on x800 and x900 printers.

Jack,
For maximum gamut I actually prefer Premium Glossy which has an even wider gamut than Premium Lustre, but unfortunately suffers more from metamerism. There are always trades-off and compromises.

Examining your test image with the Bill Atkinson 'PrmGlossy PGL2' profile applied in Proof Setup (PGL2 for 2880 dpi printing), I see that none of the yellows are out of gamut except for very slight traces in the autumn scene. It would be interesting to compare effects with the Epson profiles for the 7900, but all I can find for download is the MK profile. With regard to the yellows, the 7900 MK profile performs about the same as the 7600 PrmGlossy, but is worse in other areas.

I hope that PS gamut warning is accurate.

Title: Rant 23
Post by: Ray on October 05, 2009, 01:21:45 pm
Quote from: madmanchan
...the take-home point being that sometimes there are two different colors that can be seen on the display, which may not be seen in print, and similarly sometimes there are two different colors that appear in print, but cannot be distinguished on the display.

Eric,
Interesting article! As I understand, LCD monitors are generally technically inferior to CRT monitors with regard to color gamut and contrast ratio, which is why I'm still using my Sony Trinitron. However, it seems difficult to get any precise information on what the color gamut of various monitors actually is, except for the ARGB standard that applies to very expensive LCD monitors. I have the impression that an old CRT monitor may well have a color gamut equivalent to a very expensive LCD monitor. Perhaps that used to be the case but is no longer.

Plasma displays seem to be closer to the old CRT with regard to contrast ratio and other factors. I'm very impressed with the latest HDTV displays from Panasonic. A contrast ratio of 40,000:1 is fantastic.

Is there any reason that you know of, why plasma monitors could not be used for processing images for print?

Title: Rant 23
Post by: tho_mas on October 05, 2009, 01:37:24 pm
Quote from: Ray
As I understand, LCD monitors are generally technically inferior to CRT monitors with regard to color gamut and contrast ratio (...)
? no. CRTs have smaller gamut and less contrast than current LCD monitors.

Quote from: Ray
A contrast ratio of 40,000:1 is fantastic.
what for?
Title: Rant 23
Post by: Ray on October 05, 2009, 01:44:09 pm
Quote from: tho_mas
? no. CRTs have smaller gamut and less contrast than current LCD monitors.

 what for?


Please provide the evidence. I can't find it through a Google search, and as I've already demonstrated with my crops of sRGB versus ProPhoto RGB images, my Sony Trinitron is doing very well.

Quote
what for?

Out-of-gamut color shades are often dark. You need a monitor with a good contrast ratio to see them.
Title: Rant 23
Post by: tho_mas on October 05, 2009, 02:44:06 pm
Quote from: Ray
Please provide the evidence.
If you could provide a profile of your monitor...

Quote from: Ray
Out-of-gamut color shades are often dark. You need a monitor with a good contrast ratio to see them.
with too high contrast you are actually blind as it is intensely demanding and stressing for the eyes. It takes some time to adapt from bright to dark tones if the contrast is too high ... so basically you'll see less in dark tonal values (as the eyes adapt for the brightest tones).
Too, you can't see out of gamut colors as they are simply out of gamut - they are clipped to the monitors color space. What you can do in Photoshop is to set you monitor profile as proof profile, switch off preview but turn on color warning. This way you'll see at least if there are colors in a certain image your monitor can't display. If there are out of gamut colors check by numbers if the colors have modulations/transitions. A handy and helpful tool here is Chromix Color Think - you can load images and compare the colors of the respective images with your printer (or monitor) profile. Color Think does not only show that colors are clipped, it shows how far the clipping goes.
Title: Rant 23
Post by: JeffKohn on October 05, 2009, 02:51:50 pm
Quote
Interesting article! As I understand, LCD monitors are generally technically inferior to CRT monitors with regard to color gamut and contrast ratio, which is why I'm still using my Sony Trinitron.
The only thing CRT's are better at is black-point, and the latest LCD's have improved in that area a lot. But total contrast is much higher on LCD's because they can reach much higher luminosity. As for gamut, there were a few CRT's with extended gamut but not many. With LCD's there are many models with wide gamut support now.

Quote
A contrast ratio of 40,000:1 is fantastic.
A contrast of 40,000:1 is ridiculous. Those are marketing numbers.  You're not going to get that kind of contrast in real-world viewing, nor would you _want_ to.
Title: Rant 23
Post by: Ray on October 05, 2009, 09:18:30 pm
Quote from: tho_mas
If you could provide a profile of your monitor...


I can find no inherent profile specified for my monitor. However, my monitor is profiled with an Eye-One colorimeter. All I have to go by is that some differences between the sRGB color space and the ProPhoto RGB color space are clearly visible on my monitor, comparing Jack Flesher's test image with a copy converted to sRGB. That would imply that the gamut of my monitor is wider than sRGB, would it not?

My monitor is a budget Sony Trinitron, nothing special, and was not marketed as being capable of displaying the full ARGB gamut. I've seen no difference yet between ProPhoto RGB and Adobe RGB on my monitor.

Quote
with too high contrast you are actually blind as it is intensely demanding and stressing for the eyes. It takes some time to adapt from bright to dark tones if the contrast is too high ... so basically you'll see less in dark tonal values (as the eyes adapt for the brightest tones).

I think that's nonsense. You are confusing contrast with brightness. Two monitors can be equally bright but the one with the higher contrast ratio will reveal more detail in the deeper shadows. The eye adapts almost instantly to changes in brightness level. It takes some time for that adjustment only when moving from a prolonged period in extreme darkness to extreme brightness such as bright sunlight.


Quote
Too, you can't see out of gamut colors as they are simply out of gamut - they are clipped to the monitors color space.


Of course. The issue we're discussing is 'out-of-gamut colors' in relation to the monitor which can however be seen on the print. This is what we've been exploring during the last few posts. There are also colors that are within gamut that you can't see. The ProPhoto color space has a few.
Title: Rant 23
Post by: Ray on October 05, 2009, 09:52:30 pm
Quote from: JeffKohn
The only thing CRT's are better at is black-point, and the latest LCD's have improved in that area a lot. But total contrast is much higher on LCD's because they can reach much higher luminosity. As for gamut, there were a few CRT's with extended gamut but not many. With LCD's there are many models with wide gamut support now.

A contrast of 40,000:1 is ridiculous. Those are marketing numbers.  You're not going to get that kind of contrast in real-world viewing, nor would you _want_ to.


That doesn't sound right to me. I admit the contrast ratio of LCDs is improving all the time and the really expensive LCD monitors may well be as good as any CRT monitor was, for all I know. Whilst the greater luminosity of the LCD monitor may be very good for sales (nice and shiny), it's not necessarily good for calibration and photographic processing, as I understand. Relatively low luminosity but very high contrast ratio is best for photographic purposes. The plasma display would seem to be ideal in this respect.

Quote
A contrast of 40,000:1 is ridiculous. Those are marketing numbers.  You're not going to get that kind of contrast in real-world viewing, nor would you _want_ to.

You may be right, but what makes you think that only manufacturers of plasma displays exaggerate their specifications? The fact remains that plasma displays have a reputation for having significantly better CR than LCD displays. I'm not aware that there is any disadvantage in the monitor having a higher contrast ratio than is required to display a particular scene, just as I can see no disadvantage in a camera having a higher dynamic range capability than is required to capture a particular scene with a low SBR.
Title: Rant 23
Post by: tho_mas on October 06, 2009, 03:00:02 am
Quote from: Ray
I think that's nonsense. You are confusing contrast with brightness. Two monitors can be equally bright but the one with the higher contrast ratio will reveal more detail in the deeper shadows.
no nonsense. Contrast is the relation form absolute brightness (i.e. white) to black. With, say, a brightness of 120cd/m2 and a black point of 0.3cd/m2 the contrast is 400:1. At the same brigthness you can increase contrast by lowering the black point (but of course not below 0cd/m2). Still begs the question how well you can differentiate dark tonal values if the contrast is too high.
Title: Rant 23
Post by: Ray on October 06, 2009, 03:33:34 am
Quote from: tho_mas
no nonsense. Contrast is the relation form absolute brightness (i.e. white) to black. With, say, a brightness of 120cd/m2 and a black point of 0.3cd/m2 the contrast is 400:1. At the same brigthness you can increase contrast by lowering the black point (but of course not below 0cd/m2). Still begs the question how well you can differentiate dark tonal values if the contrast is too high.

I don't see how it's possible to have a contrast ratio that's too high. If the contrast ratio in the image you are displaying on your screen is 500:1 but your screen is capable of only 400:1, then you would have to flatten your image to see full detail in the shadows. On the other hand, if the CR of your screen is 2,000:1, then no problem. As long as the CR of the screen is higher than the 'subject brightness range' of the image you are viewing, then I see no problem.

As I understand, the advantage of the plasma display (and the old CRT) is that the luminance of any individual pixel can be completely turned off to provide a true black. LCD monitors rely upon back-lighting which is on to some extent all the time.
Title: Rant 23
Post by: tho_mas on October 06, 2009, 03:56:11 am
Quote from: Ray
If the contrast ratio in the image you are displaying on your screen is 500:1 but your screen is capable of only 400:1, then you would have to flatten your image to see full detail in the shadows.
You seem to think that the contrast ratio cuts the scope of tonal values at the low end. This is not the case due to color management - the entire tonal range is displayed (from RGB 0-0-0 to RGB 256-256-256 in a certain color space) ... just within a lower contrast ratio. Of course you are running into problems if the contrast is below 1:256. I boost the black point of my display to avoid too high contrast. I calibrate to lum. 105cd/m2 (matching my viewing conditions) and black 0.3cd/m2... which is something like 1:330. That is still a higher contrast as any print can reproduce. And I still can differentiate the entire range of tonal values (e.g. in a greyscale target with very fine steps or whatever).

edit: you can simlulate the effect of too high contrast regarding adaptation quite easy: make a big white square in Photoshop. Now make three small squares in the center of the white image side by side - first with RGB 0-0-0, second with RGB 2-2-2, third with RGB 4-4-4.
Hopefully you can differentiate the color patches. Now fill the white background with a mid grey and wait a few seconds - you will be able to differentiate the 3 color patches much better now, no?
Of course this is just a simulation... but it shows clearly how much our perception is affected by the brightest tonal values (white). Hell, even the small white planes in the tool panes of Photoshop may affect adaption as our eyes always adapt to white.
Theoretically a high contrast might be better (as there is more differentiation from, say, RGB 3-3-3 to RGB 4-4-4) ... you just do not profit of a too high contrast as you simply see less in dark tonal values due to adaption... try it out. It's basically the same if you climb on a glacier and than stop off at a chalet without windows but only candlelight... you won't see anything for a few minutes.






Title: Rant 23
Post by: Ben Rubinstein on October 06, 2009, 06:13:28 am
I don't know numbers and I don't understand the figures. What I do know is that I've never seen an LCD that can match my £20 2nd hand CRT's for brightness control or contrast. They (after calibration) are all too bright, too contrasty and too punchy. My CRT's are WYSIWYG for print in both brightness, colour and contrast when the brightness is correct. I could'nt do that with my ACD and I've got friends who say that they have to estimate because their screens are too contrasty, all using latest calibrators. Having to use extremely expensive screens which still can't show the gamut you people all like to use, work in a specific brightness working station to match the overbright screens - and still not get WYSIWYG - doesn't sound like progress to me.
Title: Rant 23
Post by: tho_mas on October 06, 2009, 06:37:59 am
Quote from: pom
I don't know numbers and I don't understand the figures. What I do know is that I've never seen an LCD that can match my £20 2nd hand CRT's for brightness control or contrast.
that says something about the LCDs you have seen but nothing about the capabilities of good LCDs. A problem with medicore LCDs is often that you can't adjust them very well below a certain luminance level... which might still be too bright for some. That's true. Then again good LCDs can be calibrated to low luminance levels.
As to standarized viewing conditions CRTs are simply too dark. The recommended 500lux for a viewing booth translates to 160cd/m2 luminosity for the display (by numbers). Then again this applies to industry standards and might not affect your particular viewing conditions at your personal working station.
Title: Rant 23
Post by: madmanchan on October 06, 2009, 07:01:12 am
Hi Ray, the plots in my article are from an old LCD with a narrow gamut. I did not mean to suggest that all LCDs behave this way. Several current LCDs have a much larger color gamut. This alone does not necessarily make them great for photo editing, but it would change the shape of the plots significantly. Even with the wider gamut LCD displays, however, the basic conclusions for the article still apply -- namely, that there are some colors that can be printed but not previewed on these (wider gamut) displays, and vice versa. Stated mathematically, neither gamut is a proper subset of the other.
Title: Rant 23
Post by: Ben Rubinstein on October 06, 2009, 08:53:48 am
Quote from: tho_mas
that says something about the LCDs you have seen but nothing about the capabilities of good LCDs. A problem with medicore LCDs is often that you can't adjust them very well below a certain luminance level... which might still be too bright for some. That's true. Then again good LCDs can be calibrated to low luminance levels.
As to standarized viewing conditions CRTs are simply too dark. The recommended 500lux for a viewing booth translates to 160cd/m2 luminosity for the display (by numbers). Then again this applies to industry standards and might not affect your particular viewing conditions at your personal working station.

The newer ACD which I owned, a friends iMac 23" and others that I don't remember about. My point is that you need to spend BIG bucks to get an LCD with which you can control the contrast and brightness to what pretty much any simple CRT can do. That stuff about CRT's being too dark is nonsense. I have mine turned down to 28/100 to match print in a brightly lit room. 160 luminosity is stupidly bright for normal conditions, funnily enough not one single one of my clients ows a 'viewing booth'. Real world people, real world....
Title: Rant 23
Post by: JeffKohn on October 06, 2009, 10:35:51 am
Quote
That doesn't sound right to me. I admit the contrast ratio of LCDs is improving all the time and the really expensive LCD monitors may well be as good as any CRT monitor was, for all I know. Whilst the greater luminosity of the LCD monitor may be very good for sales (nice and shiny), it's not necessarily good for calibration and photographic processing, as I understand. Relatively low luminosity but very high contrast ratio is best for photographic purposes. The plasma display would seem to be ideal in this respect.
Today's better LCD's have good enough black points that it's no longer an issue. My Eizo can get to 0.3cdm^2, I don't see any point in going lower than that. My LCD also has excellent grayscale tracking and a color gamut that exceeds sRGB. I held out with CRT for a long time because the early LCD's were no good. But my current display is better than any CRT I ever used.


Quote
You may be right, but what makes you think that only manufacturers of plasma displays exaggerate their specifications?
I never said they were the only ones to exaggerate. Desktop LCD's claiming 10,000:1 contrast ratios are equally stupid. In fact, displays that claim such extremely high contrast ratios are probably not going to calibrate well, because the white point will be too high. You typically want a white point in the 90-150 cdm^2 range depending on your ambient light and personal preferences (I calibrate to 100cdm^2). Many of these super-bright displays won't go that low, or if they do it seriously degrades other aspects of performance.


Quote
I'm not aware that there is any disadvantage in the monitor having a higher contrast ratio than is required to display a particular scene, just as I can see no disadvantage in a camera having a higher dynamic range capability than is required to capture a particular scene with a low SBR.
First, do some simple math. Even if a plasma/crt can hit 0.01 cdm^2, a 40,000:1 contrast ratio would mean a white point of 400 cdm^2. You don't want a display that bright. A contrast range that extreme would be murder on your eyes, especially since to differentiate the darkest tones would require working in a completely dark room. You really don't want a contrast ratio higher than about 300-500:1 for graphics work, and even that is far greater than what can be achieved in print.
Title: Rant 23
Post by: Ray on October 06, 2009, 09:33:51 pm
Quote from: JeffKohn
First, do some simple math. Even if a plasma/crt can hit 0.01 cdm^2, a 40,000:1 contrast ratio would mean a white point of 400 cdm^2. You don't want a display that bright. A contrast range that extreme would be murder on your eyes, especially since to differentiate the darkest tones would require working in a completely dark room. You really don't want a contrast ratio higher than about 300-500:1 for graphics work, and even that is far greater than what can be achieved in print.


You seem to be confusing contrast ratio with brightness, just like tho_mas. If in tandem with a high CR your monitor also has a high brightness, then it's quite likely that you will need to reduce the brightness for calibration purposes and as a consequence the CR will also effectively be reduced.

There have been numerous discussions on the forum about this issue. For photographic purposes a high CR in combination with a relatively low brightness seems to give best results. The LCD has traditionally been weak on this point. It appears to be the case that some people have been fooled by what appears to be a good CR according to manufacturers' specs, are attracted to the monitor because the image is bright in daylight, then find on calibration that the CR of their monitor is not as good as they thought it would be because they have had to turn down the brightness.

Now, I'm not sure if the following simple maths really applies to your example of the 40,000:1 plasma with a brightness of 400 nits. I'm not a calibration expert or a color scientist. However, if we reduce the brightness of that display from 400 nits to, say 80 nits, we end up with a contrast ratio of 40,000 x 80/400 = 8000:1.

Lets say the manufacturer has exaggerated the CR specs. Let's say it's really only 1/4th of what they claim. We still have effectively a monitor with a CR of 2,000:1 and a brightness or white point of 80 nits.

If this calculation is valid (and I accept that it may not be that simple) we may have a clear case of over-kill, but at least we're sure that the CR will never be inadequate. It doesn't matter if the equipment is better than it needs to be, as long as the price is not higher than it needs to be. I imagine there would be difficulties in calibrating such a display, but that's another issue. None of my calibration packages have an option for plasma displays.

Pom's point is quite correct in my view. You need a really expensive LCD monitor to match the qualities of the old CRT. I notice that there's an Eizo monitor that claims to display 97% of the ARGB gamut, the 24" CG241W at a price of A$3,545. It has a claimed CR of 850:1 and a brightness of 300 nits.

I think my 19" Sony Trinitron cost about A$1,200 several years ago. I'm not happy about spending over $3,000 for a replacement, especially now that I've discovered that my old Sony can display a good portion of the Adobe RGB gamut. What portion I don't know, but certainly wider than sRGB.
Title: Rant 23
Post by: JeffKohn on October 06, 2009, 10:58:07 pm
Quote
You seem to be confusing contrast ratio with brightness, just like tho_mas. If in tandem with a high CR your monitor also has a high brightness, then it's quite likely that you will need to reduce the brightness for calibration purposes and as a consequence the CR will also effectively be reduced.
No, I am not confusing anything. Do you know what a contrast ratio is? It's the ratio between the maximum white and the deepest black that a display can reproduce.

Quote
Now, I'm not sure if the following simple maths really applies to your example of the 40,000:1 plasma with a brightness of 400 nits. I'm not a calibration expert or a color scientist. However, if we reduce the brightness of that display from 400 nits to, say 80 nits, we end up with a contrast ratio of 40,000 x 80/400 = 8000:1.

Lets say the manufacturer has exaggerated the CR specs. Let's say it's really only 1/4th of what they claim. We still have effectively a monitor with a CR of 2,000:1 and a brightness or white point of 80 nits.
I was using an absurd example, I don't for a minute believe that any current displays can reach a black level of .01 cdm^2. And even if you cut that by 1/4, a 2000:1 ratio at 80 cdm^2 would require a black level of .04 cdm^2, which is still not realistic. And even if it was, you'd have to sit in a pitch black room to observe such black levels.

Quote
Pom's point is quite correct in my view. You need a really expensive LCD monitor to match the qualities of the old CRT. I notice that there's an Eizo monitor that claims to display 97% of the ARGB gamut, the 24" CG241W at a price of A$3,545. It has a claimed CR of 850:1 and a brightness of 300 nits.

I think my 19" Sony Trinitron cost about A$1,200 several years ago. I'm not happy about spending over $3,000 for a replacement, especially now that I've discovered that my old Sony can display a good portion of the Adobe RGB gamut. What portion I don't know, but certainly wider than sRGB.
I don't know where you got your pricing, maybe that's the official MSRP from Eizo; but I paid $2K for my CG241W about 2 years ago, and ColorMall.com is currently selling them for $1800. Not such a far cry from your $1200 Sony when you consider the Eizo is a larger display. There are also more affordable displays from NEC among others that would still exceed the performance of your Sony CRT (which BTW at $1200 is hardly representative of the CRT's most people were using back in the day).

Title: Rant 23
Post by: Ray on October 07, 2009, 12:08:16 am
Quote from: JeffKohn
No, I am not confusing anything. Do you know what a contrast ratio is? It's the ratio between the maximum white and the deepest black that a display can reproduce.


Exactly! As regards contrast ratio, Plasma displays are at the top of the list, then CRTs, then LCDs. (Although some folks still claim that the CR of a good CRT might still be the best). Of course, there may be some overlap according to price. Development of CRTs ceased long ago. As I mentioned before, both plasma and CRT are able to render blacker blacks because their pixels can be virtually turned off. LCDs have to contend with that back-light which is never turned off.

However, if you pay enough, I accept that a good LCD will be perfectly adequate for photographic purposes, just as my Sony Trinitron is perfectly adequate for my purposes, with the exception of those few subtle shades of color that exist in the ProPhoto space, that may be apparent on a print but not necessarily apparent on my monitor, or yours.

Quote
I was using an absurd example, I don't for a minute believe that any current displays can reach a black level of .01 cdm^2.

Why not? If individual pixels can be turned off so they emit no light whatsoever, then such a monitor should be capable of amazing CR. Whether your viewing conditions are suitable to appreciate such enhanced CR is another matter. I'm very pleased so far with the detail I see on black suits on my partner's 11th generation 50" Panasonic plasma TV with a mere 30,000:1 contrast ratio. I look forward to getting one of Panasonic's latest 12th generation Viera models with an even better CR of 40,000:1 and a claimed dynamic (on/off) CR of 2,000,000:1.

I'd like to get the ultra-thin, wireless 54" model and hang it on the wall like a picture, but I'll probably get the much cheaper, standard 54" model and buy a 5D2 with the price difference   . However, I might over-indulge and get both. ISO 6400 shots on the 5D2, downsampled to 1920x1080, might look okay   .

Quote
I don't know where you got your pricing, maybe that's the official MSRP from Eizo; but I paid $2K for my CG241W about 2 years ago, and ColorMall.com is currently selling them for $1800. Not such a far cry from your $1200 Sony when you consider the Eizo is a larger display. There are also more affordable displays from NEC among others that would still exceed the performance of your Sony CRT (which BTW at $1200 is hardly representative of the CRT's most people were using back in the day).

My prices are all in Australian dollars. A quick google for a best price for the CG241W came up with A$3250. My attitude is, 'if it ain't broke, don't fix it'. I think many of us in the past have struggled to get a good match between monitor and print. I have such a good match. I'll hang on to my current Sony monitor for as long as it continues to provide that good match.

I recently bought an Acer P244w 24" LCD monitor for my laptop which has a BD reader. The monitor is capable of the full HD resolution of 1920x1080 and of course has an HDMI input. I thought it might be useful to watch the occasional Blu-ray movie, but also thought it might calibrate better than my laptop screen. It probably does, but it's not a patch on my Sony Trinitron. Rather poor contrast ratio, but it's a cheap monitor of course.

Edit: By the way, if I compare apples with apples, I'd be looking at an Eizo CG19, same size and aspect ratio as my Sony, but lower maximum resolution. It has no claim for displaying the Adobe RGB gamut, has a CR of 450:1 and a brightness of 280 nits, which is good. But would it be better than my Sony? Best price from a Google search, A$2,200.
Title: Rant 23
Post by: tho_mas on October 07, 2009, 04:29:22 am
Quote from: Ray
now that I've discovered that my old Sony can display a good portion of the Adobe RGB gamut. What portion I don't know, but certainly wider than sRGB.
maybe some day you will discover the directory of your montitor profile on your computer as well ...
Title: Rant 23
Post by: Ray on October 07, 2009, 05:35:06 am
Quote from: tho_mas
maybe some day you will discover the directory of your montitor profile on your computer as well ...


Directory? Do you mean, System32/spool/drivers/color, where my profiles reside? What are you trying to say? Speak clearly.
Title: Rant 23
Post by: tho_mas on October 07, 2009, 05:59:32 am
Quote from: Ray
Directory? Do you mean, System32/spool/drivers/color, where my profiles reside? What are you trying to say? Speak clearly.
I've already asked for your monitor profile above but apparently you are not able to provide it by now:
Quote from: Ray
Quote from: tho_mas
If you could provide a profile of your monitor...
I can find no inherent profile specified for my monitor. However, my monitor is profiled with an Eye-One colorimeter
If you would post your monitor profile I could load it into Color Think and compare it to Jack's color patches and e.g. to my CG241W profile and something more.
Would be much more enlightening (for you!) as your chain of evidence.
Title: Rant 23
Post by: Ray on October 07, 2009, 08:43:12 am
Quote from: tho_mas
I've already asked for your monitor profile above but apparently you are not able to provide it by now:
I can find no inherent profile specified for my monitor. However, my monitor is profiled with an Eye-One colorimeter
If you would post your monitor profile I could load it into Color Think and compare it to Jack's color patches and e.g. to my CG241W profile and something more.
Would be much more enlightening (for you!) as your chain of evidence.


Sorry! I thought you were referring to the manufacturer's profile for my monitor. PM me your email address and I'll send you the latest profile I created.
Title: Rant 23
Post by: bjanes on October 07, 2009, 10:58:19 am
Quote from: madmanchan
Along the lines of Jack's post, I have some gamut plot examples here (two papers, plus an old ~sRGB display):

http://people.csail.mit.edu/ericchan/dp/Ep...800/gamuts.html (http://people.csail.mit.edu/ericchan/dp/Epson3800/gamuts.html)

the take-home point being that sometimes there are two different colors that can be seen on the display, which may not be seen in print, and similarly sometimes there are two different colors that appear in print, but cannot be distinguished on the display.

Eric,

An excellent demonstration of gamut mismatching between color spaces that prompted me to review some of my files and thoughts on the matter. In addition to comparing the gamuts of various color spaces, one should also consider the real world colors contained in those spaces and how those colors can be reproduced on screen and in print. Gernot Hoffmann (http://www.fho-emden.de/~hoffmann/gamuts08072002.pdf) (see page 10 of this large PDF) and Kodak have determined what colors in real world images should be reproduced and this gamut is outside of sRGB and aRGB, and when working in ACR, ProPhotoRGB is the space that can contain these colors. However, some of these colors can not be reproduced on screen (with a sRGB monitor) or in print even with a wide gamut device such as the Epson 9600 with Premium Lustre paper.

Here is an image (original in ProPhotoRGB but shown in sRGB for internet display) of a flower with moderately saturated purple and yellow:

[attachment=17020:DSC_0056_sRGB.jpg]

The gamuts of the image, sRGB, and the Epson 9600 printer with Premium Lustre paper are shown below. The purples are well outside the gamut of the printer, but are mostly contained within sRGB. Such examples as this are the basis for my previous statement in this thread that wide gamut printing is somewhat of a misnomer. The yellows are well outside the sRGB gamut, but are mostly within gamut for the printer.

[attachment=17021:056_Gamuts.png]

This pseudocolor plot demonstrates the ΔEs of the mismatch, giving an indication of how the eye would perceive such differences for the screen and printer. The screen sRGB has a much smaller ΔE for the purple than does the printer. The reverse is true for the yellows.

[attachment=17025:GamutVis..._Ep_9600.png]

[attachment=17023:GamutVison056_sRGB.png]

This image contains greens in the leaves of the lily plant which are out of gamut for both the printer and screen, but more so for the screen. As your plots showed, sRGB is deficient in low luminance greens and the printer is much better with these colors. The Delta Es are smaller for the printer and if one were rendering into sRGB, these would be lost and the gamut of the printer would not have been used to advantage. The difference is about 4 or 5 delta Es, and I am not sure how apparent this would be to the eye in a print.

[attachment=17026:0011_sRGB.jpg]

[attachment=17027:011_Gamuts.png]

[attachment=17028:GamutVis..._Ep_9600.png]

[attachment=17029:GamutVison011_sRGB.png]
Title: Rant 23
Post by: Ray on October 07, 2009, 12:10:39 pm
Quote from: bjanes
Here is an image (original in ProPhotoRGB but shown in sRGB for internet display) of a flower with moderately saturated purple and yellow:

[attachment=17020:DSC_0056_sRGB.jpg]

The gamuts of the image, sRGB, and the Epson 9600 printer with Premium Lustre paper are shown below. The purples are well outside the gamut of the printer, but are mostly contained within sRGB. Such examples as this are the basis for my previous statement in this thread that wide gamut printing is somewhat of a misnomer. The yellows are well outside the sRGB gamut, but are mostly within gamut for the printer.

Wow! That purple really is way out of the gamut of even Premium Glossy. However, colors that can be seen on the monitor, but are outside the gamut of the printer/paper, can be brought back into gamut using selections and reduction of contrast or luminance.

The greater concern would be colors that one can't see on the monitor but which appear on the print, thus creating the impression that perhaps one's monitor calibration or printer/paper profile is not adequate. As a matter of fact, I recall when using ColorEyes on my 32 bit system that I had minor trouble with certain shades appearing slightly different on the print to what my monitor displayed. I later bought an Eye-One because Gretag Macbeth were quicker off the mark to develop a driver for my new Win 64 bit system. I believe my calibration also became more accurate, but it's just an impression. I haven't got the time or the reason to do a rigorous comparison. If something works and does the job, there's little need to look for improvements.

Title: Rant 23
Post by: tho_mas on October 07, 2009, 12:32:09 pm
Thanks for the profile.

So… first with CRTs it's not possible to make hardware calibration (well, it is possible with highend monitors but you need totally different devices).
So the gray axis is linearized on the Video LUT of the graphics card.
In your case the blue channel is cut at the low end.
That means: your monitor doesn't display the full tonal range of an 8bit image due to softwarecalibration and due to a weak gray balance at the low end:
[attachment=17031:01_lut.jpg]

Here's your monitor colour space (white) compared to sRGB.
Basically your monitor matches sRGB quite well … it exceeds sRGB in yellow and green but does not reach the red and blue primaries:
[attachment=17032:02_sony_srgb.jpg]

Here's the same but in addition with AdobeRGB:
[attachment=17033:03_sony_srgb_argb.jpg]

Here's your monitor (white), AdobeRGB (colored) and my CG241W (yellow).
My monitor is quite similar to AdobeRGB. The primaries blue and green are a little less saturated, red exceeds AdobeRGB.
So after 6545 hours of use it's pretty much within the marketed specs of the above mentioned 97% AdobeRGB.
[attachment=17034:04_sony_argb_eizo.jpg]

From Jack's test image I took the portraits (as a "real world" image).
Here's the colour plot of the portraits:
[attachment=17035:04_portraits_color.jpg]

The same compared to your monitor profile:
[attachment=17036:05_portr...lor_sony.jpg]

The same compared to the Eizo:
[attachment=17037:06_portr...ony_eizo.jpg]

So regarding your post #90: evidence provided (regarding the gamut of a mid range LCD - the CG241W is not bad, but there are displays with higher gamut).
Too, due to hardware calibration the Eizo displays the full tonal range with fine transitions (so without banding) and with a clean gray axis from white to black (the worst variance is 0.5 DeltaE). You'll get similar results even with cheaper Eizo displays as the are linearized ex factory quite good (AFAIK the same applies to NEC displays).
Title: Rant 23
Post by: bjanes on October 07, 2009, 12:37:11 pm
Quote from: Ray
Wow! That purple really is way out of the gamut of even Premium Glossy. However, colors that can be seen on the monitor, but are outside the gamut of the printer/paper, can be brought back into gamut using selections and reduction of contrast or luminance.

The greater concern would be colors that one can't see on the monitor but which appear on the print, thus creating the impression that perhaps one's monitor calibration or printer/paper profile is not adequate.

Yes, when working in ProPhotoRGB in Photoshop and using current monitors, there is a problem with softproofing with images that are within the gamut of the printer but outside the gamut of the monitor. When editing the image to fit within the printer gamut, the result can not be shown on the monitor. One can use the out of gamut warning, but this is all or none and the result could be just out of gamut or out of gamut by a large amount. One work around would be to have a pseudocolor preview showing Delta Es similar to the GamutVison plots I showed in my previous post.

[attachment=17038:GamutWarning_056.png]
Title: Rant 23
Post by: bjanes on October 07, 2009, 12:45:25 pm
Quote from: michael
So my thesis isn't that prints are a superior means of showing ones images (though for me they are). I realize that many people never print. But, simply that with today's technology they can't show you everything that's in the file the way a print can. Or, if they can, they suffer from very high cost, the extremes of technology, or other impediments.

Michael

It is a mistake to think that a print can show everything in the file. With some images, even an sRGB monitor can easily display colors that are out of gamut of current printers. For example, see my post 110 (http://luminous-landscape.com/forum/index.php?s=&showtopic=37947&view=findpost&p=315590) in this thread,

Regards,

Bill
Title: Rant 23
Post by: Ray on October 07, 2009, 01:25:29 pm
Quote from: tho_mas
Thanks for the profile.

Tho_mas,
Thanks for the analysis. Are these results dependent upon the quality of the calibration? As I mentioned in my email, the profile was generated in automatic mode. I've never felt the need to mess around with manual selections, although I tried that when experimenting with the calibration of a new a plasma TV. I seemed to get errors and gave up..


Quote
Basically your monitor matches sRGB quite well … it exceeds sRGB in yellow and green but does not reach the red and blue primaries:

This sounds bad. On the other hand, bjanes has just posted an image of a purple flower with embedded sRGB profile, that is way out of gamut in relation to my widest gamut paper for my printer, Premium Glossy. Purple is a mixture of blue and red. I see no reason to be concerned about my monitor not being able to display colors that my printer is also not able to display. That my monitor exceeds the sRGB primaries in yellow and green is probably just fine because my printer also exceeds such primaries   .

For me, what counts is getting a workable and sufficiently close match between monitor and print. I have that at present.

Thanks again for your work on this.
Title: Rant 23
Post by: Jack Flesher on October 07, 2009, 03:07:16 pm
Quote from: Ray
This sounds bad.
It certainly isn't good Ray -- your monitor is basically rendering about 6-bits of data.

Quote
On the other hand, bjanes has just posted an image of a purple flower with embedded sRGB profile, that is way out of gamut in relation to my widest gamut paper for my printer, Premium Glossy. Purple is a mixture of blue and red. I see no reason to be concerned about my monitor not being able to display colors that my printer is also not able to display. That my monitor exceeds the sRGB primaries in yellow and green is probably just fine because my printer also exceeds such primaries   .
I suspect if you proof it with a profile for the x900 printer on PPPLuster, you will see a significant improvement to the blues and purples over ANY x600 profile.  (Actually, it is possible you WON'T see it on your monitor!)

Quote
For me, what counts is getting a workable and sufficiently close match between monitor and print. I have that at present.
It maybe is visually, but then your printer is 2-1/2 generations old -- the newest printers are visibly superior. And before everybody gets bent out of shape,  I'm not saying the x600 printers are bad, just that x800 and up (as well as newer HP's and Canons) are going to be visibly better...

Cheers,
Title: Rant 23
Post by: Ray on October 07, 2009, 03:23:21 pm
Quote from: Jack Flesher
It certainly isn't good Ray -- your monitor is basically rendering about 6-bits of data.


I suspect if you proof it with a profile for the x900 printer on PPPLuster, you will see a significant improvement to the blues and purples over ANY x600 profile.  (Actually, it is possible you WON'T see it on your monitor!)


It maybe is visually, but then your printer is 2-1/2 generations old -- the newest printers are visibly superior. And before everybody gets bent out of shape,  I'm not saying the x600 printers are bad, just that x800 and up (as well as newer HP's and Canons) are going to be visibly better...

Cheers,

Jack,
The issue of calibration accuracy has not been answered. I'm not a geek and generally prefer automatic solutions. The profile analyzed by Tho_mas was an automatic product.

Quote
I suspect if you proof it with a profile for the x900 printer on PPPLuster, you will see a significant improvement to the blues and purples over ANY x600 profile.  (Actually, it is possible you WON'T see it on your monitor!)

Maybe. Can we be sure? That purple flower from bjanes was very much out of gamut. With an x900 I would guess it would still be out of gamut, but to a slightly lesser degree.

Could we say we are now engaged in the pixel-peeping equivalent of color rendition?  
Title: Rant 23
Post by: tho_mas on October 07, 2009, 03:26:49 pm
Quote from: Ray
Are these results dependent upon the quality of the calibration?
in part certainly. Colorimeters are not very accurate in dark tonal values... this is another reason to boost the blackpoint to 0.3cd/m2 (of course only on displays that have a lower black point nativley and that are appropriate for hardware calibration). Spectralphotometers are even worse in dark tonal values.
As long as you can work with your equipment and are satisfied with the results that's all fine. But when you once have seen a perfectly calibrated display with wide gamut side by side with a print in a viewing booth you start to wonder if it might be time to move on.
Title: Rant 23
Post by: tho_mas on October 07, 2009, 03:32:21 pm
Quote from: Ray
That purple flower from bjanes was very much out of gamut.
it doesn't make sense to argue from a certain example to the next certain example. As long as you want to print high saturated colours (either way yellow, blue, red, magenta or whatever) a wide gamut display makes sense. As long as you want to see fine transitions without artefacts (especially for black&white) a display with hardware calibration and at least 10bit, better 12bit LUT and 16bit processing makes sense - a lot of sense. Your TV simply can't show what these displays can show. That's pure physics.
Title: Rant 23
Post by: Jack Flesher on October 07, 2009, 03:41:01 pm
Quote from: Ray
With an x900 I would guess it would still be out of gamut, but to a slightly lesser degree.

Well you are free to "guess" all you want .  

Speaking for myself, I have compared multiple prints side-by-side out of a variety of printers.  Which is exactly why I sold my x600 the second I could get my hands on an x800.  Also why I sold my larger x800 as soon as the x900 came out. (I skipped the x880.)  So to clarify, I can see differences and find them significant enough to want the newer printer, but realize another's vision could be substantially different and they may not see differences OR find them significant...

So yes, YMMV!
Title: Rant 23
Post by: Ray on October 07, 2009, 04:55:06 pm
Quote from: Jack Flesher
Well you are free to "guess" all you want .  

Speaking for myself, I have compared multiple prints side-by-side out of a variety of printers.  Which is exactly why I sold my x600 the second I could get my hands on an x800.  Also why I sold my larger x800 as soon as the x900 came out. (I skipped the x880.)  So to clarify, I can see differences and find them significant enough to want the newer printer, but realize another's vision could be substantially different and they may not see differences OR find them significant...

So yes, YMMV!

Jack,
Are you just playing politics? Have you looked at that sRGB purple flower from bjanes? It's massively out of gamut in proof colors on my terribly inferior monitor which is capable of only 6 bit color, according to you.

Let's get real. The argument in this thread is that it's the quality of the print which is the final arbiter. If my 6 bit CRT monitor can display colors that a professional Epson printer cannot, then there's something that doesn't quite gel, don't you think?
Title: Rant 23
Post by: Ray on October 07, 2009, 05:29:28 pm
Quote from: tho_mas
it doesn't make sense to argue from a certain example to the next certain example. As long as you want to print high saturated colours (either way yellow, blue, red, magenta or whatever) a wide gamut display makes sense. As long as you want to see fine transitions without artefacts (especially for black&white) a display with hardware calibration and at least 10bit, better 12bit LUT and 16bit processing makes sense - a lot of sense. Your TV simply can't show what these displays can show. That's pure physics.

I'm finding it difficult to reconcile what I see on my monitor with your test results. You seem to be implying that with a better monitor my prints will be better. Could you please elaborate.

I've mentioned more than once that my monitor produces far more vibrant colors than the print. I continually have to reduce the saturation of my images to get them into gamut before printing. In proof colors in PS I have to work hard to get the image as good as the monitor rendition without proof colors. If I then toggle between proof colors and no proof colors, after boosting contrast and saturation etc, the monitor display with proof colors unticked still looks better

It's the print which, on balance, is the weak link, not the monitor, unless one has a really cheap monitor like a laptop screen or an Acer P244w, or any low quality LCD display.

The only issue which might be a concern is a mismatch between proof colors on monitor and the final print. I don't have such a mismatch that I can distinguish from the inherent differences between the transmissive nature of the monitor and the reflective nature of the print.
Title: Rant 23
Post by: tho_mas on October 07, 2009, 05:36:05 pm
Quote from: Ray
If my 6 bit CRT monitor can display colors that a professional Epson printer cannot, then there's something that doesn't quite gel, don't you think?
you should understand that the shape of monitor profiles and printer profiles are basically different. Monitor profiles have the shape similar to a pyramid upside down - high saturated colours and lot of differntiation in bright and mid tones, less in dark tonal values. Printer profiles have a different shape - they look "bellied". Less differentiation in bright tonal values but more saturation in mid tones and dark tonal values.
Working spaces in the form of matrix colour spaces are monitor profiles as well: all matrix colour spaces represent self illuminant virtual display devices and mostly include wide ranges of nonprintable colors in bright tonal values.
So - your "observation" applies to all monitor color spaces. But that doesn't help you with your low bit display and its small color space. The smaller the gamut of a certain monitor the less you can see of a certain printer color space. Your monitor might have a wider gamut in bright tonal values... but at the same time a smaller gamut in printable colors.
Here's your monitor compared to a paper I use frequently (Innova Fiba ultrasmooth) - as you clearly see your monitor color space exceeds the printer only in the bright tones, but not in the midtones (and at the low end of course as there is no printer profile with pure black):
[attachment=17046:sony_innova.jpg]

Here's a short article. The topic is actually a tablebased working space but the article adresses the problem short and quite clear: http://photogamut.org/E_idea.html (http://photogamut.org/E_idea.html)
Title: Rant 23
Post by: tho_mas on October 07, 2009, 05:45:37 pm
Quote from: Ray
I'm finding it difficult to reconcile what I see on my monitor with your test results. You seem to be implying that with a better monitor my prints will be better. Could you please elaborate. (...)
actually already adressed in my last post.
I don't say your prints will be better. I am saying with a better monitor you could see more of the printable colours.

Quote from: Ray
I continually have to reduce the saturation of my images to get them into gamut before printing. In proof colors in PS I have to work hard to get the image as good as the monitor rendition without proof colors.
this indicates that you are working with too high saturated colours (especially in bright tonalvalues) from the start. Too, your setup is apparently not adjusted for print related editing. I literally do no editing prior to print (little of course but not a big deal) and the prints match the monitor view. This is what a monitor is for, no?
Title: Rant 23
Post by: Jack Flesher on October 07, 2009, 07:19:33 pm
Quote from: Ray
Jack,
Are you just playing politics?
SNIP
If my 6 bit CRT monitor can display colors that a professional Epson printer cannot, then there's something that doesn't quite gel, don't you think?

I believe it's more about *WHERE* your monitor and my printer colors differ inside the human visual space that can make all the difference in the print; not just the total volume of their gamuts.  But you are welcome to disagree.  And yes, I am engaging in the politics of being polite instead of arguing

Bottom line Ray, if you are getting prints you are happy with, I say good on you!

Best,
Title: Rant 23
Post by: Ray on October 08, 2009, 09:49:37 pm
Quote from: tho_mas
But that doesn't help you with your low bit display and its small color space. The smaller the gamut of a certain monitor the less you can see of a certain printer color space. Your monitor might have a wider gamut in bright tonal values... but at the same time a smaller gamut in printable colors.
Here's your monitor compared to a paper I use frequently (Innova Fiba ultrasmooth) - as you clearly see your monitor color space exceeds the printer only in the bright tones, but not in the midtones (and at the low end of course as there is no printer profile with pure black):
[attachment=17046:sony_innova.jpg]

Look Tho_mas, I understand the concept here quite well, if not the specific detail. I'm also a firm believer in the principle of using the best tool for the job, not the best tool for some future imaginary purpose.

On the basis that I am not delusional, not color blind, not given to seeing apparitions, and taking into account the fact that I've experimented with various colorimeters over the years, including the Spyder, ColorEyes, Eye-One and recently a ColorMunki, is it difficult for you to understand that my Epson Enhanced Matte prints made on an Epson 7600 wide-format printer really do not show any shades of color, whether bright, mid-tone or low-tone, that are not visible on my 10 year old Sony CRT.

Now I understand the argument, that maybe the prints I made of Jack's evaluation test image, actually do show certain shades that are outside the gamut of my monitor and that they are perhaps just too subtle for me to notice. Perhaps only someone with super keen eyesight like Jack or yourself would notice them. There's really not much I can do about that. But I can tell you, when I toggle between proof colors on, and proof colors off, there's a huge difference in vibrancy across the whole tonal range, that one would be blind not to see.

In other words, the shades that even a 10 year old CRT can produce which are not visible on the print, are far more significant than the shades that might be visible on the print but not visible on the monitor, and as you've mentioned, there's no printer profile with pure black, so on that count alone, the monitor excels, even my 10 year old CRT.

As a matter of interest, I compared those gray squares within the black frame, bottom left of Jack's test image. On both the print and the simulated print on my monitor (proof colors enabled) the darkest gray that is clearly visible is 20. The match between print and monitor (proof colors enabled) is so close that there's the barest hint of a gray square at 18, but nothing darker which is not effectively black.

If I untick 'proof colors', the monitor clearly shows several shades darker. The darkest which is clearly visible is 8, with the barest hint of a shade at 6.

Now it would be interesting if I happened to have a roll of Premium Lustre or Premium Glossy at hand that I could switch with my current roll of Enhanced Matte. Unfortunately I would also have to flush the matte black ink out of the system. This is troublesome and expensive, so you can understand why I'm not prepared to do this for the sake of a comparison. This is a major inconvenience of the x600 series.

What I've learned from this exercise, and your analysis of my profile, is that an upgrade to an x900 series Epson printer might cause some dissatisfaction with my current monitor if I start seeing different shades on my prints that are not apparent in proof mode on my monitor. My first reaction would be that the Epson profiles bundled with the new printer are not as good as the Bill Atkinson profiles I currently use. However, as a result of this conversation I would now consider that this might not be the true reason and that there's a possibility that the wider gamut of the new printer and/or paper has simply outstripped the gamut of my monitor in a practical and visible manner (as opposed to a theoretical diagram). When that time comes, I'll be ready to upgrade my monitor.  
Title: Rant 23
Post by: Schewe on October 08, 2009, 11:41:06 pm
Quote from: Ray
What I've learned from this exercise, and your analysis of my profile, is that an upgrade to an x900 series Epson printer might cause some dissatisfaction with my current monitor if I start seeing different shades on my prints that are not apparent in proof mode on my monitor. My first reaction would be that the Epson profiles bundled with the new printer are not as good as the Bill Atkinson profiles I currently use. However, as a result of this conversation I would now consider that this might not be the true reason and that there's a possibility that the wider gamut of the new printer and/or paper has simply outstripped the gamut of my monitor in a practical and visible manner (as opposed to a theoretical diagram). When that time comes, I'll be ready to upgrade my monitor.  


By all means, if you update from a 7600 to a 7900 (the 7600 being a full THREE generations old) you would do well to look at updating your image display. The fact is that the 7600 is way old tech whose gamut of color did not stress a modern LCD let alone a "good" LCD with Adobe RGB color gamut. The 7900 color gamut can NOT be contained in Adobe RGB (let alone sRGB).

But, if your images are in Adobe or sRGB, you might want to question the upgrade to the 7900. Cause small gamuts of a working space won't really push these new printers...in fact, it would be a waste, seriously...
Title: Rant 23
Post by: Jack Flesher on October 08, 2009, 11:48:49 pm
Quote from: Ray
As a matter of interest, I compared those gray squares within the black frame, bottom left of Jack's test image. On both the print and the simulated print on my monitor (proof colors enabled) the darkest gray that is clearly visible is 20. The match between print and monitor (proof colors enabled) is so close that there's the barest hint of a gray square at 18, but nothing darker which is not effectively black.
If you were to print that image on PPPLuster from an x800 or x900 printer, you would see grays down to 8 or 6, and could measure 4 and probably 2 with your spectro.  You would also see up to 253 or 254 on the high grays.

Bottom line is your chosen printer/paper combo's gamut is pretty narrow.  This is not inherently bad, especially if you like the result, but relevant to this discussion as it limits what you can lay down in on a print.  The fact that your monitor matches it simply indicates that the monitor is equal to or perhaps slightly larger than the paper/ink you use.  However it does not mean your monitor can accurately render the gamut of a larger paper/ink combo.

On another tack, given what you've stated about your satisfaction with your prints and monitor, you could probably take up a total sRGB jpeg workflow from capture to output and not lose much of anything at all to a 16-bit raw workflow; your needs simply don't require a larger gamut.   Huge workflow convenience advantage for you as it would eliminate the entire raw conversion process.  Heck, I'm actually a little jealous of that!

Cheers,
Title: Rant 23
Post by: tho_mas on October 09, 2009, 03:54:06 am
Quote from: Ray
Now I understand the argument, that maybe the prints I made of Jack's evaluation test image, actually do show certain shades that are outside the gamut of my monitor and that they are perhaps just too subtle for me to notice.
  the image contains high saturated colours of ProPhotoRGB. There is nothing "subtle" - it's a hughe difference.

Quote
the shades that even a 10 year old CRT can produce which are not visible on the print, are far more significant than the shades that might be visible on the print but not visible on the monitor
that might be true regarding your setup. Still: I think that your workflow and calibration is not accurate. Apparently your images contain too much differentiation in bright tonal values and too less in midtones. Otherwise you would see certain colours in the prints your monitor can't display.
Even sRGB is much too "big" in the brightest tonalvalues for every printer. But - again - it's far too small in the tonal range printers produce. See again your monitor compared to the Innova paper (on Epson 11880 with ultrachrome K3) which is actually not a very big printer gamut.
You are probably working with a too dark monitor and this is why you are probably shifting all the relevant data in too bright ranges. Specualtion.
Convert your images in the above mentioned color space "PhotogamutRGB". It will render the colors of your images in a tonal range that is reproducable in prints.
The color space is a bit dated as today there are printers with more differentiation especially in dark tonal values. But for your printer Photogamut should be fine as working space.
So your workflow could be like that:
- RAW (in ProPhoto... I don't mind)
- convert a copy of the TIF to PhotoagmutRGB with perceptual RI (Photogamut is tablebased)
- edit the image
- use perceptual RI for printing
- re-think this workflow when you are going to upgrade your printer and monitor

And forget about the "paper simulation" in the soft proof settings. Your display should be calibrated visually to paper white and if so the colour mangement has no reason to spin this white to the white of the paper. The blueish/grayish apperance of paper white in softproof mode has nothing to do with the color of "white" of the respective paper - this is just an error of the measurement devices in conjunction with papers that contain optical brighteners (if this is hopefully the correct term). But you should use "simulate black ink" (i.e. - a simluation of the print that is displayed relative colormetric on the monitor).

I think you have enough info now, no?





Title: Rant 23
Post by: Ray on October 09, 2009, 11:35:49 am
Quote from: Jack Flesher
If you were to print that image on PPPLuster from an x800 or x900 printer, you would see grays down to 8 or 6, and could measure 4 and probably 2 with your spectro.  You would also see up to 253 or 254 on the high grays.

Not according to my softproof setup. Even Premium Glossy (Bill Atkinson profile) doesn't show darker grays. I went to the trouble of swapping black cartridges when I switched from Premium Lustre to Enhanced Matte a couple of years ago.

Quote
On another tack, given what you've stated about your satisfaction with your prints and monitor, you could probably take up a total sRGB jpeg workflow from capture to output and not lose much of anything at all to a 16-bit raw workflow; your needs simply don't require a larger gamut.   Huge workflow convenience advantage for you as it would eliminate the entire raw conversion process.  Heck, I'm actually a little jealous of that!

Not quite   . I like those golden yellows. The stronger yellows of ProPhoto RGB have been apparent to me for a few years now. Did I also mention the lack of differentiation, in the sRGB print of your test image, of two shades of green which were clearly apparent in the ProPhoto print and clearly apparent on my monitor in both soft proof mode and out of soft proof mode?
Title: Rant 23
Post by: Ray on October 09, 2009, 11:48:58 am
Quote from: Schewe
By all means, if you update from a 7600 to a 7900 (the 7600 being a full THREE generations old) you would do well to look at updating your image display. The fact is that the 7600 is way old tech whose gamut of color did not stress a modern LCD let alone a "good" LCD with Adobe RGB color gamut. The 7900 color gamut can NOT be contained in Adobe RGB (let alone sRGB).

But, if your images are in Adobe or sRGB, you might want to question the upgrade to the 7900. Cause small gamuts of a working space won't really push these new printers...in fact, it would be a waste, seriously...

Jeff,
I've been using ProPhoto RGB for several years now. I was persuaded by the argument, 'even if one can't see any difference now, one probably would some time in the future'. On my 7600 it's the strength of the yellows that became apparent soon after I bought the machine.

Upgrading to a 7900 has now been complicated by the fact that my current monitor may not be adequate. A 7900 plus the best quality Eizo monitor will set me back quite a bit. The economic principle of 'Opportunity Cost' comes into play here.
Title: Rant 23
Post by: Ray on October 09, 2009, 12:01:51 pm
Quote from: tho_mas
the image contains high saturated colours of ProPhotoRGB. There is nothing "subtle" - it's a hughe difference.

A huge difference between what? I keep telling you, there's no observable difference between my print of Jack's test image with embedded ProPhoto profile using Epson Ehanced Matte, and the image as displayed on my monitor with proof colors enabled prior to printing. I hold the print next to the monitor and compare. There's no difference; zilch, nada.

However, there is a difference between the the sRGB version amd the ProPhoto version. Not a huge difference but a noticeable difference. Principally, the yellows in the autumn forest scene are more golden in the ProPhoto version, and two of the green squares in the lower section of the test image have merged into one shade in the sRGB print. I see no differences elsewhere. The strawberries are equally red. The sunset looks identical in both images as do the portraits of the children. However, it's possible if this test image were significantly larger, I might discern a few other differences.

Such differences as I observe, between the sRGB print and the ProPhoto print, are also apparent on my monitor with proof colors enabled.

Quote
I think that your workflow and calibration is not accurate. Apparently your images contain too much differentiation in bright tonal values and too less in midtones.

I never print my images without viewing the images first in Photoshop with proof colors enabled, and adjusting the image to taste to compensate for loss of vibrancy. (Except for Jack's test image.)

The purpose is to get a result on print which matches as closely as possible the view on the monitor. My prints give me that match, but only with proof colors enabled. They always look less vibrant than they appear on the monitor after unticking 'proof colors'. As I mentioned before, this can be frustrating if you like vibrant colors and you start off with an image that has not been adjusted for printing but adjusted for viewing on the monitor. I print only a very small proportion of my images. The vast majority have not been soft proofed.

Quote
You are probably working with a too dark monitor and this is why you are probably shifting all the relevant data in too bright ranges. Specualtion.

Why would you think that when I mentioned in my previous post that the darkest shade of grey, on Jack's test image, was exactly the same as on my monitor, in proof color mode.; 20,20,20 with the barest hint of 18,18,18.

Quote
Convert your images in the above mentioned color space "PhotogamutRGB". It will render the colors of your images in a tonal range that is reproducable in prints.

PhotogamutRGB? What's that? I've been using the ProPhoto RGB color space for years.

Quote
And forget about the "paper simulation" in the soft proof settings. Your display should be calibrated visually to paper white and if so the colour mangement has no reason to spin this white to the white of the paper. The blueish/grayish apperance of paper white in softproof mode has nothing to do with the color of "white" of the respective paper - this is just an error of the measurement devices in conjunction with papers that contain optical brighteners (if this is hopefully the correct term). But you should use "simulate black ink" (i.e. - a simluation of the print that is displayed relative colormetric on the monitor).

Selecting 'simulate black ink' (for me) produces a print which is slightly darker that the appearance of the image on my monitor. This is why I prefer 'simulate paper color'. It gives me a more exact match.

Quote
I think you have enough info now, no?

I have absolutely no problems with my current state of color management in relation to my 10 year old CRT monitor, 5 year old Epson 7600 printer, Epson Enhanced Matte paper, Bill Atkinson profiles and Eye-One monitor calibration. But thanks for your advice.  

Title: Rant 23
Post by: tho_mas on October 09, 2009, 12:42:01 pm
Quote from: Ray
Selecting 'simulate black ink' (for me) produces a print
The selection of "simulate black ink" doesn't affect the print. It affects the preview of the print softproof on the monitor.
Title: Rant 23
Post by: Ray on October 09, 2009, 01:15:33 pm
Quote from: tho_mas
The selection of "simulate black ink" doesn't affect the print. It affects the preview of the print softproof on the monitor.

Exactly correct! The print will look the same whether you select 'paper color' or 'simulate black'. But the image on the monitor with proof colors ticked, will not (necessarily) look the same as the print.

That's the difference which is a critical point. One edits on the monitor and one expects (or hopes) that the print will be the same.
Title: Rant 23
Post by: Jack Flesher on October 09, 2009, 01:21:21 pm
Quote from: Ray
Not according to my softproof setup. Even Premium Glossy (Bill Atkinson profile) doesn't show darker grays.

Try PPPLuster instead of glossy .  If you still can't see it after that, then there is something wrong with either your proof set up or your monitor.   Bottom line is printing that test image on PPPLuster with an x800 or x900 printer with a good profile, you can see at least 8 and usually 6 on the print, and measure down to 4 or 2...  

(Perhaps irrelevant since I use a different monitor than you and my own profile, but I can see down to 6 with maybe a hint at 4 and up to 253 in my proof view for Epson PPPLuster, which very closely mirrors how the print on PPPLuster looks for me after it's dried down.)    

Cheers,
Title: Rant 23
Post by: tho_mas on October 09, 2009, 01:21:49 pm
Quote from: Ray
Exactly correct! The print will look the same whether you select 'paper color' or 'simulate black'. But the image on the monitor with proof colors ticked, will not look the same as the print.
That's the difference which is a critical point. One edits on the monitor and one expects (or hopes) that the print will be the same.
so the simulation of "paper" produces a brighter preview than the simulation of "black ink" only?
(-> "Selecting 'simulate black ink' (for me) produces a print which is slightly darker that the appearance of the image on my monitor. This is why I prefer 'simulate paper color'. It gives me a more exact match.")
Title: Rant 23
Post by: bjanes on October 09, 2009, 01:27:22 pm
Quote from: Ray
Jeff,
I've been using ProPhoto RGB for several years now. I was persuaded by the argument, 'even if one can't see any difference now, one probably would some time in the future'. On my 7600 it's the strength of the yellows that became apparent soon after I bought the machine.

Upgrading to a 7900 has now been complicated by the fact that my current monitor may not be adequate. A 7900 plus the best quality Eizo monitor will set me back quite a bit. The economic principle of 'Opportunity Cost' comes into play here.

Even if you upgrade to the best Eizo monitor which covers 98% of the Adobe RGB gamut, there would still be a gamut mismatch between the monitor and printer gamuts. Shown below is a gamut map comparing aRGB and the X900 printers. There will still be colors in gamut for the printer which can not be shown on the monitor and colors that are in the gamut of aRGB which can not be printed. If you upgrade printers, you can make use of the expanded gamut and print the colors which are out of gamut of the monitor. However, soft proofing would be a problem. I don't have a profile for the Eizo, but it apparently is similar to aRGB as shown in the thread comparing Eizo and NEC.

[attachment=17070:7900_aRGB.png]

If money is a concern (as it is for most of us) have you considered getting a NEC instead of Eizo?

NEC vs EIZO (http://luminous-landscape.com/forum/index.php?s=&showtopic=36856&view=findpost&p=303052)
Title: Rant 23
Post by: Ray on October 09, 2009, 02:04:33 pm
Quote from: bjanes
If money is a concern (as it is for most of us) have you considered getting a NEC instead of Eizo?

Bill,
I've considered many options. It's so easy to get attracted by advertising spiel. I've simply got no problems with my current set-up regarding color accuracy. However, I feel I am hampered with my 7600 which doesn't allow switching between matte paper and glossy paper without an ink purge. The main attraction of the current x900 series is the facility to switch paper types without having to flush one's sytem of ink.
 
I've considered getting an LCD as a replacement for my magnificent Sony CRT, and I've bought a couple. A 17" Sony LCD for my partner, a 24" Acer for my laptop, and a an Eizo FlexScan S1010 which I bought at the same time as my Eye-One colorimeter a few years ago.

I'm ashamed to admit, the Eizo is still sitting in it's box. Why is this? It's an enigma to me also.
Title: Rant 23
Post by: Ray on October 09, 2009, 09:54:05 pm
Quote from: tho_mas
so the simulation of "paper" produces a brighter preview than the simulation of "black ink" only?
(-> "Selecting 'simulate black ink' (for me) produces a print which is slightly darker that the appearance of the image on my monitor. This is why I prefer 'simulate paper color'. It gives me a more exact match.")


No. Wrong way round. Selecting 'simulate black ink' slightly lightens the preview on my monitor, compared with 'simulate paper color'. If A is lighter than B, then it follows that B is darker than A.
Title: Rant 23
Post by: tho_mas on October 10, 2009, 10:49:48 am
Quote from: Ray
Wrong way round.
ah, okay, I've confused that. Sorry.

Problem here is what I've mentioned above.
Paper white in paper profiles is often blueish/grayish - at least if the profiles are measured by devices without UV cut filter.
This is due to optical brighteners in papers (not all papers of course, but most photographic papers contain optical brighteners).
But blueish white is not white. It is not only bluer than white it's at the time darker than white.
But your Paper is visually … white.
What happens with your softproof if use paper simulation?
I've taken the profile a FUJI paper (RGB printer / C-print) because these papers contain a lot of optical brighteners so the problem is quite obvious.

This is crop from a sky with clouds - screenshot of the original file:
[attachment=17097:01_sp_off.jpg]

This is the file with softproof relative colormetric + BPC / simulation of black ink:
[attachment=17098:02_sp_rcm_black.jpg]

This is the file with softproof relative colormetric + BPC / paper simulation.
Due to the relative colorimetric RI you can see quite well the blueish white of the paper - and it's darker.
[attachment=17099:03_sp_rcm_paper.jpg]

This is the file with softproof absolute colorimetric / simulation of black ink.
Color management spins the blueish paper white to the warmer white point of the display.
(too, the contrast range is compressed to the level of the paper contrast but is displayed at the luminance level of the monitor - so no compenstion).
[attachment=17100:04_sp_acm_black.jpg]

This is the file with softproof absolute colorimetric / paper simulation:
[attachment=17101:05_sp_acm_paper.jpg]

In the last screenshot the white point compensation works fine so far but where are the transitions of the clouds?
They are suppressed on the monitor preview - but they are visible in the print (as in screenshot #2).
The blueish (and therefore darker) white (which is nothing else than a measurement device error) compresses the transitions in bright tonal values.
You'll find the same effects of compression all over the tonal range of the image… just less obvious in less bright tones.
This effect is clearly visible with papers containing a high amount of optical brighteners and less visible in papers with a small amount of optical brighteners - but still there.
This is why it is not a bad idea to calibrate the monitor to paper white (both luminance and white point) and set the softproof to relative colorimetric RI + BPC (or perceptual RI) and simulation of black ink only - the monitor white already matches paper white so there's no reason to adapt white points for the preview.



Title: Rant 23
Post by: Eric Myrvaagnes on October 10, 2009, 01:44:08 pm
Tho_mas,

That's a very nice and effective demonstration of the effects of rel col vs abs col and especially of the "simulate paper color" effect.

Thanks for showing it.

Eric

Title: Rant 23
Post by: bjanes on October 10, 2009, 02:59:38 pm
Quote from: Jack Flesher
Bottom line is your chosen printer/paper combo's gamut is pretty narrow.  This is not inherently bad, especially if you like the result, but relevant to this discussion as it limits what you can lay down in on a print.  The fact that your monitor matches it simply indicates that the monitor is equal to or perhaps slightly larger than the paper/ink you use.  However it does not mean your monitor can accurately render the gamut of a larger paper/ink combo.

On another tack, given what you've stated about your satisfaction with your prints and monitor, you could probably take up a total sRGB jpeg workflow from capture to output and not lose much of anything at all to a 16-bit raw workflow; your needs simply don't require a larger gamut.   Huge workflow convenience advantage for you as it would eliminate the entire raw conversion process.  Heck, I'm actually a little jealous of that!

Cheers,

Jack's dismissive reply to Ray ignores a few relevant facts. In fact, the sRGB gamut will clip blues, greens and yellows that are in gamut for his printer as shown in the gamut plot below. Ray is justified in using ProPhotoRGB to preserve those colors. He previously mentioned he likes the yellows that are preserved in ProPhotoRGB; those would be lost if he used sRGB as his working space.

[attachment=17113:Epson7600vs_sRGB.png]

Of course, one is not interested in gamut plots which show what could be in an image but rather with what colors are actually in one's image. The plot below demonstrates the gamuts of the printer test image, sRGB, and the 7600 printer with Premium Lustre paper (using the profile from a local lab that uses the printer). Note the yellows that are clipped in sRGB but are within the gamut of the printer. Another 3D view could have shown the greens with a similar gamut mismatch. The printer can print saturated greens and yellows at mid-luminance that are out of the sRGB gamut. Of course, a reflection print can not print low luminances because of its limited DMax, especially with art papers with a limited DMax, but some of these can be shown on the sRGB screen because of the gamut mismatch.

[attachment=17117:7600_sRG...estImage.png]

Then one should ask which colors in the test image are out of gamut for the printer and how does affect the image. The GamutVision plot below shows the ΔEs for various parts of the image. A ΔE of one represents a just noticeable difference and a ΔE of 5 may not significantly impair the image, but this is only a rough guide and varies with images and viewing conditions. The largest ΔEs are in the patches with high luminance saturated colors which do not occur in the real world part of the image, but which could occur if you are taking night pictures of neon lights on Times Square or the Ginza or in some landscapes with bright colors in flowers. Other large ΔEs occur in the low luminance areas of the image that can't be printed with a limited DMax. In the most important regions of the image (e.g. straw berries and skin color) the ΔEs are 5 or less and even in the sunset the ΔEs are relatively low in the higher luminance colorful areas of the image.

[attachment=17121:7600_GamutVision.png]

Even if you clip the test image to sRGB as Jack suggests, there are still a lot of out of gamut colors. However, some colors that could have been printed are lost.

[attachment=17122:7600_Gam...ion_sRGB.png]

With the x900 printer and the EFP (exhibition fiber) paper, the situation is much improved, but there are still a lot of out of gamut colors. Note that the scale has changed with this analysis.

[attachment=17123:9600_gamut.png]

I hesitate to state that I am still using the Epson 2200 and this thread has convinced me it is time to upgrade. Since I don't have a MFDB, I don't print at very large sizes and the 3880 or 4880 might be a good choice for my needs. I understand that Michael will review the former soon. However, I understand that these printers waste ink if you need to change blacks. My monitor's gamut does not cover all of sRGB, so a monitor upgrade would also be a good idea. However, even the high end Eizos that would cover Adobe RGB could not show the gamut of the printer and I might defer that purchase until LED monitors which offer more flexibility in calibration are available at reasonable prices. Any comments or suggestions?





Title: Rant 23
Post by: Jack Flesher on October 10, 2009, 09:37:18 pm
Quote from: bjanes
Jack's dismissive reply to Ray ignores a few relevant facts. \

SNIP

demonstrates the gamuts of the printer test image,  sRGB, and the 7600 printer with Premium Lustre paper (using the profile from a local lab that uses the printer).

And as is usual, your post ignores facts actually pertinent to the discussion we're having -- like the one that Ray prints on MATTE paper with Mk ink on his 7600, NOT Epson Premium Luster and Pk ink...
Title: Rant 23
Post by: Ray on October 11, 2009, 12:29:58 am
Quote from: Jack Flesher
And as is usual, your post ignores facts actually pertinent to the discussion we're having -- like the one that Ray prints on MATTE paper with Mk ink on his 7600, NOT Epson Premium Luster and Pk ink...


However, Jack, as I've mentioned a couple of times, the ProPhoto print of your test image on my Enhanced Matte does show a more golden yellow than the sRGB print, and does show a clear differentiation between the first two shades of green squares and between the two cyan squares. In the sRGB print those two green and two cyan squares look like the same shade of green and the same shade of cyan, and furthermore, those differences are equally apparent in the preview on my monitor with proof colors enabled.

Since your test image was printed 'as is' with no adjustments whatsoever, it makes no difference what monitor I use. I could print the image from an uncalibrated laptop and the print should look the same. What counts for editing purposes is the closness of the match between the preview on the monitor and the resulting print.

Now the fact is, I have a sufficiently close match using Enhanced Matte, so a better monitor would serve no purpose with my current printer and paper. Whether it would serve a purpose with another paper type with a wider gamut, such as Premium Lustre or Premium Glossy, is another matter which I cannot conveniently test right now because of the ink-swap hassle. However, I have used one roll of Premium Glossy and a couple of rolls of Premium Lustre in the past, before I switched to Enhanced Matte. In fact, I recall I was using Prm Lustre when I bought the Eye-One colorimeter. I bought an Eizo FlexScan S1910 monitor at the same time because it was 'on special' and because I thought an LCD monitor would be less strain on the eyes and because I thought I might get a better calibration if my old Sony CRT proved to be lacking.

It so happened that the 'easy' automatic option in the iMatch software produced such a close match on my old CRT, with the Prm Lustre I was using at the time, that I saw no reason to set up the FlexScan.  I don't have enough time to play with toys. I have a practical orientation towards photography. What works, works, sometimes despite the theory stating it shouldn't. However, if I were to revisit the situation with Prm Lustre and 'pixel-peep' the subtle shading in your test image, I'm prepared to accept I might see subtleties in the print which are not, and cannot be, visible on my monitor.

I'm getting close to the end of my matte black ink and current roll of matte paper. After whichever runs out first, I'm going to switch to Premiem Gloss and Photo Black ink, but at present I'm rather busy doing the tiling in my new house. It's a very, very tedious job (perhaps because I'm such a perfectionist). I'd rather print test images.  
Title: Rant 23
Post by: Ray on October 11, 2009, 02:58:27 am
Quote from: tho_mas
ah, okay, I've confused that. Sorry.

Problem here is what I've mentioned above.
Paper white in paper profiles is often blueish/grayish - at least if the profiles are measured by devices without UV cut filter.
This is due to optical brighteners in papers (not all papers of course, but most photographic papers contain optical brighteners).
But blueish white is not white. It is not only bluer than white it's at the time darker than white.
But your Paper is visually … white.
What happens with your softproof if use paper simulation?
I've taken the profile a FUJI paper (RGB printer / C-print) because these papers contain a lot of optical brighteners so the problem is quite obvious.

This is crop from a sky with clouds - screenshot of the original file:
[attachment=17097:01_sp_off.jpg]

This is the file with softproof relative colormetric + BPC / simulation of black ink:
[attachment=17098:02_sp_rcm_black.jpg]

This is the file with softproof relative colormetric + BPC / paper simulation.
Due to the relative colorimetric RI you can see quite well the blueish white of the paper - and it's darker.
[attachment=17099:03_sp_rcm_paper.jpg]

This is the file with softproof absolute colorimetric / simulation of black ink.
Color management spins the blueish paper white to the warmer white point of the display.
(too, the contrast range is compressed to the level of the paper contrast but is displayed at the luminance level of the monitor - so no compenstion).
[attachment=17100:04_sp_acm_black.jpg]

This is the file with softproof absolute colorimetric / paper simulation:
[attachment=17101:05_sp_acm_paper.jpg]

In the last screenshot the white point compensation works fine so far but where are the transitions of the clouds?
They are suppressed on the monitor preview - but they are visible in the print (as in screenshot #2).
The blueish (and therefore darker) white (which is nothing else than a measurement device error) compresses the transitions in bright tonal values.
You'll find the same effects of compression all over the tonal range of the image… just less obvious in less bright tones.
This effect is clearly visible with papers containing a high amount of optical brighteners and less visible in papers with a small amount of optical brighteners - but still there.
This is why it is not a bad idea to calibrate the monitor to paper white (both luminance and white point) and set the softproof to relative colorimetric RI + BPC (or perceptual RI) and simulation of black ink only - the monitor white already matches paper white so there's no reason to adapt white points for the preview.


Tho_mas,
Interesting comparisons! Correct me if I'm wrong, but your suggestion of calibrating the monitor to the paper white would cause images edited on such a monitor to be inappropriate for other purposes, such as viewing web images or converting images to sRGB for viewing on one's plasma TV, would it not?

I've opened your original clouds' image using 'simulate paper color', and the result I get on my monitor is quite different to your version of paper simulation. Quite different! I've placed the 3 images side by side in a new document so you can compare easily. The first one, top left, is is my rendition of your original image in my proof setup. I do not see the bluish white of the paper that's clearly evident in your 'paper color simulation'.

[attachment=17126:Tho_mas_clouds.jpg]

When I try the 'advanced option' when calibrating with Eye-One, I find that extracting the paper-white values from the Bill Atkinson profiles is not possible. I get the impression I would have to buy an X-rite spectrophotometer to create my own profiles in order to do this.

If I select D65 as a white point in 'advanced mode', the monitor calibration simply does not match the print when softproofing, so what's the point of the advanced option in my situation? The easy (automatic) option produces a near-perfect match.

I mentioned that the dark grays visible on both print and monitor preview, match exactly. The same cannot quite be said for the lightest grays. In that respect, my matte paper slightly exceeds the performance of the monitor, but so slightly it's not an issue. On Jack's test image, the brightest gray, 254, is completely lost on my monitor preview, in soft-proof mode (whether simulate paper color or black ink). 253 is so faint it's effectively lost, but 252 is definitely noticeable, although faint.

However, on the print, if one holds the print at the right angle against the light, there's a slight hint of both 254 and 253. The implication here is, if I were to print out some cloud images like your examples above, the print should reveal a very small increase in subtle shading in such clouds, compared with my monitor preview.

The bottom line here, surely, is that all that counts is the match between print and monitor preview. If the print is very slightly better in some regard than the monitor preview, that's fine by me. If the print is worse than the monitor preview, that's not good. The fact that subtle shading in the sky is lost with Absolute Colorimetric rendering is not necessarily a problem. One just selects the sky, in soft-proof mode, and darkens it.

I use whatever rendering intent produces the best result to my taste. Sometimes I use Saturation Intent for photographic images, even though such rendering intent is usually recommended only for pie charts. I'm often confronted with out-of-gamut colors and I take some trouble to select such areas that are out-of-gamut and bring them back into gamut by reducing saturation or lightening or darkening the specific areas.

Incidentally, the temperature of my calibrated monitor seems unusually high at 7600K, although the luminance seems okay at 94.3 nits. The minimum luminance is 0.0 cd/m2, presumably because CRTs have no backlight.

Our monitors would clearly seem to be calibrated differently, but I repeat, what counts in the final analysis is the match between print and monitor preview.

By the way, the profile I sent you for analysis was probably a bit old, July 2009. Eye-One's reminder is every month at a minimum (edit: at a maximum. I too can get things things the wrong way round   ). I considered this too tedious and unnecessary. There should be an option for 'every 2 months', but perhaps 3 months is too much. With a re-calibration, the following image shows a more noticeable difference between the two greens in the lower half of the image.

Red, yellow and green are quite different in the upper half, on monitor and print, but bear in mind that all out-of-gamut colors in this image were brought back into gamut before printing. Beware! ProPhoto RGB profile embedded.

[attachment=17127:LL_sRGB_...to_space.jpg]
Title: Rant 23
Post by: tho_mas on October 11, 2009, 06:57:56 am
Quote from: Ray
Correct me if I'm wrong, but your suggestion of calibrating the monitor to the paper white would cause images edited on such a monitor to be inappropriate for other purposes, such as viewing web images or converting images to sRGB for viewing on one's plasma TV, would it not?
No, why? Basically you adjust the white point of the monitor to a color temperature you consider as a neutral white under given lighting conditions.
Now, if you watch TV the TV might look a bit blueish first but you're adapting to the white point of the TV after a few minutes.
Example: you are working at your monitor in the evening. Now you make a break for a cup of tea in the kitchen (with tungsten light) and read a magazine. You'll consider the paper of the magazine yellowish. But after some minutes (15minutes or so) you are adapted to the tungsten light and you'll consider the magazine paper as white. Now you return to the monitor and in the first moment you'll think - shit, what a blueish soup. After some minutes you will adapt to the monitor white again. This is the basic principle.
As to web/sRGB: the white points in the profiles don't change the appearance of an image - they are just referring to the illumination of the PCS.
If you convert white of ProPhoto (5000K) to sRGB (6500K) it's still RGB 255-255-255… so white.

But… calibrating to paper white under controlled viewing conditions implies that the calibration is okay for a certain range of papers but maybe too warm or too cold for other papers.
Within the range of photographic papers I'd say just calibrate to your preferred paper and it will be quite okay for all the other papers as well (as a long as you don't use paper simulation).
You can take a sheet of offset print paper type 1 or 2 (glossy/matte coated) as well - it's quite a good average.

When viewing printed images you're adapting as well - you're adapting to the color temperature of the brightest ambient light source (always). So you'll consider paper white as a neutral white under the lighting conditions you are viewing the images in.
Only if you compare two papers side by side you'll notice the different white points of these papers.

Quote
I've opened your original clouds' image using 'simulate paper color', and the result I get on my monitor is quite different to your version of paper simulation. Quite different!
you are trying to reproduce the "theory" and that's fine. Theory is useless if it doesn't reflect reality. But did you get what I was showing with my examples, did you get the principle? If yes, you should know why it is looking so different with your paper profile on your monitor.
Your softproof darkens the print preview too much and it looks gray. But in fact printed the sky is blueish (with a very slight nuance towards purple). This file has been printed as C-print (semi matte and glossy and metallic), on Innova Fiba ultra smooth and as CMYK print (coated paper). On all these prints the sky is always blueish - without any editing of the sky's color prior to printing. It has never been gray and it wouldn't be gray on your paper as long as your printer profile and the color management is accurate.
Actually I think it's quite easy: if the monitor white already matches paper white (white point and luminance) - why should colormanagement spin the white point and darken the image then?
And as to the contrast range of the printer/paper this is reflected in the simulation of "black ink".
The term "black ink" is a bit misleading - actually they should name it "relative colorimetric preview (of the softproof on the monitor)".

Quote
When I try the 'advanced option' when calibrating with Eye-One, I find that extracting the paper-white values from the Bill Atkinson profiles is not possible
no, no, no… forget the values!
Eye up the paper under controlled lighting conditions and adjust the white point and luminance of your monitor manually! Then calibrate to the "native white point".

Quote
Incidentally, the temperature of my calibrated monitor seems unusually high at 7600K
yes, that's very high.

Quote
The minimum luminance is 0.0 cd/m2, presumably because CRTs have no backlight
forget it! This is a mismatch of the i1-display and the i1-Match software. Colorimeters are inaccurate at very low levels.
Set up your monitor in a dark room an switch it on. It's not pure black.


Title: Rant 23
Post by: Ray on October 11, 2009, 07:22:05 am
Quote
No, why? Basically you adjust the white point of the monitor to a color temperature you consider as a neutral white under given lighting conditions.
Now, if you watch TV the TV might look a bit blueish first but you're adapting to the white point of the TV after a few minutes.
Quote

No. I'm not so much interested in adaption after a periodof time. An exact match between monitor and print is what counts. In a period of time, the print fades and all bets are off.

Again, you're not making much sense. An exact match between monitor and print is all that counts. I have it, outside of extreme pixel-peeping.
Title: Rant 23
Post by: tho_mas on October 11, 2009, 07:51:37 am
Quote from: Ray
No. I'm not so much interested in adaption after a periodof time.
you didn't get what I was trying to explain. The comparision of monitor view and print view has to match immediately - this is why you should edit and eye up prints under controlled lighting conditions.

Quote
An exact match between monitor and print is all that counts.
of course. This is why you need controlled viewing conditions - e.g. a viewing booth. And this is why it is a good thing if monitor white matches paper white in the viewing booth.
If you don't work under controlled viewing conditions that's fine... but then we don't have to discuss any further. At least not about a "match" of monitor preview and print.
Here on the forum you might search for contributions of "digital dog" who is a reputable expert of CM. I think he's working the way I do or at least similar.
Title: Rant 23
Post by: Ray on October 11, 2009, 10:46:03 am
Quote from: tho_mas
you didn't get what I was trying to explain. The comparision of monitor view and print view has to match immediately - this is why you should edit and eye up prints under controlled lighting conditions.

of course. This is why you need controlled viewing conditions - e.g. a viewing booth. And this is why it is a good thing if monitor white matches paper white in the viewing booth.
If you don't work under controlled viewing conditions that's fine... but then we don't have to discuss any further. At least not about a "match" of monitor preview and print.
Here on the forum you might search for contributions of "digital dog" who is a reputable expert of CM. I think he's working the way I do or at least similar.


No! no! no! You are being totally anal.

Peoeple don't view their prints in a viewing booth. They view them on their walls in their houses, in normal daylight and in the evening with artificial lighting.

I've mentioned before, the reflective quality of prints results in their appearance changinging throughout the day and evening depending on the ambient lighting conditions. This is one of the properties of the print which is different from the monitor.

I never adjust what my eyes tell me to a theory of what they should see. I wouldn't know how.
Title: Rant 23
Post by: tho_mas on October 11, 2009, 10:52:17 am
Quote from: Ray
Peoeple don't view their prints in a viewing booth. They view them on their walls in their houses, in normal daylight and in the evening with artificial lighting.
is that so? That is totally new to me. Now I'll rethink my entire workflow. Thanks for the insights.
Title: Rant 23
Post by: Jack Flesher on October 11, 2009, 10:55:29 am
Quote from: Ray
Now the fact is, I have a sufficiently close match using Enhanced Matte, so a better monitor would serve no purpose with my current printer and paper.

And why I said earlier, that is great for you!  Seriously, it sounds like you are content with your present printer/paper and monitor combination and no upgrade is needed, which is truly awesome for you.  However, none of that alters the fact that your particular setup is also leaving a lot of the total color and DR capability of your camera unused...  

Happy shooting and printing!  
Title: Rant 23
Post by: bjanes on October 11, 2009, 11:00:32 am
Quote from: Jack Flesher
And as is usual, your post ignores facts actually pertinent to the discussion we're having -- like the one that Ray prints on MATTE paper with Mk ink on his 7600, NOT Epson Premium Luster and Pk ink...


Even so, with Enhanced Matte and MK ink, some yellows are in gamut for the printer and out of gamut with sRGB, which comports with Ray's observations, which you have ignored. Also, some greens are in gamut for the printer and out of gamut for sRGB. Ray apparently knows what he is doing.  My original assertion holds. I don't have a profile for the 9600, so I used one for the Epson 2200, which has a similar gamut, likely not superior to the more expensive x600.

[attachment=17133:Epson220...dMatteMK.png]
Title: Rant 23
Post by: Ray on October 11, 2009, 11:10:45 am
Quote from: Jack Flesher
And why I said earlier, that is great for you!  Seriously, it sounds like you are content with your present printer/paper and monitor combination and no upgrade is needed, which is truly awesome for you.  However, none of that alters the fact that your particular setup is also leaving a lot of the total color and DR capability of your camera unused...  

Happy shooting and printing!

Watch this space when I change to Prm Gloss   .
Title: Rant 23
Post by: Ray on October 11, 2009, 11:32:09 am
Quote from: tho_mas
you didn't get what I was trying to explain. The comparision of monitor view and print view has to match immediately - this is why you should edit and eye up prints under controlled lighting conditions.

of course. This is why you need controlled viewing conditions - e.g. a viewing booth. And this is why it is a good thing if monitor white matches paper white in the viewing booth.
If you don't work under controlled viewing conditions that's fine... but then we don't have to discuss any further. At least not about a "match" of monitor preview and print.
Here on the forum you might search for contributions of "digital dog" who is a reputable expert of CM. I think he's working the way I do or at least similar.

It does match immediately, although it's better to wait a few minutes for the ink to dry and settle in. When I compare print with monitor, it's either in a day lit room or, under the same artificial lighting from an energy-saving, cool-daylight bulb.
Title: Rant 23
Post by: Ray on October 11, 2009, 11:33:54 am
Quote from: tho_mas
is that so? That is totally new to me. Now I'll rethink my entire workflow. Thanks for the insights.

No need to rethink your workflow as long as you've got that perfect match between monitor preview and print, as I have.
Title: Rant 23
Post by: Jack Flesher on October 11, 2009, 01:05:55 pm
Quote from: Ray
No need to rethink your workflow as long as you've got that perfect match between monitor preview and print, as I have.

I think Thomas was being sarcastic Ray .  Your methodology works for your particular paper, profile, printer and monitor, and since that's all you use, it works -- a classic case where your particular combination of wrongs happen to cancel each other out and make a right    

However, you'll likely be screwed with any paper change.  OTOH, Thomas' workflow is a proper CM workflow and he (along with everybody else that uses same) can switch between papers or even printers without significant issues.

Again, happy shooting and happy printing!
Title: Rant 23
Post by: Ray on October 11, 2009, 06:12:49 pm
Quote from: Jack Flesher
However, you'll likely be screwed with any paper change.  OTOH, Thomas' workflow is a proper CM workflow and he (along with everybody else that uses same) can switch between papers or even printers without significant issues.

Well, I can't argue with that, yet, because it's so troublesome for me to change papers right now and I haven't got the time. However, I don't see any better alternative calibration options that are available to me. The iMatch software doesn't allow me set the paper-white as a target, from the Bill Atkinson profiles I use. As I've mentioned already, I was using Prm Lustre without any calibration problems (that I recall) before I tried the matte paper. However, I do recall having one or two slight issues with previous equipment using ColorEyes software with an x-rite DTP94 colorimeter and a Viewsonic 19" CRT monitor on an older WinXP 32bit system.

However, far more problematic than any calibration mismatch was a tendency of the yellow ink in my 7600 printer to become contaminated with cyan which normal head-cleaning routines didn't fix. As a work-around I resorted to extending the canvas of the first print in each session and placing a pure yellow band at the top. I'm pleased the problem has now disappeared. I believe the contamination was in the nozzle caps.
Title: Rant 23
Post by: tho_mas on October 11, 2009, 06:26:50 pm
Quote from: Ray
The iMatch software doesn't allow me set the paper-white as a target, from the Bill Atkinson profiles I use.
the values would be useless in any case. Do you atually read was others write?
u-s-e   y-o-u-r   E-Y-E-S ... or can't you tell white from white?
Title: Rant 23
Post by: Ray on October 11, 2009, 07:22:09 pm
Quote from: tho_mas
the values would be useless in any case. Do you atually read was others write?
u-s-e   y-o-u-r   E-Y-E-S ... or can't you tell white from white?

Now you are becoming insulting. This is what Chavez and Blatner, in their book on CS4, have to say about Simulate Paper Color and Simulate Black Ink.

Quote
Checking Simulate Black Ink turns off the black-point compensation, while checking Simulate Paper Color makes Photoshop use absolute colorimetric rendering instead, forcing the printer to reproduce the actual "paper white" and actual "ink black" of the simulated proof.

If you're printing to a low-dynamic-range process, such as newsprint or inkjet on uncoated paper, Simulate Black Ink will give you a much better idea of the actual blacks you'll get in print.

I would not describe Epson Enhanced Matte as newsprint or uncoated inkjet paper.
Title: Rant 23
Post by: bjanes on October 11, 2009, 07:43:35 pm
Quote from: Jack Flesher
I think Thomas was being sarcastic Ray .  Your methodology works for your particular paper, profile, printer and monitor, and since that's all you use, it works -- a classic case where your particular combination of wrongs happen to cancel each other out and make a right    

However, you'll likely be screwed with any paper change.  OTOH, Thomas' workflow is a proper CM workflow and he (along with everybody else that uses same) can switch between papers or even printers without significant issues.

Again, happy shooting and happy printing!

Sarcasm and condescension are not welcome in this forum, but abound in some quarters.  

Thomas matches the monitor white point to the paper color in his viewing booth, and this is suggested as the proper workflow if I interpret Jack's comment correctly. However, there are other viewpoints. In Color Management, Second Edition, by Bruce Fraser, Chris Murphy, and Fred Bunting, it is suggested to set the white point to 6500K, adjust the luminances of the monitor and viewing booth to the same level, and deal with the color of paper white in the profile. However, they do allow that some people working in extremely color critical environments and having complete control over all variables may want to adjust the white point of the monitor to match a sheet of paper in the viewing booth. What type of paper is not specified. These conditions are not met by most non professional photo enthusiasts. I don't know what setup Ray uses, but my "viewing booth" is a Solux lamp, and I won't attempt to enter the rarefied world of Thomas and Jack.  

Bruce and colleagues note that a CRT allows changing the white point by varying the RGB channels, but it may not operate optimally at a lower white point. With LCDs, the white point has to be adjusted in a look up table unless you are using a model with separate red, blue and green LEDs for the backlight. With an 8 bit LUT, the gamut of the monitor will be compromised. I would be interested in what our resident experts think of the advice given by Bruce et al.
Title: Rant 23
Post by: tho_mas on October 11, 2009, 07:52:24 pm
Quote from: Ray
Now you are becoming insulting.
I've mentioned the visual match of paper and monitor white several times. You seem to ignore this. Too, I explained why the measurement with certain papers (especially those used for photgraphy, namely literally all semi glossy or glossy papers) leads to a blueish/grayish representation of white in those profiles.

Quote
Checking Simulate Black Ink turns off the black-point compensation, while checking Simulate Paper Color makes Photoshop use absolute colorimetric rendering instead, forcing the printer to reproduce the actual "paper white" and actual "ink black" of the simulated proof.
The first part (simulate black ink turns off the BPC) - in my reading - is nonsense. In fact it enables BPC for the preview on the monitor.
If you don't trust me, try it yourself. Look at an image (Jack's printer test) without softproof. Now enable softproof with rel.col RI + BPC and only with "black ink" for the preview: the dark tonal values will be displayed brighter (and less contrastier). Why? Because the black point of the paper (of the target profile) is reflected for the preview on the monitor. With "black ink" disabled the (rel.col+BPC) softproofed image is displayed within the contrast range of the monitor.
Too, as with "paper simulation" the "black ink" simulation is selected simultaniously that would mean - in the consequence of what they are saying - that BPC for the preview would be turned off as well. Doesn't makes sense to me... other than there a misreading due to language barriers.
Now as to paper simulation... in theory that's all great. Above all: in theory it's true. The problem is the (real world) measurement of certain media. Photographic papers often contain, I repeat myself, optical brighteners. Optical brighteners reflect UV light as bluesih. So the actual white is recorded as a light blue in the profile. And therefore at the same time darker than white. See my post #141.

You can make your own proof: adjust the display to the white of the paper - so that the monitor white matches the paper white under your prefered viewing conditions visually.
Now white is white. Not warmer nor colder but the same. Not darker nor brighter, but the same.
Now you enable paper simulation and the white changes. At the latest now you must wonder what is going on - the monitor already matched the paper white; if the white changes duee to "paper simulation", it simply can't be "true".

Title: Rant 23
Post by: tho_mas on October 11, 2009, 08:06:06 pm
Quote from: bjanes
However, they do allow that some people working in extremely color critical environments and having complete control over all variables may want to adjust the white point of the monitor to match a sheet of paper in the viewing booth. What type of paper is not specified.
too, it is possible to edit the white point of the media profiles and in fact this is what some professionals do. But it is a lot of work and AFAIK one have to really know what he is doing. So we should cancel that :-)
I'm sorry that I can't provide links or literature where the problem is discussed... it's all German.
Title: Rant 23
Post by: Ray on October 11, 2009, 09:06:12 pm
Quote from: tho_mas
I've mentioned the visual match of paper and monitor white several times. You seem to ignore this. Too, I explained why the measurement with certain papers (especially those used for photgraphy, namely literally all semi glossy or glossy papers) leads to a blueish/grayish representation of white in those profiles.

 The first part (simulate black ink turns off the BPC) - in my reading - is nonsense. In fact it enables BPC for the preview on the monitor.
If you don't trust me, try it yourself. Look at an image (Jack's printer test) without softproof. Now enable softproof with rel.col RI + BPC and only with "black ink" for the preview: the dark tonal values will be displayed brighter (and less contrastier). Why? Because the black point of the paper (of the target profile) is reflected for the preview on the monitor. With "black ink" disabled the (rel.col+BPC) softproofed image is displayed within the contrast range of the monitor.
Too, as with "paper simulation" the "black ink" simulation is selected simultaniously that would mean - in the consequence of what they are saying - that BPC for the preview would be turned off as well. Doesn't makes sense to me... other than there a misreading due to language barriers.
Now as to paper simulation... in theory that's all great. Above all: in theory it's true. The problem is the (real world) measurement of certain media. Photographic papers often contain, I repeat myself, optical brighteners. Optical brighteners reflect UV light as bluesih. So the actual white is recorded as a light blue in the profile. And therefore at the same time darker than white. See my post #141.

You can make your own proof: adjust the display to the white of the paper - so that the monitor white matches the paper white under your prefered viewing conditions visually.
Now white is white. Not warmer nor colder but the same. Not darker nor brighter, but the same.
Now you enable paper simulation and the white changes. At the latest now you must wonder what is going on - the monitor already matched the paper white; if the white changes duee to "paper simulation", it simply can't be "true".

I haven't ignored it. I'm just not clear how you are matching the paper color and monitor white in your calibration process. After I've calibrated my monitor, I wouldn't want to touch any controls. I've got options of individual R,G,B, adjustments, and a selection for 9300K, 5000K and sRGB. The monitor is calibrated at maximum contrast setting and minimum brightness setting.

I've gone to the trouble of posting a screen shot of your original cloud scene as it appears on my monitor using 'Simulate Paper Color' demonstrating that my setup does not produce a bluish/grey white with Epson Enhanced matte which quite a white paper for matte and I would therefore think it would have a lot of artificial whiteners and brighteners. Now it's true that Simulate Paper Color in relation to the Prm Lustre and Prm Gloss profiles does produce a slightly more blue/grey appearance in my monitor preview, but I assume (from memory) that's how the papers are. The whites of Matte papers, especially rag papers, tend to be more yellow, and glossy papers more blue. Is this not the case?

I don't use a viewing booth. I work on my computer in an open environment, in broad daylight with a view out of the window. In the evening I use 'cool-daylight' energy-saving lights.
Title: Rant 23
Post by: tho_mas on October 11, 2009, 09:58:36 pm
Quote from: Ray
I haven't ignored it. I'm just not clear how you are matching the paper color and monitor white in your calibration process. After I've calibrated my monitor, I wouldn't want to touch any controls. I've got options of individual R,G,B, adjustments, and a selection for 9300K, 5000K and sRGB. The monitor is calibrated at maximum contrast setting and minimum brightness setting.
ah, okay, I see.
As part of the calibration you set the white point by adjusting the RGB chanels to match a certain target (e.g. 5800K or 6500K or whatever).
Skip that!
Take your prefered paper (in your case preferably in smooth, cloudy daylight) and adjust the RGB chanels so that the monitor white matches the white of the paper.
Now reset the white point target in the calibration software, i.e. set the software to calibrate to "native white point" (i1 Match provides it as far as I remeber correctly). Now the calibration software will do the rest as usual.
After calibration, of course, don't touch the monitor menu anymore.

Quote
I've gone to the trouble of posting a screen shot of your original cloud scene as it appears on my monitor using 'Simulate Paper Color' demonstrating that my setup does not produce a bluish/grey white with Epson Enhanced matte which quite a white paper for matte and I would therefore think it would have a lot of artificial whiteners and brighteners. Now it's true that Simulate Paper Color in relation to the Prm Lustre and Prm Gloss profiles does produce a slightly more blue/grey appearance in my monitor preview, but I assume (from memory) that's how the papers are. The whites of Matte papers, especially rag papers, tend to be more yellow, and glossy papers more blue. Is this not the case?
That's true. Matte papers are more yellowish and glossy papers more blueish (basically). Cheap newspaper papers are extremely yellowish. There are some papers without any optical brighteners, e.g. from Canson and AFAIK Hahnemühle has some, too (probably other manufacturers as well, I don't know).
But the measurment errors in the profiles affect the preview on the monitor much more as you would consider the difference between those papers hanging on the wall. My prefered FUJI semi matt is slightly darker and a touch more bluesih in reality as the Innova Fiba ultra smooth. But in softproof mode with paper simulation it's a world of a difference - here the FUJI is clearly blue, the Innova is actually almost neutral (edit: hups - bad memory; the Innova is quite blueish as well)... still paper simulation darkens the image on the monitor. That would be fine if the resulting print would be darker as well... but it is not.
I.e. whilst the differences of the papers are almost negligible (as long as you don't compare the papers side by side) the white points in their respective profiles are very, very different.
So I don't change the calibration of the display either way which print I prepare. The monitor simply displays "neutral" white and as long as the papers I use are within a certain tolerance that's all fine.
I would possibly re-calibrate if I'd use noticeably warmer papers as well (with hardware calibration you just switch profiles and the software adjusts the monitor hardware to the respective values, so that's quite comfortable with HW calibration in this case).
BTW: I don't have any printer... I am prining in a lab.
Title: Rant 23
Post by: Ray on October 11, 2009, 10:42:46 pm
Quote from: tho_mas
Take your prefered paper (in your case preferably in smooth, cloudy daylight) and adjust the RGB chanels so that the monitor white matches the white of the paper.
Now reset the white point target in the calibration software, i.e. set the software to calibrate to "native white point" (i1 Match provides it as far as I remeber correctly). Now the calibration software will do the rest as usual.
After calibration, of course, don't touch the monitor menu anymore.


I'll experiment a bit and see what results I get. I notice that the RGB channels on my monitor cannot be adjusted individually. I've always left them at their maximum setting. My first colorimeter was the Spyder many years ago that I used to calibrate a ViewSonic monitor that did have individually adjustable RGB channels. I found it frustrating and impossible to get a good result. The Eye-One produces a much better result automatically. I believe when trying the 'advanced mode' I incorrectly set D65 as a white point instead of 'native white point'.

Thanks for your feed-back.
Title: Rant 23
Post by: tho_mas on October 12, 2009, 06:16:15 am
Quote from: Ray
Thanks for your feed-back
You're welcome.

Just checked a selection of the Bill Atkinson profiles I found on the web.
Here's a selection of paper profiles set as softproof for a white plane with re.col+BPC & paper simulation (screenshots converted to sRGB).
I've copied a white square in the center so that the tint shift and luminance decrease is clearer.
Now if you consider the white square in the center as the white color and white luminance level you are perceiving when you look at the papers these examples show quite well the dilemma of color management regarding the white point simulation of print media.
Yes, there is a slight difference in the tint of papers when you compare them visually (so "real world") - but these differences are clearly far less accentuated visually as the white points in the respective profiles (try to) simulate.

Bill Atkinson Epson9800 Kodak Premium Luster:
[attachment=17148:9800_kplu.jpg]

Bill Atkinson Epson9800 Premium Glossy:
[attachment=17149:9800_pgl.jpg]

Bill Atkinson Epson9800 Premium Semimatte:
[attachment=17150:9800_psm.jpg]

Epson 11880 Innova Fiba ultrasmooth:
[attachment=17155:innova_fb_us.jpg]

Epson 11880 Hahnemühle Photorag:
[attachment=17154:hm_photorag.jpg]

FUJICOLOR CRYSTAL ARCHIVE DIGITAL PAPER TYPE DPII glossy:
[attachment=17151:fujigloss.jpg]

FUJICOLOR CRYSTAL ARCHIVE DIGITAL PAPER TYPE DPII semimatte:
[attachment=17152:fujimatt.jpg]

And some standard offset profiles...

GraCol2006coated1v2:
[attachment=17153:gracol06_coated1v2.jpg]

Swop2006coated3v2:
[attachment=17157:swop06_coated3v2.jpg]

ISOcoatedV2:
[attachment=17156:isocoatedV2.jpg]

Quote
I'll experiment a bit and see what results I get
yes, I'd say figure out if you can get a closer match with paper simulation deselected in the softproof settings.
By now, as you've mentioned above, the prints get too dark compared to the display when you set the softproof without paper simulation.
If this is true, decrease the luminance level of your monitor a little bit so that monitor white matches at least the brightness level of your paper under your prefered viewing conditions.
Whether you get a closer match with or without "black ink" you have to see. In my experience on very good linearized displays it's better to select "black ink" simulation as it reproduces the contrast range of the print on the monitor.
If this doesn't turn out to be an improvement switch back to the setting you are currently working with as it obviously works good for you.

Have fun :-)

Title: Rant 23
Post by: JeffKohn on October 12, 2009, 01:25:43 pm
Quote
Matte papers are more yellowish and glossy papers more blueish (basically).
I think it has more to do with the use of OBA's, than whether the surface is matte or glossy.

The problem with matching monitor white to paper white during calibration, is that now your monitor calibration is only good for one paper. Maybe that's all some folks need, but many folks use multiple papers.
Title: Rant 23
Post by: tho_mas on October 12, 2009, 02:54:29 pm
Quote from: JeffKohn
The problem with matching monitor white to paper white during calibration, is that now your monitor calibration is only good for one paper. Maybe that's all some folks need, but many folks use multiple papers.
to reply I quote myself:
Quote from: tho_mas
Yes, there is a slight difference in the tint of papers when you compare them visually (so "real world") - but these differences are clearly far less accentuated visually as the white points in the respective profiles (try to) simulate.

does, BTW, the Colormunki solve the problem?
Title: Rant 23
Post by: Wayne Fox on October 12, 2009, 03:52:08 pm
Dang, thought I was lost and fell into the color management forum for a minute ...

(sorry, couldn't resist  )
Title: Rant 23
Post by: Alan Goldhammer on October 12, 2009, 04:45:50 pm
Quote from: tho_mas
does, BTW, the Colormunki solve the problem?
You could create a separate monitor profile for each paper if you want to use the matching white approach.  Spectraview allows this but it may be far more trouble than it's worth if you are printing on more than 2 or 3 papers (I think you would have to reboot the computer to load the new monitor calibration if you are switching papers; not to mention the need to go through the recalibration process multiple times according to your schedule (I recalibrate every two weeks)).
Title: Rant 23
Post by: tho_mas on October 12, 2009, 04:55:14 pm
Quote from: Alan Goldhammer
You could create a separate monitor profile for each paper if you want to use the matching white approach.  Spectraview allows this but it may be far more trouble than it's worth if you are printing on more than 2 or 3 papers (I think you would have to reboot the computer to load the new monitor calibration if you are switching papers; not to mention the need to go through the recalibration process multiple times according to your schedule (I recalibrate every two weeks)).
actually I was referring to the UV cut filter of the Colormunki - in my opinion the Colormunki should not have problems with optical brighteners... so actually it should be possible to create a profile with an accurate white point with the Colormunki, no?
Title: Rant 23
Post by: Ray on October 12, 2009, 06:46:54 pm
Quote from: tho_mas
yes, I'd say figure out if you can get a closer match with paper simulation deselected in the softproof settings.
By now, as you've mentioned above, the prints get too dark compared to the display when you set the softproof without paper simulation.
If this is true, decrease the luminance level of your monitor a little bit so that monitor white matches at least the brightness level of your paper under your prefered viewing conditions.
Whether you get a closer match with or without "black ink" you have to see. In my experience on very good linearized displays it's better to select "black ink" simulation as it reproduces the contrast range of the print on the monitor.
If this doesn't turn out to be an improvement switch back to the setting you are currently working with as it obviously works good for you.

Have fun :-)


Thomas,
My memory has been jogged. I bought the Eye-One with i1Match software over 4 years ago. The colorimeter comes with a plastic cap which clips over the sensors when measuring the ambient temperature. My first few calibrations would have been in 'advanced' mode, selecting 'native white point'. This option requires one to go through the procedure of taking an ambient temperature reading before sticking the colorimeter on the middle of the screen. That cap is a hassle to prise off.

I would have initially ticked the 'reminder' box. The maximum reminder period is no more than one month. That's too fussy for me. At some point, to save unnecessary hassle, I would have switched to 'easy' mode which gave me equally good results, and have stuck with that mode for the past few years, although I now don't bother recalibrating every month as recommended. You might recall the monitor profile I sent you was dated July 2009. A recalibration every 3 months I think is good enough.

Now, what happens when I adjust the monitor before calibration, in an attempt to match the paper white in my ambient lighting conditions with the pure white of a new document in Photoshop on the monitor?

Almost complete disaster!   .  Although I'm working in an open area with just the natural daylight through the windows, the ambient temperature of my environment is very warm, according to i1. As low as 3900K. I presume the reason for this is that much of my interior walls consist of varnished timber which is rather reddish yellow.

If I have a near-perfect match with my monitor settings at maximum RGB brightness, which I do, then any changes to the monitor settings prior to calibration are likely to result in a worse match, wouldn't you say?

First of all, it's not really possible in my circumstances to get a precise match between monitor white and paper white, probably because I have no adjustment of the individual RGB channels. Selecting the monitor's preset 5000K seems about as close as I can get for Epson Enhanced Matte. The resulting calibration using the monitor's 5000K preset, produces an image in proof setup, with simulate black ink, which looks in daylight like the print looks in warm, artificial light in the evening.

Now, that's not a bad appearance. I quite like it. But it's not a good match between monitor preview and print under the same ambient lighting conditions. However, if a client were to request a print which would only be viewed in the usual warm lighting conditions of the average home in the evenings, and if I were only prepared to work during the day with my current setup, then that monitor profile would be ideal for that specific purpose, would it not? I could even show the client on the monitor in the daytime what his print would look like in his requested viewing conditions in the evening.

Which brings me to a related issue. If you are producing prints under controlled lighting conditions, in a viewing booth for example, then ideally such prints should always be viewed under the same or similar lighting conditions, give or take a reasonable margin of error. Now, this may be possible in a gallery or museum, but is unlikely to be the situation in people's homes. In fact, one of the beauties of the print, as opposed to the monitor, is the way a print's appearance will change so easily in accordance with changing lighting conditions. The hues and shades of a print on your wall, viewed in the natural light of a bright day, will differ considerably when viewed in the evening; will differ again in the subdued lighting suitable for watching TV, and yet again when having a candle-light dinner. Is this not the case?

I still believe that whatever your viewing environment, the most important thing is the match between monitor preview and print, not whether you have a viewing booth, or work in the evening with D50 or D65 artificial lighting, or whether your walls are white, off-white or cream etc.
Title: Rant 23
Post by: tho_mas on October 13, 2009, 04:13:35 am
Quote from: Ray
I still believe (...) the most important thing is the match between monitor preview and print...
correct
Quote
...not whether you have a viewing booth, or work in the evening with D50 or D65 artificial lighting, or whether your walls are white, off-white or cream etc.
search for D50 and keyword "metamerism" on the web and you'll find some reasons why the above mentioned match is achieved at best under those viewing conditions.
Title: Rant 23
Post by: Ray on October 13, 2009, 08:03:31 pm
Quote from: tho_mas
correct
 search for D50 and keyword "metamerism" on the web and you'll find some reasons why the above mentioned match is achieved at best under those viewing conditions.


I shall be very interested to find out what the ambient temperature reading is when I move into my new house which has been designed with lots of windows and glass sliding doors (I have no privacy issues where I live). The ceiling and walls are mostly off-white, and the floor tiles are a pale beige. I wouldn't be surprised if the ambient temperature during the daytime is pretty close to D50. For the evening, I'll experiment with whatever lighting I can find. If the D65 of daylight streaming through my windows results in an ambient temperature of D50 throughout the house, then I should fit the house with D65 light bulbs for the evenings, right?