Topic: Rendering deep shadow detail (Read 32235 times)

mposter · « **Reply #40 on:** December 04, 2005, 06:14:32 pm »

>>Michael, I'm using a 4800 so it is not the size of the paper.

My mistake. Then I'm guessing (and it's just a guess) you're attempting to feed the paper from the tray and the epson driver won't let you feed any paper you've set to "smooth fine art" from the tray. They insist you feed that manually. It's a function of thickness in that case I believe.

>>The Smooth Fine Art setting is not available - in fact the Epson driver won't display any media that is not meant to be used with the Matte Black cartridge - cutting out playing with anything "smooth".

"Smooth" doesn't mean it's not to be used with MK ink. Both Ultrasmooth Fine Art and Smooth Fine Art are pefectly happy with MK ink, in fact it's usually preferred over PK for those and other matte cotton rag papers.

Mark D Segal · « **Reply #41 on:** December 04, 2005, 09:07:29 pm »

Thanks Michael - most likely your right about it being the tray business. I can see them trying to protect the feed mechanism by disallowing the tray for thick media -anyhow - bottom line is that it severely limits the flexibility for playing with paper types in the printer driver.

mposter · « **Reply #42 on:** December 05, 2005, 08:46:06 am »

I learned this from a friend not too long ago. Place the paper into the tray, press the "up" arrow on the printer control panel and the printer will feed the sheet. Set the driver up for manual feeding. This way you can feed thinner sheets with settings that normally won't work when you're feeding from the tray.

Mark D Segal · « **Reply #43 on:** December 05, 2005, 01:05:38 pm »

Michael, good tip on fooling the feed mechanism. Thanks.

Now, I have tested ImagePrint on an Epson 9800 using both my deep-dark greyscale and a "challenged" image. The results are that it provides very, very little improvement in separating tones within the lowest quarter of the tone curve relative to the performance of the Epson driver. ImagePrint is an excellent piece of software for other purposes, and its forthcoming resolution of the ink exchange problem will be a boon to people who need to switch media often enough in a year, but for the particular issue I was trying to resolve it does not seem to me a cost-effective solution. I may be "pushing the envelope" in terms of what the technology can do with this particular set of photographs and media, but will now attempt other approaches within Photoshop to steepen the contrast curve within this section of the tonal range for those images.

photopat · « **Reply #44 on:** December 05, 2005, 01:27:30 pm »

Mark.

What were the settings in IP(shadow point compensation/rendering intent) set to when printing these test.

Did you "softproof" these images in PS using the IP profile and epson profile(to compare the two to see which one might have been more true/close to the print)
on their system???
Did they differ from your display???Or did they show the same "over optimistic" soft proof wiew??
How did the softproof look like in IP compared to the print??

If not done.you have achived nothing in your strive for a WYSIWYG management
And you're back in the "guessing" game again.

Patrick.

Mark D Segal · « **Reply #45 on:** December 06, 2005, 12:52:57 pm »

Patrick, shadow compensation was tried at 50 (IP recommended default) and 100. Rendering intent was rel. col. There was no perceptible difference between these two results.

While the procedures underlying your questions are scientifically vanilla, it was not possible to do all that without disrupting the whole third floor of their shop because it is qute brightly lit, and had other customers buying (expensive) toys. Therefore even though the monitor had a hood, I could see straight away that the viewing conditions were not suitable for doing sensible monitor to print comparisons within the context of their own set-up.

Therefore the test consisted of comparing what came out of their 9800 using ImagePrint and the correct IP profile for EEM versus what came out of my 4800 using the Epson profile for EEM with the same image file and rendering intent. When you compare these results both for the greyscale wedges and a challenged photograph the comparative results are so close that one can come to reasonably reliable conclusions about how the two kinds of software compare for separating deep quarter-tones, despite the fact that the procedure could not be quite as rigorous as it ideally should have been.

I'm not back to square one. What I did saved me a thousand bucks, and suggested I should address the issue with Curves in Photoshop and a slight mental adjustment in how I estimate what I am seeing in the lower quarter-tones on the monitor - absent the "perfect profile", be it for the monitor or the printer or both.

Tim Gray · « **Reply #46 on:** December 06, 2005, 02:02:25 pm »

How is the Version 3.1.07 of ColorEyes Display for setting contrast and brightness? I use the Gretag Eye-One Display (not the version 2) and it is basically useless in setting contrast and brightness on my LCD monitor - is there some automation of those paramaters with your unit?

The reason I ask, is that rather than making a "mental adjustment" could you adjust the c&b to match the print output of a graduated test strip? ...but obviously that won't work if the process of setting the c&b is hardwired into the callibration process.

photopat · « **Reply #47 on:** December 06, 2005, 02:03:37 pm »

Just wanted to have all the input from your test. =)
Since WYSIWYG was as a part of the question (and as I see it has sadly for you .not been solved )

50 is not default (it's just how it's set up when you install it and this is miss understod by many as default value,and I guess in this case this IP reseller)
100 (relative intent)will print more like how the soft proof in PS will look like.

Sorry to hear that their setup was in a way that a soft proof wiew comparing(to your setup) was not possible.
Since this would most likley have shown you that the print was really close to theIP profile in soft proof.(and did not produce a "better" print...It printed what the soft proof would have shown you.......WYSIWYG)

I personally have found the IP profiles to be very accurate in softproof compared to print.(I'm less happy with some other stuff)
And I think that this is exacly what it's all about.(the epson profile "lies.".. and does not show what the print will look like)
This is something that you can test for yourself quite easily.
Just download the EDAY_profile for EEM from IP for the x800 printers(yes anybody can do that)
Put it in the folder were it's possible to use as a profile in PS for soft proofing(I don't know were in a pc since I'm on a Mac).
Compare the soft proof to the print that were printed with 100 shadow point compensation in IP

If I'm not totally misstaken (and there is always that possibility =) ........) You'll find the IP profile(in soft proof wiew) more"close" to the IP print and thereby easier to adjust..
And if that one is "way" of aswell...I can say that your display calibration is most likey not correct
Patrick.

Mark D Segal · « **Reply #48 on:** December 07, 2005, 10:57:53 pm »

Patrick, actually there are two distinct issues: (i) matching the monitor image with the printed image, and (ii) how good the printer/profile are at differentiating deep quartertones. Most of what I wrote above addresses the latter. It only addresses the former to the extent the monitor is inherently capable of differentiating lower luminosity levels than is the printer (at least using matte media). I shall try your IP suggestion to see whether that profile is better at tricking the soft proof into more accurately behaving like the printer. Thanks for the idea.

Mark D Segal · « **Reply #49 on:** December 08, 2005, 01:50:27 pm »

Patrick, I downloaded the IP colour profiles for EEM paper on the 4800 (there are five); the suffix is now R-DAY, not E-DAY. I cycled through all five of them in View>SoftProof>Custom and they create a faint impression of the image that is far, far removed from the printed output. Not useful on this set-up - perhaps one needs the IP program for them to correctly feed data from the file to the monitor. Anyhow, thanks for the suggestion - only a few minutes invested in trying it out!

I'm going to re-calibrate and re-profile the monitor once again, but I think my current monitor profile is pretty good. "ColorEyes Display" can eek viewable information out of image files that is somewhat extra-ordinary, so it is possible that the monitor is giving me inherently richer data than profiles and printers can cope with down in those very low luminosity values. I don't believe the Epson profiles are chronic liars - I think they're on the whole quite truthful, but they tell little white lies about deep black situations. For those challenged images, I may have to be content with depending more on the numbers and careful curves work to maximize printed tonal separation in that range.

photopat · « **Reply #50 on:** December 08, 2005, 03:14:51 pm »

Mark.
You don't need IP to get the correct soft proofwiew in PS as a comparisment to those prints you got from your "IP" test.
So unless you had paperwhite checked when soft proofing there is nothing wrong in what you did when trying this(as I see it)

I only check the paper white to check "overall" color changes that the paper will produce.
But mostly leave it unchecked when doing my adjustments.
I know some of the "big" guys out there recomend checking this during the whole soft proofing
procedure.
I've found that only usable if your "print wiewing setup" is sitting right next to your display.
In my case it's located 90 degrees to my left,so unchecking it gives me a a more "correct" wiew when it comes to blacks and contrast etc since I find the blacks look kinda washed out when checking this compared to the print.
I still do think that a little bit in this problem might be the epson profile.
I do think they tell a bit of a lie between soft proof and print when it comes to deep blacks

However I kinda have the feeling that this(if you tried it) would probably not change anything for you as a closer match between soft proof and print.

And since I know nothing(only what I can read on the net) of the software you're using when calibrating you display, and I'm located in Sweden and you're located in Canada.
It's not like I can swing by your place to see for myself what you see

I'm actually starting to think (as you say yourself about those deep blacks) that unless your calibration is a"little bit off"

Quote

they tell little white lies about deep black situations.

when soft proofing.

If this would be the case.
How about creating a adjustmentlayer that corrects for those dark shades as a visuall guidance while softproofing to make sooft profing a bit easier.
And just deactivate it when printing.
But then again there is always the chance to forgett unchecking it and getting a spoiled print =)

Anyway.it's been fun/frustrating/intresting trying to figure out the cause to your "problem"

Patrick.

Mark D Segal · « **Reply #51 on:** December 08, 2005, 04:05:19 pm »

Patrick,

I tried softproofing with an IP profile because you suggested it in your next to last post. However, I did have "Simulate Paper White" selected. That indeed makes a huge difference. When I deselect Paper White, the monitor image comes much closer to the print, but not any closer than the soft proof I get using the Epson EEM profile (but in this case with "Simulate Paper White" selected). So it seems in the final analysis, another round of monitor calibration followed by some creative use of curves will be the outcome. Many thanks for all your helpful suggestions.

BernardLanguillier · « **Reply #52 on:** December 08, 2005, 09:33:12 pm »

Mark,

There will be a point beyond which the cost of your testing will start to outweight the higher price of Hahnemhule papers... Just teasing you of course.

Cheers,
Bernard

Mark D Segal · « **Reply #53 on:** December 09, 2005, 12:55:58 pm »

OK Bernard,

I know you're teasing - but a comment like that won't slip by an economist like me - so here goes: my Epson 4800 costing model has a column for "technical" which includes all that stuff - accumulated cost of testing this and that over the past six or so weeks has been $ 28.47 - not even the price difference between 50 sheets of Enhanced Matte and 50 sheets of Hahnemule - so I'm still ahead of the game!

Cheers,

Mark

PS. MOAB Entrada, good as it is, also has issues in deep quartertones. I think I'm a real fuss-pot - that may be the real heart of the problem!!!!

Eric Myrvaagnes · « **Reply #54 on:** December 09, 2005, 04:00:43 pm »

Quote

OK Bernard,

I know you're teasing - but a comment like that won't slip by an economist like me - so here goes: my Epson 4800 costing model has a column for "technical" which includes all that stuff - accumulated cost of testing this and that over the past six or so weeks has been $ 28.47 - not even the price difference between 50 sheets of Enhanced Matte and 50 sheets of Hahnemule - so I'm still ahead of the game!

Cheers,

Mark

PS. MOAB Entrada, good as it is, also has issues in deep quartertones. I think I'm a real fuss-pot - that may be the real heart of the problem!!!!
[a href=\"index.php?act=findpost&pid=53150\"][{POST_SNAPBACK}][/a]

And, Bernard, don't forget that the $28.47 includes all of the 5,317 hours that Mark has devoted to the testing (at $0.0 per hour).

Eric

BernardLanguillier · « **Reply #55 on:** December 09, 2005, 10:23:12 pm »

Fair enough Mark, I think that we can allow you to spend another 2000 hours then.

Jokes aside, thanks for doing this, it is interesting for all of us.

Cheers,
Bernard

Eric Myrvaagnes · « **Reply #56 on:** December 09, 2005, 11:44:33 pm »

Quote

Jokes aside, thanks for doing this, it is interesting for all of us.
[a href=\"index.php?act=findpost&pid=53175\"][{POST_SNAPBACK}][/a]

Yes, indeed. Thank you, Mark. I have learned a great deal from your work.

Eric

Mark D Segal · « **Reply #57 on:** December 10, 2005, 09:07:32 am »

And many thanks to all of you who have contributed. It's the people and their varied insights that make the L-L Forum such a great resource.

Author Topic: Rendering deep shadow detail (Read 32235 times)