Luminous Landscape Forum
Site & Board Matters => Luminous Landscape Video => Topic started by: OleDK on September 12, 2007, 02:08:13 pm
-
First: The videos are fantastic, even better than the LR tutorial.
But the softproofing section left me confused. It works - my print are dramatically improved. But I'm confused.
I believed that the printer/paper ICC profiles would capture the paper's properties, so that printing from lightroom with the correct profile but without softproofing and without tweaking curves and saturation should give you as good a print as you can expect from that printer/paper combination. In other words, I believed that the profiles would "contain" any necessary black, curve and saturation tweaks.
But that's obviously wrong. Where did I go wrong in my logic?
BR
Ole
-
The output profiles do capture the properties you mention. What you're missing is the soft proof (simulation of this effect) on images you view. The numbers for the printer are correct, the numbers for the preview are not.
-
Softproofing merely gives you the ability to fine tune the image for the particular stock, by seeing a more accurate representation of the image on the selected stock with the chosen profile. It also gives you a way of choosing the rendering intent visually. It alone does not affect the way a file prints, it is only a tool for visualizing the anticipated output and being able to add a bit of human input to the automatic process involved in printing with an ICC profile.
-Ron
-
Andrew and Ron, thanks for your answers. It seems as it is difficult to explain what I don't understand.
I do understand that the softproofing simulates what the print will look like with a particular profile and rendering intent.
I don't understand why it's possible to improve that by tweaking. For example, Schewe improves the print by adding a slight curve to make the softproof look more like the on-screen image. I you have a good profile, both the print and the softproof should look as much like the on-screen image as the printer's/paper's gamut will allow. So how is it possible to improve that by adding a curve? Isn't that a sign that the profile is not good, because it failed to apply that curve in translating the RGB values?
BR
Ole
-
I don't understand why it's possible to improve that by tweaking.
Because profiles are not perfect nor do they know squat about images (only how a big pile of solid colors output on the printer measure). You don't see your image as a million solid square colored pixels of a specific numeric value. But a profile does.
-
I don't understand why it's possible to improve that by tweaking. For example, Schewe improves the print by adding a slight curve to make the softproof look more like the on-screen image.
BR
Ole
[a href=\"index.php?act=findpost&pid=139242\"][{POST_SNAPBACK}][/a]
It seems to me that what is happening is that, first you adjust the image so that it looks good on the screen. Then you turn on softproof and the image looks like crap. So you adjust the softproof image to look as much as possible like the non-softproof ( the good looking) image.
If you now turn off softproof, your image will appear on the screen as you originally wanted but will print as you intended (more or less )
Alan
-
Because profiles are not perfect nor do they know squat about images (only how a big pile of solid colors output on the printer measure). You don't see your image as a million solid square colored pixels of a specific numeric value. But a profile does.
Soft proofing & matte (http://forums.dpreview.com/forums/readflat.asp?forum=1003&message=24798190)
In the above link, I have read that for matte papers it is even worse. It was even advised to switch off paper simulation and black point compensation while using relative intent to help to soft proof more correctly for the matte paper. At least in this post, it seems more an HP issue.
I really like my HP B9180 but things seems not so easy for this printer.
-
As I recall, this issue is pretty well covered in the videos with jeff stressing that this will happen. I get the same with my Epson 2400 and matte papers but the screen proof is very good nevertheless (it looks c..p because it is compared to the version on scereen)
-
It was even advised to switch off paper simulation and black point compensation while using relative intent to help to soft proof more correctly for the matte paper. At least in this post, it seems more an HP issue.
[a href=\"index.php?act=findpost&pid=139337\"][{POST_SNAPBACK}][/a]
It ain't an HP issue...it's a failure on the part of the user to know how to use soft proofing...
Look, you evaluate an image on a computer display, if you are using an LCD, you'll have a contrast range of perhaps 500/1. When you soft proof, the Display Options for paper white and ink black will change your display range from 500/1 to more like 100/1 (for matte paper).
Sure, it'll look like crap–so will your print relative to the display at 500/1 contrast range–which is the whole purpose of the soft proof...to show you what your image WILL look like on paper. It's then up to you to correct for the soft proof display...
And I seriously wouldn't spend any time on DPReview and expect any really educated info...
-
And I seriously wouldn't spend any time on DPReview and expect any really educated info...
[a href=\"index.php?act=findpost&pid=139439\"][{POST_SNAPBACK}][/a]
I had been softproofing sort of like Jeff showed in the C2P video, but I picked up some really good tips there. If I didn't softproof, my prints would be 'crap' too LOL--but, you need a good paper profile. The 3800 Epson profiles seem pretty good IMO, but most other papers seem to need custom profiles.
I tend to send serious posters to this forum--I try to find a thread that will help them. I've also been recommending the C2P and LR videos. I know that the LR videos got me up to speed really quickly. The C2P brought me back to considering a workflow incorporating LR (which I had used in beta, got vs. 1.0 as licensee of RSP, and after ACR 4.1 came out had more or less given up on LR) so I wanted to be sure I had a pretty good grasp of at least the basics of LR (I didn't it turned out--I didn't realize how easily one can do the round trip to PS).
Diane
-
Thank you to everyone who answered.
I don't think I'm less confused, but I don't feel bothered by that confusion anymore :-)
And the video helped a lot on my print results.
Br
Ole
-
And the video helped a lot on my print results.
Right now. I have followed what I learned about soft proofing (amongst other things) in C2P video while printing and I got excellent result. For sure, I am only a little amateur but I really like to understand what I am doing and try to optimize the modest prints I do.
I hope that experts like Mr Schewe and Mr Reichman among some others will continue to make web/video tutorial. It offers for amateur like me a way to learn "modern" photography matters from the bests.
I wish I could go in Antartica... Is there any LL, D-65,... workshop in Belgium?
Have a good print,
John.
-
I hope that experts like Mr Schewe and Mr Reichman among some others will continue to make web/video tutorial
Amen.
-
It ain't an HP issue...it's a failure on the part of the user to know how to use soft proofing...
Sure, it'll look like crap–so will your print relative to the display at 500/1 contrast range–which is the whole purpose of the soft proof...to show you what your image WILL look like on paper. It's then up to you to correct for the soft proof display...
[a href=\"index.php?act=findpost&pid=139439\"][{POST_SNAPBACK}][/a]
Here is the problem: (and by the way thank you so much for a warm and informative tutorial..I enjoyed the company very much)
I printed on Hahnemuhle Photorag Bright White 310 and used the appropriate paper profile downloaded from the Hahnenmuhle website.
I followed the exact instructions from the c2p tutorial on a colormanaged, calibrated PC system.
The print that I hold in my hand is a very precise rendition of the image on the screen - WITHOUT softproofing. When I compare the print I am holdeing in my hand with the in CS3 softproofed version with simulate paper color checked I see a flat hazy greyish version with no blacks. This version is so unlike the actual print.
Now where is "the fault on the part of the user" ??
It helps a lot to uncheck the "simulate paper color" but then again that's not what the tutorial recommends. Could it be that this particular paper profile is not good for softproofing?
-
It's entirely possible that the profile you've downloaded has a poor interchange space to display space set of tables that prevent it from being really useful for softproofing...output profiles have two purposes...going to the printer space from your color space and going to the display space from your printer profile. Good profiles will do both well...
On the other hand, if you say the print match is "a very precise rendition" of your screen image, then I think the user error involved here is that you are looking at color (which can match easily) but are ignoring the differences in the dynamic range. There is simply no way a print, with a contrast range of oh, say, 150/1 could EVER match a computer display that may be 500/1.
You say that when soft proofing, with black ink and paper white checked, it looks grey with no actual black? That is indeed correct...compared to the black of the display, the print will be far less contrasty. The white will be dimmed as well. This is as expected...and the fact is, that "look" is actually much more accurate to the print than you are thinking. With a properly calibrated and profiled display, and an accurate soft proof set of tables for the output profile, only then is the soft proof "accurate" if you know how to look at and evaluate the soft proof.
Try hitting the F key twice and the Tab key to hide any and all white components of Photoshop. You should have a black border all around the image with no palettes visible. Then put the print on a lightbox that is 90º from the display (you simply can not put them side by side because the white point and luminance will never match between display and lightbox). Then turn on soft proofing and look away...then look to the print and then swivel your head to look at the display. Done properly, with accurate profiles, the screen to print match will be very, very close both in terms of color and dynamic range. Only then can you use soft proofing to allow you to make subtle corrections from screen to print.
If you aren't using the paper white and ink black display options, you aren't getting what you need from soft proofing...
-
You say that when soft proofing, with black ink and paper white checked, it looks grey with no actual black? That is indeed correct...compared to the black of the display, the print will be far less contrasty. The white will be dimmed as well. This is as expected...and the fact is, that "look" is actually much more accurate to the print than you are thinking. With a properly calibrated and profiled display, and an accurate soft proof set of tables for the output profile, only then is the soft proof "accurate" if you know how to look at and evaluate the soft proof.
[a href=\"index.php?act=findpost&pid=139772\"][{POST_SNAPBACK}][/a]
Thank for your reply. Of course my monitor (30" HP LP3065) is calibrated (Spyder2Pro) and of course I don't compare sreen and paperprint side by side. I studied the C2P very carefully and followed the guidelines.
I also acknowledge the differences in dynamic range.
I follow you all the way.
My only problem is that black is dark grey on screen when softproofing but pitch black on paper. The dynamic range may be similiar but it is in my opinion incorrect softproofing to show grey shades when my printer is capable of printing pitch black. And I mean totally black, not just dark.
My initial thought was that softproofing in CS3 could possibly be a little obsolete because the next generation of printers may have overcome the "printing black inadequacy"?
It is only some Fine Art paperprofiles that display softproofing inaccuracy regarding black. Otherwise I don't see that problem and have no problems softproofing.
(English isn't my native language. Sorry)
-
I hope that experts like Mr Schewe and Mr Reichman among some others will continue to make web/video tutorial
Yes their tutorials are very very nice
Really Great
-----------------------------------------------------
Canvas Print
http://www.fruit-art.co.uk (http://www.fruit-art.co.uk)
-
For people who know the terminology, Andrew and Jeff explain this well, but it's a difficult thing to explain to people new to soft-proofing.
Consider this version: your monitor, even if calibrated and profiled, will not ever display color exactly the same way that the print will, because of the inherent differences in transmissive light (monitor) vs. reflective light (ink on paper).
The range of colors a device can reproduce is its gamut, and the image file itself has its own gamut: the color values it contains. Some of these colors can be accurately printed, others accurately displayed on the monitor, but not necessarily both.
Soft-proofing in Photoshop uses your monitor to simulate a visual interpretation of what the image will look like when printed on a specific printer and paper. So you're simulating the image gamut, mapped to the printer gamut, mapped to the monitor gamut.
When soft-proofing, the black ink and paper white simulations will usually look awful on the display. But this is reality, because the color management system is translating the color values of the file through the color gamuts of the device profiles. Blacks on a monitor will appear much darker than a print because on a monitor black is the absence of light. On a print, even black ink is reflecting some light.
You have to train your eye for Soft-Proofing; especially for black ink and paper white.
And you have to use the Simulate Paper White and Black Ink options to get it right.
Put two copies of the same image side by side (use the Image > Duplicate command) - one is soft-proofed, the other is not - the goal is to visually match the Soft-Proofed version as closely as possible to the Reference (not soft-proofed).
Use adjustment layers to make them match, usually a boost in saturation (maybe around +10 in Hue & Saturation?) and Curves (usually lightening the shadows to open them up). With some papers (eg. Silver Rag, Premium Luster) you won't have to make much of an adjustment. With others (eg. Somerset Velvet, PhotoRag or any kind of canvas) the adjustments will be more extreme because of the limited color gamut of the substrate.
In my classes I've been asked the question "but since we're using a custom profile for the printer and paper, doesn't that take care of the color matching?" No, because of different color gamuts. Remember that the photo has its own color gamut (depending on the working color space) and some of these color values can't be reproduced exactly with the chosen printer/paper combination. The numeric color values are translated through the color management system and always have to be adjusted for the gamut of a specific device.
Within Photoshop, files with embedded or assigned working profiles (Adobe RGB, ProPhoto, sRGB etc.) are always displayed color-managed. If you use a good monitor calibration system (EyeOne, not Spyder), you can trust the colors in the Master file. But you still have to adjust for the output.
So when soft-proofing, you will almost always need to make adjustments to your Master file to get the print to match what you see on the screen. I have Master files with Adjustment Layers made for Museo Silver Rag, Epson Premium Luster, Hahnemuhle Photo Rag, Breathing Color Canvas... the list goes on and on. Each printer and paper combination has been soft-proofed and adjusted before printing.
And in the end, sometimes there is no substitute for a hard proof. Be willing to make test prints and make adjustments accordingly.
-
Thanks for your very clear contribution to this topic.
-
For people who know the terminology, Andrew and Jeff explain this well, but it's a difficult thing to explain to people new to soft-proofing.
SNIP
So when soft-proofing, you will almost always need to make adjustments to your Master file to get the print to match what you see on the screen. I have Master files with Adjustment Layers made for Museo Silver Rag, Epson Premium Luster, Hahnemuhle Photo Rag, Breathing Color Canvas... the list goes on and on. Each printer and paper combination has been soft-proofed and adjusted before printing.
And in the end, sometimes there is no substitute for a hard proof. Be willing to make test prints and make adjustments accordingly.
[a href=\"index.php?act=findpost&pid=142535\"][{POST_SNAPBACK}][/a]
Thanks Nat. Great explanation and I have bookmarked it to use when people want a clear explanation of softproofing.
Diane
-
In my classes I've been asked the question "but since we're using a custom profile for the printer and paper, doesn't that take care of the color matching?" No, because of different color gamuts. Remember that the photo has its own color gamut (depending on the working color space) and some of these color values can't be reproduced exactly with the chosen printer/paper combination. The numeric color values are translated through the color management system and always have to be adjusted for the gamut of a specific device.
[a href=\"index.php?act=findpost&pid=142535\"][{POST_SNAPBACK}][/a]
Nat, thanks a lot for a thorough explanation. That question which you have been asked in your classes is exactly the question I tried to start this thread with.
-
Nat, thanks a lot for a thorough explanation. That question which you have been asked in your classes is exactly the question I tried to start this thread with.
Regarding the gamuts of files and devices, another explanation I neglected to mention is that the printer profile describes the characteristics (and color gamut) of the printer/paper combination but does not alter it. So when printing from Photoshop (and using Photoshop to Manage Colors), Photoshop uses the printer profile in conjunction with the system CMS to render the print job. This is why making previous adjustments while soft-proofing is necessary.
However, if you print using a RIP (raster image processor) instead of printing from Photoshop, the RIP can actually send data that affects the behavior of the printer, so soft-proofing isn't as critical.
-
Nat, perhaps one could add to this that the trasnmitted luminance from a display is brighter and has more DR than reflected light especially from a matte printing paper, and the soft-proofing, with "Simulate Paper White" selected, helps to mimic the appearance of image luminosity on paper, allowing one to make the final luminosity tweaks under soft-proofing that will minimize disappointment about what emerges from the printer.
-
I just watched the soft proofing film.
I made a profile for Moab bright paper the same way they did in the video. My problem is when I soft proof with paper color checked, I have a big shift in color. It adds some kind of majenta in the photo.
Actually I have printed a sepia photo without softproofing and my print is near what I see in my monitor regarding color.
What am I missing?
Thanks,
Nicole
-
I just watched the soft proofing film.
I made a profile for Moab bright paper the same way they did in the video. My problem is when I soft proof with paper color checked, I have a big shift in color. It adds some kind of majenta in the photo.
Actually I have printed a sepia photo without softproofing and my print is near what I see in my monitor regarding color.
What am I missing?
Thanks,
Nicole
[a href=\"index.php?act=findpost&pid=153420\"][{POST_SNAPBACK}][/a]
A good profile? Seriously, how did you make the printer profile? What hardware, what software? Note that none of the software that uses a scanner in the process to create a printer profile will produce good profiles (in my experience). A spectrophotometer and a good software (Monaco Profiler of ProfileMaker or even an Eye-One and Eye-One Match) will be required...
-
A good profile? Seriously, how did you make the printer profile? What hardware, what software? Note that none of the software that uses a scanner in the process to create a printer profile will produce good profiles (in my experience). A spectrophotometer and a good software (Monaco Profiler of ProfileMaker or even an Eye-One and Eye-One Match) will be required...
[a href=\"index.php?act=findpost&pid=153424\"][{POST_SNAPBACK}][/a]
I have exactly the same profiler than Michael has in the video: eyeone photo.
When I softproof on photorag natural, I don't see this shift in color, only in the bright. I made the profile the same way as we see in the video. The photo I tried was a very light tone photo: a closeup of a baby in black and white with a little tone in it. I profile my papers since 3 years and the way it was done in the video was the way I was doing it. My printer is the epson 7800
Thanks for your help
Nicole
-
"I you have a good profile, both the print and the softproof should look as much like the on-screen image as the printer's/paper's gamut will allow."
Ole, you answered your own question, but you may not have realized it. "as the printer's/paper's gamut will allow" is the key phrase in your post.
If the print is your final output, you have to use soft-proofing to let you see a close approximation of what will actually come out of your printer. Then you can make final adjustments to your image to optimize for what the printer/paper combination can actually produce. With today's technology, the printer/paper combination is the weakest link in the chain, if you don't soft-proof you're playing against the house. You may luck out occasionally but, over time, the house always wins.
You won't be able to nail it down absolutely, there's a big difference in the emissive image you see on your monitor and the reflective image you get from the print. But soft-proofing will get the two a lot closer than just shooting in the dark. And, given the cost of ink and paper, that's a "Good Thing".
HTH,
Bill
-
Regarding the gamuts of files and devices, another explanation I neglected to mention is that the printer profile describes the characteristics (and color gamut) of the printer/paper combination but does not alter it. So when printing from Photoshop (and using Photoshop to Manage Colors), Photoshop uses the printer profile in conjunction with the system CMS to render the print job. This is why making previous adjustments while soft-proofing is necessary.
However, if you print using a RIP (raster image processor) instead of printing from Photoshop, the RIP can actually send data that affects the behavior of the printer, so soft-proofing isn't as critical.
Whoa, your previous post was very clear, but now you have lost me. Are you really saying that if I have Photoshop manage colors, the only effect of the printer profile is to modify the soft-profiled monitor image? So I will get the same print (from the same file on the same printer on the same paper) no matter what profile I use? I hope this is not true. I thought the whole point of a profile is to alter the values sent to the printer, to make the print match the file as closely as possible, whether printing from Photoshop or from a RIP.
-
Thanks alot. I also own Camera to print, as always beeing very confused person. This discussion made mind clear up a bit extra.
One day hopefully I will fully understand the black magic off camera to print. Thou the tutorial and this forum has helped me a great deal to understand alot more.
Cheers,