Soft Proof question in CS3

[cabasner]

Hello all,

First, I want to say that the "From Camera to Print" video done by Jeff Schewe and Mike is one of the best all around digital image primers I have seen (not that I'm an expert).  It goes into great detail and just covers so much.

That said, the chapter on soft proofing, while I understand what Jeff was talking about, brought to my mind a fundamental question.  What Jeff was doing was comparing the image that he had created in ProFoto RGB color space, to the soft proofed image that was generated using the profile for (in the case on the video) the Epson 3800 printer.  He then went on to modify, to some degree, the image so that the soft proofed image, which was the image to be printed, matched, as far as possible, the original, non soft proofed image.  What doesn't make any sense is why we would simply not edit the image under soft proof condition FROM THE BEGINNING, rather than edit it under other conditions, then try to 'backfit' the soft proof to match the image we just so painfullly worked on to get perfect.

WHAT AM I MISSING?

[picnic]

Hello all,

First, I want to say that the "From Camera to Print" video done by Jeff Schewe and Mike is one of the best all around digital image primers I have seen (not that I'm an expert).  It goes into great detail and just covers so much.

That said, the chapter on soft proofing, while I understand what Jeff was talking about, brought to my mind a fundamental question.  What Jeff was doing was comparing the image that he had created in ProFoto RGB color space, to the soft proofed image that was generated using the profile for (in the case on the video) the Epson 3800 printer.  He then went on to modify, to some degree, the image so that the soft proofed image, which was the image to be printed, matched, as far as possible, the original, non soft proofed image.  What doesn't make any sense is why we would simply not edit the image under soft proof condition FROM THE BEGINNING, rather than edit it under other conditions, then try to 'backfit' the soft proof to match the image we just so painfullly worked on to get perfect.

WHAT AM I MISSING?
[a href=\"index.php?act=findpost&pid=131819\"][{POST_SNAPBACK}][/a]

Jeff's way is the way I've been doing it for a good while.   My reason for not doing it the way you suggest is--I use that same image in a variety of different ways--both in my online gallery (sRGB), then perhaps on different media depending upon size wjocj will require using a different profile and different results for softproofing (for instance, if I create a book--and sometimes even the cover--I do some handmade/handbound books---then it would be a different media than for fineart).  And---I save the image as native---native size, working space, etc.---and may use it differently--or print it differently--in the future.  

Others may have other reasons.  Consequently, I softproof when I'm ready to print on a particular paper and make adjustments for that particular image then---and often save those adjustments since I may want to use them in the future.

I wanted to add I found the videos very informative and interesting also.

Diane

[Schewe]

WHAT AM I MISSING?
[a href=\"index.php?act=findpost&pid=131819\"][{POST_SNAPBACK}][/a]

If ALL you were going to EVER do to an image was to make prints from a single printer/paper combination, what you suggest would be ok...nothing wrong with it. But you would limit what you would be able to do to repurpose the images.

I only soft proof once I have a specific destination, including printer & paper in mind...otherwise what I'm looking for is an opimized Archive image.

[kaelaria]

Exactly as stated above.

I keep my masters, then softproof for each output separartely:

monitor
local printer gloss
local printer matte
mpix gloss
mpix matte
mpix mettalic
ezprints...you get the idea.

The whole POINT of softproofing is to see what happens to your original when it meets an output that is less than ideal, and they all are in differing ways.

[cabasner]

Thanks to all who answered, and especially Jeff, since now I have the word right from the source!  I guess I really wasn't missing anything, just didn't consider the ideas of images for multi use and optimizing for each particular destination.

Thank you again.

[cbcbell]

I just watched section 14, "Soft Proofing," and also learned a great deal — what an excellent way to explore complex subjects in depth! I love the idea of being able to create actions and assign function keys to preview the changes in rendering intent between Perceptual and Relative Colorimetric, but when I tried to create an action on my Mac that recorded the View > Proof Colors command (Cmd-Y), and stored the fact that I had a custom proof settings group saved and chosen as a default, I found that nothing was recorded. I'm quite comfortable recording a variety of Photoshop actions, but don't know what extra step needs to be taken to make it possible to record this action. Can anyone point me in the right direction?

Thanks.

[genemcc]

Christopher,

If you need to insert the CTRL+Y toggle into your action then while recording - go to the flyout menu on the actions panel and choose INSERT MENU ITEM.  You'll get a dialog instructing you to select the menu item with your mouse.  Select the VIEW->PROOF COLORS from the menu and you'll see it in the dialog box.  Press OK and now it's in your action.

I hope that helps!

[mdijb]

Even though I attempt soft proofing and use the method descrived in the video, I have never been impressed with it's accuracy.  There is just to large a difference in contrast and brightness between the screen image and the results on the fine art papers  I use.  The image on paper simply down not match--reflected light from paper vs transmitted light on the monitor.  The colors seem to match well, but the other attributes do not--I believe this comes with the territory--Soft proofing can get you in the right ballpark, but additional tweaking is still required and someitmes it takes a lot.

My procedure is now to write an action that makes a basic curves adjustment layer that thru experience I know my printer is going to need to get the image I want.  I make four strengths of this adjustment and save each one.  Then I create a proof sheet and print all four on one sheet of paper.  From there I evaluate which one matches my intent and make furhter tweaks to get the final  print.

I find that spending a lot of time worrying about trying to get a match that just won't happen, is not worth the time.  Instead I make the final judgements based on how the image looks on the paper and use that as my guide.

MDIJB

[cbcbell]

Gene, that's just the information I needed — thanks very much!

[Schewe]

Even though I attempt soft proofing and use the method descrived in the video, I have never been impressed with it's accuracy.  There is just to large a difference in contrast and brightness between the screen image and the results on the fine art papers  I use.  The image on paper simply down not match--reflected light from paper vs transmitted light on the monitor.
[a href=\"index.php?act=findpost&pid=138069\"][{POST_SNAPBACK}][/a]

User error....

If yu know what you ae doing, and do it right, the screen to print match should be 90% + match. If it isn't, you are either doing something wrong or you don't understand what you are seeing.

There are two ways of soft-proofing; one for color (which is easy) and one for dynamic range (which is tough). but only combining them both gets you 90%+ the way...

You can say it doesn't work (I've heard all the reasons) and point out where you think it fails, but in _MY_ experience, really accurate softproofing is scary accurate and once you cross that threashold, you never look back-you become a believer...the trick is learning to cross the line!

[Eric Myrvaagnes]

User error....

If yu know what you ae doing, and do it right, the screen to print match should be 90% + match. If it isn't, you are either doing something wrong or you don't understand what you are seeing.

There are two ways of soft-proofing; one for color (which is easy) and one for dynamic range (which is tough). but only combining them both gets you 90%+ the way...

You can say it doesn't work (I've heard all the reasons) and point out where you think it fails, but in _MY_ experience, really accurate softproofing is scary accurate and once you cross that threashold, you never look back-you become a believer...the trick is learning to cross the line!
[a href=\"index.php?act=findpost&pid=138133\"][{POST_SNAPBACK}][/a]

Jeff is totally correct, as usual.

IME, two of the easiest ways to screw up in a "color managed" setup are
(1)  Having your monitor too bright, and
(2)  Not carefully balancing the room illumination to the monitor to get print-viewing as close as possible to what the monitor shows.

There will always be a difference, but you have to learn to judge the small difference. It's not unlike the old darkroom days of judging a print while wet.

[thompsonkirk]

Back to the original question:

Under certain circumstances, it does make sense to work as you suggested.  If you use only a couple of papers & have consistently-made profiles, you can make proof prints & workprints by soft-proofing & printing with one paper/profile, & saving that file; & then later make only small adjustments when soft proofing & printing with the appropriate profile for fine-art paper.  You don't have to check back to an original or "Ur" version.  Only if you use a bunch of papers &/or need to soft-proof for a variety of media - papers, web, press, etc. - do you really need to check back as Jeff advises.  

In my case, I soft-proof to make workprints on inexpensive EEM paper & save that version of the image as the starting-point for later final prints.  I do about 90% of my PS work & check it out with the cheaper paper.  When changing to $6-8 sheets of fine-art paper (HPR or occasionally Innova UltraSmooth Glossy 285), I start with the EEM image & then soft-proof, readjust, & print with the appropriate profile for HPR (or IUSG).  If I'm lucky, I get from the EEM version to a satisfactory fine-print version in 2 tries.  

Switching from the EEM/workprint to the HPR/final paper & profile, you need only tinker with the small adjustments that account for differences between these 2 papers:  room for a little more saturation, because HPR can yield slightly bolder & more 'painterly' tones than EEM; & a small curves adjustment because HPR can yield a slightly richer black & a little more shadow gradation.

Kirk
www.dryreading.com/kirkthompson

[Colorwave]

I just downloaded Jeff and Michael's video this last week and already feel like it has permanently altered my approach to printing.  

Jeff's softproofing technique has taken the softproof from something I felt offered some, but limited value, to a technique that offers tremendous value for me.  No one point or tip made the difference, it was just the synergistic sum total I got from watching him work and the big picture that finally came together for me.  

After proofing and printing a handful of images this week, I already feel I can trust my softproof and do more to the image with less fear and less trial and error.  The side by side technique of comparing soft proof to original is a very simple method but often more useful than switching back and forth in one window.  Of course, without good profiles all bets are off, but so far APS has served me well in that regard.  Thanks Jeff!