On Dec 4, 2014, at 10:35 AM, Brad Funkhouser <brad.funkhouser@xxxxxxxxxxx> wrote: > Viewing a printer-space-tagged image in Photoshop seems to result in > relative mappings of printer space paper white to display white and printer > space black to display black. So it still takes soft proof settings to get > an accurate display of the expected results. This is true -- as is your description of what happens when you use Argyll to do the absolute conversion back to a working space. But I've given up, for the most part, on worrying about on-screen proofing with absolute colorimetric intent. Your visual system is too smart for that...unless you've got the most amazing workspace setup and exercise perfect discipline, adaptation and other effects cause the onscreen absolute version to just look like you've got a color filter over it, which isn't at all the same cognitive stimulus you get from viewing the print. For me, opening the image with the printer's space (and whatever Photoshop does) is a great way to get an idea of where you're going to lose details, where color shifts are going to happen, that sort of thing -- especially if you do the blink comparator thing with original and printer-profiled versions. But try to do that with an absolute rendering and _everything_ looks off, making it very difficult to spot the changes. I've also come to realize that chances are slim that I'm ever going to manually do a better job than Argyll...sure, I could fix that one spot over there, but that's just going to make things worse in this other spot, and so on. Almost always, the solution is a paper with a bigger gamut, or else an acceptance that it is what it is. And when the image lies within the printer's gamut, it's nearly impossible to tell original from copy...which is why "better paper" is my go-to solution. If you _do_ have a world-class soft proofing setup, fantastic...but an actual print (perhaps a reduced size and / or crop) is so much easier and reliable.... b&
Attachment:
signature.asc
Description: Message signed with OpenPGP using GPGMail
