Ad
Ad
Ad
Pages: « 1 2 3 [4] 5 »   Bottom of Page
Print
Author Topic: arbitrary RGB profiles for opening RAW  (Read 12745 times)
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #60 on: April 09, 2014, 03:07:28 PM »
ReplyReply

is it possible to link to those originals from Atkinson?

I have inspected the Epson profile I used for printing the ProPhoto evaluation image. It contains the B2A0,B2A1, and B2A2 intent tags which I believe are perceptual, relative, and absolute respectively. Now, why did the Photoshop manages colors path fail to honor them?
« Last Edit: April 09, 2014, 03:34:41 PM by Lundberg02 » Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #61 on: April 10, 2014, 01:37:42 AM »
ReplyReply

I have inspected the Epson profile I used for printing the ProPhoto evaluation image. It contains the B2A0,B2A1, and B2A2 intent tags which I believe are perceptual, relative, and absolute respectively. Now, why did the Photoshop manages colors path fail to honor them?

Generally speaking, BtoA0, BtoA1, and BtoA2 (or vice versa) are the Perceptual, Colorimetric and Saturation tables respectively. In practice, relative colorimetric and absolute colorimetric rendering intents use the same Colorimetric table, with the latter performing white point mapping while the former does not.
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #62 on: April 10, 2014, 02:21:17 AM »
ReplyReply

Good to know...I didn't plot the PP RGB against the ARGB image in ColorThink...that seems to be fairly conclusive except I added the ARGB image into the PP RGB image and converted the ARGB to PP RGB as a layer. So, either the original test target was done in PP RGB and transformed to ARGB or the original target was in ARGB and transformed into PP RGB. Only the author of the target would know for sure how the target was made.

Jeff, this is really interesting - I've tried doing it by your method (convert), and weirdly, the test image colors expanded when converting from Adobe RGB to ProPhoto. This is reminding me of darlingm's thread a while back.

I've mapped the test image's colors in ColorThink against Adobe RGB (wire frame) and attached it below. The "Adobe RGB" and the "Convert Adobe RGB to ProPhoto RGB" layers are named in the tiff file, layered so one can perform the hide/unhide of layers to view the changes.

So besides the issue that Jeff's method is not accurate for the kind of evaluation he was making, does anyone know why the test image colors will expand when converting from Adobe to ProPhoto? Btw I've also tried converting from sRGB to Adobe RGB with similar colors-are-expanding results.
Logged
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #63 on: April 10, 2014, 03:33:29 PM »
ReplyReply

I printed the ProPhoto test target from PhotoLine just for laughs.   I used Printer color and ColorSync, Perceptual, Absolute, and Saturation intents. All were exactly the same and no different from the results from Photoshop
So this is a farce. I can't think of any explanation. I'm going to assume that nothing works as advertised and just go ahead the way most people do, screw around until it looks decent.
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #64 on: April 10, 2014, 06:45:53 PM »
ReplyReply

I printed the ProPhoto test target from PhotoLine just for laughs.   I used Printer color and ColorSync, Perceptual, Absolute, and Saturation intents. All were exactly the same and no different from the results from Photoshop
So this is a farce. I can't think of any explanation. I'm going to assume that nothing works as advertised and just go ahead the way most people do, screw around until it looks decent.

Something is wrong. When you soft proof in Photoshop, does the preview change as you select differing rendering intents?
Logged
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #65 on: April 10, 2014, 11:56:07 PM »
ReplyReply

I'll say something is wrong. But what? There is nothing I can think of that would affect two completely different applications the same way. The only common elements are the image and the printer including its profile.

When you convert back to ProPhoto you're just doing the inverse transform. The color plot should expand. Same with sRGB.
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #66 on: April 11, 2014, 07:00:40 AM »
ReplyReply

When you convert back to ProPhoto you're just doing the inverse transform. The color plot should expand. Same with sRGB.

That was my guess. That's rare that Jeff made an error in this. I initially suspected the difference in the inverse transform (converting back to ProPhoto from Adobe) was due to the difference in white point of the color spaces and the inevitable rounding errors that would occur. However, the sRGB to Adobe RGB convention results in larger differences than the original Adobe RGB colors than I would expect. It is always the bright reds where the differences are the largest.

But what? There is nothing I can think of that would affect two completely different applications the same way. The only common elements are the image and the printer including its profile.

So soft proofing with an epson printer profile does not work except for the relative colorimetric intent?
Well you are on a Mac. Maybe colorsync is screwing up again?
Logged
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #67 on: April 11, 2014, 01:55:58 PM »
ReplyReply

Sorry, I haven't done soft proof as yet. The fact that two different apps do the same thing has me baffled, unless Photoshop uses ColorSync when "Photoshop manages colors". I don't think that's true.  I guess I can soft proof just for drill and then decide about OS and Photoshop preferences.  I may google ColorSync to see if anyone else has had this problem.

Added: Soft proof does nothing either, although I don't trust my sRGB monitor to show it. You would think Saturation would have some effect though.
I was unable to find any trouble reports concerning ColorSync.
I have no idea what to do.
« Last Edit: April 11, 2014, 04:09:27 PM by Lundberg02 » Logged
FranciscoDisilvestro
Sr. Member
****
Offline Offline

Posts: 500


WWW
« Reply #68 on: April 11, 2014, 07:21:17 PM »
ReplyReply

When you convert back to ProPhoto you're just doing the inverse transform. The color plot should expand. Same with sRGB.

This is not possible. There is no way to know if the color in the smaller gamut came from out of gamut colors in a larger space. Try a different tool.

Regarding perceptual and relative colorimetric rendering intents, I see differences both in softproof and in prints when using a proper output (printer/paper) profile. The diferrences can be very subtle, especially in the prints. I have used the reference image mentioned before and the differences in softproofing are huge (and yes, the file I have is Prophoto RGB). If you don't see any difference, either you profiles are matrix based (Then only relatilve colorimetric) or there is something wrong in you system.
Logged

bjanes
Sr. Member
****
Offline Offline

Posts: 2785



« Reply #69 on: April 11, 2014, 09:07:58 PM »
ReplyReply

Jeff, this is really interesting - I've tried doing it by your method (convert), and weirdly, the test image colors expanded when converting from Adobe RGB to ProPhoto. This is reminding me of darlingm's thread a while back.

I've mapped the test image's colors in ColorThink against Adobe RGB (wire frame) and attached it below. The "Adobe RGB" and the "Convert Adobe RGB to ProPhoto RGB" layers are named in the tiff file, layered so one can perform the hide/unhide of layers to view the changes.

So besides the issue that Jeff's method is not accurate for the kind of evaluation he was making, does anyone know why the test image colors will expand when converting from Adobe to ProPhoto? Btw I've also tried converting from sRGB to Adobe RGB with similar colors-are-expanding results.

It does not seem reasonable that colors would be expanded when converting from AdobeRGB to ProPhoto, since the purpose of color management is to keep colors the same. To test this phenomenon of expansion, I downloaded Bruce Lindbloom's synthetic color checker (which is in L*a*b). I converted to AdobeRGB in Windows Photoshop CC using the Adobe ACE color engine (the other option is the Microsoft CMM. I don't what options are available on the Mac, but colorsynch is likely an option).

I then converted the AdobeRGB image to ProPhotoRGB (relative colorimetric with dithering off) and used ColorthinkPro to assemble color lists for the two images and calculated Delta Es. A representative range of red values is shown. The RGB values are of course different, since we are using different color spaces, but the L*a*b values are virtually identical with Delta Es of less than one. This test fails to show any expansion. For information on colorlists, I would suggest referring to the Digitaldog's tutorial.

Bill

Logged
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #70 on: April 11, 2014, 09:39:47 PM »
ReplyReply

I've found some test procedures with profiles to check rendering intents and will try them tomorrow.  I lean toward believing the ProPhoto Printer Evaluation Test Image as being the problem. Hasta mañana.
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #71 on: April 12, 2014, 07:28:47 AM »
ReplyReply

It does not seem reasonable that colors would be expanded when converting from AdobeRGB to ProPhoto, since the purpose of color management is to keep colors the same. To test this phenomenon of expansion, I downloaded Bruce Lindbloom's synthetic color checker (which is in L*a*b). I converted to AdobeRGB in Windows Photoshop CC using the Adobe ACE color engine (the other option is the Microsoft CMM. I don't what options are available on the Mac, but colorsynch is likely an option).

I then converted the AdobeRGB image to ProPhotoRGB (relative colorimetric with dithering off) and used ColorthinkPro to assemble color lists for the two images and calculated Delta Es. A representative range of red values is shown. The RGB values are of course different, since we are using different color spaces, but the L*a*b values are virtually identical with Delta Es of less than one. This test fails to show any expansion. For information on colorlists, I would suggest referring to the Digitaldog's tutorial.

Bill



Hi Bill, your test using the synthetic colorchecker will not show the "colors expanding" phenomenon, since it does not contain colors outside of Adobe RGB, whether it is generated in Lab or not.

I downloaded another of Bruce Lindbloom's test images, An RGB Image Containing All Possible Colors. Then I did the following:

1. Assigned ProPhoto RGB to it.
2. Convert that to Adobe RGB.
3. Re-convert that back to ProPhoto RGB.

It exhibits the exact same "colors expanding" phenomenon. I graphed the results in 3D in ColorThink, and also in the Color Worksheet. Unfortunately, the number of unique colors is too large (always more than 60, 000, since the target is made up of gradients) but the pictorial dE map shows a significant number of colors over 4 dE2k.

Please try it yourself.

Lundberg02's explanation by way of inverse transform is not very accurate, I believe, as the difference compared to the original ProPhoto RGB colors is too huge also. I have never tried this experiment before! These are unexpected results, and I hope to know why this behavior is observed.
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #72 on: April 12, 2014, 07:31:26 AM »
ReplyReply

I lean toward believing the ProPhoto Printer Evaluation Test Image as being the problem.

I'm pretty sure that is not the problem - I have no issues with that test image on my end, with different rendering intents, both in soft proofing as well as printed versions.
Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2785



« Reply #73 on: April 12, 2014, 09:15:34 AM »
ReplyReply

Hi Bill, your test using the synthetic colorchecker will not show the "colors expanding" phenomenon, since it does not contain colors outside of Adobe RGB, whether it is generated in Lab or not.

Yes, I chose that image because it contains no illegal colors.

I downloaded another of Bruce Lindbloom's test images, An RGB Image Containing All Possible Colors. Then I did the following:

1. Assigned ProPhoto RGB to it.
2. Convert that to Adobe RGB.
3. Re-convert that back to ProPhoto RGB.

It exhibits the exact same "colors expanding" phenomenon. I graphed the results in 3D in ColorThink, and also in the Color Worksheet. Unfortunately, the number of unique colors is too large (always more than 60, 000, since the target is made up of gradients) but the pictorial dE map shows a significant number of colors over 4 dE2k.

Please try it yourself.

Lundberg02's explanation by way of inverse transform is not very accurate, I believe, as the difference compared to the original ProPhoto RGB colors is too huge also. I have never tried this experiment before! These are unexpected results, and I hope to know why this behavior is observed.

According to Lindbloom's site, 12.7% of possible ProPhoto colors are outside of the L*a*b gamut. Assigning ProPhotoRGB to the image with all possible RGB colors would include these illegal colors, and it is not surprising that anomalous results would occur, since L*a*b is used as an intermediate space in the transform.

I'm not certain that there is a real problem here.

Bill
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #74 on: April 12, 2014, 09:56:19 AM »
ReplyReply

Yes, I chose that image because it contains no illegal colors.

According to Lindbloom's site, 12.7% of possible ProPhoto colors are outside of the L*a*b gamut. Assigning ProPhotoRGB to the image with all possible RGB colors would include these illegal colors, and it is not surprising that anomalous results would occur, since L*a*b is used as an intermediate space in the transform.

I'm not certain that there is a real problem here.

Bill


You can also try assigning Adobe RGB (no illegal colors), converting to sRGB, and converting back to Adobe RGB with the same colors-expanding results.
Logged
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #75 on: April 12, 2014, 03:19:16 PM »
ReplyReply

http://www.colorwiki.com/wiki/Stunt_Profiles
This link provides output profiles for verifying intents by producing tinted prints. I'll be using this to verify my workflow.
i tried Saturation intent and the print came our blue as specified.
So now it must be the ProPhoto Printer Evaluation Image itself that isn't any different visually in the various intents. How can this be( as my control systems professor used to say after filling his blackboard with an equation and writing an equals sign)?
« Last Edit: April 12, 2014, 05:23:32 PM by Lundberg02 » Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2785



« Reply #76 on: April 12, 2014, 04:28:59 PM »
ReplyReply

You can also try assigning Adobe RGB (no illegal colors), converting to sRGB, and converting back to Adobe RGB with the same colors-expanding results.

I performed the test you suggested and got similarly large DeltaEs. This does not make sense to me and there must be some bugs in the workflow. This is a matter for Thomas Knoll or Eric Chan to address. That said, in a color managed workflow all images should be tagged with their color space and there should not be a need for assignment of profiles. One instance where one might not want to embed a profile in one's files is when sending files saved in the native space of a printer or similar output device with a chosen rendering intent. The printer profiles are large and would waste considerable space. The untagged files could be sent directly to the printer without adjustments.

Bill

Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #77 on: April 12, 2014, 08:51:59 PM »
ReplyReply

http://www.colorwiki.com/wiki/Stunt_Profiles
This link provides output profiles for verifying intents by producing tinted prints. I'll be using this to verify my workflow.
i tried Saturation intent and the print came our blue as specified.
So now it must be the ProPhoto Printer Evaluation Image itself that isn't any different visually in the various intents. How can this be( as my control systems professor used to say after filling his blackboard with an equation and writing an equals sign)?

Well, I don't know why either. Either the differences in the ProPhoto Printer Evaluation Image are too subtle for you to pick up (highly unlikely) or something is wrong elsewhere with your system.  Huh
Logged
samueljohnchia
Sr. Member
****
Offline Offline

Posts: 260


« Reply #78 on: April 12, 2014, 08:59:49 PM »
ReplyReply

I performed the test you suggested and got similarly large DeltaEs. This does not make sense to me and there must be some bugs in the workflow. This is a matter for Thomas Knoll or Eric Chan to address. That said, in a color managed workflow all images should be tagged with their color space and there should not be a need for assignment of profiles. One instance where one might not want to embed a profile in one's files is when sending files saved in the native space of a printer or similar output device with a chosen rendering intent. The printer profiles are large and would waste considerable space. The untagged files could be sent directly to the printer without adjustments.

Bill



Thanks for the confirmation of this phenomenon!

So this behavior definitely should be treated as a bug, not something normal in the way profiles are converted? It would seem almost impossible that this issue was not brought up before.

I beg to differ, profiles are assigned all the time, whether we are involved or not. Raw converters assign profiles to raw images. In a color managed workflow, one also must assign a scanner profile to a film scan, and then usually convert it to one of the many well behaved RGB working spaces for further editing. So what happens when one assigns a scanner profile (smaller gamut) and converts to ProPhoto RGB (wider gamut)? Are colors going to expand as well?
Logged
Lundberg02
Jr. Member
**
Offline Offline

Posts: 96


« Reply #79 on: April 12, 2014, 11:37:25 PM »
ReplyReply

Well, I don't know why either. Either the differences in the ProPhoto Printer Evaluation Image are too subtle for you to pick up (highly unlikely) or something is wrong elsewhere with your system.  Huh
I have looked at the various prints under supposedly full spectrum lighting and can't see any difference. Can you tell me your visual impression of the Saturation intent print? this should be the one with the most noticeable change I would think. If you could point out which part of the image to examine I would appreciate it. Maybe post a crop of side by side with relative.
« Last Edit: April 14, 2014, 12:57:42 AM by Lundberg02 » Logged
Pages: « 1 2 3 [4] 5 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad