Jump to content

Strange colour representation (and different from Lightroom)


Recommended Posts

17 minutes ago, NotMyFault said:

I would inspect your monitor profile with one of the tools listed in the post i linked before. Assuming a V4 profile could cause the issue.

Check if your Calibration SW is capable of using V2 profiles.

Okay, I just looked it up in i1Profiler. It's a v2 profile, so the version shouldn't be the culprit.

Link to comment
Share on other sites

We may chase a phantom.

The website contains 2 versions of the file.

Actually, only the "print" version seems to have a color profile that is working correctly in photo. The pink color pad gets RGB info panel value 255/0/255.

When using the "Monitor" version, the pink tab is much less saturated 248/19/239 - effectively, all your apps render this file with wrong colors!

Attached you can find the output of exiftool, which lists details of the color profiles of both images.

These files seem very "crafted", and the contained sRGB  profiles looks more like a device profile (instead of a generic profile).

Profile Creator                 : Hewlett-Packard
Profile ID                      : 0
Profile Copyright               : Copyright (c) 1998 Hewlett-Packard Company
Profile Description             : sRGB IEC61966-2.1
Media White Point               : 0.95045 1 1.08905
Media Black Point               : 0 0 0
Red Matrix Column               : 0.43607 0.22249 0.01392
Green Matrix Column             : 0.38515 0.71687 0.09708
Blue Matrix Column              : 0.14307 0.06061 0.7141
Device Mfg Desc                 : IEC http://www.iec.ch
Device Model Desc               : IEC 61966-2.1 Default RGB colour space - sRGB
Viewing Cond Desc               : Reference Viewing Condition in IEC61966-2.1
Viewing Cond Illuminant         : 19.6445 20.3718 16.8089
Viewing Cond Surround           : 3.92889 4.07439 3.36179
Viewing Cond Illuminant Type    : D50
Luminance                       : 76.03647 80 87.12462
Measurement Observer            : CIE 1931
Measurement Backing             : 0 0 0
Measurement Geometry            : Unknown
Measurement Flare               : 0.999%
Measurement Illuminant          : D65
Technology                      : Cathode Ray Tube Display

 

image.png.6256916da63d65e7b1620e7aa5c6fa36.png

P.txt M.txt

Link to comment
Share on other sites

As I wrote before, the problem occurs with all (?) files. I just had the example at hand and it was publicly available. It really shouldn't matter where the profile comes from either. If the industry standard can handle it, Affinity should be able to as well. The file is also specifically meant for matching the monitor (in the photo exposure context).

Link to comment
Share on other sites

Could you please provide a screenshot of this test image, rendered by Photo and one of the other apps. Please use full screen captures at identical positons, zoom level 100%.

https://www.burosch.de/testbilder-uebersicht/337-audio-video-equipment-check.html

The advantage of this image is that it contains numerically defined colors, so we can directly check if it is correct (for all apps).

Link to comment
Share on other sites

On 11/19/2021 at 7:02 PM, Rincewind said:

I did some more tests on this and I think the following shows the problem quite well one more time.

I have now in parallel opened the test file

Just in case: here on mac this magenta swatch shows in APub's Photo Persona the values of your 3 other apps. By the way: Though switching the monitor profile in macOS preferences does obviously change the visual appearance, it neither alters the measured value in the image nor reflects the difference in a screenshot taken in that different profile in the same way as it appears different; switching the profile back makes the different screenshot appear more like the one before the monitor switch.

1807613724_testbildmagentaAPubAPhoto.thumb.jpg.f357b029a3aa16346ac3c1d888f46e0b.jpg

macOS 10.14.6, MacBookPro Retina 15" + Eizo 27"

Link to comment
Share on other sites

5 hours ago, NotMyFault said:

Could you please provide a screenshot of this test image, rendered by Photo and one of the other apps. Please use full screen captures at identical positons, zoom level 100%.

https://www.burosch.de/testbilder-uebersicht/337-audio-video-equipment-check.html

The advantage of this image is that it contains numerically defined colors, so we can directly check if it is correct (for all apps).

No, unfortunately not. I don't spend 15€ just to prove with another example that Serif has to do something...

image.png.daf2e399635db61cb57814859a073d72.png

Link to comment
Share on other sites

I've had a look at the OP monitor profile, it is a v2.1 profile and is using a D50 illuminant. There's nothing wrong with the profile:

monitor-gamut.thumb.png.1cf20584dbf76580263911a2493d6ca5.png

From the creator tag it looks to be made using xRite software. For reference, xRite own Pantone so if they cannot create a profile correctly we should all just give up right now :)

According to the ICC spec, v2 profiles default to D65 white point. This profile is a v2, but has a D50 white point. I believe this is the part that Affinity cannot handle correctly. It is not out of spec, it is handled with a chromatic adaption table.

As the OP has seen, the profile works fine with everything except Affinity. To point the finger of blame at the profile/OS/anything else makes no logical sense.

To the OP - can you re-calibrate your monitor and choose to output a v4 profile instead? That version defaults to D50 and should I believe work for you.

Link to comment
Share on other sites

3 hours ago, BofG said:

To point the finger of blame at the profile/OS/anything else makes no logical sense.

Nobody is blaming or finger pointing.

We simply try to investigate what causes the issue, and find a solution or workaround. 

This D50 white point issue is one of the rare exceptions where Affinity Apps seem to be actually buggy. In 90% of claimed color profile bugs the issue is on other aspects, like corrupt color profiles, mismatch between color profile and image, using device profiles as document color profile, or user error. 

We do not investigate to blame anybody.

2 hours ago, v_kyr said:

See also related:

That was exactly I tried to remember (but failed). Thank you.

Here a link to a related issue rated as bug by mods.

 

Edited by NotMyFault
Added related bug report
Link to comment
Share on other sites

See also other related forum postings:

AFAIK the Affinity suite is LittleCMS based and that CMS should usually support the whole ICC 4.3 spec nowadays.

☛ Affinity Designer 1.10.5 ◆ Affinity Photo 1.10.5 ◆ OSX El Capitan

Link to comment
Share on other sites

4 hours ago, NotMyFault said:

Nobody is blaming or finger pointing.

It's just a turn of phrase, kind of like "why keep looking". Not intended to cause offence to anyone.

4 hours ago, NotMyFault said:

We simply try to investigate what causes the issue

Was it not already shown that it's down to Affinity not correctly interpreting the monitor profile?

Link to comment
Share on other sites

6 minutes ago, BofG said:

It's just a turn of phrase, kind of like "why keep looking". Not intended to cause offence to anyone.

Was it not already shown that it's down to Affinity not correctly interpreting the monitor profile?

Maybe i failed to recognize. It is clear once mods reply and assign a tag with bug id, or move it to the bug section.

And in many cases i try to understand what actually is causing the issue, and what workarounds are possible.

Link to comment
Share on other sites

10 hours ago, BofG said:

As the OP has seen, the profile works fine with everything except Affinity. To point the finger of blame at the profile/OS/anything else makes no logical sense.

That was my thought exactly.

10 hours ago, BofG said:

To the OP - can you re-calibrate your monitor and choose to output a v4 profile instead? That version defaults to D50 and should I believe work for you.

I have now recalibrated the monitor and created a v4 profile. (The X-Rite software by the way recommends using v2 for compatibility with third-party software).
The differences are almost gone (+/-1 in RGB values), now Chrome has more of a borderline deviation.
My conclusion would be: Use of a v4 profile is suitable as a workaround, the comprehensive solution would rather be a bugfix in Affinity Photo (and maybe Designer and Publisher if they share the code part). With the v2 profile, the match (apart from Affinity's Photo) was actually better or even perfect.

Thank you for pointing out the connection between v4 profiles and D50 white point! I think the bug now should be sufficiently narrowed down for a developer.

Now it only remains to hope that Serif will finally take care of the issue. Unfortunately, judging from recent reactions, I don't have many hopes. The forum reminds me a lot of Github issue discussions lately, only there it's about open source and everyone can work out and contribute bug fixes themselves...

Link to comment
Share on other sites

I just tried creating v2 profiles with i1Profiler (a couple of years old version) first with D65 and then with D50 illuminant, and could not make any meaningful difference when viewing the same images in Photoshop (CS6) and Affinity Photo (latest version), on Windows. It is hard to see this as a device profile issue, but it is hard so see this as anything else, either. I do not have LR installed on this computer though (or a later version of PS), so cannot compare directly.

Link to comment
Share on other sites

I don't think it's as straightforward as that, it seems like it's only a subset of V2/D50 profiles that have an issue. What the specific conditions are that trigger the problem I don't know.

I think this mostly shows up as the "white is yellow" issue, which the D50/D65 mixup would explain perfectly. The ICC spec isn't great, from what I recall v2 profiles can have v4 tags in therefore requiring forward compatibility.

I wouldn't want to have to debug it :)

 

Link to comment
Share on other sites

  • 2 weeks later...
  • 2 weeks later...

I just had a read of this:

https://affinityspotlight.com/article/display-colour-management-in-the-affinity-apps/

There's mention in there of another colour profile issue of white displaying as yellow. It's dismissed in a fluff of hand waving as a bad profile that affects all colour managed apps. Comically, they suggest searching for the phrase "whites are yellow" whilst that phrase a is link to Google searching for "Photoshop whites are yellow" - obviously don't want people coming across the Affinity complaints lol.

Anyway, the article provides no detail of what a "bad profile" contains, and gives no proof at all that a profile that gives yellow in Affinity does the same in all other managed applications.

All this is to say don't hold your breath for a detailed answer from Serif on this issue.

P.S. I do really like Serif and Affinity, please don't take me off your Christmas list for this :)

Link to comment
Share on other sites

1 hour ago, BofG said:

There's mention in there of another colour profile issue of white displaying as yellow. It's dismissed in a fluff of hand waving as a bad profile that affects all colour managed apps.

Actually, it says "many colour managed apps," not all of them.

1 hour ago, BofG said:

Anyway, the article provides no detail of what a "bad profile" contains, and gives no proof at all that a profile that gives yellow in Affinity does the same in all other managed applications.

Again, they did not say it affects all color managed apps. But even if you just click the PS Google link, there are many hits that explain the problem in PS is caused by a "bad" monitor profile or an uncalibrated monitor. 

Affinity Photo 1.10.5, Affinity Designer 1.10.5, Affinity Publisher 1.10.5;  2020 iMac 27"; 3.8GHz i7, Radeon Pro 5700, 32GB RAM; macOS 10.15.7
Affinity Photo 
1.10.5.280 & Affinity Designer 1.10.5 for iPad; 6th Generation iPad 32 GB; Apple Pencil; iPadOS 15.0.2

Link to comment
Share on other sites

"..specific display profiles that are installed by default for various displays. They are defective in regards to their compatibility with colour management solutions and do not work properly"

I'm interested to see how you twist that to mean defective profiles can work properly with some colour management solutions.

How about we hold Serif to task to explain their assertions instead of trying to twist words and deflect the question?

Link to comment
Share on other sites

33 minutes ago, BofG said:

"..specific display profiles that are installed by default for various displays. They are defective in regards to their compatibility with colour management solutions and do not work properly"

I'm interested to see how you twist that to mean defective profiles can work properly with some colour management solutions.

How about we hold Serif to task to explain their assertions instead of trying to twist words and deflect the question?

I made request to remove/filter those useless and confusing profiles. Your vote added there appreciated.

 

Link to comment
Share on other sites

1 hour ago, BofG said:

I'm interested to see how you twist that to mean defective profiles can work properly with some colour management solutions.

All I am saying is under "Whites look yellow" in the Spotlight article there is this:

Quote

This issue is not exclusive to the Affinity software: it actually happens with many colour managed apps...

 

Affinity Photo 1.10.5, Affinity Designer 1.10.5, Affinity Publisher 1.10.5;  2020 iMac 27"; 3.8GHz i7, Radeon Pro 5700, 32GB RAM; macOS 10.15.7
Affinity Photo 
1.10.5.280 & Affinity Designer 1.10.5 for iPad; 6th Generation iPad 32 GB; Apple Pencil; iPadOS 15.0.2

Link to comment
Share on other sites

On 12/15/2021 at 6:11 PM, deekay said:

I'm having the same issue and replied to your bug report with example screenshots. 

Thanks for dealing with this and please let us know if there's a solution out there.

Thank you! Sometimes I doubt that I have really posted the issue. This is the first problem that Serif really seems to want to say NOTHING about. I simply can't understand...

On 12/15/2021 at 8:33 PM, BofG said:

All this is to say don't hold your breath for a detailed answer from Serif on this issue.

P.S. I do really like Serif and Affinity, please don't take me off your Christmas list for this :)

I also like Affinity products a lot and have convinced quite a few people in the past to ditch the Adobe subscription. (And I wasn't even on the Christmas list!)
But this is slowly becoming just unprofessional....
I just hope they are working on a solution and don't want to make hasty false statements.
Whereas it would still be much better to say ANYTHING than nothing at all... 😞

Link to comment
Share on other sites

Part of the problem might be that the apps use a third-party colour management system (Little CMS). How much knowledge the team have of the internal workings of that system could play a part.

I've looked with a profile inspector at a "yellow is white" profile from a user here and that profile didn't have any obvious issues. I don't have Adobe apps to test it with, but just that assertion the profile is faulty should be taken with a grain of salt until genuine proof is provided.

Link to comment
Share on other sites

5 hours ago, BofG said:

Part of the problem might be that the apps use a third-party colour management system (Little CMS). How much knowledge the team have of the internal workings of that system could play a part.

Quite possibly, but in my view openness would be the better solution in this case. Maybe someone can even help Serif with the problem...
Apropos, I made another comparison and included GIMP, since it uses LittleCMS. Adobe Lightroom, Irfanview and GIMP have exactly the same colour reproduction, Affinity Photo is off.
I would have been surprised by a general error of this kind in LittleCMS anyway, but this should still be an indication for a specific problem in Affinity Photo. So far I have yet to find a graphics software that gets it wrong too.... ;-)

5 hours ago, BofG said:

I've looked with a profile inspector at a "yellow is white" profile from a user here and that profile didn't have any obvious issues. I don't have Adobe apps to test it with, but just that assertion the profile is faulty should be taken with a grain of salt until genuine proof is provided.

I also tend not to believe that my X-Rite software is not creating a correct profile, especially as every other software can handle it (at least every one I have tested).

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...

Important Information

Please note there is currently a delay in replying to some post. See pinned thread in the Questions forum. These are the Terms of Use you will be asked to agree to if you join the forum. | Privacy Policy | Guidelines | We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.