Soft proof adjustment sRGB profile not working (Affinity Photo for Windows 1.10.5.1342)

[Maarten Langeveld]

Hello developers,

 

Affinity Photo is a great product but regrettably soft proofing AdobeRGB images to sRGB gamut is not working on Affinity Photo for Windows 1.10.5.1342. 

Adding the soft proof adjustment layer with sRGB profile to an Adobe layer that contains colours that are out of sRGB gamut, nothing happens... As opposed to for example selecting Agfa: Swop Standard profile.

See attached document.

Would you please fix this in next release?

 

Kind regards,

Maarten

 

 

Affinity Photo for Windows1.10.5.1342 soft proof adjustment sRGB not working.afphoto

Edited July 15, 2022 by Maarten Langeveld
Textual correction

[Dan C]

Hi @Maarten Langeveld,

Thanks for your report!

I can confirm this is a known bug, logged with our developers - so I'll be sure to 'bump' this now using your document example.

We hope to have this fixed in a future version, I hope this helps

[PixelEngineer]

On 7/15/2022 at 10:19 AM, Dan C said:

Hi @Maarten Langeveld,

Thanks for your report!

I can confirm this is a known bug, logged with our developers - so I'll be sure to 'bump' this now using your document example.

We hope to have this fixed in a future version, I hope this helps

Hi @Dan C 

In the Adobe forums there are many users having the same question, as to why when soft proofing from AdobeRGB to sRGB, they are not seeing a change in color. There appears to be an answer for this. An expert wrote a paper on AdobeRGB and sRGB color space. In the paper he explains why many people get confused when comparing these spaces.

Source:
Solved: Soft Proofing not working for sRGB - Adobe Support Community - 12111557
 

You can read a paper that he wrote on AdobeRGB and sRGB color space in the link below.
Color Numbers and Color Gamut by Andrew Rodney author of "Color  Management for Photographers"
http://digitaldog.net/files/ColorNumbersColorGamut.pdf

FYI: In the sample files that were uploaded in both of these threads,  in the layers panel, the AdobeRGB photo on the top layer is not exactly the same as the AdobeRGB photo in the bottom layer.  There is a "shift" in a part of the photo. So it was modified. They need to be exact duplicates. This can throw a tester off. Just wanted you to be aware of that.

[Dan C]

Many thanks for letting me know, and for the extra due diligence taken here, it's certainly appreciated!

I'll be sure to pass this information through to our team who are investigating this issue further internally

[lacerto]

I am not sure about the point of elaborating discussion of something that I assume is already acknowledged as a bug or a feature omission, but as the attachment in the OP's post was poor in illustrating the issue of the topic, and the issue referred on Adobe forum was different (related to an illusory problem of not seeing difference between the soft proofed and the converted image, which there naturally does / should not exist -- unless working with Affinity apps where soft proofing with non-media based ICCs like sRGB simply does not work), here is a clip that shows the issue of the topic more clearly (though using a larger-gamut ProPhoto color space as the source, rather than Adobe RGB, to accentuate the difference when comparing with the expected behavior as shown when soft proofing in PS). To be able to reproduce the problem, the user should have a system that can display larger than sRGB color gamut.

 

The color space of the clip itself is truncated but the relative difference in behavior can still be clearly seen. In a nutshell: the expected behavior is that the Soft Proof adjustment previews a different (lower) gamut color space as reliably as possible and that the preview matches the actual conversion.

As mentioned in many posts before, Affinity Soft Proof seems to be useful only in RGB based workflows when targeting to lesser-gamut ICCs involving specific media (though lack of paper simulation and dithering options make it less useful than color proof of PS). It does reflect the change of color space also in context of CMYK workflows but the simulated output can be far away from the actual conversion, and is of little use because Affinity apps provide much better CMYK "soft proof" internally when having the document designed and viewed (but containing color definitions made in non-target, larger-gamut color spaces) in CMYK color mode. 

[pbasdf]

In my experimentation, I have found that the problem with Soft Proofing to sRGB is very specific to the sRBG IEC61966-2.1 profile.  I found that if I use a different sRGB profile, the soft proofing appears to work.  The document shown below is in Adobe RGB.  The rectangles are, from top to bottom (on the canvas and in the layer stack)

(1) a rectangle filled with a colour gradient from RGB 0,255,255 to 0,255,0.  No adjustments.

(2) an identical rectangle with a Soft Proof adjustment using the sRGB2014 profile available from the ICC (https://www.color.org/srgbprofiles.xalter).

(3) an identical rectangle with a Soft Proof adjustment using the sRGB IEC61966-2.1 profile.

(4) the first rectangle, exported to TIFF using sRGB IEC61966-2.1 (profile embedded), placed within this document.

Note that (I hope this comes out in the screenshot):

(3) appears identical to (1), ie soft proofing to sRGB IEC61966-2.1 has no discernible effect,

(2) is different from (1), ie soft proofing to sRGB2014 does have an effect, and

(2) appears identical to (4), ie soft proofing to sRGB2014 seems to have the same effect as converting to sRGB IEC61966-2.1. 

For information, I use a 2019 iMac running macOS Monterey 12.5; I assume the sRGB IEC61966-2.1 profile derives from there.  In fact, as a separate test (not shown), I have copied this profile and used the Mac's ColorSync utility to modify JUST the "desc" tag in the profile: soft proofing with this seemed to work identically to the sRGB2014 profile from the ICC.

I hope all that is useful for those investigating the issue, and also hopefully provides a workaround for those who want to Softproof sRGB in the meanwhile: head over to the ICC and install one of their sRGB profiles (even if you only use it for Softproofing, not for conversion).

[lacerto]

Thanks for further information. Yes, this seems to be correct. I also tested other RGB profiles (Apple RGB and Adobe RGB) using ProPhoto RGB as the source image color space and they do cause change in Soft Proof adjustment, so this could really be just the standard sRGB-profile related issue. I'll make some comparisons with other software to see if I get identical results there. N.B. I ran the tests above on Windows (11) and assume that the default sRGB profile came with the system (but will check this -- EDIT: yes, it is one that came with Windows 11).

UPDATE: After some initial tests, I seem to get identical results with the 2014 sRGB profile when using Photoshop (2022), and as this is a v2 profile it should be safe to use it in place of the older v2 profile, so this really appears to be a good work around. There might be point in using it also as the replacement of the default sRGB working profile in the Preferences.