ICC color management improvements

[BugFix]

Hi folks, currently, XnView MP supports basic ICC color management, but lacks transparent profile info and softproofing, which are essential for accurate color workflows (also simply for sorting as per color space).

Suggested improvements:
-> Display embedded ICC profiles, monitor profile, rendering intent, and color mode directly in the UI.
-> Softproofing to simulate target profiles (e.g., print), with selectable rendering intents and optional black-point compensation.
-> especially a comparison mode / split windows view for simulated and embedded profiles would be a unique feature
-> Enhanced export options: convert to sRGB for web, embed or exclude ICC profiles.

Benefit: Ensures consistent color on display and export, reduces surprises with images lacking profiles, and makes XnView MP more suitable for professional photo workflows, makes it unique in the "viewers" class of software.

[jkm]

You're asking for color management features on par with Photoshop. (Also you neglected to mention color space conversion, which is also essential.)

I agree that robust color management capabilities, such as the capabilities you describe, are essential for good color workflows and professional results.

However, things like soft proofing and rendering intent are in my view only useful when you can actually correct and compensate as needed to produce the desired output. Thus, these features need to be paired with equally robust color correction and editing capabilities.

XnViewMP is not that kind of app. It's primarily a viewer not an editor. The editing and color correction capabilities are rudimentary.

I'm not sure what you describe would be truly beneficial without also dramatically improving the editing capabilities.

Otherwise, you use soft proofing (as a prelude to output on a device with a different gamut, naturally, otherwise your workflow is done with viewing on screen) and you detect issues, and then what? Do you think XnViewMP's color correction is up to that task?

Apps like Photoshop/LRC/Photolab/CaptureOne are more suited to that. I go to Photoshop for color corrections for output, because it has the color adjustment capabilities necessary.

If XnViewMP could do all the things you describe, I would still do my output in PS, because it has the color adjustment capabilities necessary.

I'm not against better color management in XnViewMP, but it seems like it would need to be part of a bigger ask. I'd be interested in your thoughts on that point.

Also, a couple of the things you mention (such as displaying the embedded ICC profile, and exporting with or without the embedded profile) are already present (there's also an overlay icon to indicate images with an embedded profile, and color space/profile is a searchable criteria, and hopefully you've seen the new Color Space tree in the Catalog filter), but yes most would need to be added.

[BugFix]

Hello, first of all, I didn't demand anything in that sense, I merely made a suggestion.

Perhaps I didn't explain myself clearly enough. My suggestion is intended to help improve cataloguing and allow for more criteria to be used and see or compare results when converting to the output, and yes, perhaps I went a little too far, but if it's not possible to display, simulate or handle color spaces identically to the processing programs, then there's a gap that could be closed.

Last but not least, a good program like XNView thrives on progress and further development without straying from its original purpose. There should be a strategy; my suggestion or parts of it could possibly become a long-term development goal, couldn't it? I didn't mean any harm.

h.a.n.d.

[jkm]

Hi... I didn't think you did mean any harm, and I certainly didn't accusing you of demanding anything. The way you described the need was good and I thought quite complete, once color space conversion was added, so it was well thought out in terms of non-editing functionality.

It just seems to me that to make the improvements you suggest (and as I said, I'm not against it at all) truly useful, more changes to the editing capabilities would be required, so I wanted to explore that and see what you thought.

So I was asking if you though that was true, or if you were envisioning some use case where the editing capabilities didn't matter. I intended it as a straightforward question. If you were thinking of some use case I hadn't thought of, I wondered what it was, because I didn't understand the benefit without the editing capabilities.

I'm all for improving the cataloging of XnViewMP, and a lot of progress has been made on that front lately. Indeed, the color profile node in the catalog filter and searching on color profile metadata were just added. So as I said, a little bit of what you mentioned is there. I don't see how soft proofing assists cataloging/organizing/searching, but perhaps I'm missing something.

I certainly didn't think you went too far... Perhaps not far enough! It just depends on the objective and direction of the app.

Are you interested in the app improving as an output stage, where you could do your printing or color space conversion/exporting in the app? That's a bit different from cataloging, and obviously it's not something the app is very good at yet. It would be something of a new direction for the app to go in.

I was just trying to get an exchange of ideas... There wasn't any hidden meaning or implication in what I wrote.

Cheers...

[BugFix]

After taking a few minutes to re-think about it, hopefully you can clearly see what I suggested. maybe I should have done that right away. In addition, our discussion has led me to gain a clearer picture of what I believe would be a great feature. I wasn't quite sure what I wanted yet, so thank you very much.

I'm not interested in turning XNview mp into a micro-Photoshop or an output studio, the added value lies rather in a new role between viewer and editor:

"Color-conscious cataloging and decision-making software!" Yeah!

Non-destructive soft proof simulation as a decision-making aid, closely integrated with the catalog function and without changing the original files and without editing requirements.

View, compare, and classify images before they go into an editor or print workflow is the benefit. Identify which images are suitable for a target profile (e.g., print/web) and which are problematic. Filter series/archives according to output suitability, not just metadata.

Display and filtering according to embedded ICC profile/color space.
Soft proof view as pure view simulation to identify clipping and color/hue shift.
(Multiple [3+]) comparison/split view: embedded profile vs. simulated target profile vs. different rendering intent (<- That's going too far again, isn't it?)

Virtual catalog flags such as “soft proof ok / out-of-gamut” or similar, filterable and sortable and may be for multiple color spaces, when proofed with a flag. No color correction, no print finalization instead, transparency, comparison, and better decisions in the catalog.

This would clearly differentiate XNviewMP within the “viewer class” and close a gap between pure viewing and image editing. Color management is generally weak in viewers. I.m.h.o. there is no viewer on the market with such gadgets... additionally XN has a speed advantage, in Editors this process is also not perfect and probably slower.

[jkm]

Ok, I think I see what you mean. Some thoughts inline below...

I'm not interested in turning XNview mp into a micro-Photoshop or an output studio, the added value lies rather in a new role between viewer and editor:

"Color-conscious cataloging and decision-making software!" Yeah!

Non-destructive soft proof simulation as a decision-making aid, closely integrated with the catalog function and without changing the original files and without editing requirements.

Yes, ok, I get it. I agree there could be some value here. And you're sure you're still ok with needing to go to external software for color correction and output? I'm thinking of Lightroom users who are accustomed to being able to do both functions in one place, so is this functionality, without the editing, enough of a benefit for them?

View, compare, and classify images before they go into an editor or print workflow is the benefit. Identify which images are suitable for a target profile (e.g., print/web) and which are problematic. Filter series/archives according to output suitability, not just metadata.

Display and filtering according to embedded ICC profile/color space.
Soft proof view as pure view simulation to identify clipping and color/hue shift.
(Multiple [3+]) comparison/split view: embedded profile vs. simulated target profile vs. different rendering intent (<- That's going too far again, isn't it?)

Yes, I think this is interesting and a good idea. I would point out that there is a certain amount of "editing" implicit in this, because soft-proofing inherently requires code for gamut conversion. So since the gamut conversion code would have to be done anyway, might as well allow the effects to be exported. And if we accept the premise that editing is not required (personally I am still unsure on that point) then no, you don't go to far.

You mention showing clipping. This is something that has been requested for years, and actually is on the roadmap now. But you're really talking about gamut conversion induced clipping, which is a good idea. I had previously been thinking about just visually showing clipping, which is the norm, and is consistent with that you're envisioning above. But what might be interesting is a search tool of sorts, that essentially says "show me the list of images that clip when this gamut or profile is applied". Sort of an automated "gamut check" if you follow me.

What nags at me about this the whole soft-proofing/gamut aspect is just the audience for it. It's good to implement things that will provide benefit to a lot of people, ergo things that cater to a niche audience tend to get deprioritized. Lots of people do photography, but nowadays the vast majority output to screen and that's it. So few people are actually printing seriously. People who output to screen often don't care about gamut. So that would really be my only concern: is it going to benefit enough people? For the people who need it, certainly it is beneficial, but are there enough of those people to move the needle? And of those people, how many do we lose because they think "yeah but if I can't print from the same app, what's the point?"

But yes I like the idea. The implementation could be some significant work though, and features like gamut conversion have to be done right, or they cause more harm than good.

Virtual catalog flags such as “soft proof ok / out-of-gamut” or similar, filterable and sortable and may be for multiple color spaces, when proofed with a flag. No color correction, no print finalization instead, transparency, comparison, and better decisions in the catalog.

I suspect the flagging could be done with existing functionality, couldn't it? Color labels (the names can be changed) would accomplish the flagging aspect.

This would clearly differentiate XNviewMP within the “viewer class” and close a gap between pure viewing and image editing. Color management is generally weak in viewers. I.m.h.o. there is no viewer on the market with such gadgets... additionally XN has a speed advantage, in Editors this process is also not perfect and probably slower.

You are unquestionably right in what you say above. In fact this was the notion behind some of the recent improvements. There's no other "viewer class" software that has the metadata capabilities that XnViewMP now has. The new catalog filters and comprehensive exiftool metadata support set it apart. So these sorts of color management and awareness capabilities would definitely be another step in that direction.

[BugFix]

As far as I know, Lightroom does not support the following:

Soft-proof views are always tied to editing states, not used as a pure viewer/catalog simulation.

No:
-> simultaneous comparison of multiple soft proofs:
– Relative Colorimetric vs. Perceptual
– With and without BPC
– Original + two proof variants (true 3-way split view)
-> color-managed comparison without an editing context.
-> catalog filtering based on soft-proof results.

In Lightroom, soft proofing is always a preliminary step to correction as an editing tool and not a decision-making step without editing. My concept is intentionally limited to visual analysis and catalog decisions and from my point of view this is completely sufficient (for my needs).

Lightroom shows color problems. XnView (with this approach) would make them comparable, filterable, and actionable at the catalog level.

One small addition regarding the “niche” concern: soft-proofing isn’t only about printing. Even users who never print constantly convert from wide-gamut spaces (ProPhoto/AdobeRGB) to sRGB for web or delivery, where the same gamut-clipping and color-shift issues apply. (To be honest, that's probably the most common use case.)

The lack of editing functions is also intentional and, in this case, probably more of an advantage: no sidecars, no status management, no risk of accidental changes, just quick analyses and decisions at catalog level.

By the way, you are right with your question/comment about editing after proofing or before exporting, printing, etc. This is the path toward the functions of editor programs and not the core of my approach.

[jkm]

One small addition regarding the “niche” concern: soft-proofing isn’t only about printing. Even users who never print constantly convert from wide-gamut spaces (ProPhoto/AdobeRGB) to sRGB for web or delivery, where the same gamut-clipping and color-shift issues apply. (To be honest, that's probably the most common use case.)

Yes of course, although I think a lot of people who aren’t printing also are not shooting in those color spaces to begin with; they’re shooting sRGB and sticking with it because they consider it safe and compatible, or don’t know enough to make a decision. You must realize of course that you’re a member of a small minority that even knows what color spaces and gamut are, much less uses them regularly.

What do you think of the “automated gamut check” idea?

[BugFix]

I actually think the automated gamut check fits the concept very well. If it is understood as a non-destructive, purely analytical step. It answers the question if this image cause problems for a given target color space.

Used this way, it could flag images that clip or shift significantly for a selected target profile so it would be possible to filter by output suitability and handles large archives without requiring any editing.

That would be one strong link between color management and the catalog. useful even for users who never print but export from wide-gamut to sRGB.

I agree 100% with your assumption that “a lot of people who aren’t printing also are not shooting in those color spaces” but XNview is far beyond the stage of simply existing for viewing purposes. Why else would you need extensive metadata management or histogram view?

[BugFix]

I thought again about the aspect of the “broad masses” of users, and yes, certainly not the vast majority of users will understand or use it, BUT:

It appeals to software collectors I, too, am one of those who likes to have “the best” software, even if I don't use everything. Who can claim to have fully understood Photoshop? I certainly can't, to be honest. I like to compare this to German drivers, especially those who can hardly see or react anymore, who have the fastest cars and then only wobble along at 120km/h on the highway.

But that doesn't detract from the reputation, popularity, desirability, or price of the cars. So I think a development in this direction doesn't hurt, as long as the user can do what he want with it. Perhaps this will also make it more attractive to pros, who would have to buy the software.

We could discuss this topic endlessly, but that would only be useful if one of the developers took a look at it and thought about it. Since it seems that only the two of us are discussing this here, it probably doesn't make much impact Maybe someone important will stumble across it. Nevertheless, I am pleased to have found an ally here who understands the approach. It's still fun to chew this over, thank you!

Catch you on the flip side & stay awesome