Request/Suggestion: Fix to avoid being caught-out if NOT using Soft Proofing (with new Working Color Space)

[John-M (Melbourne, Australia - Win10)]

@Guenterm

When we talk about monitors, we also talk about the displayability of colours or colour spaces such as sRGB, AdobeRGB…
If I calibrate an sRGB monitor to sRGB, I work in sRGB and if I output a file, again in sRGB, I can assume that another user with a calibrated sRGB monitor will see the image in true colour.
So far, there is no mention of soft proofing.

Spot-on ! … All good.

@Guenterm

If I want to print a photo (myself or from a service provider) I have to use softproof to take the printer and paper specifics into account … etc … etc … eyc

Yes - no argument - Except that, this post has absolutely nothing to do with printing.

@Guenterm

I hope I have understood everything correctly up to this point, and I would like to address again that I can do this in LR and AP without endless discussions.

It appears to me that you have missed the point - it must be due to my poor communication.

I’m really sorry if this is difficult to grasp - - but I have tried really hard to explain it … I accept responsibility for not having done it well enough - or with too much detail … or something ???

@Guenterm

So now I’m getting out of this too and I’m confused.

I’m really tempted to give up too, Guenterm … I assure you; I’m not doing this for fun !

I could easily be selfish, because I (and others I’m in-touch with on this issue) do understand when & why it occurs - and we have some simple work-arounds to mitigate it. However, I’m aiming for a solution that simply works - for all those users who will have no inkling of the implications.

John

[jeffholdgate]

On my Mac, SP on or off makes no difference. I exported 2 sRGB. files on with SP on, and one with SP off. Put them in PS as layers and set the blend mode to difference. There is no difference. So it seems that this additional algorithm is not deployed in my setup. Is it different for the Windows PL?

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

@Joanna

Now - all of a sudden - PL6 dictates that I have to complicate my workflow and virtually insists that I do soft proofing, even if all I want to do is export to an sRGB JPG file.

I have been doing a lot of testing and communicating with @John-M in a private thread and believe I have come up with a solution that will work for me as well as you.

I have found that using Colour Rendering set to Generic renderings with Neutral color, realistic tonality (gamma 2.2) and SP=Off provides me with the same result as PL5 except for more vibrant colours due to the wider working colour space, which is actually very pleasing to my eye.

I saw on one of you many posts that you also use Generic renderings with Neutral color, realistic tonality (gamma 2.2) so I think you may find this to work for you. The colours will be slightly different too PL5 but that I believe is due to the wife gamut working colour space and is an improvement to my eyes.

I used to use DxO Camera profile in my Colour Rendering settings which game me horrible colours in PL6. Going to Neutral colour fixed all my confusion.

As far as soft proofing goes, you don’t need to do it unless you really want to see a “simulation” of your output. If you do want to do soft proofing then DO NOT try and bring all the out of gamut colours back into gamut as that will ruin your photos as you have demonstrated with your flying lobsters! Simply tweak your photo if it does not look the best on the SP colour space.

I have tested this on both a P3 and sRGB monitor and I get consistent output which is very similar to the same setup in PL5. If you really want to you could always force Adobe RGB as you WCS to be even closer to how PL5 works, but I think you would be missing out on the advantages of editing in a wide gamut colour space before exporting (and converting) to a smaller gamut colour space.

Give it a go and let me know how it works for you

[John-M (Melbourne, Australia - Win10)]

OK - - I’ll try to explain this once again - keeping it as simple and succinct as possible.

I do appreciate all those who have engaged in this topic - If you genuinely want to understand what I’m trying to get-across, please read to the end & consider each point without jumping to your own conclusion.

Note: Starting assumption is that we have Working Color Space (WCS) = DxO Wide Gamut (WG)

1. This issue has absolutely NOTHING to do with printing (even tho it is related to Soft Proofing !!)
   – Please just trust me on this, and read on.

2. When I was testing PLv6, I found, for a specific image, that the way it looked within PL on my sRGB monitor was not exactly how it looked, after exporting it via the sRGB ICC-Profile, when viewed on that very same sRGB monitor !! !!
   – this was something I had not experienced with PLv5
   – like most of you, I did not believe I needed to be Soft Proofing in this situation, so I had SP=OFF

3. After more tests, with more images, I found that this phenomenon occurred particularly for images with saturated colours … It still did not occur to me that this was related to SP - Why would it ?!

4. I pestered DxO about it and received the following explanation (from a very reliable source);
   – In the Export to Disk process, PL asks for a target ICC-Profile (Default = sRGB)
   – Now, when exporting to the digital file, PL applies an additional algorithm during that process.
   – There are no settings for this algorithm … (I’m calling it DxO’s “secret-sauce refinement”).
   – It’s purpose is to Protect Saturated Colors (PSC) in the conversion from WG, and the “strength” of the algorithm is specific to the target ICC-Profile

5. OK, so what ?!
   – This algorithm is applied during the Export to Disk process (as described just above)
   – This algorithm is applied when SP=ON
   – However, this algorithm is NOT applied if SP=OFF … Aha !!! … Now do you get it ???

6. My reaction, to DxO, was to ask/say; Why do you not apply the algorithm when SP=OFF ?
   – and the answer was; “Because, with SP=OFF, we don’t know the color space of your target file”

7. To which my next response was;
   – For users intending to convert to a different color space (from the one used by their current monitor) - then they ought to understand the need to do Soft Proofing in that case … These are sophisticated PL users; let then deal with that themselves.
   – For everyone else, who will have absolutely no inkling that SP is needed when simply working on Monitor-A and exporting an image for viewing on Monitor-A (or one “similar” to it) … there should be default behaviour by PLv6 that “just works” … for consistent WYSIWYG behaviour.

8. Recap;
   – In order for WYS (within PL) to = WYG (when the result is viewed on the same or “similar” monitor) in all cases - including when SP=OFF (which is the setting that typical users will have) … DxO’s “secret-sauce refinement” algorithm needs to be applied in all cases; not only if SP=ON.
   – However, currently, PLv6 is NOT applying this algorithm if SP=OFF because; an output ICC Profile is not provided in this case … BUT, I contend, it’s not unreasonable, in this case, to assume that the current monitor is the target.

9. See here for the (refined by “crowd-sourcing”) solution. In summary;

• Re-instate setting in Preferences to specify the ICC Profile of the monitor used by PhotoLab
  – OR, have PL deduce this itself … which, I am advised, is technically do-able.

• If SP=ON then no change to current PLv6 behaviour is required - because, the algorithm is applied.

• If SP=OFF then it’s reasonable to assume that the user simply expects the WYSIWYG-behaviour that we’ve always enjoyed pre-PLv6
  – So, in this case, apply DxO’s “secret-sauce refinement” algorithm for the current monitor.
  – Therefore, regardless of whether SP=ON or OFF … we will always enjoy WYSIWYG-behaviour.

---

This will be my last attempt at an explanation … I truly don’t wish to sound arrogant, or annoyed; I’m just doing my best … but, henceforth, I will respond only to questions related to the above …
Ask away !

John

[Guenterm (Win11, MAC OS, DPL 5, FP, VP, Nik4, Olympus)]

@John-M

Yes - no argument - Except that, this post has absolutely nothing to do with printing.

Hello John,

I have now read through everything again and your last post from point 5 onwards has given me a little more clarity about what the original post was supposed to say. Please excuse me if I have watered down the topic a little.
However, I now understand even less the procedure and the implementation of the functions (incl. softproof ) by DXO.
I will continue to test the trial, but will use my version 5 productively until I feel I have a largely bug-free and logical version 6.
Who can say how many surprises are still hidden.
Thanks again for your patience and no-nonsense approach to clearing up all my confusions.

best regards

Guenter

[Joanna (Joanna Carter)]

@KeithRJ

I saw on one of you many posts that you also use Generic renderings with Neutral color, realistic tonality (gamma 2.2) so I think you may find this to work for you

Thank you Keith. That is essentially where I want to end up most of the time.

Your message prompted me to do a series of tests.

Let’s start by editing the lobsters in PL5.

This is the image, totally untouched, apart from default optical corrections…

I have activated the highlight and shadow warnings but, as the image stands, no warnings are produced.

However, if I apply around 50 to the fine contrast slider, we start to get small blown highlights and even smaller blocked shadows…

In PL5, what I would do to eliminate these warnings is to slightly reduce the top and and raise the bottom levels on the tone curve…

For me, that would be a case of “job done” ready for export or printing.

---

Now we come to PL6.

Deleting the DOP file and opening with the same default settings, but with the WG selected, the same image gives…

An overall brighter and slightly more saturated appearance.

This time, even with no fine contrast added, the highlight and shadow warnings are present…

Adding 50 of fine contrast makes them even more noticeable…

However, doing the same tweaks to the end of the tone curve is just as effective as in PL5, completely removing the warnings…

---

However, now we have OOG warnings available so, let’s go on to activate them for a screen that has been profiled for sRGB…

Normally, I wouldn’t have bothered but, since PL6 now makes me aware of these, I suppose I had better do something about them.

Getting rid of the warning on the red meant adjusting the luminance on the red spot of the colour wheel…

Getting rid of the warnings for green was slightly more difficult as it required both reducing the saturation and increasing the luminance to get a rendering that was still sufficiently saturated…

What isn’t apparent is the OOG blues, which I had to zoom in to see. The warnings being blue made this particularly difficult. I would rather see something like the grey that Adobe use.

Once again, correcting these involved both saturation and luminance adjustments, which is quite fiddly…

---

Finally, I gently dropped the tone curve to give me a deeper overall coloration…

---

So, in the end, a rendering that might be deemed “acceptable”, although still not as satisfactory as that obtained in PL5 with very little effort, but with a whole load more effort.

Well, at least, I don’t have to bother with soft proofing for printing. I shall stick to exporting to ProPhoto RGB TIFF files and using the macOS print dialog for that.

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

@Joanna, what if you ignored OOG warnings and just did an export, is the output the same as with PL5?

I don’t think you should be trying to remove the OOG colours as the export will adjust them to the output colourspace. The OOG warning is there for information purposes and to let you know that some changes will be made to those colours when you export so you may see some differences in the output when compared to the screen.

In PL5 the WCS is smaller than in PL6 so the colours you work with in PL6 may be larger than in PL5 but the export should be the same for the same target profile.

Does that make sense?

[John-M (Melbourne, Australia - Win10)]

Hi Joanna.

I trust you’ve had a chance to work-thru my “simple & succinct” explanation above (?)
– I hope it helped.

Would you please point me at a download for your Lobster example … I’m sure you’ve provided one, but I cannot find it again.

John

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

@Joanna here is my edit of your lobsters. I reduced the exposure slightly and removed the clipping in the same way as you described above, then exported using sRGB profile. No soft proofing or anything other than the the changes mentioned above and the default lens corrections.

Does this match your expectations?

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

Now here is something interesting:

I loaded the above jpg into PL6 and it shows Monitor Gamut warnings in the blue areas! My monitor is P3 but this is a sRGB image so why would there be monitor gamut warnings?

Is PL6 exporting the full gamut of the photo but tagging with the sRGB profile so that other colour managed software can display the image after applying their own method of displaying OOG colours for a sRGB image.

So the question now is:

When exporting a large gamut photo to a sRGB colourspace, does PL6 export the full gamut but embed the sRGB profile to tell other software that they should convert the image to sRGB when displaying.

OR

Is the output fully converted to the sRGB colourspace?

I am sure I have seen mentioned elsewhere that even though an image is tagged with an ICC profile, it can be removed and reverted back to the full original colour space and then tagged with a different profile (maybe I am smoking something or does this make sense?)

This is all getting too hard and we really need full clarification form DxO about what is happening each and every step along the way from loading the RAW to exporting with a profile.

[Joanna (Joanna Carter)]

@John-M

I trust you’ve had a chance to work-thru my “simple & succinct” explanation above (?)
– I hope it helped.

Hmmm. Please don’t take this the wrong way, but I wanted to avoid explicitly using soft proofing, to see what the effect was.

I did also try with soft proofing, but that seems to make life even more difficult.

I’m also still not sure which gamut warnings to switch on or off below the histogram and for what purpose.

From what I can gather, the right-hand one is meant to be for a “target” gamut and, certainly, on my P3 screen, setting that to sRGB throws up all sorts of warnings that you end up with a horribly desaturated image trying to get rid of them…

It’s still too confusing because there seems to be more than one way of skinning this particular cat. PL5 was so simple and yet got me the results I always anticipated, both for posting images here and for high quality large format printing.

@John-M

Would you please point me at a download for your Lobster example … I’m sure you’ve provided one, but I cannot find it again.

I can’t find it either and am having problems posting anything but JPG files to the forums. I have sent you a DM asking for your email.

@KeithRJ

Does this match your expectations?

Sort of

@KeithRJ

This is all getting too hard and we really need full clarification form DxO about what is happening each and every step along the way from loading the RAW to exporting with a profile.

I couldn’t agree more. Which is why I am only treating PL6 as an ongoing beta product and continuing to do me real work in PL5.

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

@Joanna

I’m also still not sure which gamut warnings to switch on or off below the histogram and for what purpose.

Only use the Highlights and Shadow warnings. The Monitor warning shows what colours in the file is OOG for your monitor and the SP warning shows which colours are OOG for the SP profile selected and is only available when SP=On.

I find these new warnings are informational only because when you export to the selected profile PL6 will make some changes to move these OOG colours into the selected colour space. This is exactly what PL5 does and you did not know about it because it was all done behind the scenes.

Don’t try and fix these Gamut warnings because PL6 will do a much better job for you by using their special soup recipe that John-M talks about

[Joanna (Joanna Carter)]

@KeithRJ

Don’t try and fix these Gamut warnings because PL6 will do a much better job for you by using their special soup recipe that John-M talks about

Which leads me to ask, why complicate life by providing them?

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

@Joanna

Which leads me to ask, why complicate life by providing them?

For those who want to use SP and are interested and for keeping up with the competition and because users demand SP and higher gamut workspaces

Now we are stuck trying to make sense of all this because DxO have left us all in the dark - they have not given us ANY information about what they are doing under the covers or how to use these features properly!

[John-M (Melbourne, Australia - Win10)]

@John-M

Would you please point me at a download for your Lobster example

OK - I have a copy of your RAW file - and the affect I describe above is very clearly obvious;

1. I applied DXO’s “No Corrections” preset … and I changed Color Rendering to Generic renderings with Neutral color, realistic tonality (gamma 2.2) … which, I understand, is your preference (Not that the Color Rendering makes any meaningful difference at all to this phenomenon).

2. I have Working Color Space = Wide-Gamut … and SP=OFF
   – here’s what I see (within PL) on my sRGB monitor
   

   
   
   Tip : Click on one image and then use arrows < > to quickly compare SP = OFF versus SP=ON…

3. I export to disk - with target as JPG with export ICC-Profile = sRGB~

4. I view the exported JPG on my sRGB monitor (using Irfan Viewer - not from within PL)

5. I go back to the preview within PL – and compare what I see (within PL) versus what I see via Irfan-Viewer — Oh, dear (Fake surprise) - they look quite different !!

6. I switch Soft Proofing = ON
   – here’s what I now see (within PL) on my sRGB monitor. Note esp. saturated “oranges” on RHS.
   Ah, that’s better - It’s now the SAME as what I see when viewing the exported JPG (via Irfan)
   

   

With SP=OFF - that’s NOT WYSIWYG - as we experienced it with PLv5.

---

@KeithRJ

I have been doing a lot of testing and communicating with @John-M in a private thread and believe I have come up with a solution that will work for me as well as {Joanna}.

There are work-arounds, such as the steps that Keith outlines - I’m doing something similar … but, I’m aiming for a solution that “just works” - for all the Average-Joe users out there who will have no inkling about this issue … and will expect PLv6 to exhibit WYSIWYG behaviour in all cases.

---

John

[Guenterm (Win11, MAC OS, DPL 5, FP, VP, Nik4, Olympus)]

And what have you all done in the EA Time

Only a joke

[JoJu (Joachim)]

@Guenterm

Only a joke

But an excellent one

@Joanna

Which leads me to ask, why complicate life by providing them?

To make your life more adventurous, I believe.

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

@John-M

I view the exported JPG on my sRGB monitor (using Irfan Viewer - not from within PL)

@John-M What are your Colour Management settings in IrfanView?

[KeithRJ (Keith - Perth, Australia - Win11, PL6, VP3, Canon, DJI)]

Here are results from my test with SP=Off and the following edits:

I reduced the exposure slightly and removed the clipping in the same way as Joanna described, then exported using sRGB profile. No soft proofing or anything other than the the changes mentioned above and the default lens corrections.

Preview with sRGB jpg on my P3 monitor:

Preview on my near sRGB monitor (slightly larger gamut than sRGB):

I don’t see a huge difference!

[IanS]

Hi John

I have voted as instructed

I have read the thread, but don’t understand completely how DXO have got to this stage of complexity. I use Capture One and in C1 you are always soft proofing because the raw image colours always have to be “fitted in” to something, sRGB, a paper profile etc.In C1 you choose the Proof Profile you want to use, and that’s it.

I include below an explanation of C1’s explanation in the hope an “alternative” expression of soft proofing might help.

Additionally, I would suggest that there is an alternative way, I use in C1, of applying the editing corrections to the file in order to correct for the target display, paper profile etc.that avoids creating a virtual copy.

In C1 I add a soft proof filled layer and name it, and then make the adjustments to contrast, colour etc on this layer. Then you can simply turn off the soft proof layer for normal edits and enable the layer before you print etc. This does assume you can make all your corrections with local edit tools. In C1 there is no difference between the global or local tools as in DXO, eg you have 1 exposure slider that works globally or locally.

Currently a filled layer is created rather clumsily by pushing a Control Line outside the image boundary but I would have thought it relatively simple to implement a “filled layer” option in the layer palette.

fingers crossed your suggestion is adopted

.