Sorry, Wolfgang - but I’m still not understanding you (which seems like an ongoing “theme”, for us )
Are you suggesting that it’s a bug whereby the Protect Saturated Colors Algorithm IS applied (with default strength of 50) when the Export PSC setting is selected - even if Soft Proofing is OFF ?
If that’s it - then I disagree … I reckon that’s correct and an intentional feature.
But, if that’s not it … then please elaborate for my slow uptake !
I’m not sure if I can give you the best insight, John.
I assume shapes outlines for the Retouch tool means the target and source masks. If so, for me there is no noticeable improvement in PL6.1. It is still difficult to see the masks on many multi-colored / multi-textured backgrounds. If they could only make the masks outlines a bit thicker or use walking ants. If there is an improvement it is a fairly subtle one.
I am also not sure what number 3 means regarding tool shapes being kept. Perhaps it would be more obvious if I could do a direct comparison with PL 6.1 and the previous version of PL 6.
Please make sure to deinstall 6.0 before the upgrade. Maybe my problem was caused because I didn’t do so. If 6.1 introduces problems you can deinstall it and reinstall 6.0, this is what I did.
For now I personally will wait 2 weeks before I try 6.1 agaian since I am also pretty happy with 6.0.
DxO said in the PL6 release notes that behavior of the slider >100 is undefined. In many cases, raising it above 100 was exactly the same as setting it to 100. But in one case (Generic / neutral color with the DxO Wide Gamut color space, as I recall), I’d discovered that raising it to 200 was beneficial for an image with ridiculously saturated reds, as this recovered a lot of lost detail and color. Now another method will be needed for that particular image.
If you rename version 6 to, say “DxO PhotoLab 6.01”, you could possibly use both (one at a time) or keep the “old” one as fallback, should you not like the current release… Note: This works on Mac, Win might be against such a thing…
You have to go back to post #18, where I tested this, as well to post #20.
Assuming, that the new online user guide is not faulty, but tells us what DxO intended …(!),
.
we no more deal with a non-customizable PSC algorithm and a given result,
but a PSC slider in the softproof section, controlling oversaturation from 0 (no effect) to 100 (max).
. The softproof’s task is to show, what we are going to expect … to then of course to get it!
That procedure fully works when exporting with “Same as Soft Proofing”,
but it doesn’t with choosing a matrix-based profile (+ checkbox),
which leads to a non-existing softproof – except when the PSC slider is set to 50 (default).
[ Possibly the default’s result is congruent with the former PSC algorithm one, but that’s not the point. ]
.
DxO introduced a PSC slider to control the oversaturation while in softproof – and that’s broken.
'Tis curious that “Red” is noted as “for paper” ?!
Red is used to highlight OoG colors in the destination gamut … but, that’s not limited to printing/paper only (or is it?).
Yes, I’ve seen that and it is the ‘same’ statement as in the ?-help.
“Rendering intent” and “Simulate paper & ink” are intended for printing,
while in fact the Rendering intents apply to all imported profiles
(no use of PSC slider = non variable).
But the Destination gamut warning (red overlay) also works for matrix-based profiles.
With Soft Proofing activated and its PSC slider at something other than 50 … when Exporting-to-disk using a Matrix-style ICC Profile with its PSC checkbox selected (which applies a default PSCA strength of 50) … the exported result is not WYS-is-WYG (because the applied PSCA strengths are different).
I agree that outcome is a “gotcha” for an unsuspecting user … but I don’t agree that it’s “broken”
… that’s just how it works when Exporting is not processed the “Same as Soft Proofing”
At the moment – this export option always applies a ‘fixed’ PSC value to the outcome
. which only correlates with the (SP) PSC slider default setting (50), but all other (SP) PSC slider settings are OFF / misleading / unusable / ‘broken’.
to ‘reiterate’
You have to go back to post #18, where I tested this, as well to post #20.
Assuming, that the new online user guide is not faulty, but tells us what DxO intended …(!),
.
we no more deal with a non-customizable PSC algorithm and a given result,
but a PSC slider in the softproof section, controlling oversaturation from 0 (no effect) to 100 (max).
. The softproof’s task is to show, what we are going to expect … to then of course to get it! … …
otherwise follow your new thread / my reply in (post #25)
AH Thanks!
you bring me to the point i understand there idea.
Still i would like a more MagicWant kind of approach and a better understandable text:
“Protect Saturated Colors when export”
And 50% “strenght”?
50% drawn in by there dxo perceptual kind of compressing and 50% clipped of the out of gamut/colorspace colors?
i find this rather confusing.
how do i know it’s too much or too less?
Does it “turn off” when i adjust the image so the red mask,which tells me there are colors outside my export iccprofile, goes away?
It is a half baked “auto softproof modes with visible masking out of gamut preview” or not ?
edit: the magic want kind of PSC is only show a certain strenght when some of the colors are oversaturated and would be clipped when export in same colorspace as the working colorspace. and it shows how much/many colors are oversaturated.
if they build this also in the Soft proofing you can see in preview how much of the colors are out of gamut and thus compressed. 0% none, 10% some, 50% quite a few, 100% geesh too much to handle correct. which helps you to decide if some re-editing is needed on the file in order to contain the colors better inside the desired export colorspace.
The first three sentences describe, what the PSC slider in the SP subpalette does.
The default setting is “50” (maybe congruent with the former PSC algorithm … ?).
At the left “0” the slider maintains saturation at the cost of texture and vice versa at the right “100”.
Then the fourth sentence should clair it up:
“Note that the Export menu has a checkbox for Protection of Saturated Colors,
which lets you apply (or not) the slider effect to the exported images.”
No. You have to use your eyes to discern what is enough, not too much. The out-of-gamut warning indicators don’t change as you adjust the protect saturated colors slider in Soft Proofing. And I find they don’t reliably indicate what is truly out-of-gamut anyway.
Fortunately, this only pertains to matrix based ICC profiles (display profiles) and the automatic setting (a checkbox in the export options, equivalent to 50 on the Soft Proofing slider) is supposed to be sufficient to protect saturated colors:
Saturated Color protection: this slider only applies to
matrix based ICC profiles and, so only display profiles,
not printer profiles.
By reducing the value, the slider maintains saturation
to the detriment of the details and, To the right, the details
are preserved to the detriment of the vivacity of the colors.
Note that the Export menu has a checkbox for Protection
of Saturated Colors, which lets you apply (or not) the slider
effect to the exported images.
Reading the above, I get the impression that “Protect Saturated Colors” is a poor choice of words. Pushing the slider to higher values, reduces saturation in favor of detail, so the tool could be called “Preserve Detail in Saturated Colors” or “Prefer Detail to Saturation” …instead.
The middle position is the compromise between saturation and detail.
I don’t read it that way. If DxO’s original statement (what I linked to, which hasn’t changed with PL6.1) about what the protection algorithm automatically does (the equivalent of a setting of 50 in Soft Proofing) is true, then going above 50 drops saturation below the max saturation of the target color profile. It might also increase local contrast but won’t add new detail. My own quick test with a magenta bag suggests this is the case. Maybe someone else can run a test that proves otherwise, but it is reasonable to me to sometimes want to “overprotect” saturated colors to reduce clipping on any devices or media that would tend to do that (calibrated or not).
I suggest you have identified a mistake in the documentation - rather than a bug in implementation.
If it truly should behave as you have noted (with the checkbox for Protection of Saturated Colors determining whether to apply the PSCA according to the “slider effect”) - then there’d be no reason for the “Same as Soft Proofing” option … it would just be a duplication of functionality.
I reckon we’ll see that documentation (eventually !) corrected.
)
