The relevance, or not, of Soft Proof options/sliders to different “types” of ICC Profiles is confusing.
Intent and Simulate Paper & Ink options are not relevant when the ICC Profile selected is for a display / non-matrix profile … However, the Intent option is still enabled (but, Paper & Ink is not) !!
On the other hand;
When the selected ICC Profile is for a printer, the Preserve Color Details (PCD) slider is disabled - and its intensity setting is set to default = 50 (which, perhaps ???, implies that its related algorithm IS still applied, but at fixed strength of 50). This is ambiguous.
Here are some examples of where this is leading to confusion for users;
I suggest/request a change to the user interface for the Soft Proofing tool … along the lines of the clear differentiation that’s provided by the Smart Lighting tool.
That is, when the selected ICC Profile identifies a display / matrix-based profile;
the Intent and Paper & Ink options should be completely hidden (or both disabled) … making it apparent to users that they have no relevance in this context.
Conversely, when the selected ICC Profile identifies a printer profile;
the Intent and Paper & Ink options should be enabled … and either;
– the Preserve Color Details slider should be completely hidden
– its default value should be set at ZERO (to be clear that its related algorithm is NOT applied).
Similarly, for the Export to Disk user-interface; when the selected ICC Profile identifies a printer profile;
the Preserve Color Details check-box should be disabled (as it is for the option “Same as Soft Proofing ”) … making it apparent to users that it has no relevance in this context.
Edit: I’ve just realised - This is already the case (for the Export to Disk UI) …
Note: Improvements have been made to this request according to comments/feedback by @Egregius … See below.
Preserve color detail: this slider applies only to matrix based ICC profiles, so only display profiles – not printer profiles. Set at 50 by default, the slider lets you alter saturation and details in saturated portions. By reducing the value (to the left), 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 Preserve color detail checkbox which gives you the option to apply the slider’s effect to the exported images.
The help text in PhotoLab’s Soft Proofing palette isn’t as clear.
Thanks, Greg … In which case, there are some ambiguities that I’d like to see resolved.
That answers my basic question ! (I didn’t think to consult the manual )
Therefore, when the selected ICC Profile identifies a printer profile; the Preserve Color Details slider should be completely hidden … making it apparent to users that it has no relevance in this context.
Therefore, when the selected ICC Profile identifies a printer profile; the Preserve Color Details check-box should be disabled (as it is for the option “Same as Soft Proofing”) … making it apparent to users that it has no relevance in this context.
About Perceptual vs. Relative when a display ICC profile is selected: there are typo errors in the user guide and while most of what’s written about this implies it’s for printing only, some posts in this forum indicate there are indeed differences in display profile exports, too - and that this might be buggy behavior.
I believe it’s still visible but disabled - which I prefer to completely hidden, as that will have some people wondering where it is. grayed-out is standard indication that a feature exists but is not usable under present conditions.
My understanding (from an explanation by a member of DxO technical staff, provided during beta testing stage) is that the concept of Intent (Perceptual vs. Relative) is handled as part of the PCD-algorithm (as a unique DxO feature)… Which, as you point out above, is relevant ONLY for display / matrix-based profiles (not printer profiles).
Yes, I did think about that behaviour too, Greg … I agree your concern about options that seem to “go missing” - but, it’s the related behaviour of auto-resetting the slider to 50 that I find confusing ('cos it implies some relevance) … I figured this would be resolved by hiding the slider altogether.
I would like to hear from DxO why Intent isn’t grayed out sometimes. I don’t recall seeing any documentation saying it isn’t relevant for matrix ICC profiles. I do recall some forum comments to that effect, but am not sure that info proved accurate. Could there be an intent to make this function more useful than it is now?
Zero intensity is a misleading value. I remember it being demonstrated that a value of zero does not mean no algorithm is applied. DxO’s revised documentation for PL6.3 doesn’t make it any more clear.
I want to vote for greater clarity, but am thinking your proposals aren’t the way to go.
Did you mean to say; I don’t recall seeing any documentation saying it isn’t relevant for NON-matrix ICC profiles.
From your reference to the user guide (above), I thought we had this question clarified … (???)
Yeah - that was me !
For others wishing to check this out … (You’ll need a test-image with saturated colours - esp. reds)
Enable SP with PCD=ZERO - Observe the image and histogram … and hold that thought
Now disable SP … and the image & histogram both change !
That is; SP enabled with PCD=ZERO is NOT the same as SP disabled … which is inconsistent with all other PL tools … Eg. The result of Color Rendering activated with Rendering Intensity=Zero IS the same as with Color Rendering disabled.
I’m not sure what to think about this either (?) … Like you, I’d like to understand whether it’s on-purpose or just a UI glitch (I’ve had it suggested to me that it’s the latter).
I’d be happy to support any better suggestions, Greg (not locked-in to my own) - - but I certainly do reckon this needs urgent attention … It’s currently too confusing - esp. for such an important tool.
I agree John that it could be made clearer as to which profiles PCD applies. I now understand that it only applies to display profiles, however despite trying a number of display profiles I have yet to find one where PCD is active. All the profiles I have have an ICM extension. Does that make a difference? Also the Paper/Ink simulation option has a white checkbox beside it even when a display profile is selected but it is not selectable. The PCD option has no such box, Is this correct? See image. Thanks to you & Greg for your help & comments.
Unchecked (default): no preservation of color detail. → is correct
Checked box and softproofing profile: saturated color protection is applied, with the default value (50) applied to the color protection slider in the softproofing subpalette. → would becorrect *)
Checked box and softproofing profile: Checked box and ICC profile: Same as Soft Proofing: the ICC export profile is the same as the ICC profile selected in the Soft Proofing subpalette and color detail is preserved according to the intensity slider setting. → is correct
I have checked this intensively.*) Regardless of
Rendering intent set to perceptual / relative colorimetric
and / or
Preserve color detail set to 0, 50 or 100 …
with the Export / checked box
always a default value of 50 is applied,
which – if intended to simplify matters – I think is fine,
as otherwise one can use Export with Same as Soft proofing
and then apply whatever looks appropriate for the pic in question.
The user interface is improved, but still confusing.
As it already has been said, the choice of Rendering Intent is only applicable for printer profiles (see also the ?-help)
No. What I’m referring to is the belief that relative intent isn’t relevant for display profiles. Users have noticed it does things - and there’s concern that it behaves wrongly. If it isn’t relevant for display profiles, it should do nothing, mirror the behavior of perceptual intent (which is what DxO claimed at one point early on), or be grayed out.
X-Rite seem to say that profiles can be either matrix or LUT
Matrix or LUT based Profiles
Should I create a profile using Matrix or LUT based setting?Profiles themselves can be matrix-based or LUT (lookup table) based, both of which include the white point of the device. But matrix-based profiles are very small, while LUT profiles are much larger and more complex.
A matrix profile is a mathematical model made up of the three primary colorants of the device and some simple tonal curves, referred to as a 3 x 3 matrix.
A LUT-based profile contains much more information, consisting of a table of numbers that allows you to find an input value and its corresponding output value.
Matrix based profiles are used for simpler devices, such as scanners and monitors, while LUT-based profiles are used for more complex devices, such as printers. There are many different color spaces available, and most of them are matrix based. The advantages of a matrix-based profile are conversion speed and the ability to convert back and forth.
While there isn’t a right way or wrong way to select this setting, for general-purpose work, X-Rite recommends creating a matrix-based ICC display profile. For greater color precision, you may wish to select the table-based option.
I’ve not seen any assertion that Intent (or Mode - as it’s referred to in the user-guide) has any effect when a matrix-based, display (non-printer) profile is active. I have not been able to observe any effect .
And, altho it does not explicitly say that Intent/Mode applies only to printer profiles, the user guide refers only to “paper” and “printing” in its explanation of this setting (as quoted above by Wolfgang);
So, personally (FWIW !), I’m satisfied that the Intent/Mode setting has no relevance for display profiles.
On the other hand, if your point (that perhaps I didn’t properly appreciate ?) is around the question of whether the concept of “relative intent” (and “perceptual intent” too, for that matter) is relevant for display profiles - then that’s a (subtly) different question … to which the answer is Yes, it does / they do.
However, the reason these are not relevant for matrix-based, display profiles in DxO’s implementation of Soft Proofing is that the concept of “intent” is included in DxO’s proprietary Preserve Color Detail (PCD) algorithm … as a DxO staff member explained during the beta-testing stage.
In which case;
Yes - that’s what I have proposed in this suggestion/request;
Is that clearer, Greg - - or have I missed some additional nuance ?