And another colour management mystery

DxO PhotoLab
42

Yes !

This now clearly differentiates between “protect saturated colours” (as it applies to Color Rendering) and “preserve colour detail” - which applies only to Soft Proofing and/or when Exporting to disk (and appropriate options are selected).

John M

43

Note: A couple of key points from DxO’s newly published White Paper.

For DxO PhotoLab 6, we’ve worked to ensure that all of the luminance details captured by the sensor are maintained throughout your workflow.

For the best possible quality, our reengineered algorithm is designed to act in two stages:

  • first when converting from sensor native color to working color,

  • and then when converting from working color to output color.

  • The first stage (Protect saturated colors in the Color Rendering palette) has been reworked and improved compared to PhotoLab 5,

  • the second stage (Protect color details’ in the Soft Proofing palette) is entirely new [to PLv6]

There’s also good, clear explanation for DxO choosing to go with their own version of “Wide Gamut” - rather than using the ProPhoto RGB color space. See section; “How we designed our new working color space”

John M

1 Like
44

…still, “psc” does not really say what it does. Saturated colours cannot be protected if they are out of gamut. If no sensor data is out of “DxO Wide” gamut, there should be no need to protect something imo.

Is DxO implying that sensor data can be lost, even with DxO Wide?

:exploding_head:

1 Like
45

Here’s my understanding, from experimentation & observation.

  • PSC seems to be applied for the current monitor profile

Whereas;

  • We know that PCD applies to the profile selected for soft-proofing.

For example, using my “very saturated reds” test image - with Monitor gamut warnings switched ON.
image … Showing saturated reds as being out-of-gamut for my monitor.

  • changing the PCD slider (on the Soft Proofing tool) has absolutely no impact on OoG warnings

  • whereas, changing the PSC slider (on Color Rendering tool) can have significant impact.
    image

So, in practical terms;

  1. Use the PSC slider to get your image looking as you want it to be, within the constraints of the monitor you’re using with PL.

  2. Use the PCD slider to tweak the result that you’ll get when you Export to Disk for the selected ICC Profile … using Destination OoG warnings as a guide (instead of visually) if your monitor is not capable of rendering the ICC Profile you’ve selected for Soft Proofing.

Simples ?! … Perhaps not !

John M

PS. I’m “thinking aloud” here, somewhat - so, please don’t hesitate to disagree.

46

It can be in the process from source to destination.

George

47

If you think end to end then yes.

If, as quoted above by @John-M to which I related my post, PSC is used in the step from sensor to working colour space, there should be no necessity for such a function, unless the working colour space is smaller than what comes from the sensors.

Assuming that PSC is necessary (or DxO would not have introduced it) we must therefore accept to either lose colour information or change it before we even start editing.

1 Like
48

This is the conclusion I came to too :slightly_smiling_face::+1:t2:

1 Like
49

I think there is a case here because a RAW file needs to be demosaiced and colour balanced and there might be a need/opportunity to PSC here. Just throwing ideas around!

50

I didn’t intend to give that limited impression …

There’s a process-flow diagram in the whitepaper that identifies the steps in DxO’s RAW conversion pipeline.

image
image972×221 50.1 KB

I take it that the PSC-Algorithm is applied at the stage labelled “Color Rendering” - - and the PCD-Algorithm is applied during “Conversion based on ICC Profile”.

I reckon that’s a fair point. There’ll be color information contained within DXO’s Wide-Gamut working color space that cannot be represented on our monitors (esp. in the case of sRGB monitors like mine and, probably, the majority of PL users) … So, we need to make a decision about how we’d like to have this color info “squeezed” into the gamut that our monitor is capable of rendering;

  • There’s a “magic wand” option to determine the degree of PSC adjustment to apply (So, I’ll probably tend to go just with what it auto-recommends).

And/Or

  • We can adjust the PCD slider (for Soft Proofing) to determine how much color-saturation versus fine detail we’d prefer to retain. [# note]

Here’s an (extreme) example of the latter - based on my test-image with very saturated reds;
image - PCD slider = 0 … Full saturation to detriment of detail (a la PLv5)

image - PCD slider = 100 … Allows us to balance saturation vs detail.

# note: This reflects my workflow (when my export target is for rendering an image to a display device) to have Soft Proofing activated at all times during my editing session.

John

51

… or simplify – and go the PL6 Classic-Legacy route, especially when on a sRGB screen

2 Likes