Comparing filmic color science v5/v6

The thing is that after all the tricks to prevent hue shifts the perceptual impression for many users is … hue shift. “stupid” tools, whilst causing other problems, hide or avoid this issue quite well. This effect, whatever it is, can be seen in quite a few other samples as well and seems to affect colours more subtly in less extreme photos.

2 Likes

The white parts are looking completely blue, why is that?

DSC09445.ARW.xmp (12.9 KB)

@flannelhead , please could you explain a bit more about using rec709 regarding the out-of-gamut colours, in particular, why not just use the gamut checking button?

1 Like

I have the same question :slight_smile:

questionable edit decisions on uncalibrated monitors probably.
(I cranked the colorfulness of near-achromatic colors too high, giving the windowsill too much of the “blue-sky spill” color)

how about this?


DSC09445-1.jpg.out.pp3 (15.1 KB)

the main point, CAM16 not showing those color shifts in the petals, remains.

Cheers, and thanks for asking!

3 Likes

Aaah, I see! That’s better in my opinion. There’s always a blue cast in those shadows but it was a bit extreme and had a strong transition in the other rendition.

It maybe along the lines of the issue with the display profile coming before the histogram profile. So at one time unless it was fixed …the only way to truly check gamut was to set your working and display profile to the same thing…in the past I would say you set both to rec2020 and then if you have not violated it the output mapping should handle it…but @flannelhead may have another reason… I will dig up the git link…

EDIT

1 Like

See my reply in this issue. The details are laid out there.

1 Like

Bit off topic but was that issue with the gamut and display profile resolved… is that what was corrected here ??

Nope, that PR contained mainly a refactoring and a small fix. The actual issue is still present.

1 Like

Too much desaturation could look like an artifacts,saturation is probably more important than hue for the finished image.
In this particular photo the hue shift is not visibile at all.

prophoto to srgb conversion with simple rgb clipping

Gamut clipping to srgb boundaries


DSC09445.pfi (37.6 KB)

I think that a tone mapper module should not doing gamut mapping too.

I would point to this image for further testing

My new tonemapping , the goal is to have a good overall fidelity to the hdr scene:
I’ve used gamut clipping to srgb but again it’s not necessary in this case
5DSR2210.pfi (55.8 KB)

Simple rgb (hue shifted to yellow)

Darktable filmic (pink hue shift, strange desaturation, banding artifacts…)

4 Likes

Thank you @flannelhead. Your response here and in the linked issue in github were very insightful for me.

But I still have some questions: It’s possible to avoid those out of gamut colours? What is the origin of the problem? In the camera itself that is maybe using a wider color space than sRGB? Or it’s recommended to use another input color profile? Using the Sony standard color profile delivered by Capture One I do not appreciate any hue shift.

Thanks. Tricky stuff!

The in camera settings only impact the out of camera jpg…nothing impact the raw from the camera wrt color space

1 Like

So your post just reinforces my comment…why can you call this a “filmic” edit…its your filmic edit with some interesting choices…taking your edit and doing nothing but resetting filmic…gives this…not the ghost that you provided… these sorts of comparisons are really not that useful in my opinion…
image

2 Likes

It’s just filmic with raised exposure
5DSR2210.cr2.xmp (7.0 KB)

Edit: the culprit is middle-gray luminance slider

1 Like

Still @age’s .pfi looks a lot better than the image in your post, I think.

What I’m showing here is a series of colour-picker values from the two areas shown on the image.
The first pair is without exposure and filmic; the 2nd is with exposure raised; the third with filmic v6 (reset to defaults):

As you can see, the H values were restored by filmic (rounding to the closest integer, 30 and 36 became 32, 59 after exposure, and 31, 35 after filmic), and saturation was dropped.

LCh tells a different story, though:


The darker tone started from h=70 and went to 73 after exposure, then back to 71 after filmic. The brighter one, however, went from 77 to 97; I guess that was clipping, showing the ‘rat-piss yellow’:
image

With filmic enabled, the darker tone went from 73 to 71, but the brighter one fell to 68 (using max RGB).
Setting chroma preservation to no puts it at 66, luminance Y to 62, power norm to 67, Eucledian norm to 62, legacy Eucledian to 70:
image
Adjusting the white relative exposure to retain the highlights:
image

Maybe this (all but legacy Eucledian norm shifting the hue) would be worth reporting as an issue.

1 Like

5DSR2210.cr2.xmp (7.0 KB)
I think this is the best I could do with filmic only (and I really mean filmic only, I did not even adjust exposure), without going obscenely out of gamut. I used v5, for better results in this scenario.

1 Like