My personal color space nightmare...

Interesting to learn that someone experimented with it :laughing: I think my own days of doing so are over, except perhaps images dominant in cyan and aqua. It comes at the expense of colour accuracy. Still, the colour gradations are pleasing in your edit.

What does that rendition show exactly, and how can I produce it?
I have installed the DNG converter, but I don’t think it’s just that.

If one follows the DNG spec for raw conversion there are several checkpoints during the process where one needs to block negative values to zero. At each of such points I add the relative pixels to an out-of-gamut RGB image. Values are set to either 255 or zero in the appropriate color plane. I usually carry clipped pixels over then just after projection to the destination color space, but before any further processing, I add in those as well. It’s usually obvious by comparison to the rendered image which are clipped vs blocked.

The resulting image above doesn’t show you by how much values are blocked/clipped, just that they are. I find it is useful to quickly see how much clipping/blocking is due to somewhat objective ‘gamut’ issues vs subjective ones by further processing, as in this discussion. Not many options for the operator with the former, many more for the latter.

Assuming the DNG spec process, the checkpoints I use are basically in linear space just after matrix multiplication:

  1. conversion to XYZ
  2. conversion to ProPhoto (if LUTs are in the dcp)
  3. conversion to destination color space.

These are the stats I get using the DNG tags for the capture in this thread with ‘basic’ table:

VioletAreBlueaRGB_out

Ant these are for the same with the ‘look’ table applied:

VioletAreBlueaRGB_out_look

I don’t know what part of this functionality is available in raw converters discussed around here.

Jack

3 Likes

My take. PhotoFlow WB, HL blend, RCD, lens corr, OCIO filmic.
gmic hotpixel, afre_contrastfft, resize, afre_sharpenfft.

3 Likes

I was not sure what to do with the white balance.
I ended up picking the colors I liked more, probably not the real ones.
I used darktable 3.4.

PB130017.ORF.xmp (16.0 KB)

1 Like

Filmulator renders the colors extremely saturated (especially the blue) but I find it doesn’t look bad in this particular situation; it doesn’t clip colors harshly and doesn’t artifact at the bright spots on the magenta wall. That said, it’s unlikely it looked that blue in real life.

Only three changes from default:

  • Highlight Recovery 2
  • Exposure Compensation -1.33
  • White Clipping Point 0.604
2 Likes


PB130017-1.jpg.out.pp3 (14.2 KB)

3 Likes

That sounds smart.
I’m sure I haven’t fully understood everything, as I’m only just getting into colour management, but I understand that as something like the “Raw overexposed indication” but for the conversion into other color spaces, right?

And all additional blocking/clipping is then user-made by processing steps…

And that shows that I could do a better job even in sRGB space. Also with a monitor that is limited to srgb…

Well, it comes very close! The colours are indeed very saturated, especially the blue.
However, the notes on the door are black and white :wink:

Simplifying somewhat, the intuition is that once we are in a colorimetric color space like XYZ (or even sRGB) we can look at the span of possible tones in the space as being contained in a cube with the origin at the minimum possible value (zero, blocked black) and the opposite apex at the maximum possible value (clipped white, normalized to 1). Negative and clipped values cannot be rendered, so they are swept under the carpet.

To get to the final color space you need to project from cube to cube via matrix multiplication (WB, camera->XYZ->ProPhoto->XYZ->xRGB). Every projection can and often does result in some negative and clipped values that need to be discarded. There is very little one can do about this if one wants to keep things as linear as possible, and we often do (right Glenn?). What little one can do is often quite a subjective fudge. So I like to know what tones those are.

… which is however under the direct control of the user so easier to subjectively guide into final ‘gamut’. For instance some form of contrast, which is pretty well needed in every single image in order to better squeeze its tones into the smaller contrast ratio of the typical display medium today. If not done properly (no way is perfect nor necessarily more or less desirable) it will substantially shift chromaticity and increase saturation, with consequences for the final tone range.

2 Likes