Hi, I tried using HALDCLUT on RT 5.4. :hile everything is installed and seems to be fine (HALDCLUTs appear in the Film simulation menu), it isn’t working. Nothing happens when I click on it.
Any ideas why this is happening?
Have you turned the film simulation tool on? There is a power button symbol in the header line of the toolbox right in front of “Film Simulation”.
HTH,
Flössie
3 Likes
Hi @George_D & welcome!
When you click on what?
Apart from Flössie’s advice (above) also check that the
value for Strength is much greater than 0.
Have fun!
Claes in Lund, Sweden
1 Like
That’s it! Thank you so much!!! 
1 Like
I’m so glad I came across this forum… you wouldn’t believe how long I’ve been trying to find this exact solution… many thanks
Just a quick question that I haven’t been able to find the answer to. Are the HALDCLUT film simulations in RawTherapee intended to used with the auto matched tone curve or a linear tone curve?
Thanks in advance…
I might be wrong, but as HaldCLUT was there first I would assume the linear tone curve. Alberto (@agriggio) knows the details.
I think the film simulation cluts are supposed to be applied over a linear curve.
I would say it depends on the specific HaldCLUT you use. Some are quite contrasty and are best used with a linear tone curve, some others don’t change the contrast that much and can be used with a tone curve (auto-matched or custom).
Thanks @sguyada I think what you say matches my findings, it depends…
hi,
what @sguyader writes is correct, it definitely depends in general, since haldcluts allow for very general manipulations. but if you ask about the “film simulation” cluts, for sure they are unaware of the auto matching, and, if you want the look they are supposed to emulate, I think you should start from linear. but nothing prevents you to use other tone curves if you like the end result…
1 Like
Well said.
Clearly put, thanks for your help
Good question.
The “Auto-Matched Tone Curve” has nothing to do with this.
A tool which uses HaldCLUT reference images changes the colors of a photograph by the same amounts and in the same directions as were changed between the before/after images based on which the HaldCLUT image was created. If that’s confusing, read about how HaldCLUTs are created to know what I mean by “before/after” images. By asking whether the Auto-Matched Tone Curve should be used, you’re actually asking whether your photograph in RawTherapee should resemble the “before” image based on which the HaldCLUT was created. Well you cannot know what state the “before” image was. For example, if a digital raw photo of a film slide was used to conjure the curves which were then applied to the HaldCLUT identity image, then if these curve adjustments were created on top of a linear basecurve, then you would probably need to use a linear basecurve in your photo when applying the HaldCLUT to get the same deviations.
The RawTherapee Film Simulation Collection contains HaldCLUTs made by @patdavid, @michaelezra and Pavlov Dmitry. Only they know what they used, what they based the adjustments on, what the starting point was.
Furthermore, there is a good chance that the images used for creating these HaldCLUTs already had multiple adjustments applied - for example a basecurve, exposure, brightness, black level, etc.
Finally, we don’t know what gamma (linear, perceptual, sRGB, other) or color space was used during the original conjuring of the curves which the Film Simulation Collection HaldCLUTs reproduce.
For these reasons it’s quite impossible to know what state your photo should be in, and also quite irrelevant and pointless. These HaldCLUTs are not scientifically correct, they are just for looks. Apply them to your photos, don’t worry about what state things were originally in, and adjust to taste.
3 Likes
This sounds sensible. My experience is that applying the HaldClut to an image with the tone curve applied results in an image with too much contrast whereas applying it to a linear curve results in a flat looking ‘digital like’ image. Adjusting to taste seems the way to go…
good point @Morgan_Hardwood. OTOH, for some cluts the info is available, e.g.
https://discuss.pixls.us/t/new-fujifilm-xtrans-film-simulations
By the way, does anyone have a forum name or website URL for the same Pavlov Dmitry who made some of those HaldCLUTs?