RAW files out of gamut

I am using DT 2.6.2 and took some outdoor shots on a sunlit day using my Sony A900 DSLR and a Minolta 100-300 mm zoom. When opening in DT, some of the shots show the green indicators for out of gamut.

These shots were taken using the camera’s P mode, and, to my eye, they look fine. What would cause them to show as out of gamut in DT?

Also, although checking the camera’s rear LCD, it indicates that the exposure is neither over or under, but when I open the file in DT, large areas show to be over exposed. I can correct these areas such that DT’s overexposure indicators disappear, but, again, I am curious why this should be.

I seem unable to correct the out of gamut situation.

I appreciate any advice.



What is your working color space and what is your softproof color space?

Softproof and display profiles are both SRGB (web safe). Not certain where to check the working profile. Can you advise?
. . . and thanks for the response.


You should be able to check the working profile with the button that also has the softproof profile in it. You need to right click it.

Your colours are merely out of gamut? I get colours that are outside the CIE horseshoe! So they are not real colours at all!

Well, that’s what the software says. For explanations, see the discussion at Camera gamut outside horseshoe?


Raw files from modern cameras start with a colorspace significantly larger than what display or print devices are capable of depicting. That’s the whole point of color management in raw processing, transforming the original camera image down to a color space that can be regarded on a particular device. I think specifying sRGB as the softproof profile is just telling you that. I’d remove it from there; that mechanism is built to tell you how far colors are from being depictable on a specific printer.

This might help: Article: Color Management in Raw Processing (disclaimer: I wrote it, no feedback (yet) that I’m out to lunch…)

When I right click the softproof icon, two items appear, “softproof profile” and “display profile”. Both are reading sRGB. Is that the information for which you seek. I do not see a working profile.



I read your article. It is interesting, and I sort of understand the concepts discussed, but I didn’t find an answer as to why some in this group of photos are out of gamut when I open them in DT, yet have not made any DT adjustments.

. . . and also, I am getting overexposure warnings in the highlight areas, also before I have used DT to make any adjustments. My camera (Sony A900) has a button that I can press during preview after taking a shot that shows the histogram and a small thumbnail of the image. If the camera thinks the image is over or under exposed, blinkers flash a warning. These images in the camera show no over/under warnings, yet, when opened in DT, highlights are showing as significantly overexposed.

This problem is easy for me to correct in DT, and I admit that the photo looks better afterward, but I am curious if I have accidentally changed something in DT that is causing DT to up the exposure automatically when I open a RAW image in the Darkroom.

I am using the “Sony like” base curve profile.

I love DT, have enjoyed learning it (still learning, of course), so, I hope you can put up with my ignorance.

Thanks to all who have replied.


My camera (Sony A900) has a button that I can press during preview after taking a shot that shows the histogram and a small thumbnail of the image. If the camera thinks the image is over or under exposed, blinkers flash a warning.

I have no specific knowledge on your camera model but I know some cameras DO NOT display an overexposure (or, underexposure) warning if only one of the channels in the internal jpeg of the image you see. DT displays a warning if any of the channels in your raw file has clipped values with your current settings. Could this be the difference? No expert here, as well.

Take care with this endeavor, as the camera is most likely using the internal JPEG processing to assert overexposure. Depending on the particular scene’s light, it could be that the camera indicator is flagging pixels where the raw data may still be lower than the camera sensor’s saturation point.

The only good way to determine sensor-saturation overexposure is with a histogram of the raw data, where you can see the pile-up of data at the sensor’s maximum resolvable value. If the raw data is 14-bit, for instance, this value can’t be higher than 16383, and may be just a bit lower depending on how the manufacturer characterizes the performance of the individual R, G, and B channels. I know RawTherappe has a raw mode for its histogram, don’t know about darktable.

I’ve been digging through this stuff for about 3 years, and I’ve found the most telling insights have come with understanding the real raw data as a starting point. To that, a variety of things are done to get the data to an image you can edit, and some of those things are likely influencing the particular questions you’ve asked here. Good on you to ask, but realize it can be a “rabbit hole”, down which there lie beasties… :smiley: