Histogram difference - camera vs darktable.

Sure, please find

The raw white points are (dt |exif specular | exif normal):
*.804 |14338 | 13035 | 11448
*.805 (D+) | 11606 | 11435 | 11606
*.806 (D2+) | 11606 | 11435 | 11606

Looks like DT reads the WP correctly.

For me the message is to be careful with using the modes because something is happening very early in the pipeline, most probably before saving the raw file to the memory card.

1 Like

Thanks, I see one thing wrong. With my 6D, HTP 200(D+) is ISO 100, not ISO 200. HTP ISO 200 (D+) should be compared with ISO 100.

I don’t know about (D2+) and I will not be able to check your raw files before mid September.

That’s quite a big difference 14338 vs. 11448 ?

Thank you! Very interesting results and relation to the histograms. This modes (D+ and D2+) appear to be indeed quite aggressive. But the camera LCD is far from painting the whole picture.

Appreciate the comprehensive test and files!

I know this is not darktable, but rawproc will display a histogram corresponding to the image at the state of the tool selected for display. I’ve found this very helpful for following the state of the image from tool to tool in the chain. Also, it makes it very evident which data is the source for the histogram construction. And, any step in the tool chain can be selected for display, whether it looks good or not. But, by selecting the input image at the top of the chain, that allows for looking at the histogram of the raw data right out of the file, with no processing.

I bring this up because it’s very important to know the exact place in the tool chain from which the histogram is constructed, in any software…

3 Likes

Yes, yes, yes. If we don’t know what data a histogram is showing, then the histogram tells us nothing, and is useless.

1 Like

Checked your files now. Seems like 200 (D+) and 200 (D+2) are just ISO 100. And the first file you took was a normal ISO 200 file.

If you take ISO 100, ISO 200 (D+) and 200 D(+2) with same exposure settings and clip the highlight somewhere in the picture it would be easy to check if the clipped area is same for them all three.

You are completly right. One needs to compare ISO100 shots with ISO 200D+ shots to see wether the raw files are affected. In my lunch break I took some pictures found that the histograms of the shots mentioned are identical (I didn’t had a tripod and clouds etc are moving, so there is a small but neglectable deviation). The ISO200 image without highlight tone priority has a completely different histogram and white level.

From this, it looks like the raw file is not modified and we can just ignore this setting.

2 Likes