Trying to process with RawTherapee, raw images from my brand new Olympus TG-7, I experienced several problems, described therafter :
First with RawTherapee 5.8, the version that comes with my Ubuntu 22.04 : thumbnails and opened images show black (cf. screenshots 1 and 2)
Finding on this forum that ORF files are recognized since 5.9, I tried with RawTherapee 5.9 Appimage, said to incorporate Olympus TG-6 raw files processing : some thumbnails are OK, some turn purple, and opened images show black (cf. screenshots 3 and 4)
Then I tried with RawTherapee5.10 Appimage : some thumbnails are OK, some are purple, and opened images are cast with a magenta filter (cf. screenshots 5 and 6) ; note that after processing, images keep this magenta overcast.
Does it come from some features in TG-7 ORF files, absent from TG-6, and not yet implemented in RawTherapee ?
Or is there something I can do in RawTherapee 5.10 to fix this problem ?
Thanks in advance,
AL
There’s no real difference between “good” and tainted images, since a “good” image becomes tainted as soon as opened (cf. screenshot 7).
The WB parameters seem good, as i may guess by comparing the raw image (screenshot 7) and the jpeg (screenshot 8).
I think using the default Ubu repositories for RT (or DT, for that matter) is seldom a good idea. Just like with many apps, they tend to be very outdated. Now I go straight for the latest stable RT AppImage and voilĂ .
They look really different in those screenshots to me The Teinte being at less than 0.6, in particular, seems weird and may be a major factor causing the weird color, I suppose.
By the way, when [Questions] How to choose a DCP profile (if at all) got sidetracked about a white balance bug, I did a embedded-jpg-vs-raw comparison and I got the impression that JPGs could not be trusted in that regard. I mean, they seem to yield very different temperature and tint in RT than their raw counterpart, even when the camera data is read properly. So beware. I may be wrong, though; I’m no expert. Just empirical observations from the other day.
Oh and perhaps the workaround I currently use for my camera would work for you as well: try to see if ART (or any other app, actually) is able to read the missing info better. Then if it looks good enough you can just steal its temperature and tint and write them in RT.
For sure it’s got to be a misread tint… The true value won’t be 1 but setting it to 1 on one of the tainted images will likely almost completely fix it so clearly it’s just not read correctly…
@priort : Unfortunately, setting tint is not enough to fix the problem (see screenshot 9).
As you have noticed I’m all but an expert of raw processing, so I’m quite puzzled bay all that stuff…
Well one thing you can check is looking in the meta data of your files… if you share one here we can look… I don’t think RT lets you directly adjust the wb multipliers but it shows them…
see here
And they should match what is in the exif data for your files… A big mismatch would likely show that the proper support is not yet there for your camera and as such erroneous wb values can happen…it is funny it happens for some but not all…
Good news and the bad news… using the latest version of DNG converter your images come in fine. I downloaded a bunch of samples and tried them. Also there is a DCP profile for the TG-7 and it does change the result quite a bit from whatever profile RT is selecting…
Bad news is until the camera is supported I think that is your only options as the wb multipliers are the same or show as the same in RT despite resulting in quite different temp and tint numbers and changing temp and tint don’t really fix the images so for sure there is some camera support required…
But is you can run Adobe DNG convertor then you are off to the races…
Thumbnails show the embedded jpeg, which explains the difference. I’m guessing there is a problem with the black level. I will investigate when I have the time.
What represent wb multipliers ? And which exif parameter(s) should show their values ? I’m unable to find them in the metadata…
Here is a raw file if you have time to have a look on it.
Thanks in advance. P3170049.ORF (12.0 MB)
Unfortunately I didn’t succeed in installing wine to run DNG converter for windows. I’m afraid I’m bound to wait for a RT release supporting TG-7…
It maybe that your file seems to order the values as RBGG and not the normal RGGB
If you see the values from the exif…
If you divided R and B by 256 you get the wb values that you see in RT but at some point I wonder if the channels are not correctly interpreted?? Maybe this is common for ORF and has nothing to do with it… I don’t have the knowledge to say…
I tried darktable 4.6.1 that is said to support TG-7 ORF : it looks better than RT, isn’t it ? (see screenshot)
So my next duty is training to darktable
To be continued…
The Teinte being at less than 0.6, in particular, seems weird and may be a major factor causing the weird color, I suppose.
