I had also issue with the generation of the noise_result.pdf that was produced using pdftk. This is in my Ubuntu 22.04 installed using snap and if the raw files are on symlinked drive, pdftk was not able to find the files. The solution is either here: java - Why is PDFTK saying it can't find files that are right there? - Unix & Linux Stack Exchange or in subr.sh - checking for gs first and only later for pdftk.
Finally got this going on openSUSE Tumbleweed, running the tool, and getting the following error:
===> Checking profiling RAW images correctness + Jpeg export
--> ISO 100:
/home/zingerpb/Pictures/NoisePhotos/DSC01496.ARW
RawSpeed:Unable to find camera in database: 'SONY' 'ILCE-6700' ''
Please consider providing samples on <https://raw.pixls.us/>, thanks!
0.4646 [rawspeed] (DSC01496.ARW) bool rawspeed::RawDecoder::checkCameraSupported(const rawspeed::CameraMetaData*, const std::string&, const std::string&, const std::string&), line 171: Camera 'SONY' 'ILCE-6700', mode '' not supported, and not allowed to guess. Sorry.
0.4646 [libraw_open] extensions whitelist: `cr3'
0.4665 [temperature] failed to read camera white balance information from `DSC01496.ARW'!
0.4682 [dt_imageio_export_with_flags] mipmap allocation for `/home/zingerpb/Pictures/NoisePhotos/DSC01496-large.jpg' failed
0.4683 [imageio_storage_disk] could not export to file: `/home/zingerpb/Pictures/NoisePhotos/DSC01496-large.jpg'!
However, I know the camera profile has been updated and included in the latest DT release, as well as being on rawspeed.
Thanks Peter, that helped. The noiseprofile tool was pulling from a native install of darktable which didn’t have that profile, but I’ve been using the flatpak (to maintain consistency between two computers/os installs). I was able to copy - paste the cameras.xml file and that fixed it.
I’m planning to create a noise profile for the Nikon Z8, which still seem to be missing. I tried printing out a gradient on my printer, which is an (Epson 7750) inkjet printer, on normal paper.
However, the dark parts are not dark enough and with backlight, exposing so that the white part burns out, the darkest part of the gradient is perhaps only mid-gray or so.
Have anyone else printed gradients on inkjet printers? Or is it better to use laser printer?
OK, thanks – that sounds manageable then. I was assuming I would need to create an as even as possible gradient from dark to overexposed, but what you did sounds more like you relied on lens blurring to create the gradient? The only problem here is that are lots of clouds currently, so I cannot focus on infinity – then I’ll get sharp images of clouds