Ah yes, the pixii, the camera I want so badly to be open source… Alas…
If you’re still interested in this camera, there’s currently an active thread on cameraderie that has a few honest user experiences.
Too locked down for me.
Yep. It would be the perfect option for an open source firmware… We are already paying premium, it’s the least they could do
So if I’ve forked and changed the LibRaw library, do I then have to delete the original submodule and then add the modified one in my GitHub repository?
Depends on how you want to build. You can build libraw, turn off dt’s embedded version and use that.
Sorry, it seems that each answer spawns more questions. How do I build libraw (unlike DT, I couldn’t find build directions), and how do I then turn off DT’s embedded version.
If this is getting to be too much I’ll try to do more homework on my own.
there are two ways:
- you change the libraw submodule source to your GitHub fork and pull that into your darktable/src/external/libraw directory so darktable can be built with this modified libraw
- You simply change the files within your darktable/src/external/libraw directory and build darktable yourself.
don’t forget to merge also the 3 R5MII/R1 related darktable pull requests …
Thanks, that makes it clear
Darktable 5.4. will provide official support
Everything has been merged and libraw finally provides official support.
That is great news but I gave up and bought Dxo Photolab 9. They have a month trial I planned to try. Then move back to darkTable. Knowing the darkTable update was soon arriving. After a few days I changed my mind. I bought Photolab 9.
There is no excuse for libraw to have the RAW converter available a year earlier and not release it. If darkTable wants to survive they need to find another way to handle RAW conversions. This a new camera issue and not just a one model issue.
libraw is just the fallback solution since rawspeed as the standard raw decoder lacks cr3 support for several years and doesn’t seem to make any steps in that direction ![]()
This is an amusing statement. There are other raw loading libraries like rawspeed. But it also needs developer resources to reverse engineer the format or somehow get Canon to release some specification that can be implemented. without such a thing there is little progress in opensource.
And reverse engineering those format is not always fun.
there has been an unmerged PR for CR3 support in rawspeed for a “while” https://github.com/darktable-org/rawspeed/pull/271
Saying they are in need of developer resources is a very true statement. This is why we are understanding of the situation. With an exception. They had this RAW converter figured out a year ago.
This has nothing to do with limited resources. This is a complete disregard for the community. It goes both ways. We accept delays knowing the developers may have other full time jobs. We appreciate the work they do and want to support the efforts of the open source community. When a group knowingly withholds the RAW converter for a year. Then we should speak out.
Is it rawspeed that said they purposely delay for their commercial customers because they want stability? Which explains why Adobe, Dxo, Capture One wait years before they add a new RAW converter…oops!
Having support for rust and rawler would be nice. But Unfortunately I don’t know rust nor c very good to start this
That only seems for the moment: one forgets rawler is a one-man-show; libopenraw is a one-man-show; rawloader is a one-man-show; RawSpeed is for all intents and purposes a one-man-show. None of them (including LibRaw) is feature complete (and might only temporarily look like a “better” option) and can stall at any time, for whatever reason.
A one man show deserves leeway. Not a whole year. Especially since it was ready a year ago. Making the one man show not valid in this case.
@pilgrim I wasn’t taking about LibRaw and this particular case, but a more fundamental problem. There is no end solution here.
I think this statement comes across as entitled. This is FOSS. People share their work for free and don’t owe anyone their work, on any timeline. If libraw wants to wait 10yrs to share their work, that’s fine. They are free to do so. We might not like it, but that’s part of FOSS.