-
Notifications
You must be signed in to change notification settings - Fork 316
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Canon R6 CR3 support #5901
Comments
I think I answered my own question by going through the |
The only other thing, apart from the comments in If you're interested in doing the analysis on your own and come up with a json entry, I would be happy to verify it. If so, please provide a downloadable folder with the necessary shots. |
It seems someone else has already made the necessary shots as well. https://discuss.pixls.us/t/canon-eos-r6-photo-sets/20094 |
Cool! I appreciate it. I looked into it a bit more and it seems like I'd have to pay for RawDigger and run it in wine, which kind of tamped down my enthusiasm for figuring this out. |
After analysis, I get the following - quite unsurprising - results: Some explanation |
What program did you use to gather the data, and to visualize it? It looks great. This issue indicates that reading image data from Canon EOS R6 is supported, but I tested |
I'm using Wolfram Mathematica, commercial software that I got to know during my PhD and absolutely love for data processing. It should be very similarly possible to make these plots in Python. As for not showing up in the File Browser, did you enable the CR3 extension in the Preferences? |
@Thanatomanic yes, tried passing it as an argument too. Does that file open in your RT?
|
@Beep6581 The file you link to is a compressed RAW format according to ExifTool. Apparently, that is not supported yet in the CR3 decoder. ART shows the file in the browser, but outputs garbage too. Edit: In verbose mode RT outputs the following when loading the file browser:
|
@Beep6581 & @Thanatomanic It would appear that exiv2 now support the ISO BMFF, the base format for CR3s after PR Exiv2/exiv2#1229. AFAICT the main impediment to supporting R5/R6 metadata in RT (including funky whitebalance issues, #6089) is getting the rtexif -> exiv2 PR (#5889) merged. Is there anything external devs like myself (with some CV, exiv2, and C++ experience) can do to assist with this? lack of metadata support has been a significant issue for me (w/ an R5), and I'm desperate to see a fix included in the next version of RT. |
@kdm9 branch |
Still not supported. Loading an EOS R6 (FW 1.40) .cr3 file results in a plain white image. Version is 5.8. It happens on both macOS Big Sur 10.16 and Linux Ubuntu 20.04 (both Intel 64). |
There's a fork of RawTherapee called ART that does support CR3 files I think. It seemed to work fine last time I tried it, but I can't vouch for things like color accuracy. Only that it wasn't a plain white image :) |
@freebrowser1 The R6 files can be opened perfectly fine in the nightly builds. I would advise to download the dev AppImage and use it from here https://github.com/Beep6581/RawTherapee/releases/tag/nightly or use an unofficial build http://rawpedia.rawtherapee.com/Download |
Thanks. |
My new Canon R6's CR3 raw files seem to be parsed just fine in the RawTherapee file browser, but when I attempt to edit them I see an entirely white image. I noticed here that the program seems to just have a place where it stores information about how to decode raw files, and I'm assuming that's just missing.
I can shoot the necessary frames, but is there documentation on what to like, do with them? I know folks are probably busy and if there are docs on how to process the white frames and create the actual data on white levels for this sensor, I can probably follow them and submit a PR with the changes. Or at least publish a
camconst.json
so other folks with R6's can process their files without converting them to DNG first. If docs on this don't exist, maybe they should? If nothing more it might lighten the load on the developers to have drive-by contributors be able to process their own white frames. :)The text was updated successfully, but these errors were encountered: