September 2026 - Cin - Cinelerra-GG Mailing Lists

OT: Capture 576i50 analog S-video with v4l2 and PCIe Magewell Pro Capture AIO in 10-bit 4:2:2?
by terjejhanssen＠gmail.com 30 Oct '26

30 Oct '26

It's been quite a while since my last post, but I've been keeping up a bit in the background ....... In the past, we had some posts about capture devices, some few standalone or pcie based, but most usb of course. After earlier S-video to DV capturing (which was in 8bit 4:2:0 using Datavideo DN300), my plan has been to recapture all my S-video tapes to 10bit 4:2:2 in Prores LT for the final archive. In fact I tested this 10 years ago for 5 Hi8 tapes (which still work using ffmpeg) using a Blackmagic Analog to SDI mini-converter and a Hyperdeck Shuttle-2 SSD recorder. Some months ago I tried to resume this capture project, but had to conclude that the BMD converter didn't work. I ordered a new one, 2nd gen (EOS), but unfortunately its new chips seemingly require a studio-quality video signal. The Shuttle-2 didn't get a valid signal to record, even though it works perfectly fine when recording HD video from my HDV camcorder via HDMI. So at last I had to give up on this BMD analog route. And I don't wont to trust Blackmagic Design Intensity Pro 4K on Linux -- a card I previously mentioned in an old post https://tinyurl.com/2nuj7hfk There is however another option, —costly, yes—for Linux that I was not aware of and therefore didn't mention in my old post above Magewell Pro Capture Aio - One-channel HD capture card But I have yet to find user posts confirming it is capable to record SD video in 10bit 4:2:2 with v4l2, and especially on openSUSE Tumbleweed-Slowroll. Any tip or thoughts about how risky this Magewell option may be with regards to success or failure ? Here follows just my collections of Magewell urls: https://www.magewell.com/products/pro-capture-aio https://www.magewell.com/kb/detail/000020047/pro-capture https://www.magewell.com/kb/pro-capture?page=1 https://www.magewell.com/downloads/pro-capture https://www.vision-dimension.com/en/blog/video-capture/magewell/magewell-is…

4 3

reference patch in MLT tree for ffmpeg 9
by Andrew Randrianasulu 26 Oct '26

26 Oct '26

https://github.com/mltframework/mlt/pull/1281/changes/68bceba12a3c3278ce690… of course ffmpeg devs invented yet another way to get supported sample/pixel formats out of codec internals .... Will try to make mplayer work first, then may be cinelerra.

5 19

Silly question
by Andrea paz 01 Oct '26

01 Oct '26

Foreword: My friends have been telling me enthusiastically about a CAD program for Linux that’s on par with AutoCAD and uses AutoCAD’s standard file formats. According to its author, this program—OpenCAD Studio ( https://github.com/HakanSeven12/OpenCADStudio)—was developed primarily using AI (Claude). Silly question: Can AI be used for some development in CinGG? It's a silly question because I don't know anything about programming or AI, but I'm curious about it.

10 22

Info on vaapi failure at ffmpeg 9.0.2
by Phyllis Smith 29 Sep '26

29 Sep '26

The error I am getting when I render a hardware vaapi loaded video with audio using h264_vaapi.mp4 is: > Assertion pic->display_order == pic->encode_order failed at > libavcodec/vaapi_encode_h264.c:421 > The code in libavcodec/vaapi_encode_h264.c is exactly the same for 8.1 as 9.0.2. I am continuing to investigate. The errors Andrew reported are: > RenderFarmClient::main_loop: client started > [AVHWFramesContext @ 0xafddc0c0] Unsupported format: vaapi. > Failed to initialize HW frame context. > err: Invalid argument > [h264_vaapi @ 0xa7ddaf00] A hardware frames reference is required to > associate the encoding device. > FFMPEG::open_encoder err: Invalid argument > int FFMPEG::open_encoder(const char*, const char*): > open failed h264_vaapi:/dev/shm/dnxhd.mkv > Render::render_single: Session finished. > Render::render_single: Session finished. > ** rendered 1 frames in 2.310 secs, 0.433 fps > FFMPEG::open_decoder: some stre > >

2 1

patching ffmpeg 9.0.2
by Andrew Randrianasulu 26 Sep '26

26 Sep '26

--log-- guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch1 | patch -p1 patching file fftools/cmdutils.c guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch2 | patch -p1 patching file libavformat/mpegtsenc.c Hunk #16 succeeded at 1893 with fuzz 1. Hunk #17 succeeded at 1912 (offset -2 lines). Hunk #18 succeeded at 2381 (offset -2 lines). Hunk #19 succeeded at 2410 (offset -2 lines). patching file libavformat/mpegts.h patching file libavformat/bluray.c Hunk #1 succeeded at 28 (offset 1 line). patching file doc/muxers.texi Hunk #1 succeeded at 3174 (offset 20 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch3 | patch -p1 patching file libavformat/avformat.h Hunk #1 FAILED at 495. Hunk #2 FAILED at 558. 2 out of 2 hunks FAILED -- saving rejects to file libavformat/avformat.h.rej patching file libavformat/dv.c patching file libavformat/matroskadec.c Hunk #1 succeeded at 5024 (offset 174 lines). Hunk #2 succeeded at 5037 (offset 174 lines). patching file libavformat/seek.c guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch4 | patch -p1 patching file libavformat/avidec.c Hunk #1 succeeded at 2039 (offset 8 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch | patch -p1 ffmpeg-8.1.patch1 ffmpeg-8.1.patch14 ffmpeg-8.1.patch5 ffmpeg-8.1.patchC ffmpeg-8.1.patch10 ffmpeg-8.1.patch2 ffmpeg-8.1.patch7 ffmpeg-8.1.patchD ffmpeg-8.1.patch11 ffmpeg-8.1.patch3 ffmpeg-8.1.patch9 ffmpeg-8.1.patchZ1 ffmpeg-8.1.patch12 ffmpeg-8.1.patch4 ffmpeg-8.1.patchB ffmpeg-8.1.patchZ2 guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch | patch -p1 ffmpeg-8.1.patch1 ffmpeg-8.1.patch14 ffmpeg-8.1.patch5 ffmpeg-8.1.patchC ffmpeg-8.1.patch10 ffmpeg-8.1.patch2 ffmpeg-8.1.patch7 ffmpeg-8.1.patchD ffmpeg-8.1.patch11 ffmpeg-8.1.patch3 ffmpeg-8.1.patch9 ffmpeg-8.1.patchZ1 ffmpeg-8.1.patch12 ffmpeg-8.1.patch4 ffmpeg-8.1.patchB ffmpeg-8.1.patchZ2 guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch5 | patch -p1 patching file libavfilter/formats.c Hunk #1 succeeded at 112 (offset 1 line). Hunk #2 succeeded at 127 (offset 1 line). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch7 | patch -p1 patching file libavcodec/vdpau_mpeg12.c guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch8 | patch -p1 cat: ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch8: No such file or directoryguest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch9 | patch -p1 patching file libavformat/mpegenc.c Hunk #1 succeeded at 1000 (offset 11 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch10 | patch -p1 patching file libavutil/hwcontext_cuda.c Hunk #1 succeeded at 374 (offset 2 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch11 | patch -p1 patching file libavcodec/hevc/hevcdec.c Hunk #1 succeeded at 698 (offset -14 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch12 | patch -p1 patching file libavcodec/hevc/ps.c Reversed (or previously applied) patch detected! Assume -R? [n] n Apply anyway? [n] n Skipping patch. 3 out of 3 hunks ignored -- saving rejects to file libavcodec/hevc/ps.c.rej guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch1 | patch -p1 ffmpeg-8.1.patch1 ffmpeg-8.1.patch11 ffmpeg-8.1.patch14 ffmpeg-8.1.patch10 ffmpeg-8.1.patch12 guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch14 | patch -p1 patching file libavcodec/hevc/refs.c Hunk #1 succeeded at 191 (offset 2 lines). Hunk #2 FAILED at 463. 1 out of 2 hunks FAILED -- saving rejects to file libavcodec/hevc/refs.c.rej guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch | patch -p1 ffmpeg-8.1.patch1 ffmpeg-8.1.patch14 ffmpeg-8.1.patch5 ffmpeg-8.1.patchC ffmpeg-8.1.patch10 ffmpeg-8.1.patch2 ffmpeg-8.1.patch7 ffmpeg-8.1.patchD ffmpeg-8.1.patch11 ffmpeg-8.1.patch3 ffmpeg-8.1.patch9 ffmpeg-8.1.patchZ1 ffmpeg-8.1.patch12 ffmpeg-8.1.patch4 ffmpeg-8.1.patchB ffmpeg-8.1.patchZ2 guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patch | patch -p1 ffmpeg-8.1.patch1 ffmpeg-8.1.patch14 ffmpeg-8.1.patch5 ffmpeg-8.1.patchC ffmpeg-8.1.patch10 ffmpeg-8.1.patch2 ffmpeg-8.1.patch7 ffmpeg-8.1.patchD ffmpeg-8.1.patch11 ffmpeg-8.1.patch3 ffmpeg-8.1.patch9 ffmpeg-8.1.patchZ1 ffmpeg-8.1.patch12 ffmpeg-8.1.patch4 ffmpeg-8.1.patchB ffmpeg-8.1.patchZ2 guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patchB | patch -p1 patching file libavutil/hwcontext_vdpau.c guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patchC | patch -p1 patching file libavcodec/encode.c Hunk #1 succeeded at 357 (offset 52 lines). Hunk #2 succeeded at 377 (offset 52 lines). Hunk #3 succeeded at 554 (offset 53 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patchD | patch -p1 patching file libavcodec/pcm-dvdenc.c Hunk #2 succeeded at 187 (offset -1 lines). guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patchZ1 | patch -p1 patching file libavcodec/wrapped_avframe.c guest@slax:/dev/shm/ffmpeg-9.0.2$ cat ../new-git/cinelerra/cinelerra-5.1/thirdparty/src/ffmpeg-8.1.patchZ2 | patch -p1 patching file libavformat/yuv4mpegenc.c so basically only patch3, patches 12 and 14 failed. Everything else seems to be applicable.

1 0

Few more patches for cingg on Linux/ppc
by Andrew Randrianasulu 05 Sep '26

05 Sep '26

As promised long ago I tried to make adding -latomic and disabling vsx in our x264 for case where it does not work (older PPC processors) automatic. I think I hit problem with our lv2 (lilv/suil) not picked up if system ones do not present at compile time, but this is another problem I'll try to solve. For now can you test attached four patches and make sure they do not break anything else (x86/amd64)? I tried them on Termux/aarch64 and here they seems to be harmless.

3 2

OT: whisper.cpp works on powermac G5!
by Andrew Randrianasulu 03 Sep '26

03 Sep '26

I was trying to find any use for those large language models and found whole site with autotranscribed WWDC videos. I looked into whisper.cpp and after some bugreporting it installs and works even on 20 years old processor! https://files.mastodon.online/media_attachments/files/117/206/767/104/645/6… I used base model: wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin\… and "jfk.mp3" sample (real classic!): wget https://github.com/ggml-org/whisper.cpp/raw/refs/heads/master/samples/jfk.m… I think in my case without any acceleration and just 2 2GHZ cores it processed 11 seconds in 126 or so seconds,so def. not realtime, but a) there are turbo variants of model b) usually people have more powerful machines with gpu, so may be it can be used for auto-transcribing some long lecture or somesuch. This speech-to-text was promised LONG ago and I guess only now, after serious sacrifices in terms of energy and other damages by Big Bro Tech, actually becoming usable ... :/ But I figured if I used OCR for all those 25 years I can use this STT engine too, at least for testing.

2 1

Video editor on browser with AI
by Andrea paz 02 Sep '26

02 Sep '26

If anyone is interested in trying out a browser-based video editor that uses AI: https://github.com/MartinDelophy/ai-video-editor I don't think it's very reliable when it comes to privacy...

2 1