- Cin - Cinelerra-GG Mailing Lists

dav1d 0.5.1 patch fix?
by Andrew Randrianasulu 03 Dec '21

03 Dec '21

it does not work on non-x86 at the moment (need more work) but at least on x86 patching should be ok now. anyone want to test? (patch with git am please :-} )

2 11

Debian 11 build good but problems on execution
by Phyllis Smith 02 Dec '21

02 Dec '21

This is what I did: - installed Debian Bullseye from 18 CDs - installed the minimum amount of stuff I thought I needed but still not enough... - ran blds/bld_prepare.sh debian to install the rest of necessary packages - executed ./bld.sh and there were no errors - built with the debugger - Cinelerra comes up successfully - a video .mov file with no audio loads successfully - a .jpg fails and backtrace below - a big buck bunny .mkv file fails and backtrace below Just Loading a x.jpg file: -------------------------- Thread 48 "ci" received signal SIG32, Real-time event 32. [Switching to Thread 0x94afdb40 (LWP 23149)] 0xb7fd2559 in __kernel_vsyscall () (gdb) bt #0 0xb7fd2559 in __kernel_vsyscall () #1 0xb75c03ed in __GI___clock_nanosleep_time64 (clock_id=<optimized out>, flags=0, req=0x94afd1cc, rem=0x94afd1dc) at ../sysdeps/unix/sysv/linux/clock_nanosleep.c:52 #2 0xb75c04be in __GI___clock_nanosleep (clock_id=0, flags=0, req=0x94afd238, rem=0x0) at ../sysdeps/unix/sysv/linux/clock_nanosleep.c:92 #3 0xb75c6a20 in __GI___nanosleep (requested_time=0x94afd238, remaining=0x0) at nanosleep.c:27 #4 0x00d83dbc in Timer::delay (milliseconds=1500) at bctimer.C:100 #5 0x00d97d45 in BC_Repeater::run (this=0x5e2f3e0) at bcrepeater.C:172 #6 0x00d832da in Thread::entrypoint (parameters=0x5e2f3e0) at thread.C:68 #7 0xb7c920b4 in start_thread (arg=<optimized out>) at pthread_create.c:477 #8 0xb7600296 in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:108 Just loading bigbuckbunny...mxf ------------------------------- Thread 19 "ci" received signal SIGSEGV, Segmentation fault. [Switching to Thread 0xa44fab40 (LWP 23194)] BC_Xfer::xfer_rgb888_to_rgb888 (this=0xa44f9ffc, y0=<optimized out>, y1=64) at xfer_bc_rgb888.C:38 38 int r = *inp++, g = *inp++, b = *inp++; (gdb) bt #0 BC_Xfer::xfer_rgb888_to_rgb888 (this=0xa44f9ffc, y0=<optimized out>, y1=64) at xfer_bc_rgb888.C:38 #1 0x00d95c45 in BC_Xfer::xfer_slices (this=0xa44f9ffc, slices=1) at ../bcxfer.C:304 #2 0x00d9609a in BC_Xfer::xfer (this=0xa44f9ffc) at xfer.C:840 #3 BC_CModels::transfer (output_ptrs=0x96311760, out_colormodel=9, out_x=0, out_y=0, out_w=64, out_h=64, out_rowspan=192, input_ptrs=0x96300d30, in_colormodel=9, in_x=0, in_y=0, in_w=808, in_h=808, in_rowspan=2424, bg_color=0) at ../bcxfer.C:205 #4 0x00d8a099 in VFrame::transfer_from (this=0x96311690, that=0x963007a0, bg_color=0, in_x=0, in_y=0, in_w=808, in_h=808) at vframe.C:1236 #5 0x009c115f in VFrame::transfer_from (this=0x96311690, frame=0x963007a0, bg_color=0) at /mnt0/cintoday/cinelerra-5.1/cinelerra/../guicast/vframe.h:147 #6 0x00c1b3aa in ResourceVideoThread::do_video (this=0x583a130, item=0x5a986a0) at resourcethread.C:428 #7 0x00c1b01e in ResourceVideoThread::draw_item (this=0x583a130, item=0x5a986a0) at resourcethread.C:351 #8 0x00c1ad78 in ResourceThreadBase::run (this=0x583a130) at resourcethread.C:303 #9 0x00d832da in Thread::entrypoint (parameters=0x583a130) at thread.C:68 #10 0xb7c920b4 in start_thread (arg=<optimized out>) at pthread_create.c:477 #11 0xb7600296 in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:108 is=0x583a130) at resourcethread.C:303

2 1

udfclient 0.8.11 patch testing?
by Andrew Randrianasulu 29 Nov '21

29 Nov '21

udfclient is suite of udf-related utils developed for NetBSD Available for debian-based distros and arch, but notably not for rpm-based distros according to https://repology.org/project/udfclient/versions can you download orig. tarball and patch it with attached patch and try to './configure / make' it on Linux box with normal {for us} gnu make? And run resulted udfclient from build dir? ./udfclient file.img (-W for writing) Also, can you try to compile my hack on newfs_udf from NetBSD current? just make in source tree.. might fail due to no configure at all. this is slightly different newfs_udf command, creates udf up to maximum revision 2.60 (with some limitations). My hack most likely will not work directly on optical drive - but it worked for me on small image file. Can you test it on large file too? both hacks already on my github (but I am not sure I uploaded files correctly via their webinterface)

1 0

udf verifier
by Andrew Randrianasulu 27 Nov '21

27 Nov '21

Can anyone try to test bdwrite's output with this tool: https://www.lscdweb.com/registered/udf_verifier.html for Netbsd created 2.50 udf on hard drive file it requires -blocksize 512, but for real bluray image blocksize must be 2048 (default) $ ./udf_test -blocksize 512 ~/udfimg.img > log there is linux (386) binary, but of course I compiled it anew, on termux. no scsi support, but othervise seems to work

2 6

Testing HDV on a Blu-ray Disc Without Re-encoding
by Terje J. Hanssen 26 Nov '21

26 Nov '21

I pick up again and continue another thread on this topic started a year ago: https://www.mail-archive.com/[email protected]/msg02145.html In this post I start carefully with three questions regarding manual definitions and procedure in this section: https://cinelerra-gg.org/download/CinelerraGG_Manual/HDV_on_Blu_ray_Disc_Wi… > An MTS file is a video file saved in the high-definition (HD) MPEG > Transport Stream video format, commonly called AVCHD. It contains HD > video compatible with Blu-ray disc format and is based on the MPEG-2 > transport stream. MTS files are often used by Sony, Panasonic, Canon > and other HD camcorders. ========== Q1. My HDV 1080i video file transfered from tape to HDD via Firewire, get a file extension M2T, which then is recognized as format MPEG-TS by Mediainfo. AFAIK the next generation camcorders after HDV used H.264/AVC (AVCHD)? ========= > Legal input for > Video – MPEG1VIDEO, MPEG2VIDEO, H264; > Audio – MP1, MP2, AC3, AC3PLUS, DTS, TRUHD. To verify this with FFMpeg for the previous HDV test-file "20081103140154.m2t" available on internet: ffmpeg -i 20081103140154.m2t 2>&1 >/dev/null | egrep "Audio|Video" Stream #0:0[0x810]: Video: mpeg2video (Main) ([2][0][0][0] / 0x0002), yuv420p(tv, bt709, top first), 1440x1080 [SAR 4:3 DAR 16:9], 25000 kb/s, 25 fps, 25 tbr, 90k tbn, 50 tbc Stream #0:1[0x814]: Audio: mp2 ([3][0][0][0] / 0x0003), 48000 Hz, stereo, fltp, 384 kb/s According to the manual, both the HDV MPEG2Video stream and MP2 audio stream should be legal input to create a Blu-ray disc. However, MP2 isn't on Sony's list or other I have seen over audio and video codecs supported on Blu-ray Disc media. https://www.sony.com/electronics/support/articles/00029663 Q2. Can someone clarify if MP2 rather should be transcoded to i.e AC3 instead, possibly also remux to BDAV/M2TS if required (below)? (On my first attempt with "MTS" format, Mediainfo reported this FileExtension_Invalid: ts m2t m2s m4t m4s tmf ts tp trp ty) ========= Q3. I also wonder if the manual command procedure creates a BDAV or a BDM structure, and optional if a DVD-RW disc can be used to create and playback this "Blu-ray disc" using a Blu-ray burner and a standalone UHD-Blu-ray player device? (At the time I have no free BD-R/RE discs left) ========= Transcode to AC3 and remux to M2TS with FFMpeg if required: ffmpeg -i 20081103140154.m2t -acodec ac3 -vcodec copy 20081103140154.m2ts ffmpeg -i 20081103140154_m2t-ac3.m2ts 2>&1 >/dev/null | egrep "Audio|Video" ......... Stream #0:0[0x1011]: Video: mpeg2video (Main) ([2][0][0][0] / 0x0002), yuv420p(tv, bt709, top first), 1440x1080 [SAR 4:3 DAR 16:9], 25000 kb/s, 25 fps, 25 tbr, 90k tbn, 50 tbc Stream #0:1[0x1100]: Audio: ac3 (AC-3 / 0x332D4341), 48000 Hz, stereo, fltp, 192 kb/s mediainfo 20081103140154_m2t-ac3.m2ts .......... General ID : 1 (0x1) Complete name : 20081103140154_m2t-ac3.m2ts Format : BDAV Format/Info : Blu-ray Video File size : 42.2 MiB Duration : 13 s 360 ms Overall bit rate mode : Variable Overall bit rate : 26.4 Mb/s Video ID : 4113 (0x1011) Menu ID : 1 (0x1) Format : MPEG Video Commercial name : HDV 1080i Format version : Version 2 Format profile : Main@High 1440 Format settings : CustomMatrix / BVOP Format settings, BVOP : Yes Format settings, Matrix : Custom Format settings, GOP : M=3, N=12 Format settings, picture structure : Frame Codec ID : 2 Duration : 13 s 440 ms Bit rate mode : Constant Bit rate : 25.0 Mb/s Width : 1 440 pixels Height : 1 080 pixels Display aspect ratio : 16:9 Frame rate : 25.000 FPS Standard : Component Color space : YUV Chroma subsampling : 4:2:0 Bit depth : 8 bits Scan type : Interlaced Scan order : Top Field First Compression mode : Lossy Bits/(Pixel*Frame) : 0.643 Stream size : 40.3 MiB (95%) Color primaries : BT.709 Transfer characteristics : BT.709 Matrix coefficients : BT.709 Audio ID : 4352 (0x1100) Menu ID : 1 (0x1) Format : AC-3 Format/Info : Audio Coding 3 Commercial name : Dolby Digital Codec ID : 129 Duration : 13 s 440 ms Bit rate mode : Constant Bit rate : 192 kb/s Channel(s) : 2 channels Channel layout : L R Sampling rate : 48.0 kHz Frame rate : 31.250 FPS (1536 SPF) Compression mode : Lossy Delay relative to video : -5 ms Stream size : 315 KiB (1%) Service kind : Complete Main Menu ID : 256 (0x100) Menu ID : 1 (0x1) Duration : 13 s 360 ms List : 4113 (0x1011) (MPEG Video) / 4352 (0x1100) (AC-3) Service name : Service01 Service provider : FFmpeg Service type : digital television ============= Terje J. H

2 9

Fwd: arm/termux series (possibly broken on x86)
by Phyllis Smith 25 Nov '21

25 Nov '21

Checked into GIT, additional TERMUX/ANDROID patch related to mjpegtools + the libwebp patch3 from the other day. This takes care of randrik19 -- 0012, 0020, 0035, and 0054. Andrea, could you build because the libwebp patch is in too and just run a couple of videos through? Andrew, could you verify checkin? Next working on the libbthread mods (Andrew, thanks for the latest releated email on these).

3 2

truehd patches from ffmpeg git
by Andrew Randrianasulu 24 Nov '21

24 Nov '21

so.. patches mostly applicable to ffmpeg-4.4 while I slightly edited 'last frame' fix due to other ffmpeg changes apply 'last frame' and '24-bit' for encoding truehd hopefully lossless apply two mlpdec patches for more correct decoder flush/crc check try to apply mlp.h and libavformat patches for timestamps in demuxer... stop-returning-empty for mlpenc not tested. after all this patching both ffmpeg and cin should work better with truehd streams (still require 'strict -2' in profile) there were other fixes and code movements but i hope those patches can fix real bugs and allow losless re-encoding between pcm and truehd (check crc's!)

1 0

ffmpeg's e-ac3 encoder
by Andrew Randrianasulu 24 Nov '21

24 Nov '21

according to some tests it works, but some hw decoders (in tv) migh refuse stream ( https://www.reddit.com/r/ffmpeg/comments/k3z9i9/how_bad_is_ffmpegs_eac3_enc… it also does not support more than 6 channels https://trac.ffmpeg.org/ticket/3595?cversion=0&cnum_hist=7 command line and bitrates possible here: https://ffmpeg-user.ffmpeg.narkive.com/JJ0ohMKA/eac3-encoding -acodec eac3, up to 6 mbps... testing should be as simple as copying ffmpeg/audio ac3 profile with new name and replacing ac3 with eac3

3 3

mpeg2 encoding with ffmpeg (for dvd, from doom9 forums)
by Andrew Randrianasulu 23 Nov '21

23 Nov '21

I found few ffmpeg/mpeg2 related threads but nothing about mpeg2/bluray encoding https://forum.doom9.org/showthread.php?t=173332 ==== Quote: ffmpeg.exe -i # -aspect # -dc 10 -g 12 -maxrate 8000k -q:v 2 -b_strategy 2 -brd_scale 2 -bf 2 -profile:v 4 -intra_matrix "8,16,19,22,26,27,29,34,16,16,22,24,27,29,34,37,19,22,26,27,29,34,34,38,22,22,26,27,29,34,37,40,22,26,27,29,32,35,40,48,26,27,29,32,35,40,48,58,26,27,29,34,38,46,56,69,27,29,35,38,46,56,69,83" -inter_matrix "16,17,18,19,20,21,22,23,17,18,19,20,21,22,23,24,18,19,20,21,22,23,24,25,19,20,21,22,23,24,26,27,20,21,22,23,25,26,27,28,21,22,23,24,26,27,28,30,22,23,24,26,27,28,30,31,23,24,25,27,28,30,31,33" -pass 1 -passlogfile "ffmpeg" -an -c:v mpeg2video pass1.m2v" ffmpeg.exe -i # -aspect # -dc 10 -g 12 -lmin 0.75 -mblmin 50 -qmin 1 -qmax 31 -maxrate 8000k -b:v 3500k -bf 2 -pre_dia_size 5 -dia_size 5 -qcomp 0.7 -qblur 0 -preme 2 -me_method dia -sc_threshold 0 -sc_factor 4 -bidir_refine 4 -profile:v 4 -mbd rd -mbcmp satd -precmp satd -cmp satd -subcmp satd -skipcmp satd -intra_matrix "8,16,19,22,26,27,29,34,16,16,22,24,27,29,34,37,19,22,26,27,29,34,34,38,22,22,26,27,29,34,37,40,22,26,27,29,32,35,40,48,26,27,29,32,35,40,48,58,26,27,29,34,38,46,56,69,27,29,35,38,46,56,69,83" -inter_matrix "16,17,18,19,20,21,22,23,17,18,19,20,21,22,23,24,18,19,20,21,22,23,24,25,19,20,21,22,23,24,26,27,20,21,22,23,25,26,27,28,21,22,23,24,26,27,28,30,22,23,24,26,27,28,30,31,23,24,25,27,28,30,31,33" -pass 2 -passlogfile "ffmpeg" -an -c:v mpeg2video out.m2v Just replace the # characters with your own values, maybe play with the bitrate... === I never used those custom matrix params - someone with better eyesight (and low error-tolerance :-}) can check them.. about dvb-t muxing http://www.waveguide.se/?article=creating-dvb-t-compatible-mpeg2-streams-us… http://forum.doom9.net/showthread.php?t=174620 === new-ish dvd encode with ffmpeg thread making use of negative sc_threshold value for less artefacting in areas with high motion. Not sure if our ffmpeg 4.4 still respects this. @REM Threshold for scene change detection (sc_threshold) [−1000000000;1000000000] (Negative values indicate that it is more likely to insert an I-frame when it detects a scene change) set E_TSD=-30000 === also, dc precision should be higher for higher bitrates (dc 12 for max?) .. As starting point for first shot just increase bitrates to BD-levels?

3 11

Formats and Codecs in CinGG
by Andrea paz 22 Nov '21

22 Nov '21

I present a first draft to create Appendix section about formats and codecs supported by CinGG. I don't know DV; DVD; BLU RAY and RAW formats. If anyone uses them and can provide better info, it would be my pleasure. Also the info on MOV as a good format for editing I have only read and not understood well. Are they correct? Corrections, additions and advice are welcome. The bigger question is: is it worth stretching the already big manual with this section? [NOTE: is not what Terje requested; I'm not sure how to implement those tables. If you can write them, then I'll transpose them into Latex...]

4 12