well, case 2 is for usual 2.0 stereo (two channels) .. but it has identifier symbolically defined as bluray_audio_format_stereo and value of this constant is '3' .... so, for truehd you were getting 12 'channels' for usual 5.1 sound? so this check failed..? On Tuesday, April 19, 2022, Phyllis Smith <[email protected]> wrote:
For Truehd, in bdwrite.C getting error message as shown in second Bold below when 004 ffmpeg truehd new 4 patches added. In addition it shows "case 2" as STEREO although earlier in the code it is "3" as shown immediately below (and COMBO 12 not handled).
BLURAY_AUDIO_FORMAT_MONO = 1, *BLURAY_AUDIO_FORMAT_STEREO = 3,* BLURAY_AUDIO_FORMAT_MULTI_CHAN = 6, BLURAY_AUDIO_FORMAT_COMBO = 12, // Stereo ac3/dts, ... static int bd_audio_format(int channels) { int audio_format = 0; switch( channels ) { case 1: audio_format = BLURAY_AUDIO_FORMAT_MONO; break; case 2: audio_format = BLURAY_AUDIO_FORMAT_STEREO; break; case 6: audio_format = BLURAY_AUDIO_FORMAT_MULTI_CHAN; break; default: fprintf(stderr, *"unknown bluray audio format %d ch\n", channels*); exit(1); } return audio_format; }
On Mon, Apr 18, 2022 at 10:33 PM Andrew Randrianasulu < [email protected]> wrote:
On Tuesday, April 19, 2022, Phyllis Smith <[email protected]> wrote:
Forgot to add that this patch also applied and result checked in even though not yet functional: 0012-truehd-experimental-bluray-profile.patch
On Mon, Apr 18, 2022 at 6:21 PM Phyllis Smith <[email protected]> wrote:
Checked into GIT from megapile_8 the patches related to creating Bluray media by Andrew:
0005-EXPERIMENT-update-bdwrite-for-lpcm-hevc.patch 0006-EXPERIMENTAL-add-variable-chapter-interval-to-bdwrit.patch 0007-EXPERIMENTAL-add-bd_profile-array-to-bdcreate.C-so-i.patch 0008-EXPERIMENTAL-bdcreate-add-m2ts-profiles-part-1.patch 0009-Fix-bdcreate.C-layout.patch 0010-EXPERIMENTAL-attempt-to-get-value-from-popuptextbox. patch0012-truehd-experimental-bluray-profile.patch 0013-EXPERIMENTAL-add-truehd-audio-profile-for-bdcreate.C.patch 0014-EXPERIMENTAL-add-tsmuxer-checkbox.patch 0015-EXPERIMENTAL-fix-layout-in-bdcreate-after-tsmuxer-ch.patch 0016-NOT-YET.patch 0017-EXPERIMENTAL-shell-script-run-tsmuxer-conditionally-.patch 0018-Attempt-at-fixing-bdwrite-stream_type-coding_type-co.patch 0019-EXPERIMENTAL-attempt-at-fixing-bdwrite-on-Android-af.patch 0020-EXPERIMENTAL-real-fix-for-bdwrite.patch 0021-EXPERIMENTAL-try-to-add-lpcm-to-tsmuxer-meta-file.patch 0022-EXPERIMENTAL-really-fix-audio-names-fir-tsmuxer-meta.patch 0037-freebsd-in-bdcreate-bdwrite.patch and from megapile_10: 0077-Better-layout-in-bdcreate.C-after-my-changes.patch
Already in from a previous checkin as related to Bluray from megapile_8 were: 0001-ffmpeg-4.4-bluray-lpcm-encoder.patch 0002-Fix-ffmpeg-4.4-bluray-lpcm-patch-for-5.1.patch 0003-Add-simple-m2ts-lpcm-profile.patch 0011-Rename-m2ts_pcm.m2ts-to-bluray_lpcm.m2ts.patch But because I forgot to test 0004-Improve-truehd-decoder- encoder-from-ffmpeg.git.patch the choice in bdcreate.C for truehd (line 865) was commented out until I can test a smaller set tomorrow. Although *I am confused why the following lines of code (around line 70) in bdcreate.C do not include truehd as an option???* I looked in megapile_10 to see if I missed a related patch but did not see anything. static struct bd_profile { const char *name; } bd_profiles[] = { {"bluray.m2ts"}, {"bluray_lpcm.m2ts"}, };
I think I just did things a little bit different, so may be this piece if code no longer needed - need to look again (just wake up)
and around line 350 (and maybe other places too?): fprintf(fp,"MUXOPT --blu-ray --hdmv-descriptors\n"); fprintf(fp,"V_MPEG4/ISO/AVC, bd.m2ts, track=4113\n"); if(!strcmp(use_profile, "bluray.m2ts")) fprintf(fp,"A_AC3, bd.m2ts, track=4352\n"); if(!strcmp(use_profile, "bluray_lpcm.m2ts")) fprintf(fp,"A_LPCM, bd.m2ts, track=4352\n"); fprintf(fp,"\n"); fclose(fp);
yes, not yet for tsmuxer - not quite sure how it works there (latery I read truehd must come with ac3 'core' embedded, not sure if ffmpeg can output this in one pass? - https://forum.doom9.org/ showthread.php?t=176668&page=61 - comment 1202 and around...)
it will be very interesting to hear from your experiment!
oh YES, you are right, I swapped those unintentionally!
probably explains your bug with chapters?
YOU ARE RIGHT ! now chapters work (hopefully you understand why it takes me so long to test and check in changes as I do not want to add problems -- the oxen is slow, but the earth is patient).
yes, very big thanks for NOT rushing things out!
"lpcm" works too and when I press the "info" button on the remote control on the LG bluray player, it clearly shows it is lpcm -- this not working was my mistake due to not getting the current GIT head.
\0/