Opened 4 years ago
Closed 3 years ago
#10029 closed defect (fixed)
avc audio quality regression
| Reported by: | ami_stuff | Owned by: | |
|---|---|---|---|
| Priority: | important | Component: | avcodec |
| Version: | git-master | Keywords: | on2avc regression |
| Cc: | Blocked By: | ||
| Blocking: | Reproduced by developer: | yes | |
| Analyzed by developer: | no |
Description
decodes much better with 4.2.7
./ffmpeg_g -i 32khz_mono_56kbps.avi out3.wav
ffmpeg version N-109064-gc124981b79 Copyright (c) 2000-2022 the FFmpeg developers
built with gcc 9 (Ubuntu 9.4.0-1ubuntu1~20.04.1)
configuration:
libavutil 57. 42.100 / 57. 42.100
libavcodec 59. 52.101 / 59. 52.101
libavformat 59. 34.101 / 59. 34.101
libavdevice 59. 8.101 / 59. 8.101
libavfilter 8. 50.100 / 8. 50.100
libswscale 6. 8.112 / 6. 8.112
libswresample 4. 9.100 / 4. 9.100
Input #0, avi, from '32khz_mono_56kbps.avi':
Duration: 00:00:30.08, start: 0.000000, bitrate: 212 kb/s
Stream #0:0: Video: msvideo1 (MSVC / 0x4356534D), rgb555le, 320x240, 144 kb/s, 25 fps, 25 tbr, 25 tbn
Stream #0:1: Audio: avc ([1][5][0][0] / 0x0501), 32000 Hz, mono, fltp, 56 kb/s
Stream mapping:
Stream #0:1 -> #0:0 (avc (on2avc) -> pcm_s16le (native))
Press [q] to stop, [?] for help
Output #0, wav, to 'out3.wav':
Metadata:
ISFT : Lavf59.34.101
Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 32000 Hz, mono, s16, 512 kb/s
Metadata:
encoder : Lavc59.52.101 pcm_s16le
size= 1876kB time=00:00:29.98 bitrate= 512.6kbits/s speed= 947x
video:0kB audio:1876kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.004060%
Attachments (2)
Change History (6)
by , 4 years ago
| Attachment: | 32khz_mono_56kbps.avi added |
|---|
by , 4 years ago
| Attachment: | 24khz_stereo_56kbps.avi added |
|---|
comment:1 by , 4 years ago
| Component: | undetermined → avcodec |
|---|---|
| Keywords: | on2avc regression added |
| Priority: | normal → important |
| Reproduced by developer: | set |
| Status: | new → open |
| Version: | unspecified → git-master |
comment:2 by , 3 years ago
Wait a second. How is that possible? New lavu/tx is more accurate. But yeah, it does happen.
comment:3 by , 3 years ago
comment:4 by , 3 years ago
| Resolution: | → fixed |
|---|---|
| Status: | open → closed |
It now decodes mostly the same, but some bytes are quite different like before there were 0xFF06 and now 0x0007. See f296c765905ca110a05d77a7e919d992800d2e11 and 2eb7df1cbd30d7c104face1ac93bc32aba070d53.
Note:
See TracTickets
for help on using tickets.



Regression since 6ba0aa1770ba29eb4126c6a706f6b0cd3809648f