Page 1 of 1

h264 video regression between 0.8.2-test3 and 0.8.2

Posted: 20 Aug 2005 21:38
by disposable2004
first...i have to say that VLC is great! 0.8.2 has fixed my G5 audio issues.

problem:
a. 0.8.2 has no video output while playing h264 mp4 encoded using handbrake 0.7.0-beta3. (see messages below)
b. the exact same file works fine with vlc 0.8.2-test3 and also with quicktime 7.
c. the exact same symptoms happen on windows XP SP2 (fully patched) and with mac osx 10.3.9.
d. tried test with 3 different sets of videos with exact same results

note:
e. h264 videos encoded by handbrake 0.7.0-beta2 work fine in vlc 0.8.2
f. simple (non-h264) mp4 videos playback fine in all vlc versions
g. i am assuming that 0.8.2 is "newer" than 0.8.2-test3. please advise if this is not the case.

i can email a short snippet to a developer if needed. please PM me with an email address if so desired.

thanks

i get the following messages under 0.8.2. i stopped playback and the lines beginning with "main debug" showed up:
fmpeg warning: cannot decode one frame (1535 bytes)
<tons snip>
ffmpeg warning: cannot decode one frame (1149 bytes)
ffmpeg warning: cannot decode one frame (1001 bytes)
main debug: control type=0
main debug: control: stopping input
main debug: closing input
mp4 debug: freeing all memory
main debug: unlocking module "mp4"
main debug: unlocking module "access_file"
ffmpeg warning: cannot decode one frame (1020 bytes)
ffmpeg debug: ffmpeg codec (h264) stopped
main debug: unlocking module "ffmpeg"
main debug: thread times: real 0m11.109375s, kernel 0m0.000000s, user 0m0.000000s
main debug: thread 3648 joined (src/input/decoder.c:191)
main debug: killing decoder fourcc `avc1', 2 PES in FIFO
main debug: unlocking module "faad"
main debug: thread times: real 0m11.109375s, kernel 0m0.125000s, user 0m0.015625s
main debug: thread 3668 joined (src/input/decoder.c:191)
main debug: killing decoder fourcc `mp4a', 0 PES in FIFO
main debug: unlocking module "bandlimited_resampler"
aout_directx debug: CloseAudio
aout_directx debug: DirectSoundThread exiting
main debug: thread times: real 0m11.000000s, kernel 0m0.031250s, user 0m0.015625s
main debug: thread 4208 joined (directx.c:605)
main debug: unlocking module "aout_directx"
main debug: unlocking module "trivial_mixer"
main debug: thread times: real 0m11.218750s, kernel 0m0.000000s, user 0m0.062500s
main debug: thread 3580 joined (src/input/input.c:386)

Re: h264 video regression between 0.8.2-test3 and 0.8.2

Posted: 20 Aug 2005 21:51
by disposable2004
c. the exact same symptoms happen on windows XP SP2 (fully patched) and with mac osx 10.3.9.
to clarify, all combinations of files and versions fail and succeed exactly the same on both windows XP SP2 and mac osx 10.3.9.

Posted: 23 Aug 2005 17:56
by The DJ