Age | Commit message (Collapse) | Author |
|
CVS patchset: 6826
CVS date: 2004/07/20 21:38:34
|
|
* make some helper functions non-inline that Bastien had to move because of
their inlining (I realized that some helpers are a bit too large for inlining,
so they might worsen the I-cache usage.)
* properly use logging macros in all DXR3 code
* prepare SPU decoder for new (and hopefully correct) SPU forcing; this
should fix some longstanding DXR3 SPU problems, but the code needs more testing
so it is disabled for now
CVS patchset: 6824
CVS date: 2004/07/20 16:37:44
|
|
CVS patchset: 6811
CVS date: 2004/07/18 17:36:25
|
|
CVS patchset: 6809
CVS date: 2004/07/18 15:34:57
|
|
CVS patchset: 6808
CVS date: 2004/07/18 15:31:29
|
|
Fix "hurry_up" behavior (keep the metronom happy).
Enable DR1 for the mpeg12 decoder.
Remove all the mpeg-es parsing from here, use the new parser instead.
Handle frame format changes (width, height and aspect ratio)
Tested with all my mpeg streams, and with some DVDs with still menus.
Enjoy ;)
CVS patchset: 6805
CVS date: 2004/07/18 00:50:02
|
|
CVS patchset: 6766
CVS date: 2004/07/03 16:47:55
|
|
CVS patchset: 6762
CVS date: 2004/07/01 20:56:15
|
|
CVS patchset: 6761
CVS date: 2004/07/01 20:16:24
|
|
Let's see tomorrow how many segfaults are gone ;-)
CVS patchset: 6759
CVS date: 2004/06/28 22:44:57
|
|
* remove two unneeded files
CVS patchset: 6727
CVS date: 2004/06/21 16:37:00
|
|
CVS patchset: 6681
CVS date: 2004/06/13 13:23:58
|
|
CVS patchset: 6680
CVS date: 2004/06/13 00:10:47
|
|
gcc 2.96 does not. this causes a bunch of undefined symbols:
undefined symbol: ff_init_me (../.libs/xineplug_decode_ff.so)
undefined symbol: ff_estimate_p_frame_motion (../.libs/xineplug_decode_ff.so)
undefined symbol: ff_fix_long_p_mvs (../.libs/xineplug_decode_ff.so)
undefined symbol: ff_fix_long_mvs (../.libs/xineplug_decode_ff.so)
undefined symbol: ff_h263_encode_motion (../.libs/xineplug_decode_ff.so)
undefined symbol: h263_encode_init (../.libs/xineplug_decode_ff.so)
Mike, can you merge this back to ffmpeg cvs?
CVS patchset: 6677
CVS date: 2004/06/12 22:14:13
|
|
initializations
CVS patchset: 6672
CVS date: 2004/06/11 02:25:56
|
|
CVS patchset: 6669
CVS date: 2004/06/10 20:46:25
|
|
CVS patchset: 6664
CVS date: 2004/06/10 04:46:18
|
|
set, it's optional
CVS patchset: 6649
CVS date: 2004/06/06 16:13:30
|
|
Daniel: you don't even need libw32 decoder to play your stream now. ;)
http://naboo.homelinux.org/~daniel/media/p_length.avi
CVS patchset: 6643
CVS date: 2004/06/05 16:37:37
|
|
CVS patchset: 6637
CVS date: 2004/06/04 13:29:29
|
|
the pattern of dummy in alpha, mmx etc.
CVS patchset: 6625
CVS date: 2004/06/01 14:36:34
|
|
CVS patchset: 6612
CVS date: 2004/05/30 19:24:17
|
|
CVS patchset: 6582
CVS date: 2004/05/23 16:34:42
|
|
developers have been informed.
CVS patchset: 6541
CVS date: 2004/05/15 14:23:15
|
|
CVS patchset: 6538
CVS date: 2004/05/14 13:36:03
|
|
Tested with mpeg1 and mpeg2 streams.
I've not checked if it's slower than libmpeg2...
Change the ffmpeg deocoder priority to test it.
CVS patchset: 6522
CVS date: 2004/05/12 07:41:43
|
|
quick fix for wma static noise.
(a more general solution should be employed in configure.ac)
CVS patchset: 6520
CVS date: 2004/05/11 21:31:55
|
|
CVS patchset: 6511
CVS date: 2004/05/09 23:31:59
|
|
no memcpy is needed if the fifo buffer contains a full frame, it's true most of the time.
Tested with avi, asf, mov, mpeg1/2 (broken for years) streams.
CVS patchset: 6502
CVS date: 2004/05/09 14:22:54
|
|
CVS patchset: 6447
CVS date: 2004/04/26 19:33:47
|
|
CVS patchset: 6446
CVS date: 2004/04/26 18:31:57
|
|
CVS patchset: 6445
CVS date: 2004/04/26 18:03:01
|
|
please everybody watch out for spelling (and other) errors
CVS patchset: 6442
CVS date: 2004/04/26 17:50:06
|
|
CVS patchset: 6441
CVS date: 2004/04/26 15:38:18
|
|
CVS patchset: 6437
CVS date: 2004/04/25 18:57:04
|
|
CVS patchset: 6416
CVS date: 2004/04/19 21:18:01
|
|
playback of some mov's that only use part of the endoded frame
CVS patchset: 6415
CVS date: 2004/04/19 21:03:02
|
|
CVS patchset: 6406
CVS date: 2004/04/15 15:49:56
|
|
CVS patchset: 6284
CVS date: 2004/03/17 20:30:27
|
|
CVS patchset: 6283
CVS date: 2004/03/17 20:24:16
|
|
systems. The problem was little-endian systems were doing byte
swapping which big-endian systems do not need to do.
CVS patchset: 6282
CVS date: 2004/03/17 17:03:26
|
|
CVS patchset: 6280
CVS date: 2004/03/16 23:31:30
|
|
* increase ffmpeg from 5 to 6
* increase some others from 6 to 7 (they should still override ffmpeg)
* increase liba52 from 2 to 5 (since 5 is somewhat default)
* decrease win32 DLL Qt from 5 to 1 (don't worry, this is the only decoder
to handle these types, so this is not really a change; it's just that
all the other win32 decoders use prio 1)
CVS patchset: 6272
CVS date: 2004/03/16 14:12:03
|
|
CVS patchset: 6262
CVS date: 2004/03/15 01:22:22
|
|
CVS patchset: 6259
CVS date: 2004/03/14 21:37:52
|
|
CVS patchset: 6255
CVS date: 2004/03/14 21:24:21
|
|
CVS patchset: 6254
CVS date: 2004/03/14 21:21:11
|
|
CVS patchset: 6253
CVS date: 2004/03/14 21:14:07
|
|
CVS patchset: 6252
CVS date: 2004/03/14 21:07:01
|
|
CVS patchset: 6247
CVS date: 2004/03/13 19:42:22
|