1a771ae671
Merge the stream struct into the parent decoder
12 years ago
5fff1c4e47
Update the ffmpeg decoder to use avcodec_decode_audio4
12 years ago
5f4c33f896
Only store one packet at a time
12 years ago
9d86890d9d
Only use one stream for the ffmpeg decoder
12 years ago
a3bd3a40ca
fix 2 unrelated leaks
12 years ago
18623a5efe
Merge branch 'videoplayback' of https://github.com/ChrisKCat/openmw into videoplayback
12 years ago
edf18a7d6e
change destruction order to fix crash on exit when a video is playing
12 years ago
a6c3e06e54
fixed fog in some cells
12 years ago
c869444dcf
Don't leak the IO context if avformat_open_input fails
12 years ago
63e86555b6
use sample_aspect_ratio if available
12 years ago
5ed04ae53e
added black bars
12 years ago
e6d27cd6d0
Merge remote-tracking branch 'scrawl/videoplayback' into videoplayback
12 years ago
9b3cf5c159
Use a looping thread to trigger refreshes
12 years ago
fa1ad381da
Make sure packets are cleaned up properly
12 years ago
7477345481
fixed video material
12 years ago
db23c8152e
Only duplicate AVPackets as needed
...
Packets that don't have a destruct method are using static memory, which will
only be valid until the next av_read_frame call. Otherwise, it's already
dynamically allocated and will remain valid.
12 years ago
6008cf0d15
Remove unneeded video_current_pts field
12 years ago
da44141b95
Avoid creating extra texture unit states on the video material
12 years ago
d50698d7d1
Clean up the rectangle and scene node used for displaying the video
12 years ago
eb0e8d9e37
Simplify PacketQueue::get
12 years ago
b41a77648e
Avoid re-reading the source sample to duplicate, to avoid pointer aliasing
12 years ago
62a995d492
Calculate audio sync once per read
12 years ago
32b0a5d96e
Update the queued sample count immediately
12 years ago
cbdf6ad595
Move more stuff to where it should be, and improve cleanup
12 years ago
71ff90aaee
Don't use sub-frame timing for the video clock
12 years ago
582efcdb9b
Always try to resync if the clock difference is large
12 years ago
f97eaec7ab
Consolidate some code
12 years ago
d66d8a3118
Don't assume we can write beyond the end of the sample buffer
12 years ago
7332ffb0f8
Let the wanted sample size go down to 0
12 years ago
05c6483257
Fix external clock
12 years ago
2f37d31108
Move some definitions into the source file they're used in
12 years ago
a6e627001a
Avoid a for(;;) construct
12 years ago
4d6c05f6cc
Tighten audio skew allowance
12 years ago
8db5d10f10
Avoid showing a video picture if we're late
...
Ideally we should skip decoding, or at least YUV->RGB conversion, too.
12 years ago
157cb10f56
Fix a 16-bit audio assumption
12 years ago
c2b711d195
Move some more methods to the class they're part of
12 years ago
26a09ee7ba
Move some methods into their respective class
12 years ago
5221298a7f
Move a couple packet queue methods into the struct
12 years ago
90294c589b
Use a volatile bool for the refresh
12 years ago
f555dc60eb
Reduce some indentation
12 years ago
6cedd64509
Fix audio sync correction sizes
12 years ago
e291d0943d
Merge remote-tracking branch 'scrawl/videoplayback' into videoplayback
12 years ago
606fb982a8
Update to use avcodec_decode_audio4
12 years ago
82564e07c7
fix crash when video file doesn't exist
12 years ago
27cd9ff732
Revert "Use the decoded frame pts when available"
...
This reverts commit cab68df257
12 years ago
9d6f656822
fixed ogre resource functions
12 years ago
43481ad117
Use the external clock by default
12 years ago
f95e72166c
hide type column in non-mixed type tables
12 years ago
faa5ef0874
rewrote column class
12 years ago
0a5ab977b7
Use the decoder's sample offset for calculating the stream offset
12 years ago
Chris Robinson
scrawl
Marc Zinnschlag