[FFmpeg-devel] H.264 CABAC Decoder on IBM Cell

Erich Marth destinationcloud
Thu May 7 13:39:30 CEST 2009


Dear ffmpeg Team,

my name is Erich Marth and I'm currently doing some research on porting
ffmpeg or to be more preciously the h.264 codec to the cell architecture.
Till now I do have a good code status, but i do suffer big problems in
doing an optimized version of the cabac entropy decoder.
F.e.: Playing Pirates III (1080p) with CAVLC, an I frame takes
about 18ms to decode, but trying to decode a single slices I frame from
a 720p cabac video takes about 100ms to decode. So I was taking a look
at the current ffmpeg implementation why cabac takes so long on the cell,
and in both files cabac.c and cabac.h there are a lot of asm tweaks for
current
intel and amd processors. Are there any tests on doing the same asm
optimization
for the powerpc port. Are there any other optimizations concerning cabac.
What
was the speedup before the asm routines in asm?

Thx.
Erich Marth



More information about the ffmpeg-devel mailing list