[MPlayer-dev-eng] [PATCH] fast linear resample in af_resample

[Balatoni Denes]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Anders (and everybody)!

I attempted to fix the sample omition, and I put everything in af_resample.

If you notice anything stupid, or have comments do tell me!

btw: here are 3 benchmarks  (normal, sloppy, fast) on my zaurus /w intel 
sa-1100. Fast is 140% faster, but it is still just 3.5 % cpu improvement - but
an improvement none the less. Also with sloppy it takes 3 secs to build 
filter, with normal it takes 70 secs, with fast 0 :)

Building audio filter chain for 11127Hz/1ch/8bit -> 44100Hz/2ch/16bit...

BENCHMARKs: VC:   0.431s VO:   0.604s A:   2.149s Sys:  26.506s =   29.690s
BENCHMARK%: VC:  1.4531% VO:  2.0329% A:  7.2378% Sys: 89.2762% = 100.0000%
BENCHMARKn: disp: 241 (8.12 fps)  drop: 0 (0%)  total: 241 (8.12 fps)

BENCHMARKs: VC:   0.423s VO:   0.604s A:   1.714s Sys:  27.029s =   29.770s
BENCHMARK%: VC:  1.4210% VO:  2.0293% A:  5.7563% Sys: 90.7935% = 100.0000%
BENCHMARKn: disp: 241 (8.10 fps)  drop: 0 (0%)  total: 241 (8.10 fps)

BENCHMARKs: VC:   0.422s VO:   0.600s A:   0.704s Sys:  28.268s =   29.995s
BENCHMARK%: VC:  1.4085% VO:  2.0006% A:  2.3484% Sys: 94.2426% = 100.0000%
BENCHMARKn: disp: 241 (8.03 fps)  drop: 0 (0%)  total: 241 (8.03 fps)

bye
Denes



- -- 
"Use the source Luke !"
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE929SMaQToeq3jgogRAmyFAJ4zspa4kcWao8sRpVSxjaHVBQBxqgCeOSFb
z8qltABMSeNf0VYBBaaeeF8=
=eJh/
-----END PGP SIGNATURE-----
-------------- next part --------------
A non-text attachment was scrubbed...
Name: aflinintv1.diff
Type: text/x-c++
Size: 3294 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/attachments/20021120/c40830eb/attachment.bin>