Mercurial > mplayer.hg

changeset 25195:1aec672af2d2

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Fix ao_null with float samples ao_null accepts float input, but the code calculating ao_data.bps only checked for 1-byte formats and used samplesize 2 for everything else. Because ao_null uses the bps value in its timing calculations this effectively made "playback" advance at half the correct speed. Fixed by calculating samplesize with af_fmt2bits() instead.
author uau
date Sat, 01 Dec 2007 05:17:08 +0000
parents e816d546c4fe
children 5487d363a9ac
files libao2/ao_null.c
diffstat 1 files changed, 1 insertions(+), 1 deletions(-) [+]
line wrap: on
line diff
--- a/libao2/ao_null.c	Sat Dec 01 01:39:39 2007 +0000
+++ b/libao2/ao_null.c	Sat Dec 01 05:17:08 2007 +0000
@@ -50,7 +50,7 @@
 // return: 1=success 0=fail
 static int init(int rate,int channels,int format,int flags){
 
-    int samplesize = (format == AF_FORMAT_U8 || format == AF_FORMAT_S8) ? 1: 2;
+    int samplesize = af_fmt2bits(format) / 8;
     ao_data.outburst = 256 * channels * samplesize;
     // A "buffer" for about 0.2 seconds of audio
     ao_data.buffersize = (int)(rate * 0.2 / 256 + 1) * ao_data.outburst;