[vorbis] Performance tests

[Erik Olofsson]

I thought so to. The same machine used to play mp3s at 10 % a couple of
years ago. But the new winamp decoder seems to rock pretty much. I measure
the util in task manager, and when I play mp3s winamp cpu time is lower than
ICQ for example (cpu time being the amount of time spent in winamp). Also I
have a vortex 2 soundcard, so the mixing should be done in hardware. Using
wave out as output cpu util goes way up.

-----Original Message-----
From: Michael Smith [mailto:msmith at labyrinth.net.au] 
Sent: den 16 oktober 2001 15:19
To: vorbis at xiph.org
Subject: Re: [vorbis] Performance tests

At 03:12 PM 10/16/01 +0200, you wrote:
>I did some testing of vorbis codec (Winamp plugin) on a Pentium 166 MMX,
and
>found that the performance is pretty good. I got 3 % cpu utisization if I
>turned dithering off, and 5 % otherwise. This was with a 138 kbit song.
It's
>pretty long to mp3 performance that was 0 to 1 % on that machine for a 128
>kbit song. 

These figures (both for mp3 and for vorbis) are way, way too low to even
be vaguely correct. Your measurement techniques are probably suspect,
playing
more things at once probably just happens to make them get closer to
reality.

Michael

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.