[vorbis] Decoder speed: Test on slow machine

[Maik Merten]

"David K. Gasaway" schrieb:
> For comparison, how well does this machine play MP3s?  I recall forcing WinPlay3 downmix to
> mono to play songs on my 486 DX4-100.  Of course, I suppose today's decoders are bit more
> optimized. :)

Hello David,

I only did some MP3 tests with 128 Kbit/s files. No problem. There even
was enough CPU power to keep the Winamp-GUI alive - and simple
Visualization. The Windows-Tool "Systemmonitor" (It may be
called different in non-localized Windows-versions) showed
about ~50% CPU-usage without Visualization.
Playback of 192 Kbit/s MP3 files could be possible.

Playback of a ~128 KBit/s Vorbis file used ~70-90% CPU-time
(no Vis). But as Vorbis is VBR CPU usage varies.

My conclusion:
The RC2 decoder is fast. It surely is faster than average/old
MP3-decoders. But at the time being it is not as optimized
as the best optimized MP3-decoders (I am speaking of Winamp
2.76 MP3-decoder, not Nitrane!). This is nothing new.

On the other hand: A ~96 KBit/s Vorbis file sounds as good/better
than a 128 KBit/s MP3 file. Decoding of "equal quality files"
could already be at "equal speed". (I hope you know what I mean...)
>From this point of view decoding speed isn´t an issue anymore.
(I don´t say further optimization is pointless!).

I´m looking forward to do some tests with the RC3 decoder!

bye,

Maik Merten


-- 
  MetalUnleashed: http://www.metalunleashed.de
DHTML-Bibliothek: http://www.fortunecity.de/wolkenkratzer/flick/403
             ICQ: 45648209

--- >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.