[Vorbis] Re: [vorbis] Different average bitrates on different machines

[Cameron Simpson]

On Sun, Dec 09, 2001 at 11:10:49PM -0700, Jack Moffitt <jack at xiph.org> wrote:
| > I have access to two machines
| > 	Pentium 3 450mhz, 192mb, Windows 2000 (SP2)
| > 	Athlon 1400mhz, 256mb, Windows 2000 (SP2)
| > When I encode the same wave file on each machine at the same nominal
| > bitrate (96kbps), I get output files with different average bitrates.
| > 107kbps on the Athlon, as opposed to 84kbps on the Pentium 3.
| 
| That does seem a little strange.  Normally you could attribute some
| difference to the compiler the OS etc.

Care to elaborate on this second sentence.  I'd have imagined that
the same input data and the same source code should give exactly the
same encoding and bitrate regardless of the compiler or OS, unless the
compiler has bugs.

The only things I can imagine affecting this would be:

        - quality of data obtained from the CD (if that's the source)

        - floating point rounding effects, which I would hope
          would be VERY VERY small

Is there something I'm missing here?

BTW, has anyone done a scan of the source with gcc 3.x with the
-Wsequence-point warning enabled? There's been a fair bit of discussion
recently about dodgy code doing things like:

        a[i]=f(i++)

which gives undefined results. That really could screw up encoding algorithms.

-- 
Cameron Simpson, DoD#743        cs at zip.com.au    http://www.zip.com.au/~cs/

Indeed! But do not reject these teachings as false because I am crazy. The
reason that I am crazy is because they are true.
	- Malaclypse the Younger

--- >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.