On Sun, 2003-11-09 at 21:00, Jean-Marc Valin wrote:
On recent x86 processors, floating point is faster than fixed-point.
Jean-Marc
This left me something shocked. Please, can you tell me what kind of
processors are showing this behaviour? Are you referring to speex codec
performance or to general performance?
Sorry to drive this a little off-topic but my XP1800 seems to be much
faster with integer (when this is applicable).
Thank you
Massimo
<p>--- >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
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body. No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.
Le dim 09/11/2003 à 21:32, Rodrigo Parra M a écrit :> I'm developing a Personalized system of news audible for PDAs(Pocket PC > and Palm OS), I have somes PDAs with differents processors(Samsung > (2410) 266 MHz, Intel Xscale PXA 250 to 200 and 400 MHz, OMAP1510) with > Pocket PC 2002 and 2003, and Palm OS 5 operating system. I offer myself > to test any application of speex that runs in these machines, only you > must send me the source code.No problem. Get it here: http://www.speex.org/download/speex-1.1.1.tar.gz If you want bleeding edge, you can try what's in CVS. Jean-Marc -- Jean-Marc Valin, M.Sc.A., ing. jr. LABORIUS (http://www.gel.usherb.ca/laborius) Université de Sherbrooke, Québec, Canada -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 190 bytes Desc: Ceci est une partie de message numériquement signée Url : http://lists.xiph.org/pipermail/speex-dev/attachments/20031109/37346ef5/signature-0001.pgp
Le dim 09/11/2003 à 14:33, Steve Kann a écrit :> Just out of curiosity, has anyone profiled the difference between the > floating point and fixed point implementations on processors with > decent floating point support? (i.e. x86, PPC).On recent x86 processors, floating point is faster than fixed-point. Jean-Marc -- Jean-Marc Valin, M.Sc.A., ing. jr. LABORIUS (http://www.gel.usherb.ca/laborius) Université de Sherbrooke, Québec, Canada -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 190 bytes Desc: Ceci est une partie de message numériquement signée Url : http://lists.xiph.org/pipermail/speex-dev/attachments/20031109/fbff66dd/signature-0001.pgp
I'm developing a Personalized system of news audible for PDAs(Pocket PC and Palm OS), I have somes PDAs with differents processors(Samsung (2410) 266 MHz, Intel Xscale PXA 250 to 200 and 400 MHz, OMAP1510) with Pocket PC 2002 and 2003, and Palm OS 5 operating system. I offer myself to test any application of speex that runs in these machines, only you must send me the source code. Regards. Rodrigo Parra M. MSc.(c), Computer Engenieer. Pontific Catholic University of Chile -----Mensaje original----- De: owner-speex-dev@xiph.org [mailto:owner-speex-dev@xiph.org] En nombre de Jean-Marc Valin Enviado el: Domingo, 09 de Noviembre de 2003 17:01 Para: speex Asunto: Re: [speex-dev] XScale realtime encoding possible? Le dim 09/11/2003 à 14:33, Steve Kann a écrit :> Just out of curiosity, has anyone profiled the difference between the> floating point and fixed point implementations on processors with > decent floating point support? (i.e. x86, PPC).On recent x86 processors, floating point is faster than fixed-point. Jean-Marc -- Jean-Marc Valin, M.Sc.A., ing. jr. LABORIUS (http://www.gel.usherb.ca/laborius) Université de Sherbrooke, Québec, Canada --- >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 'speex-dev-request@xiph.org' containing only the word 'unsubscribe' in the body. No subject is needed. Unsubscribe messages sent to the list will be ignored/filtered.
Le lun 10/11/2003 à 11:19, Massimo a écrit :> On Sun, 2003-11-09 at 21:00, Jean-Marc Valin wrote: > On recent x86 processors, floating point is faster than fixed-point. > > Jean-Marc > > This left me something shocked. Please, can you tell me what kind of > processors are showing this behaviour? Are you referring to speex > codec performance or to general performance? > Sorry to drive this a little off-topic but my XP1800 seems to be much > faster with integer (when this is applicable).There are two reasons. First, integer multiplications tend to take a little longer than float. The second reason is that in order to convert from float to int, there are several cases where one float operation is replaced by more than one integer operation. Jean-Marc -- Jean-Marc Valin, M.Sc.A., ing. jr. LABORIUS (http://www.gel.usherb.ca/laborius) Université de Sherbrooke, Québec, Canada -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 190 bytes Desc: Ceci est une partie de message numériquement signée Url : http://lists.xiph.org/pipermail/speex-dev/attachments/20031111/5a00fbbe/signature-0001.pgp