<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
<title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
Alfred E. Heggestad wrote:
<blockquote cite="mid1109662472.30842.3.camel@pluto" type="cite">
<pre wrap="">On Mon, 2005-02-28 at 19:42 -0500, Jean-Marc Valin wrote:
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">jean-marc: i think we can remove spx_sig_t *orig.
but am not sure about exc2Buf. is it for extension?
</pre>
</blockquote>
<pre wrap="">orig is already removed in SVN (which you should probably use). As for
exc2, it can be removed, but I'm not sure if you can just use exc
instead (maybe yes).
</pre>
</blockquote>
<pre wrap=""><!---->when removing "spx_sig_t *orig;" in the encoder, the stack usage
went from 17091 to 16451 bytes,
I have already got a working decoder running on my Ubicom IP3K,
using 10-15% of the CPU, with a decoder stack size of 4058 bytes.
The encoder is next..
</pre>
</blockquote>
<br>
It will probably be very close, if you can do it.<br>
<br>
I did a bunch of tests of encoder performance (on an Athlon XP 1700+) a
while back, and found that, for 8kbps CBR complexity 1, I got:<br>
<br>
(encode/decode speed)<br>
basic compile: 29.2x / 168x<br>
<br>
-O3 -ffast-math -funroll-all-loops<br>
-march=pentium3 -fprefetch-loop-arrays -fsingle-precision-constant:
53.4x / 361x<br>
<br>
With -msse: 65x/386x<br>
<br>
<br>
The actual numbers don't matter much, but it does mean that encoding is
almost 6x as expensive as decoding, so expect that you're going to use
60-90% of your CPU for speex encode..<br>
<br>
-SteveK<br>
<br>
<br>
<br>
</body>
</html>