MAD 0.12.4b is now available.
This version incorporates a number of performance improvements for all platforms.
It also includes (untested) native fixed-point math support for the PowerPC platform, contributed by David Blythe. Since I've made some significant changes to the way the fixed-point math routines are maintained, I'd appreciate feedback from anyone using MAD on a PPC to make sure I haven't broken anything.
Finally, this version changes the way mixed short blocks are handled in Layer III with respect to alias reduction, correcting another apparent ambiguity in ISO/IEC 11172-3. Since mixed blocks are rare (many encoders don't support them) this has until now gone unnoticed. Thanks to H.O. for reporting this.
The source code release is here:
ftp://ftp.mars.org/pub/mpeg/
Winamp users should notice improved performance in the MAD plug-in:
http://www.mars.org/home/rob/proj/mpeg/mad-plugin/
The MAD home page is, of course:
http://www.mars.org/home/rob/proj/mpeg/
Cheers,
Rob Leslie wrote:
MAD 0.12.4b is now available.
This version incorporates a number of performance improvements for all platforms.
It certainly does. The decoder thread CPU usage went down from an average of around 8.5% to 7.5% while playing back stereo 160kbps files on my Celeron 400.
I got these numbers using Another Task Manager (http://web.tiscalinet.it/TbEnrHP/higheng.html) in case anyone wants to give numbers for different processors.
Nice little optimizized release, thanks Rob.
naw
Hello,
On Fri, 9 Feb 2001, David Shin wrote:
It certainly does. The decoder thread CPU usage went down from an average of around 8.5% to 7.5% while playing back stereo 160kbps files on my Celeron 400.
I got these numbers using Another Task Manager (http://web.tiscalinet.it/TbEnrHP/higheng.html) in case anyone wants to give numbers for different processors.
Using the same "setup" (ATM + a 160 kbps file), I get 16.5% instead of 18.5% on a PPro 233.
Nice little optimizized release, thanks Rob.
Sure ! MAD is getting closer and closer to Nitrane (13.5%) ;-)
Bye,
Pierre
It certainly does. The decoder thread CPU usage went down from an average of around 8.5% to 7.5% while playing back stereo 160kbps files on my Celeron 400.
Using the same "setup" (ATM + a 160 kbps file), I get 16.5% instead of 18.5% on a PPro 233.
Nice little optimizized release, thanks Rob.
Sure ! MAD is getting closer and closer to Nitrane (13.5%) ;-)
You're all going to like the next release then too...
Cheers, -rob
-
David Shin -
Pierre Darbon -
Rob Leslie