[MPlayer-dev-eng] Re: PATCH [0/12] CoreAVCDecoder support

[Alan Nisota]

Michael Niedermayer <michaelni <at> gmx.at> writes:
> > Here are some benchmarks on Linux using a 2.8GHz P4 w/ HT enabled (adds a 
few 
> > fps with the dual-cpu patch).  I don't have a dual-cpu system to benchmark 
> > against, but others have confirmed that a dual-cpu system just about 
doubles 
> > performance with this codec.  I also have a 1.8 GHz Pentium-M which tested 
> > anywhere from the same to 20% slower on Windows compared to the P4.
> > Using CoreAVC 1.2.0
> 
> yes but what about ffh264 on Pentium M ? knowing CoreAVC is slower is one 
thing
> but it would be more interresting to also know how fast/slow ffh264 is on 
> Pentium M
Why?  either you didn't understand the benchmarks I posted, or you are asking 
an uninteresting question to me.  The benchmarks compare ffh264 to coreAVC 
(since I'm advocating CoreAVC) on my linux system.  I mentioned in passing that 
the Pentium M at 1.8GHz was slower in some cases and the same in others, which 
is just an FYI.  CoreAVC has similar performance gains on the Pentium M in 
Windows as it does on the P4 in Linux, and you'll just have to take my word on 
that.  If you want ffh264 benchmarks comparing the Pentium M and 
P4...well...I'm not going to do them.  They aren't very useful (too many 
variables)...Benchmarking mplayer on different machines is something anyone can 
do.

Back on topic, I have a new set of patches that address many of the criticisms 
I received the 1st time around.  There is now nothing in them specific to 
CoreAVC (well, you could make a case that the PECompact stuff is).  I am still 
working on figuring out why the current stub implementation doesn't work 
properly, but I'll repost once I have that figured out.