[ptx] x64 benchmarks

Pablo d'Angelo pablo.dangelo at web.de
Tue Jan 17 08:27:08 GMT 2006


Marek Januszewski schrieb:
> Hello,
> 
>  So how about we set up some big benchmarking project and we all post
> results here?

I always had the feeling that the AMD64 bit version run a bit faster.

While I haven't set up a complete benchmark project, but I have remapped a
single image from a project (with lots of black, empty space). No crop
(nona) or acceleration (PTmender) was used.

This are the user times:
             x86 32 bit (s)    AMD64 (s)      speedup
PTStitcher	35.495
PTmender	35.675		28.995		23%
nona            27.588		22.179		24%

So at least for panoramic stitching, actually using the 64 bit instructions
on that the AMD64 provides seems to improve remapping speed by ~ 20%

Note: if fast transform or cropped tiff output would have been used, the
times would differ a lot. For smaller panos, the fast transform code used by
PTStitcher and PTmender would blow nona away, while for larger panos, nona
would be faster

If there is interest, I can provide the example. However, for very fast
machine, the IO bandwith will probably the limiting factor

Details:

I'm running ubuntu breezy on and AMD64 (almost 2 years old) machine with 1.5
GB of memory. All programs have been run during the same session. cpu
frequency was fixed to 2000 MHz during the test.

> gcc -v
Target: x86_64-linux-gnu
gcc-Version 4.0.2 20050808 (prerelease) (Ubuntu 4.0.1-4ubuntu9)

> gcc -v
Target: i486-linux-gnu
gcc version 4.0.2 20050808 (prerelease) (Ubuntu 4.0.1-4ubuntu9)


This is what /proc/cpu said about my processor:

vendor_id       : AuthenticAMD
cpu family      : 15
model           : 4
model name      : AMD Athlon(tm) 64 Processor 3000+
stepping        : 8
cpu MHz         : 2002.608
cache size      : 512 KB

ciao
  Pablo


More information about the ptx mailing list