MOVNTQ on Athlon

From: CHK (b_at_b.biz)
Date: 05/31/04


Date: Mon, 31 May 2004 01:48:33 +0000 (UTC)

I have a bunch of image processing functions optimized for Pentium III in
which I use MOVNTQ instruction quite often. The goal is to prevent cache
pollution.

The size of each image is much bigger than cache size, so apparently there
is no point in caching output of the functions.

This code also runs on dual Athlon MP system, but looks like the Athlon is
choking on this instruction. MOVNTQ takes lot more cycles on Athlon compared
to Pentium-III. In fact I get better speed on Athlon when I use MOVQ.

Can somebody enlighten me what's going on?



Relevant Pages

  • Re: MOVNTQ on Athlon
    ... > I have a bunch of image processing functions optimized for Pentium III in ... > which I use MOVNTQ instruction quite often. ... > This code also runs on dual Athlon MP system, but looks like the Athlon is ... MOVNTQ takes lot more cycles on Athlon compared ...
    (comp.lang.asm.x86)
  • Re: Running D2005 on a laptop
    ... The Pentium M is superior to any P4 based laptop. ... Stay away from P4 based solutions, stick with Pentium M or Athlon 64. ... >> spend all your money on RAM. ... > The higher end CPU's have more RAM in their cache, ...
    (borland.public.delphi.non-technical)
  • Timings issue
    ... I've got access to some crypto code mostly C, ... Pentium IV, and 2.4.18 for Athlon. ... one would expect from their clock frequency difference. ...
    (sci.crypt)
  • What did I screw up?
    ... I messed up some configuration, somewhere, and I hope that someone can tell ... I have a 2 computer network. ... I have a menu option which telnets from the Athlon to the Pentium and dials ...
    (Debian-User)
  • Re: CPU benchmark for Xilinx PAR
    ... done a comparision recently but we have found on previous benchmarking is ... It would be interesting to try a single core Pentium Extreme against the ... > Here's a benchmark for PAR running on two different ... > then an Athlon X2 4800+ would really scream. ...
    (comp.arch.fpga)