Author: Gian-Carlo Pascutto
Date: 10:28:16 08/09/03
Go up one level in this thread
On August 07, 2003 at 15:08:29, Eugene Nalimov wrote: >Unfortunately, "clock per clock" rating methodology is wrong. It's only 'wrong' when you don't understand what is being talked about. I don't have access to the latest and greatest of both the Athlon, P4 and Opteron. I can do a clock for clock measurement on the ones I _do_ have AND look at the fastest available clockspeed for each CPU and base my conclusion on that. Which is exactly what I have done. My conclusion is that the Opteron 2.0Ghz is faster than anything else, even though the Pentium 4 runs at 3.2Ghz in its fastest incarnation. The only problem with this measurement is that performance may not scale linarly. For the things discussed here, it scales linarly enough. >PS. I believe you did not provide another interesting data: speed of your >program running *on the same CPU* when compiled with 32- and 64-bit compilers. >AMD64 runs 32-bit applications natively with high speed (unlike Itaniums), so >this way you can measure just effect of 64- vs. 32-bit data and 16 vs. 8 >registers. As I already said, I didn't measure it because it is irrelevant for me. I can run natively at 64 bits, no need for outdated 32 bit support. -- GCP
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.