Author: Carsten Kossendey
Date: 19:56:04 01/27/98
Go up one level in this thread
On January 27, 1998 at 22:41:12, Richard A. Fowell (fowell@netcom.com) wrote: >On another topic, I have another result in the >Crafty 14.5 "a3" test, that is to say: >(book off, ponder off, hash 6M, hashp 2M, sd 9, a3) > >84654 nps - Mac G3/300 (cpu 100%) >41608 nps - Mac 7300/180 (604e) (cpu 100%) The Mac port will always report 100% CPU usage no matter what. I didn't feel like going into the details of the PPC "Performance Manager" implementation. 100% is usually correct however. Crafty/Mac isn't quite multitasking friendly (actually it simply hogs the CPU) so the only slowdown should be caused by DMA accesses and such - Apple's Video Player chops somewhat off of those 100% for example. >Last I checked, the Alphas were doing the best on this test - what's >an Alpha number like these days? The Alphas were doing best in raw performance, but not in nps/clock ;) Anyhow I guess it's somewhere between 150k and 225k nps. >(terminal output for the G3 was:) >time: 35.81 cpu:100% mat:0 n:3031491 nps:84654 > ext-> checks:51451 recaps:19782 pawns:24 1rep:44871 > predicted:0 nodes:3031491 evals:1220247 > endgame tablebase-> probes done: 0 successful: 0 > hashing-> trans/ref:20% pawn:92% used:w96% b92%
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.