Author: Gian-Carlo Pascutto
Date: 22:57:21 07/19/05
Go up one level in this thread
On July 19, 2005 at 17:58:48, Anson T J wrote: >Or measuring if an engine scales better on an Intel or AMD chip based on the NPS >given that its the same engine? I think this is somewhat valid, for pretty much the reason Vincent mentioned: you won't get a good speedup without good scaling. >If you would like to do 10+ tests and post the NPS and time-to-depth please be >my guest. Its a lot more valid, but more to ask for than a simple test with 1 >then 2 threads. Scaling can also vary from positions to position, 1 position is not enough. There is no easy and quick way to measure things with deep engines, forget about that. -- 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.