Author: Jorge
Date: 12:07:02 07/11/02
Go up one level in this thread
On July 11, 2002 at 08:09:52, pavel wrote: >On July 11, 2002 at 05:49:38, Coxwell Strange wrote: > >> It is readily agreed that nps are not a very good guide for strength >>comparisons of different programs. However, nps do not directly translate to >>depth/time. Depth/time is not necessarily useful when comparing different >>programs, but usually is with the same general programs. With a program that >>displays nps, it has been seen that higher nps can be displayed with smaller >>hash, but the smaller hash increases the time to depth, so in at least that >>case, nps can be inversely related to time/depth. I am not sure if this is true- lower Hash displays Higher Depth/Time. But assuming that it is, are you saying that decreasing hash, increases strength, or am I missing something?? You may take it a little further and find the correlation between Time/Depth and Hash size, and also nps vs. Time/Depth. Is it negatively (inversely proportional to a certain degree) correlated? jorge > >Your above statement just gave us the reason not to take your comments about CM >seriously because you obviously don't make any sense. > >Read what Graham wrote, again. >No point of testing something when you know the result. > >pavs
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.