Author: Robert Hyatt
Date: 11:19:25 05/22/99
Go up one level in this thread
On May 22, 1999 at 11:41:21, José de Jesús García Ruvalcaba wrote: >On May 22, 1999 at 00:25:52, Robert Hyatt wrote: > >[snip] >> >>I'm not really assuming anything at all, because search loss is _not_ >>constant. It is easily possible (and 100% probable) that many searches >>with N processors run N times faster than with 1. I see this regularly. >>Yes, there are cases where it is less. But in terms of NPS, they 'delivered' >>what they said. Because when I quote NPS figures for Crafty (as does everyone >>with a parallel search) I give "raw nps" numbers. Because 'effective nps' is >>impossible to calculate. >> >[snip] > > Is it really impossible to calculate 'effective nps'? It seems that you can >count duplicate nodes, and also nodes from branches which were not needed, and >the remaining ones are the 'effective nodes'. Is there some difficulty in >counting 'effective nodes? it is really impossible. Because some of those extra nodes are good nodes that get used... sometimes they don't. And it is essentially impossible to figure out what is used and what is overhead... thanks to hashing...
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.