Author: Dann Corbit
Date: 15:04:08 03/12/01
Go up one level in this thread
On March 12, 2001 at 16:09:46, Robert Hyatt wrote: [snip] >OK. I can't imagine how a program could be written to behave like that, but >I assume it is possible since you saw it somewhere... > >In the case of Crafty, .1% is the right number. In the "main search loop" there >is a single test/branch that is executed needlessly, once per node, for the >SMP version when there is no extra processors being used. It depends a lot on what library functions are called. Sometimes, the SMP link libraries are simply slower than the single threaded versions. I have definitely seen this effect from time to time. Your program also has a very smart use of interlocked exchange which is a very inexpensive form of synchronization object. If you used something else, there could be a big penalty.
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.