Author: KarinsDad
Date: 06:55:03 12/24/98
Go up one level in this thread
Roberto, I think your idea on parallel searching during table lookup is time better spent than your idea of reading directly from the disk. For example, during disk cache, would you really want to be looking at the hard drive when the data is already cached (soemthing that would happen when searching on similar nodes in your tree and something which would also happen when you do not use the standard i/o calls)? I don't know if anyone has mentioned it here before (they probably have since this is fairly obvious), but yanking out irrelevent startup and other processes will result in memory and CPU resource savings on a system as well. You just have to be careful when using that excess memory to not set up your hash table so large that either your program or the OS starts paging memory. Happy Holidays, KarinsDad
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.