Author: Amir Ban
Date: 02:53:11 04/02/98
Go up one level in this thread
On April 02, 1998 at 01:58:24, blass uri wrote: >I apologize about my english but I am sure you will understand me >1)why do junior4.6 waste time about illegal moves? It doesn't, and this is an optical illusion. Junior indeed considers pseudo-legal moves, but the search discards illegal ones in zero time. Junior spends some lead time on every new depth iteration, in which time the current move & depth indicators do not change. If by chance the last move from the previous iteration was illegal, it appears that Junior is thinking about it a lot. >2)what program earn most from hash tables? Good question, that will not get an answer. To answer this, every programmer should describe in detail how he uses the hash table. Little chance of that. Junior is not very sensitive to hash table sizes. The programs that are expected to be sensitive to hash size are the node evaluators that hash things like pawn value and king-safety value, programs whose move ordering rely heavily on hash-table info, and of course those that use top-secret, yet-unknown-to-mankind ways of using memory. >3)I want to know if the version of junior4.6 that I have is slower then >the originial program that won in the wmccc championship and if yes >what is the ratio. It's about 20% slower. >4)I want to know if the windows system cause program to be slower >(the fact that the computer should check all the time if I want to >go to another window must slow the program by some factor). >5)what happened to fritz5 opening book in the wmccc in the >last round(my program does not do the same mistakes)? You mean 10.Qb2 ? I wondered about that myself. Amir
This page took 0.01 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.