Computer Chess Club Archives


Search

Terms

Messages

Subject: Is the Depth directly proportional to the program's strength?

Author: Jorge Pichard

Date: 15:51:05 02/02/02


As processors are getting faster we realized that out of the leading programs
the best ones are often those capable of searching the deepest in a given time.
In any given position, specially position resulting from the match between Deep
Blue Vs Mr.Kasparov if you give the same position to the 5 leading SSDF programs
to calculate, you will notice that after a certain amount of time let say 30
minutes per move, usually Fritz and Chess Tiger are the programs that always
search the deepest into any given position. It is no coincidence that programs
capable of searching the deepest are getting better result in the SSDF.
Therefore, there seems to be a direct proportional curve in strength, which
clearly indicate that the Depth is directly proportional to the program's
strenght, as long as the program is free of bugs. Another observation that
clearly indicate that programs also benefit from longer time control was the
match Gambit Tiger vs the Web. Some of the players even with the help of good
chess programs were claiming a victory for the web, but as Ed Panek and I were
observing, Gambit Tiger after reaching depth 23 and only then was able to see
that it could eventually draw the game. I noticed that not too many programs
were capable of reaching beyong depth 22, probably due to their limit, since
most programmers didn't think that it was necessary for their programs to reach
that depth.



Jorge






This page took 0.02 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.