Author: KarinsDad
Date: 09:28:40 07/20/99
Go up one level in this thread
On July 20, 1999 at 12:12:14, Dann Corbit wrote: >I think it would be interesting to benchmark chess algorithms: >0. Move generators -- all types >1. Alpha-Beta vs MTD(f) >2. Bitboards vs 0x88 >3. etc. > >Prepare a large crosstable and do a large number of runs with as many >implementations as possible and under as many different conditions as possible. > >Change the search time from very short searches (10 sec or less) up to half an >hour to find the bit O(f(n)) properties of the algorithms. > >A systematic study might eliminate a lot of guesswork or even tell us *where* >certain algorithms work better than others. For instance, we might use one >algorithm at a certain time control and a different algorithm at a longer time >control and yet another at correspondence chess time controls. Sounds good in theory, but the main difficulty comes in with the control. In order to get valid results, you have to change only one element in a program at a time. Since chess programs have so many elements which are linked to one algorithm or another (because I am using Alpha-Beta, I can also use lazy evaluation), you would have to do a lot of work to ensure that you are comparing apples with apples. I am not saying that it cannot be done, I just foresee that it would require a lot of work (similar to what Aske Plaat and some of the other researchers have done). 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.