Author: Don Dailey
Date: 09:32:47 12/16/97
Go up one level in this thread
On December 15, 1997 at 09:34:30, Walter Ravenek wrote: >On December 13, 1997 at 16:50:22, Don Dailey wrote: > >>I have an idea for generating a positional problem set to >>measure our chess programs against. This will take some >>cooperation and should involve many programs. >> >> 1. Start with (n) positions from Grandmaster games and make >> a note of the move the grandmaster played. >> >> 2. Run each participating program for some length of time (t) >> on each position. Note the move the program chooses at >> the end of the allotted time. >> >> 3. Determine which positions there is a "consensus" on. The >> best move must be agreed on by all programs. >> >> 4. Throw out the "easy" ones. Consider this our problem set. > >Seems like an idea worth trying to me. I'll run the problems in >your reduced set with Arthur. > >I wonder whether CilkChess is able to determine the second best >move (I do this by throwing out the best move and researching). >If so, you could detect at least some of the tactical positions >by requiring that the gap between the best and the second best >move be less than a given value, say half a pawn. > >Walter Ravenek Hi Walter, This a good idea, and I actually considered doing this, especially if I don't get enough programs to participate. I would probably use a fairly small margin since I'm trying to pick up positional problems, but I will take any tactical ones that fall through in the process. I'll start processing the set and post my canditate positions. I will probably post them in small batches since the whole process is liable to take some time (on my end at least.) I'm hoping to get as many as 100 problems but this might be optimistic. Thanks. Anyone who has more than 1 program can run them both. They should be different programs and not just upgrades of an older one. For example my Cilkchess 2.0 is a rewrite, everything including the evaluation is different. -- Don
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.