Author: Dann Corbit
Date: 16:53:42 02/01/99
Go up one level in this thread
On February 01, 1999 at 19:18:54, Reynolds Takata wrote: > Thanks for trying to set up this match. > >However, i think it would be of most importance if the GM only agreed to look at >Hiarcs games, that have already been played. The GM should have no access to >actually play it before the match. After all Hiarcs will have absoloutely no >chance whatsoever to practice against the GM. Him playing H7 before the match >is way too great of a handicap, i'd expect any 2200 master to beat H7 if he was >allowed this advantage. So if the GM is allowed to do this, then even if Hiarcs >was 2750 ELO(which it isn't), would stand almost no chance in the match, thus i >see the whole effort as being in someways wasted. This because hiarcs would be >lost before they ever actually play a move in the real match. There is no >legitimite match anywhere, where you can go up to an opponent and say "hey what >would you play if i play this in the match?". Though it may be hard, but since >it's your dime, i would expect that in the contract that he agree not to play >actual games against Hiarcs in the interim between now and the match, though he >should have acces to games that have been played already. Aren't you tired of >looking at our comments :) hehehe. Does Hiarcs learn? Could Hiarcs be built with a special opening book? It seems to me that the advanced preparation could go both ways. The opening book used should be published so that other Hiarcs users can benefit from it, if customization is allowed. How deterministic is Hiarcs approach? If the GM does discover a weakness, that would also be incredibly valuable to the Hiarcs programming team, since they could plug that hole in the software, making the program that much stronger for the next go-round. I would suggest, in fact, that having a GM try to find holes on a routine basis would be an excellent way to strengthen any computer chess playing program. So, the "warpage" of the experiment depends on a lot of factors and what the real end-goal is.
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.