Author: Chris Whittington
Date: 15:05:04 12/26/97
Go up one level in this thread
On December 26, 1997 at 16:40:48, Don Dailey wrote: >On December 26, 1997 at 14:16:30, Thorsten Czub wrote: >>>Perhaps they are doing heavy pruning on the computers moves? If you >>>modify the selectivity such that much heavier pruning takes place >>>on the computers side you might arrarnge extra depth that results in >>>play that is safe but not opportunistic for the computer. >> >> >>Right. Pruning is ONE explanation. > >Agreed, it is just a thought. > > >>>I remember Richard Langs programs USED to have the characteristic >>>that they would see anything you could do to them. In some experiments >>>me and Larry Kaufman did, it would quickly see that you could win >>>a piece and avoid the loss. But if you forced the piece losing move >>>and let it think for the winning side, it could not find the win without >>>a very long think! Very strange. But this was the best program in the >>>world and pehaps still is. >> >>Exactly. But Richard did it with asymmetrie. >>The plies 1,3,5,7,9,... were very much pruned and genius followed only >>the best 4 or 5 branches and the moves 2,4,6,8,... were looked >>brute-force. >>This gave him some extra plies and more look-ahead than his opponent >>programs at this time. >>With the invention of null-move his advantage was melted because the >>null-move algorithm was as effective as his asymmetrie ! Also his >>assymmetrrie caused boring style due to the fact that genius played in >>the plies 1,3,5,7,... often NOT THE BEST move. > >How did you get this information? It seems very likely to be accurate >to me. Larry Kaufman is the master at figuring out which algorthms >a program uses by playing with it and perfoming special experiments >with problems and such. We spent a lot of time trying to figure >out what Richard does, I think a lot of people have. Once there were americans doing their research,a nd europeans doing theirs. They didn't talk (until recently) > One thing >that was clear though is that the evaluation is very good too. But >we were more interested in the search. > >Other observations? His Pv's were almost always odd length. I had >the idea he only did evaluation on a given parity, for instance only >after the opponents move. I figured the search was always fixed to odd ply, because of doing static exchange evaluations. Safer for the assymetry if you stop with the decision that a piece is en prise when its your pieec, than to stop when its the opponent's piece (the opponenst piece may never get won, perhaps), so deducting it from his score and giving it to yourself is a dangerous thing to do ..... If evrybody uses capture search, this is no longer relevant. The words "used to" and "was" above imply that the recent Genius's no longer act this way, true ? Chris Whittington > >What else can you tell me so we can figure out his search algorithm?! > >I have had lots of conversations with Richard but he never tells you >anything very important about his program. Do you blame him? He >leaks small details occasionally but just enough to confuse you! But >he's actually a very nice person. > > >-- 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.