Author: Will Singleton
Date: 17:16:28 07/12/99
Go up one level in this thread
On July 12, 1999 at 10:54:05, walter irvin wrote: >On July 12, 1999 at 04:01:34, Will Singleton wrote: > >>On July 12, 1999 at 00:02:42, KarinsDad wrote: >> >>>On July 11, 1999 at 18:56:30, Andrew Williams wrote: >>> >>>>On July 11, 1999 at 13:54:47, Terry Presgrove wrote: >>>> >>>>> >>>>> Nicely done! I particularly like your use of metaphors. I find it hard to >>>>> see how anyone can expect those who post here to avoid references to chess >>>>> events and chess in general. How can you seperate chess from computer chess? >>>>> TP >>>> >>>>I don't understand your last comment. Why do we need people announcing >>>>general chess events here? Surely that is done much better at a site >>>>like TWIC? I honestly think that this stuff is out of place here. This >>>>is an excellent forum for computer chess. It will become useless if we >>>>all start to discuss whether player A is better than player B or how well >>>>player C is going to do in a tournament. There are many other places for >>>>this sort of thing on the Internet, but there are hardly any other (civilized) >>>>places for computer chess. I personally wish people would restrict themselves >>>>to the topic for which this forum was originally formed. This place not called >>>>the Computer-Chess Club for nothing! >>>> >>>>Regards >>>> >>>>Andrew Williams >>> >>>That's ok with me. As long as people stop discussing the various merits of >>>processors and motherboards and other computer components as well. You cannot >>>drop out the chess half of computer chess without also dropping out the computer >>>half of computer chess. >>> >>>Let me know when you find THE TOPIC for which this forum was originally formed. >>>:) >>> >>>KarinsDad :) >> >>I would think the computer component is key. That is, the merits of various >>cpu's and other computer related items are more relevant to computer-chess than >>what Karpov and Kasparov are doing. Correct? Human tourneys and discussions of >>same have zero to do with our topic, while computers have everything to do with >>it. >> >>Will >well its thinking like that , is a big part of the reason chess programs get >reamed by grandmasters at slow time controls .you think you can just put faster >and faster cpu's with as much chess knowledge as a monkey. but if you really >understood chess you might have already gotten there????????? by the way you >dont play chess do you . That's a pretty revealing comment, there, about chess knowledge, monkeys, etc. Are you familiar with the history on the topic? It's an interesting read, you might pick up some books on it. There's this fellow Botvinnik, he says he developed a pretty good program. Check that out. He said back in 1983 that further development on algorithms based on the brute force method could not hope to improve play <g>, that the only approach left to solving complex planning problems were "high-level" algorithms (chess-knowledge). Many experiments over the years have shown that there's a strong correlation between search depth and rating. Even more recent experiments show that even at very large search depths, play continues to improve. The fundamental reason that GM's continue to beat up (occasionally) on chess programs is that they have brains, which can process non-linearly, do pattern recognition, intuition, specific recall, whatever. Computers aren't well-suited to operate like that. Hence, the brute-force method has proved the most effective means we have to play good chess. There's a nice overview of this topic, and of the history of chess computing, by another chess guy named Levy. It appears in the June 1997 issue of the ICCAJ, under the title "Crystal Balls." In fact, there's all kinds of neat stuff in the journal, back issues are available. Will
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.