Author: Robert Hyatt
Date: 10:54:32 09/06/98
Go up one level in this thread
On September 06, 1998 at 13:27:21, Don Dailey wrote: >On September 06, 1998 at 02:22:21, Ed Schröder wrote: > >>>I have always been interested in this question of intransitivity >>>between humans and chess programs. A lot of people claim, almost >>>always based on some anecdotal evidence, that various programs >>>are much better or weaker against humans or other programs and >>>that huge intrasitivities exist. My own thinking is that this >>>is minor, but I know I'll get a lot of disagreement here. >> >>>I can't tell you how many times I have heard this type of >>>conversation: "I played a game with program >>>X and got crushed, then I played a game with program Y and >>>won easily. But when I played the two programs together, >>>program Y demolished program X." From a measly 3 games >>>can you come to the conclusion that program X is not very >>>strong against other programs but is "crushing" against >>>humans? >> >>>But once you form a conclusion, then you start noticing the >>>events that reinforce your conlusion and you minimize the >>>events that do not. Usually, the opinion propogates to other >>>people if it gets stated enough times. >> >>>Here are the ones I have heard but have serious doubts whether >>>they are true, or at least crystal clear: >> >>> 1. Novag machines a much better against people but no good >>> against other programs. >> >>> 2. Genius is not very good against people but crushes other >>> computers. >> >>> 3. Genius is particularly good aginst people, but just so-so >>> against other programs (yes, I've heard both cases stated >>> as facts.) >> >>> 4. Deep Blue will crush any computer but is not much better >>> that micro programs against humans. >> >>> 5. Any kind of forward prunning or selectivity will help a >>> lot against humans but is not so good against computers. >> >>> 6. Same as 5 but reversed. >> >> >>Hi Don, >> >>That's why I asked for the statistics of chess programs playing humans so >>we can compare and discuss. I have given mine. On my home page I have >>a database with the following items: >> >>- Unique database of the most famous Man vs Machine events. >>- AEGON 1991-1997 complete! >>- Harvard Cup 1989-1995 >>- Special matches like Deep Blue-Kasparov, Rebel-Yusupov etc. >>- Total 1900 (!) games >> >>The direct download address is: http://www.rebel.nl/mvsm.zip >> >>- Ed - >> >> >>>- Don > > >Hi Ed, > >I am in no way discouraging this kind of research. As I have stated, >I DO believe it's possible to emphasis a certain kind of intransitivity >as you are trying to do with Rebel. After all, chess players study >their specific opponents in an attempt to increase their winning >ability (even if temporarily) against that single opponent. Since >human and computers have known strengths and weaknesses it should be >possible for each to take advantage of the other. > >Having actual data as you suggest is certainly the way we should >approach this if we really want to understand this phenomenon (or >lack of.) The very nature and difficulty of getting solid data >on man vs machine makes this type of discussion more likely to >involove lots of subjective opinion and speculation rather than >facts and logical thinking. > >- Don Note that obtaining the games is not as hard as you'd think. IE how many thousands of blitz game scores would you like between Crafty/Ferret and GM players? :) Yes, mostly blitz... but there are things that can be learned there. IE computers still need work on preventing blocked pawn positions, something that I am finally having very few problems with. But there are still other cases where the GM's simply know too much instinctively, while a program has to "grind it out." But getting data on the servers is trivial... you just need an interface, a machine, and get ready for some interesting battles...
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.