Author: Uri Blass
Date: 08:09:47 07/02/01
Go up one level in this thread
On July 02, 2001 at 10:01:29, Christoph Fieberg wrote: >On July 02, 2001 at 07:48:13, Uri Blass wrote: > >>On July 02, 2001 at 05:34:51, Frank Quisinsky wrote: >> >>>Hi there, >>> >>>at the moment I make a little experiment and created a webpage to this >>>experiment. At the moment is the webpage only in German available. >>> >>>Amateure contra Profis under: >>>http://amateurschach.in-trier.de/schach/cbase/experiment.htm >>> >>>The question is: >>>Can an amateur program win a world championchip? >>> >>>I will here plays 20 tourneys swiss system with 16 programs and 9 rounds. >>>The same programs / versions with 40 moves in 10 minutes. Furthermore, without >>>tablebases (I believe this is also interesting). >>> >>>In the last days I play a test tournament, only to see different configurations. >>> >>>T-01: >>>http://amateurschach.in-trier.de/dia/t-01.jpg >>>http://amateurschach.in-trier.de/dia/t-01.zip (games in CBH). >>> >>>Crafty 18.10 with a very good result but the first 3 programs are professionals. >> >>I believe that the answer is clearly positive. >> >>It is a good idea to give every program the best hardware that it can use. >> >>If Crafty can use a good alpha when it can search 7M nodes per second then it is >>a good idea to give it to Crafty. >>If other programs cannot use the alpha then it is not Crafty's fault. >> >> >>>I think that maybe in 5% the amateurs can win the WM, but the reality is that >>>maybe 0.1% amateurs can win a tournament. >> >>The reality is that Crafty won one of the ICC tournament and ferret was one of >>the winners of another ICC tournament. >> >>Both programs are amatuers. >> >>The reality is that Junior4.5 won one WMCCC when it was amatuer. >> >>Uri > >The question I answered was: >What is the likelihood for any of the 2450 rated programs to win clear in a >tournament with players 2625, 2625, 2625, 2575, 2525, 2450, 2450, 2450? > >With the help of a simulation-tool I developed I found out that the likelihood >is a mere 0.1425% in a double round robin (14 games each player) and 0.6685% in >a single round robin (7 games each player). Based on simulation of 200.000 >tournaments. > >Uri, could you give me the ratings of players of the ICC tournament that Crafty >won. I could analyse what the likelihood for him was to win it. I cannot do it but it is clear that there has nothing between the question that you were asked and the probability of ametuers to win tournaments. It is clear that most of the players in most of the computer tournaments are weaker than Crafty. Crafty was clearly a candidate to finish in the first half in the tournaments that it won. Uri > >Best regards, >Christoph
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.