Author: Eelco de Groot
Date: 07:55:03 02/03/02
Go up one level in this thread
On February 03, 2002 at 09:24:52, Uri Blass wrote: >On February 03, 2002 at 09:11:09, Eelco de Groot wrote: > >>On February 02, 2002 at 16:19:36, Jeremiah Penery wrote: >> >>>On February 02, 2002 at 15:51:24, Mike S. wrote: >>> >>>>[D]RNBQKBNR/PPPPPPPP/8/8/8/8/pppppppp/rnbqkbnr b - - 0 1 >>>>Sensational results! Obviously, 1...Na3 is a quick win. >>> >>>The position you have set up is known as Wild5 on the chess servers, except with >>>a different side to move. Computers pretty much suck at evaluating this, since >>>there will quickly be many queens on the board. It may be that in this case >>>black wins, but I wouldn't trust Fritz in this type of position for even a >>>second. >> >> >>I get the following after a night running: >> >>[D]RNBQKBNR/PPPPPPPP/8/8/8/8/pppppppp/rnbqkbnr b - - >> >>Rebel Engine for ECTool. (c) Ed Schröder >> >>Engine version : REBEL CENTURY 2 >>Rebel personality : "Personality Q5T" >>Opening book file : RebelC.mvs >>Hash table size : 4 MB >>Analysis mode : Analyzing next move >>Refresh interval : 500 ms >> >>Game begin >> >>00:00 08.00 1.63 1..Nc3 2.Nf6 b1=Q 3.g8=Q bxb6 4.Bh6 Rb1 >>00:00 09.00 1.63 1..Nc3 2.Nf6 b1=Q 3.g8=Q bxb6 4.Bh6 Rb1 >>00:01 10.00 1.74 1..Nc3 2.Nf6 b1=Q 3.g8=Q Ba3 4.Bh6 c1=Q >>00:05 11.00 1.61 1..Nc3 2.Nf6 b1=Q 3.g8=Q Nb5 4.Bh6 Nd6+ >>00:15 12.00 2.08 1..Nc3 2.Nf6 b1=Q 3.g8=Q Nb5 >>00:49 13.00 2.30 1..Nc3 2.Nf6 Nb5 3.Ne4 b1=Q 4.g8=Q bxb4 >>03:02 14.00 3.57 1..Nc3 2.Nf6 Nb5 3.Ne4 b1=Q 4.Na6 Ba3 >>05:51 15.00 3.11 1..Nc3 2.Nf6 Nb5 3.Ne4 b1=Q 4.g8=Q bxb4 >>15:13 16.00 3.57 1..Nc3 2.Nf6 Nb5 3.Ne4 b1=Q 4.Na6 Ba3 >>04:52 17.00 4.72 1..Nc3 2.Nf6 Nb5 3.Ne4 b1=Q 4.Na6 Ba3 >>31:42 18.00 4.00 1..Nc3 2.Nf6 Nb5 3.Ne4 b1=Q 4.Na6 Ba3 >> >>19 ply looked like a fail-high but no more patience for that. >> >>It could be that the evaluation is off but that the computer is still hard to >>beat by humans. I found old Crafty-list mail from Michel Langeveld saying that >>he had much success with Crafty at Wild5: >> >> >>--------------- >>Modification to let Crafty play Wild 5, 16-07-00 >> >>I already posted this on CCC. But maybe not everybody reads that. It's a >>very small modifcation... >> >>I registered a new Crafty on ICC called Wild5Crafty. >>It hasn't lost a single time yet :-P. So which expert can let it flame. > > >2 questions: > >1)How many comp-comp games did it play? >2)How many draws and how many wins did it get? > >> >>Ofcourse Crafty's evaluation can be more optimized for Wild 5, but why >>making things stronger if nobody can win already from this... > >I doubt if nobody can win against it. >I guess that you did not try comp-comp games. > >If I remember correctly Junior could beat it in engine-engine games under >chessbase gui but I am not sure if I remember correctly. > >Uri I don't know how many games against computers Uri. You would have to ask Michel about that. I suspect that when he started his experiment, it was practically only human players playing Wild 5 on the server. And of course there are not so many real human experts at this, certainly at fast timecontrols for most it would just be fun games. Computers don't know about fun, they only know about counting bits and beans. But with so many pawns right up on the seventh and second rank, the right to move could be important I think, although not four pawns? I have no idea how many games are won by White if it can move first in Wild 5. This has to be investigated right away! Groeten, Eelco
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.