Author: Ernst Walet
Date: 11:11:26 02/17/98
Go up one level in this thread
On February 15, 1998 at 13:45:39, Ed Schröder wrote: >>Posted by Ernst Walet on February 15, 1998 at 06:27:54: > >>I've analysed some games from de Paderborn tournament with Rebel9 and >>came across the following problem, given the following game. > >>[Event "7th IPCCC"] >>[Site "Paderborn(Germany)"] >>[Date "1998.2.13"] >>[Round "5"] >>[White "Diep"] >>[Black "Zugzwang"] >>[Result "1/2-1/2"] >>[ECO ""] >>[Opening ""] > >>1.e4 e5 2.Bc4 Nf6 3.d3 Be7 4.Nf3 Nc6 5.0-0 0-0 6.c3 d6 7.Bb3 Na5 8.Bc2 >>c5 9.d4 cxd4 10.cxd4 Qc7 11.Nc3 Bg4 12.d5 Rac8 13.Bd3 a6 14.Qe2 Nc4 >>15.b3 Qa5 16.Na4 b5 17.bxc4 bxa4 18.Rb1 Rb8 19.Be3 Nd7 20.h3 Bxf3 >>21.Qxf3 Nc5 22.Bc2 a3 23.Bc1 Rxb1 24.Bxb1 Rb8 25.Qxa3 Qb4 26.Qxb4 Rxb4 >>27.Bd2 Rxc4 28.Rc1 Rxc1+ 29.Bxc1 Bd8 30.Kf1 Ba5 31.Ke2 Na4 32.Kf3 Nc5 >>33.Bc2 Kf8 34.Bd1 Bb6 35.Be2 Ke7 36.Bg5+ f6 37.Bd2 a5 38.g4 Ke8 39.Bb5+ >>Ke7 40.h4 h6 41.h5 Kd8 42.Bc3 Ke7 43.Be1 Kd8 44.Bd2 Kc8 45.Bc3 Kd8 >>46.Bb2 Ba7 47.Bc1 Bb6 48.Bf1 Na4 49.Bd2 Nc5 50.Bb5 Kc8 51.Bc3 Kd8 52.Be1 >>Kc7 53.Bc4 Kd8 54.Bc3 Kd7 55.Bb5+ Kd8 56.Bd2 Kc8 57.Bc4 Kd8 58.Be2 Ke7 >>59.Bb5 Kd8 60.Be3 Kc7 61.Bc1 Kd8 62.Bd2 1/2-1/2 > >>Rebel 9 produces illegal moves at move 55, and crashes at move 56. >>Walking manually through the game at gives a draw by repetition at move >>56, and at move 62. Is this correct, and are Diep and Zugzwang >>incorrect, or is Rebel9 incorrect, and how come the illegal moves? > >The repetition is after: >45.. Kd8 (first time) >51.. Kd8 (second time) >55.. Kd8 (third time) > >Perhaps somebody can double check? It wouldn't be the first time the >repetition routine is in error (Rebel7!) and I find it suspicious that >2 other programs do not see the repetition. You're right about these repetitions, Ed, but maybe while analyzing a game it would be better not to check this rule, see the remarks Robert Hyatt made on this. > >I got no crashes and no illegal moves but after 55..Kd8 Rebel >considers the game as over. I don't know what happens with moves >after a repetition as this in principal is not possible in a game. > > >>Futhermore, i noticed that when analyzing a game and Rebel is leaving >>the book before about move 6, the time control is inncorrect. Normally >>when you set the time to 1 minute per move, it adjust the time between >>0:52 and 1:17, and it takes more time when it feels to. However when >>analyzing before move 7, it putes the time between 0:33 and 1:07, but in >>many cases just using exacly 1 min. You can see that cause when >>rethinking the time remains 0:33 and 1:07 and does not change to 99:99 >>and 99:99. Could someone shine a light on this? Best regards, > >Correct. Rebel uses less time in the opening. This to prevent spending >lots of time on normal opening moves like Nf3, Bc4 etc. This could be all right when playing a game, but when analyzing time is not that important, I think. However, when playing a game the time control is not restricted to lets say 1 min. per move, but will adjust it self all the time! So the restriction in time only happens while analyzing. Thist seems to me like a contradiction, no? You can easily check it by playing 1. h4 i.e or by disabling the book, play a couple of moves, watch the time control and then analyze the game and watch thetime control. I also had another reproduceable lockup in Rebel9. Go to the database screen and open a database. Then search on 'no score' and then search on 'search text' and rebel locks up with me. Please give your comment on these nd kind regards, Ernst-Jan. > >- Ed - > > >>Ernst-Jan.
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.