Author: Ed Schröder
Date: 09:37:26 11/05/01
Go up one level in this thread
On November 05, 2001 at 10:02:57, Thomas Mayer wrote: >Hi Vincent, > >> if not then improve winboard. Note that against Quark i lost more time >> than he lost against me in that game he mentionned. I walk away nearly >> every move. I lose on average 30 seconds a move in the first 30 moves >> of the game, *only* because i walk away. > >But I have informed you ALWAYS when Quark has moved. I even do not press the >clock early so that you wouldn't loose to much time. I am totally sure that you >did not loose nearly as much time as I loose in this position where you didn't >inform me that Diep has moved - I just see it because my clock was running and >not yours... Anyway - game was lost for Quark because of it's own stupidity at >that point... You are a real gentleman Thomas. >> I adjust time however. > >That is funny - you haven't informed me at ANY time that you adjust time - I >have informed you like the rules in Paderborn are... Next time I should look a >little bit more what your fingers are doing... Operators are only allowed to adjust the clock with the permission of the tournament director. After the WCCC Maastricht I have send the ICCA email because IMO the rules aren't practised well enough by the tournament director. Actually it is something that has my attention since many years but this year I took the time and effort to bring 2 subjects under the attention of the ICCA. Below is the initial email, I have snipped the first subject which was about book-incest which doesn't fit this subject. Ed ================================================================== Hello all, During "Maastricht" several rule-issues were discussed among programmers which I think the ICCA at least should be aware off. #1. Book incest: [ snip ] #2. Time control: the "flexible" attitude of Jaap concerning time control towards programmers in case of all kind of problems with the clock during the game is much appreciated however there are some drawbacks. As it is now *some* programmers take advantage of that. Here is what I have witnessed: Example-1: - leave the playing table when it is your move; - stay away for some specific time; - when you return your program has moved (say) 2-3 minutes ago; - in those 2-3 minutes your program has pondered; - when the ponder move is good you are able to answer immediately. Of course the clock on the table is not in balance anymore with the clock of the chess program if you do this a couple of times. This "little problem" is solved by asking Jaap to correct the clock. A clear misuse of Jaap's goodness and ignorance what is really happening. Example-2: When there is an unclear situation on the board leave the board when your program is pondering and stay away for a long time. In the meantime your program is able to take much more time for the position than normally. Of course also here the clock on the table is not in balance anymore with the clock of the chess program and this is solved again by asking Jaap to correct the clock. Example-3: Some programmers set their time control very high when the game starts to ensure the program takes more time than usual. So far so good. But (say) at move 20-30-40 they need a change of the internal clock otherwise the program will lose on time and they pick the right moment for that judging the current board position. Jaap is called again. In all 3 examples it is the programmer who influences the game which is not allowed. I am aware of the ICCA rule that an operator is obliged to enter the move immediately but that is a difficult and not feasible rule in practice as it could create very unpleasant situations among programmers. I think improved ICCA ruling is needed. Here is my simple suggestion: allow only one change of the clock (time control). For the Maastricht time control that would mean: you are allowed to change the clock or level only on move 60. It would mean the operator must have a very good reason to demand a change on the internal clock (or level). Good reasons are: a) an urgent phone call b) feeling not well and so on. Before the game starts you simply must make a good estimation of the operator time and bare the consequences in case you estimate wrong. I think such an extension to the rules would solve the above mentioned 3 examples as the risk is too high the program will lose on time. Of course I am open to other solutions as well as long as they solve the above mentioned tricks. I don't like to mention names, all I want to say is that the programmers addressed in this email are not subject to the 3 given examples. Good luck and thanks for organizing another great tournament. Ed
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.