Author: Uri Blass
Date: 00:58:49 11/04/01
Go up one level in this thread
On November 04, 2001 at 00:06:14, Thomas Mayer wrote: >Hi Bas, > >> Against Diep Winboard's clock was slightly out of sync with the real clock. >> Don't ever let that happen! Amazing how fast the difference grows and you >> CAN'T adjust the clock in Winboard. A couple of minutes difference is >> *deadly*. And, though it is allowed to adjust the engine clock, apparently it >> is not allowed to restart Winboard+engine with a corrected time. Vincent was >> the first to point that out to the tournament directors. "Not allowed! If I >> can't restart my engine for table bases, HE can't bla bla bla..." (Tao +0.90 >> Diep +0.001 but drawish IMO). Whatever I would have done in this situation, >> *not* this pityful "no no, not allowed!". The 30 nullmoves from Diep that >> followed to push Tao through the flag I can forgive, but not the "nono". > >That's why I do it a little bit different for tournaments in Quark. >Tournamenttimecontrols in Quark are hard coded, but need anyway the time >information of Winboard. > >That's how it work: >I set Winboard always to the full time, e.g. in Leiden to 90 minutes (if there >is a cut somewhere, it is hardcoded, that Quark must play the first 40 moves in >xx minutes) >Now what is Quark doing ? There is a setable variable in its ini file, called >operatortime - a good value for this is 12. Internally it had a correction time, >this grows each move (move, not half move) about this value. >When it gets the winboard-time, it substracts the correction time from the >winboard-time and has now a prediction about what is on the clock. >Before every move, the engine writes something to the the .debug - file: Time >from Winboard, Correction time, Time it thinks must be on the clock. I check the >.debug file through game with the tail-utility. (Works fine under NT/2000/XP) >Because the opening moves are played without clock you get initially a quite >nice bonus which you can add later - there are 2 commands in Quark to correct >the internal correction time by a value, so when Quark thinks it has to much >time on clock then I must add something to the correction time, else I can >substract something. If the game will be very long, you must change the >operatortime later in the game... (But thats only interesting, when you have >much over 100 moves) - a quite playable value is 6-8 seconds if your >concentration is high enough. >With that system I was so far NEVER in time troubles, and I have always an >overview what Quark thinks what is on the real clock, so it's easy to correct. >Oh, only once I had a time problem - Vincent did not tell me, that he has moved >already and my clock runs for about 10 minutes without that I know it... But I >correct internal time and didn't miss the cut at all - I spent one minute on the >idea to miss the cut anyway, because I was very unhappy about the game, but that >wouldn't be very sportsmanlike... a "no, no" is also not very sportsmanlike... I >would understand this from commercial programs - they fight for every point and >every point is money - but from amateurs ? Well... How about pushing the >reset-button by accident to have a crash ??? :) I believe that Vincent does not consider Diep as an amateur so he fights for every point. I also do not like the behaviour that was described and I consider the game to be a draw for deciding if to buy diep. It does not mean that conidering the game as a win for Diep could convince me to buy diep because the results in the other games are not good enough. Uri
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.