Author: Sune Fischer
Date: 15:31:23 07/30/04
Go up one level in this thread
On July 30, 2004 at 10:19:57, Robert Hyatt wrote: >On July 29, 2004 at 18:56:12, Sune Fischer wrote: > >>On July 28, 2004 at 17:39:14, Robert Hyatt wrote: >> >>>I should add that I fudge near time controls anyway. IE when I get close, I >>>won't try to burn off my saved time. I just let it carry over to the next T/C >>>to help there. IE there is no sense in burning 30 minutes on move 40... >> >>I don't have that fudge. >> >>If that fudge was needed I'd say something was wrong with the time manager, why >>else would one want to save up that much time in the first place? >> >>-S. > > >You are not paying attention. I have 30 minutes left on my clock. I am trying >to use it, but I keep predicting correctly and not using any. I am paying attention :) I just don't do it like that. The ponder time is dependent on the target time. If I have a lot of time left my target time will get larger, and it becomes harder to get a pure ponderhit. More searching will still be needed even after a ponderhit to burn all the target time. I choose to do it this way because I want my time management to be independent of the opponents. > As I near the >time control, it would be easy to try to use _all_ the time. If you don't have >this "fudge" then that is what you will do. And it makes no sense to search >move 40 for 30 minutes, then drop back to 3 minutes on move 41 since you just >smoked all your extra time. That will never happen, time cannot accumulate like that due to the way the manager works (see above). >This is _the_ point that you are overlooking. You can not predict how much time >you will have left, because you can not predict either (a) how well you will >predict your opponent's move and (b) how long you will get to search it for >free. I assure you I'm not overlooking anything, I can and will control my own time 100%. Should I move faster just because my opponent does? Of course not, impatience is his problem, not mine :) > In real games there are distinct cycles. You make a move, I move >instantly. Repeat N times. Then you make a move I don't expect and break the >cycle and now after I make a move you move instantly. Repeat M times. Then >start over. If you don't handle this possibility of having extra time left over >near a time control, you will likely burn it up for nothing. No way! :) >There are other >things I do as well, such as I intentionally try to build up a known time >surplus so I have time to handle fail lows properly. I haven't tried that. It sounds a bit "lose" to me, but unlike some people I won't say it's a bad idea when I haven't tested that exact variation of an implementation. ;) -S.
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.