Author: Robert Hyatt
Date: 19:17:19 07/06/99
Go up one level in this thread
On July 06, 1999 at 16:07:36, blass uri wrote: > >On July 06, 1999 at 15:19:06, Robert Hyatt wrote: > >>On July 06, 1999 at 15:00:15, blass uri wrote: >> >>> >>>On July 06, 1999 at 13:17:48, Robert Hyatt wrote: >>> >>>>On July 04, 1999 at 17:29:35, blass uri wrote: >>>> >>>>> >>>>>On July 04, 1999 at 17:12:02, Bo Persson wrote: >>>>> >>>>> >>>>>>Not quite. >>>>>> >>>>>>If you run under Windows, a program can behave badly and be a CPU hog. It can do >>>>>>a number of "tricks", like increasing its own priority, to get more CPU time >>>>>>from the system. >>>>>> >>>>>>This will be unfair to "the nice guy" who's program runs "properly" - share and >>>>>>share alike. >>>>> >>>>>I do not suggest thinking and pondering at the same time. >>>>>The only reason that the game is going to be twice longer is that instead of >>>>>thinking and pondering at the same time I suggest to do it not at the same time >>>>>so instead thinking and pondering for 2 minutes on the same time I need 4 >>>>>minutes(2 for one engine to think and 2 for the second engine to ponder without >>>>>knowing the move of the first engine) >>>>> >>>>>Uri >>>> >>>>I've explained this several times. "ponder=off" (crafty terminology) is _not_ >>>>the way to play engine vs engine games. I do _all_ of my testing with >>>>ponder=on, and only use ponder=off for test suites and debugging. My time >>>>allocation code is tuned to run with ponder=on. Running with it off will >>>>most definitely cause some timing difficulties that are not normally seen. >>>> >>>>I'd bet that if you ask, most programmers test with ponder=on and feel very >>>>comfortable with their code. But if you ask them to play a serious tournament >>>>with ponder=off, I'd bet you would see a _lot_ of testing going on to make sure >>>>that this doesn't break anything. >>>> >>>>For _my_ program, "out-of-the-box" is the best way to run it, other than >>>>customizing hash table size for your specific hardware. Everything else is >>>>_exactly_ as I run it on ICC, which means that the 'defaults' are the best that >>>>I know how to do... >>>> >>>>Changing anything will very likely weaken it. Perhaps significantly... >>> >>>I explained that there is no problem to do something eqvivalent to ponder=on in >>>1 computer. >>>The only difference is that the games will be longer because the actions are >>>going to be not in the same time instead of the same time. >>> >>>Uri >> >> >>Uri: >> >>listen _carefully_. If you run a program with pondering disabled, it will >>_screw up_ things. It doesn't matter whether you double the time control or >>not. The program has to be told about the extra time. > >I did not suggest to disable pondering but to do something that leads to the >same results as pondering in one computer. >I agree that chessbase does not do it. > >It is simple to do it in one computer by the following steps >We start with step 1 when A is out of book. > >step 1: Engine A "thinks" about a reply to engine B and does the move. > >step 2: Engine B gets the following information:Engine A played a move and used >x seconds for the move(Engine B does not get the move of Engine A). > >Engine B ponder for x seconds and only after x seconds get the information about >the move of Engine A(The time per game for Engine B does not change during these >x seconds). > >steps 3,4 are similiar to steps 1,2 >The deatails of these steps: > >step 3: Engine B "thinks" about the reply to Engine A and does a move ( now the >time per game for engine B is changed) > >step 4: Engine A gets the following information:Engine B played a move and used >y seconds for the move(Engine A does not get the move of Engine B). > >Engine A ponder for y seconds and only after y seconds get the information about >the move of Engine B(The time per game for Engine A does not change during these >y seconds). > >step n+4 can be described by the same words as step n. > >Uri But what about an engine that is _aware_ of time passing? I use wall-clock time _always_ because CPU time is _dangerous_ in a tournament and can cause you to lose on time. And I will _definitely_ see more time going by than I expect, and it will totally confuse my time measurement code. In tournaments, I do _not_ get a clock update from xboard and use it. I do _all_ my own internal time-keeping, because _that_ is the time that is important...
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.