Author: Uri Blass
Date: 01:08:13 09/16/00
Go up one level in this thread
On September 15, 2000 at 23:32:43, Vincent Diepeveen wrote: >On September 15, 2000 at 03:11:22, Uri Blass wrote: > >>On September 14, 2000 at 13:42:49, Ed Schröder wrote: >> >>>In the upcoming Rebel Century 3.0 I have implemented a little >>>statistic routine that reveals something about the nature of SEARCH >>>that could be important for the future of computer chess in the sense >>>that it says "something" one may expect in the near future because >>>of faster and faster PC's. >>> >>>Research on this issue have already been done by Bob and Ernst and >>>it has made me curious so I have spend a little time on it. The >>>statistic shows 2 things: >>> >>>a) number of fail-low's for each depth; >>>b) number of "changed moves" for each depth. >>> >>>(a) is not so important as often fail-low's do not mean anything but I >>>wanted to know anyway. >>> >>>(b) is extremely important as it shows for each depth how many times >>>Rebel changed its mind. As you can see in the below statistic the % >>>diminish and diminish the deeper Rebel goes. >>> >>>How to read the overview: >>>- first column: iteration depth; >>>- second column: number of times the depth was reached; >>>- third column: number of fail-low's; >>>- fourth column: percentage of fail-low's; >>>- Fifth column: number of changed moves; >>>- Last column: percentage of changed moves; >> >>1)I think that the interesting question is if there is a diminishing return from >>time and not if there is a diminishing return from plies. >> >>I think that the following data may be interesting when you use fixed time of >>2^n seconds per move(you can choose n) >>column 1:number of changed moves between 1 seconds and 2 seconds >>column 2:number of changed moves between 2 seconds and 4 seconds >>column 3:number of changed moves between 4 seconds and 8 seconds >> >>column n:number of changed moves between 2^(n-1) seconds and 2^n seconds > >You're using a power of 2 here to measure where the branching >factor hardly will have a power of 2, so you're falsifying research >by using time measurements using a power of 2. > >Of course the whole idea to measure principle variation changes >i find complete nonsense in advance, but if you measure it, then >measure plydepth based, not second based! I believe that ply depth is not important. I guess that at least 10% of the moves will be different in the same ply depth if you use pruning(null move or other types of pruning). All top programs use pruning because the advantage is bigger than the demage but the demage is not the same for all programs so ply depth is not relevant. Time is the only thing that is the same for all the programs(assuming the best hardware) and if the best hardware is not available you can multiply the time by constant to get similiar results(not the same because of the fact that it is possible that the best hardware does not help in the same way in all the positions). 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.