Author: Robert Hyatt
Date: 07:45:49 10/30/03
Go up one level in this thread
On October 30, 2003 at 09:54:05, Vincent Diepeveen wrote: >On October 29, 2003 at 09:01:28, Anthony Cozzie wrote: > >>On October 29, 2003 at 04:47:31, Daniel Clausen wrote: >> >>>On October 29, 2003 at 03:15:23, Jorge Pichard wrote: >>> >>>>"Experiments in Chinook show that there comes a point where increased search >>>>depth provides diminishing returns." >>> >>>Many chess programmers agree that the search- and the eval- part of an engine >>>have to be tuned so they work optimal together. (like you can throw out certain >>>parts in the eval since they're now covered with a better/faster search etc) >>> >>>Now you take an engine, which is optimized for todays hardware to reach a >>>certain depth in typical middlegame positions and make the experiment of >>>increasing search depth. Why can't the effect of "diminishing returns" not be >>>explained by the fact that search and eval are no longer working together >>>optimal? >>> >>>It seems to me that in all these experiments which try to prove the effect of >>>deminishing returns, the errors bars are bigger than the effect they want to >>>prove. >>> >>>Sargon >> >>One experiment that both Hyatt and Heinz did was to run their programs through a >>"deep" (14 ply) search and record the frequency of new moves, that is, what % of >>the time the program changed its mind. Their conclusion was that we still >>haven't reached the point of diminishing returns. Of course, this was some time >>ago, and evals/pruning/extensions/etc have all improved since then . . . > >This experiment was imperfect of course, most likely driven by a desire to >publish without having done proper research. 'new moves' don't need to be >better. basically it was just the odd/even depth considerations and in many >positions more than 1 move are equally good after which odd/even searchline >depths are very important as a decision factor. > >Any chess knowledge to take a look at the positions was not used at all. > >Already a lot better are the experiments performed by Don Dailey who clearly >showed that evaluation quality is more important than extra search depth at >these depths and later investigations done after this phenomena. Note that Heinz >was involved in those later investigations and concluded the opposite of the >Crafty goes deep experiment. I don't know what you read, but in the JICCA article he found _exactly_ the same thing as Newborn and myself in "Crafty goes deep" (also published in the JICCA). I can give you the exact quote from his conclusions if you need to be reminded... He even went one ply deeper, as he did his tests about 2-3 years after we had done ours. And his conclusions were the same. Another ply is always better. > >>anthony
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.