Author: Uri Blass
Date: 03:52:14 01/11/02
Go up one level in this thread
On January 10, 2002 at 21:39:07, Vincent Diepeveen wrote: >On January 10, 2002 at 17:50:46, Dan Andersson wrote: > >>One reason that search is discussed so much is, IMO. That the static scores of >>evaluators are 'always' wrong. It means that an efficient and intelligent search >>(including extensions) will trumph a less efficient search allmost all the time. >>Due to the fact that the search essentially 'mines' the search space for a more >>accurate evaluation. A much better approach is to tailor your evaluator to your >>search. Granted that a good evaluator is preferable to a bad one. But making it >>behave consistently inside your search framework is the number one priority. > >Not exactly the truth. A simple alfabetasearch + nullmove + hashtables >+ simple qsearch is going to beat any other program if evaluation is >real good and the opponents is no good. > >Whatever your search, remember that the pro's are 99% busy with just >evaluation and testing. > >Search is like 0.001% of the time invested. I am sure that programmers waste more than 0.001% of their time about search rules. wasting 0.001% of you time about search meaning wasting less than an hour about search. only writing null move+fail high+fail low takes more than 1 hour. > >the reason why most like to fiddle with search is > - lossless speedups are easy to measure > - it is easy to modify something and test > - there are great tactical testsets to see whether your > search is finding tactical more (also at the same time saying > that for tournament results this says nothing about engine > strength) I do not believe that the tactical tests are great. A good tactical test should be avoiding tactical errors. Tactical errors are often done without sacrifices and existing test suites do not reveal it. In order to build a good tactical test it is better to analyze a lot of comp-comp games in order to find tactical errors when the target of program should be simply to avoid a tactical error. part of the tactcial errors can be seen as positional errors by humans. A change in the search rules is probably positive if the program can avoid tactical errors faster. 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.