Author: Robert Hyatt
Date: 14:08:20 03/16/00
Go up one level in this thread
On March 16, 2000 at 16:10:12, G. R. Morton wrote: >On March 16, 2000 at 01:34:06, Jouni Uski wrote: > >>For me personally performance in test suites is more important, than playing >>strength. Because I am not GM I can't play against them. So about 80% of my >>computer use is test positions. And I removed even Junior 6 from my hard disk, >>because it's too weak to find combinations or endgame moves... >> >>Jouni > >How can Junior 6 be poor at combinatinos, and also mediocre at positional tests >, and still be the highest rated software? If this is true it is very puzzling >since every move of a computer game can be thought of as the software?s solution >to a tactical or positional test position. There should be a very strong >correlation one would think. As a comparison, people who score very high on >I.Q. and SAT tests are nor just good on such tests - which is, of course, why >these tests are given. Their strong correlation with high performance or >success in many other activities are well noted (see ?The Bell Curve? book for >instance). Did not Larry Kauffman once claim that his test suites could be used >to fairly accurately estimate the software?s rating? What else is going on >here? It is easy to explain. In test suites, there is usually a tactical answer that must be found by search. It requires _no_ knowledge of any kind, just a fast search. Which is why I don't care much about test suites in general, as they only test one aspect of the game. Tactics are important. But in many cases, tactics is not enough until we get to machines that are in the petahertz speed range. I could easily see any program (Junior in this case) doing poorly on some tactical test, but being an outstanding chess player...
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.