Author: Christophe Theron
Date: 10:40:05 05/07/02
Go up one level in this thread
On May 06, 2002 at 23:57:13, Peter McKenzie wrote:
>On May 06, 2002 at 22:31:28, Christophe Theron wrote:
>
><snip>
>>I don't think the proportion is measured in elo points.
>>
>>My unit for the 90%/10% estimation is subjective. It's something like the amount
>>of reward for a given programming effort.
>
>I've been following this discussion, and it is certainly food for thought.
>
>I think that evaluation changes are even harder than search changes to test.
>This makes the development effort more difficult.
That's right.
Changes in the evaluation are not rewarding in term of winning percentage.
These changes result in improvement on test suites generally, on playing style
certainly, but not in test games.
The exception is endgame knowledge. Most of the time it works very well.
The funny thing is that people still believe that we are removing endgame
knowledge because our programs are supposed to count more and more on endgame
databases! :)
Christophe
>I also think that many programs do not have a good framework for adding
>evaluation knowledge, so there are big limitations on what knowledge can be
>implemented.
>
>All in all, I don't really know the answer ... what was the question again? :-)
>
>Peter
>
>>
>>A successful effort in search get a reward 9 times bigger than a succesful
>>effort in positional evaluation.
>>
>>Not to say that work on positional evaluation can be ignored.
>>
>>I notice that some chess players tend also to agree that chess is essentially a
>>matter of search (tactics).
>>
>>
>>
>> Christophe
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.