Author: Albert Silver
Date: 15:38:50 07/20/99
Go up one level in this thread
On July 20, 1999 at 03:45:51, Ed Schröder wrote:
>>>This is way too pessimistic. Tuning the engine might be the most
>>>difficult part of the job but with common sense and good test-
>>>positions you come very close to the best settings.
>>>
>>>Next, by just playing a lot of games with a narrow (pre-defined) book
>>>it is possible to check if a given parameter is wrong or right.
>>
>>I agree it is possible to come closer to the best setting
>>but the question was not
>>how to come closer to the best setting but:
>>
>>how the heck is Ed is going to be able to
>>decide which set of adjusted settings are the best? :-)
>
>Ok, I will bite... I think Rebel's default settings through the years
>vary from 90-95% being optimal.
>
>I claim that the same also counts for very other chess program.
>
>HINT: count the number of reactions from my colleagues :-)
>
>I think that with Rebel's new engine concept it is possible
>to come to 95-98% which can mean an elo improvement
>from 30-50-70 points. I think that more is not realistic and
>I will be very satisfied if Rebel will gain 30 elo points.
>
>I also think it is possible NO improvement will be found.
>
>Well... if you don't try you never know of course.
>
>Ed Schroder
>
>
>>Uri
I took a look at the log file you placed on your site so as to show what would
be subject to tweaking and found that what interested me most was all grouped
up. I am thinking particularly of Chess Knowledge which is said to increase or
decrease the quality of positional play according to its value. This is quite
vague, and I would have thought that it was precisely in the various elements of
chess knowledge that one could achieve the most interesting tweaking. Is this
left like this due to a desire to not divulge all of Rebel's cogs and screws?
Albert Silver
This page took 0.01 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.