Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Programmers: Best direction to further develop Fruit code?

Author: David Mitchell

Date: 21:35:25 06/19/05

Go up one level in this thread


On June 19, 2005 at 02:24:52, Roger D Davis wrote:

>Since Fruit is just a notch down from Shredder and Fritz, it looks like that
>maybe with some further improvements, the strongest chess program could be open
>source. That would be a tremendous victory for the open source movement, and
>would prove the superiority of open source over proprietary development.
>
>I'm curious about the directions that programmers might pursue to develop Fruit
>to the point of being strongest in the world. Can it be done? How can it be
>done? What's the best way to do it? Would it have to be done in steps? What are
>they?
>
>Roger

If Fruit were a notch ABOVE Shredder and Fritz, you might at least have ONE
datapoint to base your hopes on - for now.

But that's not the case, and you're argument has nothing but speculation behind
it. Certainly there is no proof of "open source being superior", in sight.

I'm not intimately familiar with Fruit, but Theron's advice of "Test, test, and
more testing", does sound right to me.

Whatever can be done to strengthen any good engine, will need to be done in
steps. Some will be larger than others, and many will yield absolutely no
improvement. Unless you have a working crystal ball, you have to make your best
guess, and test that change out thoroughly. Then, try your next best guess,
based on your understanding of the program, and test that out.

Just for fun, my guesses to strengthen Fruit would be: 1) Improve the opening
book, and 2) Make Fruit automatically "page" load the optimum amount of EGTB
into RAM, to help prevent it from being i/o bound during the search.

Dave



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.