For experienced chess programmers only, I like to snoop some 5-10 minutes of your time by submitting your remembrance what you considered essential in your EVAL in the 2005-2006 area, via the link:
http://www.top-5000.nl/eval_quest.htm
Background: Mark Watkins document COMP_EVAL compares the EVAL features of Fruit and Rybka with a number of open sources. One point of criticism is of course the elo gap between Fruit and Rybka with the open source engines and another valid point is the lack of volume of engines.
This poll can change that.
Thanks in advance for your contributions.
Ed
Opinion poll for experienced chess programmers
Moderators: hgm, Rebel, chrisw
-
- Posts: 6995
- Joined: Thu Aug 18, 2011 12:04 pm
-
- Posts: 1822
- Joined: Thu Mar 09, 2006 11:54 pm
- Location: The Netherlands
Re: Opinion poll for experienced chess programmers
No comments.Rebel wrote:For experienced chess programmers only, I like to snoop some 5-10 minutes of your time by submitting your remembrance what you considered essential in your EVAL in the 2005-2006 area, via the link:
http://www.top-5000.nl/eval_quest.htm
Background: Mark Watkins document COMP_EVAL compares the EVAL features of Fruit and Rybka with a number of open sources. One point of criticism is of course the elo gap between Fruit and Rybka with the open source engines and another valid point is the lack of volume of engines.
This poll can change that.
Thanks in advance for your contributions.
Ed
-
- Posts: 20943
- Joined: Mon Feb 27, 2006 7:30 pm
- Location: Birmingham, AL
Re: Opinion poll for experienced chess programmers
I answered, but note that your question is not exactly on point. Mark's paper doesn't just consider which terms are common to both Fruit and Rybka, but the methodology behind those terms as well. There are lots of ways to evaluate backward pawns, for example, so just asking "did you do backward pawns" is not exactly revealing. More interesting would be the details each used in those terms. Because that is where the eerily similar approaches are important in the Fruit/Rybka stuff Mark did. His "100% match" for a term did not mean "both do this". It mjeant "both do this using the same conditions and adjusting the score in the same way...Rebel wrote:For experienced chess programmers only, I like to snoop some 5-10 minutes of your time by submitting your remembrance what you considered essential in your EVAL in the 2005-2006 area, via the link:
http://www.top-5000.nl/eval_quest.htm
Background: Mark Watkins document COMP_EVAL compares the EVAL features of Fruit and Rybka with a number of open sources. One point of criticism is of course the elo gap between Fruit and Rybka with the open source engines and another valid point is the lack of volume of engines.
This poll can change that.
Thanks in advance for your contributions.
Ed
-
- Posts: 6995
- Joined: Thu Aug 18, 2011 12:04 pm
Re: Opinion poll for experienced chess programmers
Thank you for your contributions. I got 7 in a couple of hours, thereafter for some reason the script stopped working and contributions did not reached my mailbox any longer. A 3 hour online chat with my provider did not bring a solution so the page lost its meaning.
-
- Posts: 20943
- Joined: Mon Feb 27, 2006 7:30 pm
- Location: Birmingham, AL
Re: Opinion poll for experienced chess programmers
Those conversations usually go:Rebel wrote:Thank you for your contributions. I got 7 in a couple of hours, thereafter for some reason the script stopped working and contributions did not reached my mailbox any longer. A 3 hour online chat with my provider did not bring a solution so the page lost its meaning.
(1) reboot your cable modem / DSL modem
(2) reboot your router
(3) reboot your PC
(4) Hold until I can get someone in technical support to talk with you.
(5) go to (1)
-
- Posts: 6995
- Joined: Thu Aug 18, 2011 12:04 pm
Re: Opinion poll for experienced chess programmers
They really really tried. But instead of saying "sorry, we can't help you right now" they could not resist to blame it on me in the end
-
- Posts: 1342
- Joined: Wed Mar 08, 2006 9:41 pm
- Location: Morgantown, WV, USA
Re: Opinion poll for experienced chess programmers
Is this why Fabien stopped working on Fruit? I have fond memories of when it was announced here and all the buzz and enjoyment many people had; myself included. Definitely left a soft spot in my heart.
I wasn't around when the Fruit-Rybka debate went haywire.
I wasn't around when the Fruit-Rybka debate went haywire.
-
- Posts: 6995
- Joined: Thu Aug 18, 2011 12:04 pm
Re: Opinion poll for experienced chess programmers
I don't why Fabien stopped working on Fruit in 2006/7. But the poll is indeed about the R/F controversy. Speaking about the poll, my provider apparently (silently) fixed the problem after all, the script is alive and kicking again. Received 3 new contributions. Thanks.
-
- Posts: 2949
- Joined: Mon May 05, 2008 12:16 pm
- Location: Bordeaux (France)
- Full name: Julien Marcel
Re: Opinion poll for experienced chess programmers
AFAIK, he stopped working on Fruit because he had started a new chess project: Chess64, a new program that would have been much stronger... And then nothing... Has anyone heard something new about Chess64 or did Fabien totally left the computer chess scene?jshriver wrote:Is this why Fabien stopped working on Fruit? I have fond memories of when it was announced here and all the buzz and enjoyment many people had; myself included. Definitely left a soft spot in my heart.
I wasn't around when the Fruit-Rybka debate went haywire.
"The only good bug is a dead bug." (Don Dailey)
[Blog: http://tinyurl.com/predateur ] [Facebook: http://tinyurl.com/fbpredateur ] [MacEngines: http://tinyurl.com/macengines ]
[Blog: http://tinyurl.com/predateur ] [Facebook: http://tinyurl.com/fbpredateur ] [MacEngines: http://tinyurl.com/macengines ]