Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Deep Blue clone plays Tao at FICS right now!

Author: Bas Hamstra

Date: 16:44:30 06/21/01

Go up one level in this thread


On June 20, 2001 at 23:09:01, Robert Hyatt wrote:

>On June 20, 2001 at 13:20:26, Jason Williamson wrote:
>
>>If these results are promising any chance you will implement them yourself?
>>
>>JW
>
>
>Of course.  I just haven't seen anything that I really like so far. I have
>tried this several times myself already.  Sometimes it looks promising, but
>when I try it in real games, it seems worse.
>
>This most recent attempt to tweak old code I wrote a couple of years ago looks
>good at times, but also looks bad at times.  I'm still trying to find a case
>where it appears to play better than the non-SE version...

I talked with GCP during our match. Based on what he told me, I fiddled a little
myself. It does do something, but it clearly costs depth too. Also I don't have
the feeling "forced" on the basis of a D-3 search is very accurate.

To test it I disabled all extensions, to see if any long PV's showed up. And it
occasionally happened, so it does do something. What I did *not* see is
extremely long PV's (which I hoped for of course). But it did solve a couple of
wac's sooner than with no extensions, so I don't seem to have it all wrong.

To me it seems a cheapo version, which has is pro's and it's cons, of course.
IMO Crafty certainly played different, putting pressure immediately. Normal
Crafty is only worrying about outside passsers :-)

Best regards,
Bas.























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.