Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Interesting that Tiger 14 can't find this endgame solution into 21 ply

Author: Terry Ripple

Date: 21:30:43 01/17/02

Go up one level in this thread


On January 17, 2002 at 18:13:22, James T. Walker wrote:

>On January 17, 2002 at 15:23:02, Terry Ripple wrote:
>
>>On January 17, 2002 at 14:36:31, James T. Walker wrote:
>>
>>>On January 17, 2002 at 14:35:20, James T. Walker wrote:
>>>
>>>>I have just tried Gambit Tiger 2.0 which should play the same endgame as Chess
>>>>Tiger.  Gambit Tiger found Kg4 in the 18th ply at 4:02 on my Athlon 1.4G.
>>>>Jim
>>>
>>>Just as I typed and posted that Gambit changed back to fxg5 in the 19th ply ! ??
>>-----------
>>Jim,
>>This is what i'am experiencing and like i said, i was into 21 ply when i stopped
>>the search!
>>
>>The Rebel page claims Century can find this solution but my Rebel Century was
>>into 18 ply with still wanting to play fxg5.
>>
>>Earlier in the search Rebel Century was looking at the correct move (Kg4) but
>>then a "Failed" appeared aside of Kg4!. Then it went on to thinking once again
>>that fxg5 was correct. Do you know what this "Failed" message means?
>>
>>Best regards,
>>Terry
>
>Hello Terry,
>My Century 4 does take untill ply 18.06 to find the right move at 3:09.  At that
>point however it holds on untill way past 30 minutes.  This is on my Athlon 1.4
>G machine with 200 M hash and Rebel hitting about 1.25M/nps.  After Kg4 and Be7
>the machine must still avoid fxg5 and play Kf5 I think to make this a win.  Are
>you sure you're using Century 4 and not Century 3.2?
>Regards,
>Jim
-----------
You are correct about me using Century 3.2 and not the 4.0! Do you feel it's
worth the upgrade to 4.0? I been debating if i want to upgrade or just wait
until the Windows version comes out, but i know that i'll probaly have a long
wait for this to happen any time soon.

Regards,
Terry



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.