Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Testing with List. What's wrong?

Author: John Merlino

Date: 16:16:40 08/28/02

Go up one level in this thread


On August 28, 2002 at 15:37:48, Tomas Casanovas Martinez wrote:

>On August 28, 2002 at 02:49:03, Ulrich Tuerke wrote:
>
>>On August 27, 2002 at 23:37:22, Dann Corbit wrote:
>>
>>>List does not understand G/<time> time control.
>>>
>>>Try 40/2 or some control like that instead.
>>
>>I think, that's not sufficient. In addition you have to edit its ini file and
>>adapt the time control defined there.
>>Finally, List seems just to select the time control, which it found in its ini
>>file.
>
>I have exactly the same problem. In my case, List 4.61 (operating in different
>Chessbase environments with Winboard adapter) reaches only about 600 Nodes PS
>both in a Pentium3 933 and in a Pentium 4 1,4. The program seems to stop
>thinking when it arrives at a deepth of 10.
>
>After modifying the Ini.file the problem remains. What a pitty not to be able to
>test this strong program in a standard conditions!
>
>Regards,
>
>Tom.

I seem to have gotten SOME improvement by making sure that the time control in
list.ini is the same as the time control that the GUI is using. I played a
10-game match between CM9000 and List at 40 moves in 2 minutes. Chessmaster won
7 games, and drew 2, for a score of 8-2.

List was reaching, on average, depth 6 or 7 in the middle game, but was not
using much of its time at all. At the end of a particularly long game, it had
almost four minutes on its clock after receiving four bonus minutes. At least it
is not getting checkmated in less than 20 moves like it was before. But it is
still not playing anywhere near what other people seem to be experiencing.

I think there's still something wrong with either my settings or with the files
that I downloaded.

jm



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.