Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Chess Tiger 14.9 Palm at 54 mhz vs Movei at 450 mhz 1-0

Author: Uri Blass

Date: 19:02:09 04/04/02

Go up one level in this thread


On April 04, 2002 at 18:59:33, Rafael Vasquez wrote:

>On April 04, 2002 at 14:32:55, Uri Blass wrote:
>
>>On April 04, 2002 at 12:47:19, Jorge Pichard wrote:
>>
>>>On April 04, 2002 at 11:29:10, Uri Blass wrote:
>>>
>>>>On April 04, 2002 at 10:37:28, Jorge Pichard wrote:
>>>>
>>>>>On April 04, 2002 at 01:04:06, Rafael Vasquez wrote:
>>>>>
>>>>>>Hi, Jorge and Uri:
>>>>>>
>>>>>>This is the Arena output for you:
>>>>>
>>>>>
>>>>>I just replayed Movei moves using my Celeron 433 Mhz and even if my Celeron is
>>>>>slower than your K6 III 450 Mhz it was able to reproduce 98% of the moves for
>>>>>Movei.
>>>>
>>>>I also replayed Movei moves in my PII800 and
>>>>it needs 19 minutes/40 moves to avoid the tactical mistake
>>>>16...Qc7
>>>>
>>>>At that time control it plays 16...0-0 and at 18 minutes/40 moves
>>>>it plays 16...Qc7.
>>>>
>>>>
>>>>Uri
>>>
>>>My Celeron still plays 16....Qc7 even at 20 minbutes/40 moves, but at 21
>>>minutes/40 moves it plays 16...0-0.
>>
>>I played all the game at 18 or 19 minutes/40 moves so practically it had clearly
>>less time before 16...Qc7.
>>
>>My p800 is almost twice faster than your the k6-III 450 and your celeron has
>>similiar speed.
>>
>>Thinking history may influence the choice of the moves at the same time
>>control(the reason is that it may change the order of the pieces in the memory
>>of movei and it can cause some change in the order of moves) and I wanted to see
>>what is the faster time control that movei needed to avoid the blunder and the
>>only way to check it seemed to be playing all the game.
>>
>>  I believe that at 30 minutes/40 moves it
>>>could give Chess Tiger 14.9 a better challenge using a K6-III 450  or a Celeron
>>>433 Mhz; but again Chess Tiger also would be playing much better at that time
>>>control.
>>
>>The game was played at 30 minutes/40 moves but I think that movei used most of
>>the 30 minutes in the first 15 moves.
>>
>>
>>>
>>>PS: What Rafael needs to do is to de-activate the Opening book for CTP 14.9 and
>>>movei would be playing with the same disadvantage since both do not use TB.
>>>
>>>Pichard.
>>
>>The question is how much advantage gives the opening book.
>>There are openings that movei does not play well
>>and I hope to fix it by changing the evaluation
>>but it does not happen in every game and movei could do 3-3 against genesis
>>inspite of the fact that genesis use opening book and hash tables when movei
>>does not do it.
>
>So I assume that Tiger with book will be more instructive for you to find which
>opening positions to fix. I can continue using book then. Tiger had always less
>time than movei (except for the first four moves). If I use no book then movei
>will have more chances to win, but you want to win because of time advantage or
>improve the engine and win by quality of play?

You can continue to use book for tiger.
I do not complain about using book and I believe that at least in part of the
games movei is not going to have significant problems in book positions unless
tiger use aggresive learning and repeat the same opening that movei blunders
again and again but if you do not repeat the same time control again and again
then there is a hope that even aggresive learning is not going to help tiger.

The main problem of movei is the fact that your palm is 70 elo better that the
palm that Jorge used when movei does not use a significantly faster computer.

I expected movei to have some chances after seeing that faile could score 1.5/2
against palm tiger inspite of the fact that palm tiger beated stronger programs
and the fact that movei won a tournament when monarch and the baron played also
encouraged me to send movei for testing but even on Jorge's hardware I evaluated
palm tiger as favourite to win a match.

movei may have more chances after I do a significant improvement in the search
rules and some improvement in the evaluation but I am afraid that at that time
the palm is going to be significantly faster because of better hardware relative
to the hardware that it is using today so it can win the match again.

Uri



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.