Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: An amazing Tiger-2-result!

Author: Harald Faber

Date: 07:32:27 09/24/00

Go up one level in this thread


On September 24, 2000 at 10:18:26, pavel wrote:

>On September 24, 2000 at 02:26:26, Harald Faber wrote:
>
>>Hi all,
>>sorry for not posting frequently but my time is limited...
>>Besides some improvements in the GUI one expected a stronger engine for the new
>>Tiger. You can imagine how astonishe I was when I played several games with the
>>new Tiger/K7-500 against Shredder4/K6-200 (g/120) and Shredder led by 4.5-2.5
>>after 7 games. Although I know how strong Shredder is even on tha slow
>>200MHz-machine, but this result was unbelievable.
>
>it is not unbelievable, shredder 4 is still strong enough to be a threat to any
>chess program in the world.


No doubt, but on a much slower hardware (2.3x-2.6x) it is VERY unlikely that
Shredder could have a chance to win even such a short match.


> I was not sure if this was due
>>to strange engine progress or due to the book. Because of some trouble getting
>>the London-book started, in these 7 games the "original", small book was used.
>>After these 7 games I stoped that match, I have seen enough. I managed to get
>>the London-book to work finally successful. I started the match again, and what
>>I say is just WOW! I have stopped the match at a standing of 5-0 for Tiger!
>>Not even Tiger 12e crushed Shredder that clear. And look at the games, there are
>>wins in the endgame although Shredder is said to be very strong in there and had
>>access to all 5-man TBs!
>>
>>See the games below and enjoy!
>>(And get eager for the new Tiger :-))
>
>yes and also get eager for the new shredder....


Be sure I'll get the new Shredder and CM8000 too. :)


>your 'experiment' proves only one thing. that is, tiger won only because of its
>better book, since according to you tiger played much better when book has been
>changed.


Don't forget that the book counts to the program as well as the engine does.
The book I used first was certainly a 2nd-class-one coming with the
LChess-engine.
And if you take a look at the games you will see that Tiger did not win the
games because of winning lines out of the book, no way.


>also note that shredder is one of those programs which plays *significantly*
>better with faster processors, ( I think I read that several times somewhere in
>this forum). In same hardware things could be differant.


I didn't say that Tiger crushed Shredder on same hardware.
However, it is not common that Shredder4 got crushed by 5-0 on my 200MHz
machine. Shredder always managed to hold matches even, but surprisingly not this
time.
That 5-0 makes me say WOW!
Tiger and no other program can do better than winning X - 0.


>then ofcourse there is this question of not having enough games....
>thats why we have something called "SSDF"  ...:))
>Pavel


I am not SSDF, SSDF will test Tiger2 in the future, but not now. And
beta-testing is not restricted neither wished to only play auto232-games in an
amount SSDF does, that would be something Ed and Christophe could do on their
own.
My and other Tiger-2-results just give an impression, statistical value will
come later by SSDF.




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.