Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: A comment

Author: Enrique Irazoqui

Date: 04:56:04 10/15/00

Go up one level in this thread


On October 15, 2000 at 07:17:21, Uri Blass wrote:

>On October 15, 2000 at 06:10:30, Enrique Irazoqui wrote:
>
>>On October 15, 2000 at 05:08:53, Harald Faber wrote:
>>
>>>On October 14, 2000 at 06:00:33, Mogens Larsen wrote:
>>>
>>>>The match is certainly very interesting and it suggests that the Tiger 13.0 beta
>>>>is a very strong engine. Not a big surprise considering the amount of successful
>>>>games posted here recently. The same naturally goes for Gambit Tiger.
>>>>Congratulations to Christophe Theron for that achievement.
>>>>
>>>>However, I do find the exchange afterwards slightly dubious from the perspective
>>>>of objective information. Especially since everyone, except Jorge Pichard, are
>>>>Rebel beta testers by own admisson. Don't you have a mailing list or something
>>>>similar where you can share your feelings, impressions and general well-being?
>>>>
>>>>This thread, and other similar "private" discussions among Rebel beta testers
>>>>here, is borderline commercial exhortation IMO.
>>
>>No. It is sheer enthusiasm, in my opinion quite justified.
>>
>>>We, the Beta-testers, cannot make it right. If we post games and results, guys
>>>like you complain. If we don't, others demand them.
>>
>>True. As Howard puts it, "damned if you do, damned if you don't."
>>
>>On the other hand, you also complained a year ago when I posted results of Tiger
>>11.75. Remember your comments about my "hype"? And I only posted results and
>>games, all of them, not a selection. So I find more or less understable if some
>>people get irritated for what may look like propaganda, even when it is not.
>>
>>>>Bombarding the unsuspecting
>>>>consumer with biased comments én masse seem to be the intention.
>>>
>>>
>>>Biased? Because of posting only good results and games? What can we do when
>>>there are no bad ones?
>>
>>There are no bad ones, but not everything is glamourous either. Gambit crushes
>>Fritz 6a/b, Shredder 4, Nimzo 7.32, Goliath light, but has trouble with Junior
>>6a and Hiarcs 7.32 (scoring about 50%), and in my games Tiger beats Gambit
>>neatly. This last result I find particularly interesting.
>
>It depends on the definition of bad results.
>50% against Hiarcs7.32 and Junior6a can be described as bad results.
>
>The question is what is your expectation before the testing.
>If you expect gambit tiger to be slightly better than Fritz6a/b than 50% against
>weaker program is certainly a bad result when crushing fritz6a/b is certainly a
>good result.
>
>I am also interested in the time control.
>I am interesting in 2 hours/40 moves because I read that the impression of
>testers is that gambit tiger is better at slow time control.
>Does the default tiger beat gambit tiger also in 2 hours/40 moves?

I almost never play 40 in 2. 40/40 is my usual time control and when I must rush
I play 40/20, game/30 or so.

So far there is no evidence about Gambit scoring better at slow time controls. I
thought it would, tried it and found out that its performance was roughly the
same at Game/10, 40/20 and 40/60. I only played 20 game matches, so we may need
more games and more opponents to find out.

Enrique

>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.