Author: Christophe Theron
Date: 12:58:10 11/01/00
Go up one level in this thread
On November 01, 2000 at 08:38:30, Peter Skinner wrote:
>On November 01, 2000 at 07:04:12, Jouni Uski wrote:
>
>>On November 01, 2000 at 06:56:27, Chessfun wrote:
>>
>>>Current completed match scores;
>>>Gambit 1.0 +13 -11 =16
>>>Fritz 6a +11 -13 =16
>>>
>>>Tiger 13.0 +14 -11 =15
>>>Fritz 6a +11 -14 =15
>
>>>Current
>>>Tiger 13.0 113.5/200
>>>Gambit 1.0 113.0/200
>>>Fritz 6a 112.5/200
>>
>>I have got impression, that Tiger/Gambit are superior to Fritz6a. But this
>>isn't indicating it at all.
>>
>>Jouni
>
>The match scores between the engines is a better focus. Both Tiger, and Gambit
>Tiger, won their matches vs Fritz6a. Sure by a small margin, and even I expected
>better results.
>
>Here is a result from ICS play of mine:
>
>Lilkikr/Species: PIII 500 196 megs ram Tiger 13.0
> vs
>Pecable: Athlon 700 128 megs ram Fritz 6a/b
>
>Lilkikr/Species is ahead by +18 games.
>
>Now here is a little tid bit of info about hardware with Tiger 13.0.
>
>I tried Tiger on my "work" computer being a PIII 933 512megs of PC133 ram, and
>found hardly any significant difference is play. The knps were higher, but the
>same moves in the games that the 933 played, were able to be replicated on the
>PIII 500. This to me is positive, as I use relatively small hardware vs the big
>boys, yet I have a superior record vs them. I honestly didn't see much of an
>improvement using the PIII 933, so I have not switched to the bigger hardware.
>
>This is a great selling point.
>
>I have noticed that Tiger did not run as well on my older K6-3 500 system. The
>knps were lower than the PIII 500, and I found the depth Tiger reached was not
>close to what the PIII 500 got. About 2 ply difference.
Peter you should check this carefully, because it cannot be.
A 2 plies difference would mean that one of the computers is 5 to 9 times
slower. Of course there is not such a difference.
The difference must be 10% or less in speed, that means it is not possible to
see the difference in ply depth without doing a study on 100 moves or more.
Christophe
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.