Author: Christophe Theron
Date: 23:58:57 11/18/99
Go up one level in this thread
On November 18, 1999 at 21:25:12, Dann Corbit wrote:
>On November 18, 1999 at 20:06:37, Christophe Theron wrote:
>[snip]
>>OK. I understand.
>>
>>So maybe we can find a way to agree somehow.
>>
>>Let's say the list is not perfect due to lack of physical resources, that it is
>>necessarily unfair because every program cannot be tested on every hardware, but
>>that the people doing the list are trying to make it as fair as possible given
>>the physical limitations they have to deal with.
>>
>>If we at least agree that they are trying to do their best and that it's not
>>easy for them to please everyone, well that's a good compromise.
>
>I happen to agree with your assessment.
>
>An interesting subplot is that the SSDF is supposed to be in league with the
>evil Chessbase empire (or so the story goes).
>
>But Chess Tiger is a Rebel product in direct competition with Chessbase.
>Shouldn't Chess Tiger have been tested on a Commodore 64 in order to make it
>look bad? How is it that an upstart should suddenly shine like the sun and get
>the good hardware? I suspect that the following is true:
>0. A tester has a copy of C.T. (Hmmm... call me dense -- I just noticed that
>the initials match with Christophe Theron) and is also enamored with the product
>and so decided that this was the test they would like to attempt.
>1. C.T. provides a simple testing interface so that it is not too difficult to
>run the tests (some sort of autoplayer ability)
>
>I could be wrong on both counts but it seems plausible to me.
>
>I don't see any evil corruption by the SSDF at all -- but maybe I'm just naiive.
>;-)
Actually you got it right.
Paul Petersson is a SSDF tester. He has read some of my posts last year about
the idea of a slow hardware tournament.
He wanted to organize one (I think he still has some 386s). So he asked me for
my program and I have send it to him for this purpose.
After quite a while, he told me he had done some test games for private use and
found that Tiger was rather good. So he asked me if I was interested to have it
listed in the SSDF list. But he told me from the beginning that it would only be
tested on P90, because they were the only computers available for the test
(faster computers were running the top programs).
So I accepted the idea, and sent Tiger 11.8 to him. I thought it was a good mean
to measure how much work was still needed to equal the best programs.
A while back, he came back to me and told me that Tiger was doing very well in
the test, actually leading the P90 list with a clear margin!
This was unexpected. I had not planned seriously to enter the SSDF list before,
because I considered my program to be far from acceptable. I knew Tiger was good
at blitz but thought it was not as good at long time control games. Also it had
only a ridiculous book and no learning at all.
That's how Tiger arrived on the SSDF list.
Then Thoralf Karlsson noticed Tiger's ranking, and deduced from it that it could
be near the top on fast hardware.
That's why they are testing Tiger for their next list.
I think this should stop any suspicion. What are they interested in? They are
interested in testing, ranking and publishing a list of the best chess programs.
Don't look for anything else.
You know, they could have very easily refused to test Tiger 12.0. But they
didn't. They could have very easily refused to test CM6000. But they didn't, and
CM6000 was above ChessBase programs on P200MMX.
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.