Author: Christophe Theron
Date: 09:05:45 11/11/98
Go up one level in this thread
On November 11, 1998 at 08:27:06, Micheal Cummings wrote:
>I can understand your situation, but you put out a list in which alot of
>programmers refer to on their web sites for their chess programs to show how
>strong their programs are compared to the other programs on the market.
>
>I have also seen SSDF referred to in magazines when chess programs are
>advertised. But personally even knowing your situation after you explained it, I
>would be unhappy knowing that I am puttng out a list knowing that the rankings
>are incomplete.
As you say the list is not wrong, but it may be partly incomplete.
Note that ChessMaster is not totally absent from the list, so that you are not
left in the dark about this program.
It has been tested on P90, and most of the top programs have been tested on this
hardware too. So you have some good points of comparison.
I know that some are going to claim that it would gain more than the other
programs on P200MMX. In this case, you could try the following:
1) browse the list and find the program that gains the more points when
switching from P90 to P200MMX.
2) Add this to the elo ranking of CM on P90.
3) Compare this to the list of programs tested on P200MMX.
You will have an optimistic idea of what the ranking of CM5000 would be on
P200MMX.
Next claim will be: but CM6000 is much stronger than CM5000.
If it is the case, then why Mindscape does not provide the Auto232 testing
facility? It takes maybe 2 or 3 days to implement this in a program. It may be a
little harder for Windows programs, but source code for doing this is available
from Stefan Mayer-Kahlen (up to them to negociate the use of these sources).
The blame should be put on Mindscape, not on the SSDF.
SSDF is doing a great work, since several years. They only get critisism for
this. It's really unfair. Their task is HUGE and the least programmers that can
do is to provide a little help.
Reasons for not providing Auto232 support may be:
1) you don't care about the list
or
2) you prefer that your ranking stays unknown
or
3) you don't have time to implement Auto232
I don't believe in (3)...
So in any case, the SSDF should not be blamed. We should all be thankful for the
great job they are doing. If they make mistake, let's help them. We need them!
>It is just like me going and selecting programs that fit my criteria and saying,
>"okay this program I rank as the best, with the criteria I set".
This is wrong. And again the blame is to put on Mindscape.
>Yes SSDF does a great job in what they have, but I am just sad that there is no
>way to get some sort of indication of ranking for this program on your list.
>
>If I had the time I would try and get a few games played manually with
>chessmaster and send them in, but alas I do not.
This is exactly the problem. You want them to do the job, and you blame them.
And in the same time you admit that the task is not easy.
>It is just an unfortunate situation that chessmaster can pull of some great
>games and we can post them and say wow what a move or game, and in the end we
>are still blind as to how it rates against other programs.
>
>All round it is sad that SSDF cannot get a ranking for this program and that
>Mindscape will not put and autoplayer into the program.
It would be fair that Mindscape help the SSDF testers. Maybe there are more
programmers on ChessMaster than testers in the SSDF!
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.