Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SSDF rating list

Author: Jan Nyström

Date: 07:38:02 06/08/98

Go up one level in this thread


Hi !

In my other message i fu.... up and deleted some of ed's
letter,(sorry) so if u want to read 100 % of the letter
 try this page.....
this computer ........ su.....and......**##¤¤¤¤###   :-(

Jan :-)

On June 08, 1998 at 06:57:29, Vincent Diepeveen wrote:

>
>And Ed what has happened with your proposal, who joins in?
>
>Chessbase?
>Weiner?
>
>Greetings,
>Vincent
>
>On June 07, 1998 at 17:22:58, Komputer Korner wrote:
>
>>On June 06, 1998 at 01:46:07, Ed Schröder wrote:
>>
>>>>Posted by Jan Nyström on June 05, 1998 at 20:43:01:
>>>
>>>>Hi !
>>>>In the CCC Opinion Poll about the ssdf rating list, what was the
>>>>result? was it 100 % against ssdf ? Some time when i read the e-mails in
>>>>ccc, it looks like ewerybody are against fritz and ssdf.....
>>>>but the only program i can read about in ccc is fritz...??
>>>
>>>>Where are we now in the "ratinglist caos" ??? ask the people
>>>>who start the "my program is the no 1 !! no my program is 14,65 points
>>>>better.... no way my dosn't use the auto 232 this and that.....
>>>>bla... bla... bla....." (MABY SOME OVER REACTION BUT....   :-)  )
>>>
>>>>what do u Schröder, Lang, etc... think about the ssdf list ?
>>>
>>>>i want to keep the list !!!!!!
>>>
>>>Same here Jan.
>>>
>>>Two weeks ago I sent the below email to the main programmers or people
>>>in charge. Since Ossi Weiner in his wisdom decided to quote the whole
>>>email to 50-60 people I can better make it public as well. It was an
>>>unpleasant experience for me that I did not receive one single reply!
>>>Bottom Line, better forget automatic testing. We are going back to the
>>>stone age :(
>>>
>>>I didn't manage, perhaps you the customer can convince them after all
>>>I am just a competitor. It's my opinion, without AUTO232 no SSDF and
>>>CCL anymore. The latter is 100% true for CCL as I will quit CCL for
>>>sure as it makes no sense to me to test Rebel10,11,12... against old
>>>and out-dated versions.
>>>
>>>This doesn't mean *I* will remove auto232 from Rebel10. Rebel 10.0
>>>will support auto232 as usual. I am not so afraid of all the book
>>>preparations as it is happening already since day one of SSDF.
>>>
>>>Here is the group-email...
>>>
>>>- Ed Schroder -
>>>
>>>------------------------------------------------
>>>Date: Fri, 22 May 1998 16:33:12 +0200 (CEST)
>>>From: Ed Schroder
>>>Subject: AUTO232 and SSDF
>>>Cc: Marty Hirsch
>>>    Stefan Meyer Kahlen
>>>    RICHARD LANG
>>>    Mark Uniacke
>>>    Helmut Weigel
>>>    Ossi Weiner
>>>    Matth. Wuellenweber
>>>    Eric Hallsworth
>>>    Thoralf Karlsson
>>>    Frederic Friedel
>>>
>>>Dear colleagues,
>>>
>>>For the next version of Rebel I will (still) release the program
>>>including the possibility of auto232. I hope all of you will do
>>>the same.
>>>
>>>It's my opinion that removing auto232 from our programs will result
>>>in a dying or uninteresting SSDF list. The same applies for my own
>>>started CCL, I assume you heard of it.
>>>
>>>It's also my opinion that WE need a list because our customers WANT
>>>a list. If there is no list I fear the topic computer chess will
>>>slowly die.
>>>
>>>Bottom line, we have a common interest to keep computer chess alive
>>>and healthy and compete with each other. Too many good things has
>>>disappeared (or are disappearing), Aegon, Harvard Cup, will there
>>>be a WCM this year? I doubt it. This tells people (our customers)
>>>that computer chess is dying and they might lose interest with the
>>>predictable result, lower sales.
>>>
>>>So my proposal is let's have a gentleman agreement, we ALL release
>>>our next version with auto232. The gentleman agreement is *ONLY*
>>>valid if *ALL* have promised by responding on this email with YES,
>>>see below.
>>>
>>>If one (of the list below) doesn't want or does not reply I consider
>>>my try as fruitless and there will be no gentleman agreement and
>>>everybody simply does what he thinks is best.
>>>
>>>But I do hope you all will see the importance of the subject as we
>>>have a common interest here to cooperate. My vote is a clear YES.
>>>
>>>Now I can understand the problem you might have with the Fritz5
>>>testing on SSDF. Perhaps you have heard of my official and also
>>>public complaints and even that I felt forced to withdrawn from
>>>SSDF completely.
>>>
>>>Also here I like to make a proposal, why don't you Thoralf promise
>>>us (the programmers) for the future (a) not to accept auto players
>>>that are not public and (b) not to accept special hardware demands
>>>again? I am sure that it would help to convince us to say YES to
>>>the gentleman agreement. In that case I would be happy to participate
>>>on your list again.
>>>
>>>Below you can sign and then return the email to me or in this group.
>>>
>>>  I agree to the stipulations of this email,
>>>
>>>  Thoralf Karlsson         ...
>>>  Richard Lang             ...
>>>  Stefan Meyer-Kahlen      ...
>>>  Marty Hirsch             ...
>>>  Frans Morsch             ...
>>>  Mark Uniacke             ...
>>>  Chrilly Donninger        ...
>>>  Ed Schroder              YES
>>>
>>>In order to keep this going I want to suggest a deadline of June 1,
>>>1998. Before this date I would like you to react to make this
>>>gentleman agreement valid.
>>>
>>>Kind regards,
>>>
>>>Ed Schroder
>>>
>>>PS, excluded from the CC-list are Frans and Chrilly as they don't
>>>want to receive group email. I trust that Fred and Ossi will make
>>>sure they get the message. Also to Eric my request is to ensure
>>>that Marc gets the message.
>>
>>I would like to add that without automatic testing, the progress of
>>computer chess slows. Testing a new idea will be horrendous once the old
>>programs with autotest drivers become too old to test against.
>
>This is true, but just think of what Chessbase did. Send in a special
>version, and the old version to play it takes many many hundreds of
>hours. So where they can develop their program
>autoplaying you at x computers without taking them time,
>same time, you are busy playing computer versus computer to see
>how well your prog is doing.
>
>Also you can cook against other programs where cooking for your program
>takes say 4000 hours manual (for 1000 test games, which otherwise gets
>done
>in 2 weeks automatically).
>
>>--
>>Komputer Korner
>
>Vincent



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.