Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Second example for a bad opening of the opponent in the tuning of Rebel

Author: Jonas Cohonas

Date: 11:09:52 09/15/02

Go up one level in this thread


On September 15, 2002 at 13:39:19, Rolf Tueschen wrote:

>On September 15, 2002 at 13:03:14, Jonas Cohonas wrote:
>
>>Rolf you are getting a bit carried away here, if shredder or it's book makes a
>>bad move you should not blame thorsten for it, the only thing you could argue
>>here is the validity of publishing such a game in terms of accuracy.
>>
>>BTW i have read your posts for a while now and it seems you are looking for
>>mistakes no matter who posts, do you have some past issues that was left
>>unfinished?
>>
>>Regards
>>Jonas
>
>I see that your question has a bad bias but I will try to answer your question -
>although it's totally off topic of course. But I will put it in the right
>context. Hope this helps.

I don't see how it is biased?

>To give you a general answer, we should leave this specific thread here. And
>it's good to do it because what I do has nothing to do with Thorsten in
>particular but with what you observed very correctly.
>
>You write that I'm looking for 'mistakes' no matter who posts. So that is almost
>the truth. And I am thankful that you noticed that I did not go for particular
>posters. Because I never do that. But what is my interest then?
>
>You detected it but wrote it down in a bit biased fashion.
>
>I'm not looking for 'mistakes' others make. I had too much to do and in the end
>I had to correct myself too often! :)
>
>When you use the term mistake you are close to mean nitpicking and not far away,
>but of course not near too, the idea of Thorsten that I threw "mud" ans nothing
>else. This is both completely wrong.
>
>Perhaps I can explain you my intentions with concrete examples.
>
>Thorsten is now making propaganda about a new Macheide style for Rebel and he's
>doing it for months now. You know I am interested how Thorsten could achieve to
>put LASKER (!), the great Wch of human chess, into Rebel! And making Rebel
>stronger this way.

Propaganda might be a strong word here, as far as i know he is not making any
money on developing his new styles...

And if he want's to put Lasker into rebel then so be it (provided he asks Lasker
first) :)

>Now I could believe Thorstens games and record, and also Ed's record. Then
>version 51 would be proven stronger. Period.

Nobody can _prove_ anything from a personal match or matches (unless they of
course have 100's of computers to test with and even then...) and i don't think
that Thorsten is trying to prove anything, however i am certain the he believes
that his efforts are improving on the default Rebel, and apparantly Ed seems to
think that too.
So when Thorsten says that the new style is stronger, it is up to you and me for
that matter to understand that it is an estimate....his estimate.

>That I could only check if I entered the autoplayer tests myself. But no one
>will ever get me into such things. I have my own reservation for the statistical
>activities in CC. You know perhaps my preference for the apples and beans motif.

Science is good, but only to a certain point, subjectively observing the
progress of any experiment will give you valuable unscientiffic information that
you can use to scientifficly reach your goal sooner.

And no i am not familiar with your apples and beans motif, please spill the
beans, no pun intended.

>And I had a second method to take a closer look into the tests and tunings.
>I take a look into the games itself!
>
>And now I gave you a first impression. Almost every win with Black in Thorstens
>data Macheide Rebel wins against Shredder on P 400 because of Shredders book! As
>a scientist I know that the results can't prove that the tuning of Thorsten had
>this effect! I know that these wins mean nothing at all. And Thorsten has just
>confirmed that he has the same opinion!!

Maybe you need to tune your scientiffic values to a point where you can relax
and enjoy the _human_ effort Thorsten is making here, although from a
scientiffic view we can't know for sure when your settings are quite right,
might take years to test ;)

>I made that point and nothing else.

And you made it a couple of times, but do you see anything positive in what
Thorsten is doing?
As a scientist you must be familiar with success and faliure rates based on
positive and negative exposure...

>Now the next question could be why Thorsten is presenting data with no meaning
>for his main intention. And more, when he knows it himself!

The why is easy (if you fine read his posts) he feels that it is his duty (not
sure if that is the right word here) to present _all_ the material gathered from
his experiment in order to keep it scientiffic.

>Perhaps you can better understand me by now. Perhaps it would be much easier for
>you to understand me if you had my experience in science.  Because there you
>should make sure that your new data should mean a thing! But in a hobby I would
>agree that it's not such great disaster.

Well would you call it anything than a hobby?

>I confess that it's interesting to help people to understand the difference
>between disaster and good data. That's all.

As long as you willing to be subjected to the same scrutiny.

>Don't let your Sunday being darkened by such worries. :)

Oh i opened a whole can of worms when i wrote about positive thinking in CTF, i
even got an award in fuzzy thinking :)) so don't worry about my level of
positivity.

Regards
Jonas



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.