RE: [computer-go] Pattern matching - example play

[Vincent Diepeveen <diep@xxxxxxxxx>]

At 12:01 2-12-2004 +0100, Persson, Magnus wrote:
>
>> -----Original Message-----
>> From: Don Dailey [mailto:drd@xxxxxxxxxxxxxxxxx]
> 
>> I guess  I'm saying that if  I wanted to  become a master in  chess, I
>> could  do  this on  my  own  by  playing tournaments,  reading  books,
>> studying games and especially analyzing my own games.  But if I wanted
>> to become a  master FAST, I would  have to do all of  those things and
>> additionally hire  the right teacher.   (Find the "right"  teacher may
>> not be easy.)
>
>Good that we agree!
>
>After reaching 2 dan 10 years ago I have not improved (perhaps I am
>stronger in the fundamentals of the game - compensated by forgetting a
>lot of stuff that were simple memorization).
>
>I think the strong teacher option is what I need.

First of all the chess level in USA is below all standards in the 'common'
regions (there are some russian GM's there which dominate the top 100 in
USA). A National Master in USA which means 2000 USCF rating, would have
1800 rating in europe.

With 1 exception of an exceptional game against a 1880 rated youth talent
who was underrated and a few rating lists later had a high rating (2100+),
I have never in my life lost from someone in a slow game from what is
called Master level in USA.

Yet i have taught several persons and some of them were rated like 1200.
One of them is still rated 1200. Which means he has a 0% chance to win from
a 1800 player (draw occurs more regular in chess). 1800 is very comparable
to dan-2 of course. 

It's just wishful thinking that you can train persons to International
Master level. Majority will not even get halfway. And it sure isn't the
teachers fault.

One of the big problems to overcome to raise yourself into a higher level
is a good life&death in go, and good tactics in chess.

If you always keep losing your groups, forget it. You must make 0 mistakes
there, or you will always soon stop to grow.

In go by playing games in a regular way i can get easily dan-levels in a
few years time, and a professional dan grade go player will be able to play
> 1800 level in chess too without any problems. Matter of a few years of
practicing. 

The most important reason for such statement that wherever someone is
professional, go or chess, we just see so many of those players join both
go and chess competitions, and they get > 1800 in chess and some strong dan
level in go. You just know such players can play without blundering and
real strong.

However the majority of players will never ever achieve such levels and
keep blundering away groups as if it is Sahara sand. So not a single type
of lesson, selflearning and whole life playing go will be able to change that.

An ex-world champion draughts who is weekly playing chess here in the
region, has a chessrating of 1880. The important rating borders are at this
moment :

1200, Average club player in USA
1400,
1500  
1600  Average club player in Europe
1800  Regional club champions
2000  National level in Europe and Master in USA
2200  Masterclass player
2300  FIDE Master titled player
2400  International Master
2500  Grandmaster
>= 2600  world top 100
>= 2700  world top 10
>  2800  Kasparov



Vincent

>Still my point here is that there is always a bottleneck in what a
>teacher can communicate to a student. The teacher is not aware of all
>processing his brain actually does. And furthermore it is not enough for
>the student to simply remember what is communicated, it is more a slow
>reorganization of how go positions are processed.
>
>--
>Magnus Persson
>_______________________________________________
>computer-go mailing list
>computer-go@xxxxxxxxxxxxxxxxx
>http://www.computer-go.org/mailman/listinfo/computer-go/
>
>
>
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/