Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

Re: Chess monitoring

__/ [mbstevens] on Tuesday 13 September 2005 06:44 \__

> Roy Schestowitz wrote:
>> __/ [Charles Sweeney] on Monday 12 September 2005 21:08 \__
>> 
>> 
>>>GreyWyvern wrote
>>>
>>>
>>>>And lo, Charles Sweeney didst speak in alt.www.webmaster:
>>>>
>>>>
>>>>>"Automatic adjudication may occur if the World Chess Network detects
>>>>>that you or your opponent is using computer assistance, since
>>>>>computer assistance is prohibited on the World Chess Network"
>>>>>
>>>>>Like I said, I don't know how they do it.
>>>>
>>>>I would suspect that they have several of the most popular chess
>>>>algorithms playing along each game and if you matched a number of
>>>>moves in  a row of any one program, you would be terminated.
>>>>
>>>>No process monitoring required.
>>>
> I havn't been following computer chess closely for a while,
> so I could certainly be be _wrong_ about the state of the
> technology, but:
> 
> Don't a lot of programs play 'book' openings for a number of
> moves in?  Many human players also memorize book openings, so
> you couldn't be faulted there by the program.


That's what thing that I thought about. There are certain move that any good
engine would go for, as well as an experienced player. Likewise gameends.
Moves made in the middle are the ones that count, but even then there is a
certain pattern recognition involved, which means that a good player is
likely to spot all that is computationally perceived as plausible and wise.


> And, many moves in, by the time either side makes a move that
> goes out-of-book it would just be surprising if they had a
> stored game that covers a particular position.  So, I'm
> wondering just how much 'automatic adjutication' may be an empty
> threat?


I am assuming that there is a more cunning algorithm here, whereby you
match/lock onto a certain algorithm and then trail a 'chain' of moves.

For example, let us say you moved the rook from A1 to A2 at some for some
given board state. This move was also suggested by algorithmX and
algorithmY. You later moved the rook to A7, which is something algorithmX
would do, but not algorithmY. You begin to suspect that the player is
'motored' by algorithmX and after 5 moves that correspond with X, the
player gets disqualified. This is all just a speculation of mine...

Roy

-- 
Roy S. Schestowitz      | "On the eighth day, God created UNIX"
http://Schestowitz.com  |    SuSE Linux    |     PGP-Key: 74572E8E
  6:45am  up 19 days  0:19,  3 users,  load average: 0.42, 0.63, 0.75

[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index