gnugo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gnugo-devel] pattern proposal [was]Twogtp test replay result


From: max-d
Subject: Re: [gnugo-devel] pattern proposal [was]Twogtp test replay result
Date: Thu, 19 Feb 2004 00:40:26 +0100

----- Original Message -----
From: "SP Lee" <address@hidden>
To: "GNU Go development" <address@hidden>
Sent: Wednesday, February 18, 2004 6:29 PM
Subject: Re: [gnugo-devel] pattern proposal [was]Twogtp test replay result


> >
> > Pattern EC220
> > ......
> > OoO..*
> > ------
> >
> > :8,OCe
> >
> > md
> >
> >
>
> This pattern could be used in some cases, but in the game it could
> introduce a big ko fight. A possible sequence is B:T8 W:S11 B:T10 W:T12
> B:S12 W:S13. My opinion is such long jump will make the later variations
> too complicated for gnugo.


i agree

???????
..o....
oOoO..*
-------

??d????
..b....
aOcO..*
-------

is there any function which says something like that:
;if d=X then a||c||b must be O

subsidiary question:
When a pattern vertex [x,y]  is marked "o" and the pattern is constrained O
and there is actually a friendly stone on the board , does this stone must
belong to an alive dragon ?

in advance thanks
md
>
> SP Lee






reply via email to

[Prev in Thread] Current Thread [Next in Thread]