I just looked at this position and it looks like a win for black in the
first position. Many Faces evaluates it as a win for black, and plays c1 to
save the lower left black group with almost no thinking time.
Mogo is correct because the lower left black group is not dead.
David
-Original
David Fotland: [EMAIL PROTECTED]:
I just looked at this position and it looks like a win for black in the
first position. Many Faces evaluates it as a win for black, and plays c1 to
save the lower left black group with almost no thinking time.
Mogo is correct because the lower left black group
I think that the original description of the position should have said
'killable' rather than 'dead', and that David missed the fact that
it is White to move.
At 08:06 27/03/2008, Hideki wrote:
David Fotland: [EMAIL PROTECTED]:
I just looked at this position and it looks like a win for black
Quoting Petr Baudis [EMAIL PROTECTED]:
Please note that pachi1 had a rather embarassing bug of starting the
random playouts with wrong color (so if the last tree node was black,
the playout would start with black as well). pachi2 has this bug fixed;
the ELO rating is still not settled, but so
Magnus Persson wrote:
Quoting Petr Baudis [EMAIL PROTECTED]:
Please note that pachi1 had a rather embarassing bug of starting the
random playouts with wrong color (so if the last tree node was black,
the playout would start with black as well). pachi2 has this bug fixed;
the ELO rating is
Mark Boon wrote:
What I have now takes 10-15 microseconds on a 2Ghz Intel computer to
find all the patterns on a board (on average for 9x9, for 19x19 it's
more like 15-20 usec)
From your difference between 9x9 and 19x19 I assume that you are updating
the patterns of the cells after a
Santiago wrote:
... Oops, wrong name ...
(Santiago is my official name, because I was born in a dictatorship that
did not allow foreign names. After that, I was too lazy to change it.
Jacques, like my French grandfather, is my real name.)
Jacques.
On Wed, Mar 26, 2008 at 08:50:28PM -0700, David Fotland wrote:
You are right.
Well, I did compete for this prize about 15 times, so I hope so :)
Are there any current prized computer tournaments or does anyone know
about Ing foundation or anyone else planning to resume the challenge?
What
Mark Boon wrote:
Thanks for the pointer Don, that might be worth looking into at some
point.
When you say 'amazing accuracy despite this speed and simplicity' I
still have to ask: how fast?. I think 10usec is actually pretty
fast. But when talking about using it during MC playouts for
Jacques,
Yes, I do an incremental update so the board-size should be almost
irrelevant. I'm not sure why I see any difference between 9x9 and
19x19 but it may have to do with the fact that the edge cuts a lot of
pattern seach short.
On 27-mrt-08, at 10:13, Santiago Basaldúa wrote:
--- David Fotland [EMAIL PROTECTED] wrote:
I just looked at this position and it looks like a
win for black in the
first position. Many Faces evaluates it as a win
for black, and plays c1 to
save the lower left black group with almost no
thinking time.
Mogo is correct because the lower
I've been thinking a bit about the collection of
patterns from games, whether of professionals or of
programs.
It is possible to get some remarkably high correlation
between the moves played by pros and a predictor - yet
still not have a good program. Why?
One possible answer is that many moves
On Thu, Mar 27, 2008 at 12:14:06PM -0700, terry mcintyre wrote:
Suppose a group can be defended - four liberties in a
row, for example. If the opponent plays inside those
four liberties, you play to divide the area into two
eyes - unless the situation is such that the group has
a second eye
David Fotland wrote:
I just looked at this position and it looks like a win for black in the
first position. Many Faces evaluates it as a win for black, and plays c1 to
save the lower left black group with almost no thinking time.
Mogo is correct because the lower left black group is not
terry mcintyre wrote:
I've been thinking a bit about the collection of
patterns from games, whether of professionals or of
programs.
It is possible to get some remarkably high correlation
between the moves played by pros and a predictor - yet
still not have a good program. Why?
In my
Actually, a better way to put this:
King and Pawn positions are generally understand so well by Grandmasters
that they know what the outcome of the game will be once it appears on
the board.
Therefore, at least 1 player is actively trying to avoid this ending,
because the game is essentially
--- Don Dailey [EMAIL PROTECTED] wrote:
terry mcintyre wrote:
I've been thinking a bit about the collection of
patterns from games, whether of professionals or
of
programs.
It is possible to get some remarkably high
correlation
between the moves played by pros and a predictor -
Oops, yes you are right. I shouldn't try to look at a position when I'm
half asleep.
David
-Original Message-
From: [EMAIL PROTECTED] [mailto:computer-go-
[EMAIL PROTECTED] On Behalf Of Matthew Woodcraft
Sent: Thursday, March 27, 2008 12:34 PM
To: computer-go
Subject: Re:
David Fotland wrote:
I just looked at this position and it looks like
a win for black in the
first position. Many Faces evaluates it as a
win for black, and plays c1 to
save the lower left black group with almost no
thinking time.
Mogo is correct because the lower left black
The Ing prize stopped when Mr Ing died. He was very interested in computer
go. His foundation still funds many go tournaments, but none for computers.
The current computer go tournaments I know of are:
European go congress (Late July)
US go congress (August, small prize this year)
Gifu
David Fotland: [EMAIL PROTECTED]:
The Ing prize stopped when Mr Ing died. He was very interested in computer
go. His foundation still funds many go tournaments, but none for computers.
The current computer go tournaments I know of are:
European go congress (Late July)
US go congress (August,
On 27, Mar 2008, at 3:39 PM, David Fotland wrote:
US go congress (August, small prize this year)
Since I announced that HSRF will supply $1,000 total of prize money for
computer Go at the US Congress this year, another person contacted me
and has agreed to add a minimum of $250. The offer
On undefined, Don Dailey [EMAIL PROTECTED] wrote:
In some of my pattern learning experiments, I discovered that only a
very small subset of possible patterns occur on the real board, and yet
for a game tree searcher it would be pretty important to understand
those patterns that are
23 matches
Mail list logo
