Author: Robert Hyatt
Date: 06:49:04 01/15/98
Go up one level in this thread
On January 15, 1998 at 09:19:45, Robert Hyatt wrote:
>On January 15, 1998 at 08:53:27, Amir Ban wrote:
>
>>On January 15, 1998 at 08:01:10, Robert Hyatt wrote:
>>
>>>On January 15, 1998 at 05:35:18, Amir Ban wrote:
>>>
>>>>
>>>>Maybe we can break the problem into two easier parts:
>>>>
>>>>- Who sees a plus score for Qxb3 ?
>>>>- Who sees a draw score for only shuffling the queen ?
>>>>
>>>>Amir
>>>
>>>
>>>I disagree. I have seen Cray Blitz and Crafty win many such positions.
>>>It
>>>takes a lot of time, but being ahead +12 is *not* going to draw if the
>>>program
>>>understands the 50 move rule and repetitions. IE sooner or later it is
>>>going
>>>to see that another couple of queen moves draws. And it is going to
>>>then
>>>search for a way to push a pawn, or capture something. Being ahead +12
>>>is
>>>enough to allow the queen sac and still remain far enough ahead in
>>>material
>>>to win.
>>>
>>>In 1986 we had a game somewhat like this at the WCCC in Germany. Except
>>>we were only up 1.5 pawns or saw. After about 30-35 moves of shuffling,
>>>Cray Blitz found what we already knew... it had to give up its pawn, but
>>>then could win. It did, but only when urged on by the threat of a draw.
>>>I think many programs would win this, assuming the game is played for
>>>another 45 moves. +12 is simply not going to draw. Now if we are into
>>>"elegant" wins, and if this has to be solved *now* or never, then we
>>>have
>>>a problem. But I've never seen a +12 position that wasn't winnable,
>>>except for a few that are close like Q vs bp...
>>
>>
>>Very well. So how about this discount: Set the 50-move counter to
>>something close to 50 and find the winning PV.
>>
>>Amir
>
>
>this I'll run... results soon...
Heres the results run on Cray Blitz (Crafty's machine seems to stay busy
playing ICC games...)
I set the 50 move counter to 6 plies away from draw. I snipped a lot of
noise around the point where it started seeing draw, as it was changing
it's mind, then failing high, then failing low, etc. I stopped after 10
seconds as it seemed to be "on track". I want to try this on Crafty
later,
as I'm not sure the two connected passed pawns are valuable enough to
make
it want to do this, period, unless the search can go deep enough to see
them
advancing. CB is a little more aggressive on evaluating connected
passers,
because it searches so deeply...
depth time score variation (1)
1 0:00 7.441 Qg2
1-> 0:00 7.441 Qg2
2 0:00 7.364 Qg2 Bc6
2-> 0:00 7.364 Qg2 Bc6
3 0:00 7.364 Qg2 Bc6 Qg3
3-> 0:00 7.364 Qg2 Bc6 Qg3
4 0:00 7.336 Qg2 Bc6 Qg3 Kb6
4-> 0:00 7.336 Qg2 Bc6 Qg3 Kb6
<noise deleted>
8 0:00 -- Qf1
8 0:00 0.000 Qf1 Bc6 Qg2 Ba4
8 0:00 0.234 Qd1 Bc6 Qxb3 cxb3 Kxb3 Kb6 c4 Kc5
8-> 0:00 0.234 Qd1 Bc6 Qxb3 cxb3 Kxb3 Kb6 c4 Kc5
9 0:00 0.234 Qd1 Bc6 Qxb3 cxb3 Kxb3 Kb6 c4 Kc5
Kc3
9-> 0:00 0.234 Qd1 Bc6 Qxb3 cxb3 Kxb3 Kb6 c4 Kc5
Kc3
10 0:00 0.302 Qd1 Bc6 Qxb3 cxb3 Kxb3 Kb6 c4 Kc5
Kc3
Bd7
10-> 0:00 0.302 Qd1 Bc6 Qxb3 cxb3 Kxb3 Kb6 c4 Kc5
Kc3
Bd7
11 0:00 0.388 Qd1 Be8 Qxb3 cxb3 Kxb3 Kb6 Kc4 Bf7+
Kd4 Kb5 b4 Bc4
11-> 0:00 0.388 Qd1 Be8 Qxb3 cxb3 Kxb3 Kb6 Kc4 Bf7+
Kd4 Kb5 b4 Bc4
12 0:00 0.388 Qd1 Be8 Qxb3 cxb3 Kxb3 Kb6 Kc4 Bf7+
Kd4 Kb5 b4 Bc4 Ke5
12-> 0:00 0.388 Qd1 Be8 Qxb3 cxb3 Kxb3 Kb6 Kc4 Bf7+
Kd4 Kb5 b4 Bc4 Ke5
13 0:00 0.293 Qd1 Be8 Qxb3 cxb3 Kxb3 Ba4+ Ka3 Bd1
c4 Kb6 b4 Kc6 c5 Bf3
13-> 0:00 0.293 Qd1 Be8 Qxb3 cxb3 Kxb3 Ba4+ Ka3 Bd1
c4 Kb6 b4 Kc6 c5 Bf3
14 0:00 0.293 Qd1 Be8 Qxb3 cxb3 Kxb3 Ba4+ Ka3 Bd1
c4 Kb6 b4 Kc6 Kb2 Bf3 c5
14-> 0:00 0.293 Qd1 Be8 Qxb3 cxb3 Kxb3 Ba4+ Ka3 Bd1
c4 Kb6 b4 Kc6 Kb2 Bf3 c5
15 0:00 0.269 Qd1 Be8 Qxb3 cxb3 Kxb3 Ba4+ Ka3 Bd1
c4 Ka6 b4 Kb6 Kb2 Bf3 c5+ Kc6 Kb3
15-> 0:01 0.269 Qd1 Be8 Qxb3 cxb3 Kxb3 Ba4+ Ka3 Bd1
c4 Ka6 b4 Kb6 Kb2 Bf3 c5+ Kc6 Kb3
So it isn't hopeless, but needs some help from the 50 move rule. This
was
run on a 1 processor C90. I suspect the 32 processor T90 could reach 20
plies in 1 second, and that this could be searched to a depth of 30 or
so,
with no trouble. In that case, I'm sure the scores would be much higher
as
by then the pawns would be well on their way...
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.