Subject: Sam Sloan Almost No-Show at World Open
From: (Sam Sloan)
Date: 7/14/2003 12:18 PM Eastern Daylight Time
Message-id:

On Mon, 14 Jul 2003 00:37:43 GMT, "George John"
wrote:

I don't see how R+B vs R being clearly wrong and R vs N being unclear would
cause one to conclude that N+R vs. B+R is unclear as well. To me it's
clearly correct. I think, given AMPLE time and nothing unusual about the
position, a 1500 could hold against a 2200 better than 90% of the time.


To bad you do not play chess. If you could play our game, you would
know that a 2200 player would beat a 1500 player almost every time in
this endgame.

You would be advised to limit your comments where chess playing skill
are not a factor.

http://www.64.com/uscf/ratings/12679293

Sam Sloan

Maybe I should play George 10 games with some time on our clocks in some random
position where I get R+B vs. R+N?

John Fernandez

"Fifiela" wrote in message
...
I ask you once again, if both players had ample time, do you think a
1500
would hold against a 2200 more than 90% of the time in a K+R+N vs. K+R+B
endgame and nothing unusual in the position?

I think that a 1500 would lose this position at least 25% of the time;
Maybe
50%. That's why their a 1500.

Why do you say this? I have been playing this end game with Fritz. The
moves are incredibly easy to find quickly, and I draw after 50 moves each
time. I'm a very rusty 1300 something player. I would be astonished if a
1500 could not find them just as easily given ample time to make each move.

George

"Sam Sloan" wrote in message
...
On Mon, 14 Jul 2003 00:37:43 GMT, "George John"
wrote:

I don't see how R+B vs R being clearly wrong and R vs N being unclear
would
cause one to conclude that N+R vs. B+R is unclear as well. To me it's
clearly correct. I think, given AMPLE time and nothing unusual about the
position, a 1500 could hold against a 2200 better than 90% of the time.


To bad you do not play chess. If you could play our game, you would
know that a 2200 player would beat a 1500 player almost every time in
this endgame.

I have been playing the Knight side of this K+R+N versus K+R+B endgame
against Fritz. I have found it very easy to draw against Fritz every time.
Please explain to me why you think it likely that a 2200 is going to outplay
Fritz in this endgame? If it's very easy for me to draw this endgame, why
would it be difficult for a 1500?

Please explain to me how the player with the Bishop can force a win against
the player with the Knight. Given my experience with Fritz, all the player
with the Bishop can do is hope that the side with the Knight will badly
blunder.

-snip-

George

"John Fernandez" wrote in message
...

Maybe I should play George 10 games with some time on our clocks in some
random
position where I get R+B vs. R+N?

John,

Do you think you can do better than Fritz 6 on a 512MB, 1.6GHz system? If
not, please send me some random positions to try. Maybe the positions I'm
trying are too easy. Also, the position that was on the board from the
actual game in question when the first draw claim was made would be of
interest.

I have been making certain my N is on a color opposite the B, and I am
careful to put my K and R on opposite colors, too, whenever there is nothing
better to do. I try to cut off the K with the R, or check the K or chase
the B with the R to burn up moves. So far, these very simple ideas have
worked well against Fritz.

George

I'm a very rusty 1300 something player. I would be astonished if a
1500 could not find them just as easily given ample time to make each move.

A 1500 over the board with the clock ticking FASTER AND FASTER and specators
CROWDED around the board and DREAMS OF GLORY clouding his judgment would
blunder it away at least 25% of the time. Also fatigue at the end of a long
game is a factor with 1500's vs master.

A certainly wouldn't expect any master to give me a draw on this position until
50 moves were made.

A full tournament game for keeps is a very different enviroment from home alone
with the computer just playing the endgame for fun.

"George John" writes:
Please explain to me how the player with the Bishop can force a win against
the player with the Knight. Given my experience with Fritz, all the player
with the Bishop can do is hope that the side with the Knight will badly
blunder.

It is a book draw, but it's quite likely Fritz isn't playing it very
well. I think there is a database for it but it would be rather
large. Even given a database, I don't know what strategy a computer
opponent should use to maximize the chance of a human blunder.

Computers can be very bad in endgames where they cannot calculate far enough
to see the win. Computers also do not make threatening complicated moves if
it is not the best move in the position - they kind of move that increases
the likelihood of a mistake by a lower rated player.

I asked in a previous post if you have Fritz set to draw = loss and you
never responded. Otherwise Fritz will give you a relatively easy draw.

You have also been practicing this ending for several days based on your
posts, so your results at this point aren't very relevant to the question of
whether or not there are insufficient losing chances. The rule book says
with ample time not ample practice.

"George John" wrote in message
. ..

"Fifiela" wrote in message
...
I ask you once again, if both players had ample time, do you think a
1500
would hold against a 2200 more than 90% of the time in a K+R+N vs. K+R+B
endgame and nothing unusual in the position?

I think that a 1500 would lose this position at least 25% of the time;
Maybe
50%. That's why their a 1500.

Why do you say this? I have been playing this end game with Fritz. The
moves are incredibly easy to find quickly, and I draw after 50 moves each
time. I'm a very rusty 1300 something player. I would be astonished if a
1500 could not find them just as easily given ample time to make each
move.

George

"Douglas L Stewart" wrote in message
...
Computers can be very bad in endgames where they cannot calculate far
enough
to see the win.

What sort of ply depth do you think is required for this endgame?

Computers also do not make threatening complicated moves if
it is not the best move in the position - they kind of move that increases
the likelihood of a mistake by a lower rated player.

I asked in a previous post if you have Fritz set to draw = loss and you
never responded. Otherwise Fritz will give you a relatively easy draw.

Sorry, I missed that. Sometimes not all posts make it to my reader, and I
sometimes read too quickly, and miss details. I will try your suggestion.

You have also been practicing this ending for several days based on your
posts, so your results at this point aren't very relevant to the question
of
whether or not there are insufficient losing chances. The rule book says
with ample time not ample practice.

Well, the first time I tried I had zero practice, and easily drew.

I did ask my 2275 rated son about this a bit earlier today, and he thinks
the 1500 will easily draw against a 2200.

George

"Paul Rubin" wrote in message
...
"George John" writes:
Please explain to me how the player with the Bishop can force a win
against
the player with the Knight. Given my experience with Fritz, all the
player
with the Bishop can do is hope that the side with the Knight will badly
blunder.

It is a book draw, but it's quite likely Fritz isn't playing it very
well. I think there is a database for it but it would be rather
large. Even given a database, I don't know what strategy a computer
opponent should use to maximize the chance of a human blunder.

I'm playing with the endgame database, but I think that's for 5 piece
endgames. This is a 6 piece endgame that seems to not be included.

George

"George John" writes:
Computers can be very bad in endgames where they cannot calculate
far enough to see the win.

What sort of ply depth do you think is required for this endgame?

There is really no way for a computer to see that far ahead. It doesn't
understand enough strategy. The only way a computer can be sure to see
a win in that kind of position is with a database. A 6 piece database
is several gigabytes, IIRC.