Star Diamond v London Pro-Commonwealth Computer Chess Cup!

Post by **Steve B** » Wed Oct 31, 2007 2:23 pm

Larry wrote:...and black's move 17 is....the envelope please...

17.... Bg6

Incidentally, LP gives -1.03 for black here, so with black being a
pawn down, white need only force some exchanges and will win.
Well, according to Hoyle.
regards
Larry

another "backward "move by one of the BPL's bishops
giving the SD time to regroup

Petrosian was famous for making "backward" moves like this

Advantage dissipating regards
Steve

Wardy · Post by **Wardy** » Wed Oct 31, 2007 9:11 pm

Hi Guys,

Well after 8.5 hours in the car I'm back!

Bg6 seems too slow to my eyes, the sd believes it is now at +1.88 after

Nf2

The village wifi internet connection still isn't working correctly, which is a bit rubbish and means I can't post a fen until I'm at work in the morning.

Can the SD swindle a win regards

Paul

Post by **Steve B** » Wed Oct 31, 2007 10:21 pm

Wardy wrote:Hi Guys,

Well after 8.5 hours in the car I'm back!

Bg6 seems too slow to my eyes, the sd believes it is now at +1.88 after

Nf2

The village wifi internet connection still isn't working correctly, which is a bit rubbish and means I can't post a fen until I'm at work in the morning.

Can the SD swindle a win regards

Paul

[fen]r4rk1/4bppp/p1p3b1/1p1n4/3P4/1BP2PPq/PP2RN1P/R1BQ2K1 w - - 0 18[/fen]

looks like yet another backward move coming up for the BPL

ResII now shows White ahead +.58 Regards
Steve

Larry · Post by **Larry** » Thu Nov 01, 2007 2:39 am

This game seems to illustrate a point. An opening sequence is only good
if both the processor and the program have the horsepower to be able
to benefit from a position reached. Perhaps a powerful software program
would do better, but it seems LP got his bishops and queen in place, then
sort of did'nt know how to continue and just became passive. So, black
continues here:
17 f3 Bg6
18.Nf2 Qd7

....with a score of -1.03 for black, accepting that the sacrifice pawn
in the opening yielded nothing. Now LP will struggle a pawn down.
regards
Larry

Post by **Steve B** » Thu Nov 01, 2007 4:55 am

Larry wrote:This game seems to illustrate a point. An opening sequence is only good
if both the processor and the program have the horsepower to be able
to benefit from a position reached. Perhaps a powerful software program
would do better, but it seems LP got his bishops and queen in place, then
sort of did'nt know how to continue and just became passive. So, black
continues here:
17 f3 Bg6
18.Nf2 Qd7

....with a score of -1.03 for black, accepting that the sacrifice pawn
in the opening yielded nothing. Now LP will struggle a pawn down.
regards
Larry

excellent point Larry
we see this many times with the older computers with smaller books
one fairly common occurrence is the fianchetto

we will see a computer play g3 or b3 (conversely b6..g6) because it is in its book,, but if the book line isn't long enough to actually finachetto the bishop the computer happily develops the bishop to e3 or d3 once out of book ..leaving it with an awkward pawn position and a poorly developed bishop

the case here with the BPL is more dramatic though
once out of book it simply retreated not happy to have saked the pawn for the attack

RESII would have played differently at move 15 developing its Bishop to g4 instead of f5 and it was out of book after 14..Qh3

still i think there is plenty of fight left in this game

the SD will not play to exchange down to a won pawn endgame
at least not with some many pieces still on the board

All Is Not Yet Lost Regards
Steve

Wardy · Post by **Wardy** » Thu Nov 01, 2007 8:50 am

Steve's right there is a long way to go yet. I'd even expect the BLP to win the endgame from a pawn down to be honest.

Play continues

[White "Star Diamond"]
[Black "Berlin London Pro"]

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 O-O 8.c3 d5 9.exd5
Nxd5 10.Nxe5 Nxe5 11.Rxe5 c6 12.d4 Bd6 13.Re2 Qh4 14.g3 Qh3 15.Nd2 Bf5 16.Ne4 Be7
17.f3 Bg6 18.Nf2 Qd7 19.Nd3 *

[fen]r4rk1/3qbppp/p1p3b1/1p1n4/3P4/1BPN1PP1/PP2R2P/R1BQ2K1 w - - 4 19 [/fen]

4 moves out of five for the knight........

is there nothing this piece can't do regards

Paul

Wardy · Post by **Wardy** » Thu Nov 01, 2007 9:41 am

Just as an after thought Larry, what kind of search depths are you getting on the BLP?

The SD is clearing 10 ply and getting a good way into the 11th but at this depth it's probably not enough to have seen the potential black attack after Qh3.

The SD is now showing just over +2 which along with the earlier high scores leads me to speculate that the king safety penalties around pawn structure and weaknesses on the diagonals are rather rudimentary in the SD.

Plucky underdog regards

Paul

Larry · Post by **Larry** » Thu Nov 01, 2007 11:34 am

Wardy wrote:Just as an after thought Larry, what kind of search depths are you getting on the BLP?

The SD is clearing 10 ply and getting a good way into the 11th but at this depth it's probably not enough to have seen the potential black attack after Qh3.

The SD is now showing just over +2 which along with the earlier high scores leads me to speculate that the king safety penalties around pawn structure and weaknesses on the diagonals are rather rudimentary in the SD.

Plucky underdog regards

Paul

Hi Paul,
regarding search depths, I've just now shut the comp down after it's
hour search, I'll let you know on it's next move.

LP is showing -1.12 for black, I think SD's +2 is a little optimistic, given
that it's only a pawn up, but with no positional advantage that I can see.
I'm predicting a fairly long game here, with LP hanging in for a draw.

Here's the score to date:
18. Nf2 Qd7
19. Nd3 Bd6

White has a good opportunity to make a couple of minor piece exchanges which I think it should do.
all the best
Larry

Wardy · Post by **Wardy** » Thu Nov 01, 2007 12:14 pm

Hi Larry,

I totally agree that the SD is over confident in it's position. I guess it comes from having a d pawn where black has no central pawns, it's control of the open file and very little penalty for blacks connected rooks and little fear of the drafty white kings position. I always find it interesting to try and work out the relative values that the machines appear to value. Certainly I think I might try some sacrificial attacks in my own games against the SD as it would appear that it's happy to weaken the pawn structure around its king.

I think the SD will swap it's useless white square bishop for the strong black knight on d5 for starters. If I were playing I would probably try the longer Bc2....., Nf4........ bringing the Queen to c2 and the dark square bishop to f4 after potential swops take place...........but then maybe that's why I'm a.....

patzer regards

Paul

Wardy · Post by **Wardy** » Thu Nov 01, 2007 8:16 pm

Well the SD isn't ready for a series of exchanges just yet.....

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 O-O 8.c3 d5 9.exd5
Nxd5 10.Nxe5 Nxe5 11.Rxe5 c6 12.d4 Bd6 13.Re2 Qh4 14.g3 Qh3 15.Nd2 Bf5 16.Ne4 Be7
17.f3 Bg6 18.Nf2 Qd7 19.Nd3 Bd6 20.Ne5 *

[fen]r4rk1/3q1ppp/p1pb2b1/1p1nN3/3P4/1BP2PP1/PP2R2P/R1BQ2K1 w - - 6 20 [/fen]

The score is a mind boggling +2.46. I think the BLP has little choice but to swap the bishop for the knight, bringing the SD rook out into open play at e5, pressurising the knight on d5, I'd expect the BLP to then play f6 driving the rook back again......but will the SD play Re1 this time?

Back and forth regards

Paul

Post by **Steve B** » Thu Nov 01, 2007 9:33 pm

Wardy wrote:
[fen]r4rk1/3q1ppp/p1pb2b1/1p1nN3/3P4/1BP2PP1/PP2R2P/R1BQ2K1 w - - 6 20 [/fen]

The score is a mind boggling +2.46.

mind boggling indeed!
im getting + .57 for White here

Optimism is good

Sometimes Regards
Steve

Larry · Post by **Larry** » Thu Nov 01, 2007 11:21 pm

[fen]r4rk1/3q1ppp/p1pb2b1/1p1nN3/3P4/1BP2PP1/PP2R2P/R1BQ2K1 w - - 6 20 [/fen]

20.Ne5 Bxe5

-1.06 for black. Depth 8/20.... this means full width 8 ply and
selective search 20 ply, as I understand.

regards
Larry

Wardy · Post by **Wardy** » Fri Nov 02, 2007 8:24 am

Well I got the Bxe5 for the BPL right but I haven't got the SD's reply correct .......

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 O-O 8.c3 d5 9.exd5
Nxd5 10.Nxe5 Nxe5 11.Rxe5 c6 12.d4 Bd6 13.Re2 Qh4 14.g3 Qh3 15.Nd2 Bf5 16.Ne4 Be7
17.f3 Bg6 18.Nf2 Qd7 19.Nd3 Bd6 20.Ne5 Bxe5 21.dxe5 *

[fen]r4rk1/3q1ppp/p1p3b1/1p1nP3/8/1BP2PP1/PP2R2P/R1BQ2K1 w - - 0 21 [/fen]

This just feels plain wrong to me. The e pawn is just too advanced to defend long term and this now allows the BLP to play Rad8 putting great pressure down the d file while also reducing the power of the e2 rook. Given that the SD still hasn't developed it's queenside rook and bishop I just don't think it can gamble with moves like dxe5.

The score has at least dropped to a more slightly more reasonable +1.76

Larry · Post by **Larry** » Fri Nov 02, 2007 11:06 am

"The score has at least dropped to a more slightly more reasonable +1.76

"

Hi Paul, yes that score is more reasonable. BLP is showing -1.18 for
black, depth 8/20.
White's 21. exd5 transferred it's queen side majority to a king side
majority, which would be considered less of an asset for white. That might
account for the score drop. Still early days.

21. dxe5 Rad8 (no surprise there)

patzer regards
Larry

Post by **Steve B** » Fri Nov 02, 2007 11:14 am

Larry wrote:"The score has at least dropped to a more slightly more reasonable +1.76 "

Hi Paul, yes that score is more reasonable. BLP is showing -1.18 for
black, depth 8/20.
White's 21. exd5 transferred it's queen side majority to a king side
majority, which would be considered less of an asset for white. That might
account for the score drop. Still early days.

21. dxe5 Rad8 (no surprise there)

patzer regards
Larry

i am liking Whites position more and more

King Side Pawn Storm Regards
Steve