Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Behind Deep Blue: 3rd print with new Hsu afterword

Author: Sune Fischer

Date: 08:15:58 05/08/04

Go up one level in this thread


On May 08, 2004 at 10:50:57, Anthony Cozzie wrote:

>Let's take a look at some of the moves the annotator didn't like:
>
>r2q1rk1/pp1n1ppp/2pbpn2/3p3b/8/1P1PPNPP/PBPN1PB1/R2Q1RK1 b - - 0 10
>
>Zappa plays the obvious 10 ...e5.  Deep Blue played 10 ...h6.  I won't call this
>a bad move, but it's clearly a pass move.
>
>1... e6-e5 2. e3-e4 Rf8-e8 3. Rf1-e1 Ra8-c8 4. a2-a4 h7-h6 5. Bb2-c3 Qd8-c7 6.
>a4-a5 Bd6-c5 7. Qd1-b1
> = (0.25)       Depth: 12/34    00:01:09.00     41299kN

Yes, but after h6 you can still play e5.
I think it has more to do with style, I imagine the score differences are minute
 and a single tiny evaluation term can be critical for liking the move or not.

>r4rk1/pp1n1pp1/2pbpn1p/q2p3b/8/PP1PPNPP/1BPN1PB1/R3QRK1 b - - 0 12
>
>Once again Zappa wants e5.  Deep Blue played Bc7, which is a pass move at best,
>and I would think the bishop is actually better on D6.

Yes that's one of those typical "ugly" computer moves, it seems a bit pointless
unless DB wanted to transfer the bishop to b6 or something.

But again in a position where there are no obvious best moves and it's all about
style it's not easy to point fingers.

>1... e6-e5 2. c2-c4 Qa5-a6 3. d3-d4 e5-e4 4. Nf3-h4 Ra8-e8 5. Nh4-f5 Bd6-c7 6.
>Ra1-c1 Nd7-b6 7. c4xd5 Nb6xd5
> = (0.24)       Depth: 11/32    00:00:49.38     30722kN
>
>r4rk1/ppbn1pp1/2p1pn1p/q2p3b/7N/PP1PP1PP/1BPN1PB1/R3QRK1 b - - 0 13
>
>And Zappa is still dying for e5 :)  Deep Blue played g5?, which cannot be
>considered anything but a blunder.
>
>1... e6-e5 2. c2-c4 Ra8-d8 3. c4xd5 Nf6xd5 4. d3-d4 f7-f5 5. Nh4xf5 Rf8xf5 6.
>e3-e4 Rf5-g5 7. e4xd5 c6xd5
> = (0.26)       Depth: 11/32    00:00:42.90     26529kN

If it's a blunder it's not an obvious blunder, Ruffian consideres g5 playable
for a while:

Ruffian 1.0.5  UCI1 15 MB:
 6   00:00 -0,01   Qa5c5 Ra1c1 Bc7d6 b3b4 Qc5b5 Nh4f3
 7   00:00 -0,05   Qa5c5 Ra1c1 e6e5 g3g4 Bh5g6 Nh4xg6 f7xg6 Kg1h1
 8   00:01  0,00   Qa5c5 Ra1c1 Qc5e7 Rc1a1 Qe7d6 Nh4f3 Qd6c5 Ra1c1 e6e5
 8   00:02  0,03   g7g5 Nh4f3 g5g4 b3b4 Qa5a4 Nf3d4 g4xh3 Bg2xh3 Nf6g4
 9   00:02  0,04   g7g5 Nh4f3 g5g4 h3xg4 Nf6xg4 b3b4 Qa5a4 Nd2b3 Nd7e5 Nf3xe5
10   00:04  0,05   g7g5 Nh4f3 g5g4 h3xg4 Nf6xg4 Nf3d4 c6c5 Nd4e2 Bc7e5 Bb2xe5
Nd7xe5 Ne2f4
11   00:08 -0,04   g7g5 Nh4f3 g5g4 h3xg4 Nf6xg4 c2c4 d5xc4 Nd2xc4 Qa5f5 e3e4
Qf5c5 Kg1h1
11   00:16 -0,02   Qa5c5 Ra1c1 g7g5 Nh4f3 g5g4 d3d4 Qc5b5 c2c4 d5xc4 h3xg4
Nf6xg4 Rc1xc4

How deep is the fail-low for Zappa on g5?
A "blunder" should be >1 pawn, IMO.

>Lets be clear: no one will ever know exactly how strong Deep Blue was.  All we
>have are the games, and those are not convincing.

Agreed, so far I haven't seen any conclusive evidence to tip the scale either
way. Speculation is fun of course, but it will only take us so far.

-S.
>anthony



This page took 0.01 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.