New Vitruvius?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

noctiferus
Posts: 364
Joined: Sun Oct 04, 2009 1:27 pm
Location: Italy

Re: New Vitruvius?

Post by noctiferus »

Thx, Eelco.

Your line, at depth 28, finds the second tactical shot Bxe6, and avoids what in the analysis is considered a bad black move, bxc3.

At the greatest depth, it considers a move not taken into account in the analysis,

Later, I'll try to post the lines analyzed in the video, for those interested in the game.
Enrico
Werewolf
Posts: 1795
Joined: Thu Sep 18, 2008 10:24 pm

Re: New Vitruvius?

Post by Werewolf »

I am still seeing an advantage for black after:

1.Ng7 Kg7
2. Qh5 Rg8

What are you going for now, Be6 or Bc6?
User avatar
Eelco de Groot
Posts: 4561
Joined: Sun Mar 12, 2006 2:40 am
Full name:   

Re: New Vitruvius?

Post by Eelco de Groot »

noctiferus wrote:Thx, Eelco.

Your line, at depth 28, finds the second tactical shot Bxe6, and avoids what in the analysis is considered a bad black move, bxc3.

At the greatest depth, it considers a move not taken into account in the analysis,

Later, I'll try to post the lines analyzed in the video, for those interested in the game.
Enrico
Would be interesting to see some more analysis Enrico! That last line I think you referred to with another move was only a Fail Low so it is total rubbish for the moves, maybe I should have waited for the 42nd iteration to finish, it did after some 1240 minutes. See below, the depth 43 lines can be ignored again.

The _037 version of Scid Serpent used unfortunately did not do well in its first match later, so I have abandoned it in the mean time, for version _038 compiled just now. The match result was so bad that I did not yet verify it in any way by more matches;

SfappropximateEval (slightly weaker I think than the standard Stockfish, tested in SF Test-framework Stage I) vs Sf20131124_037, 100 games in Shredder GUI, Shredder book, 1 minute per game per player 61.5 - 38.5 (+32, = 59, -9) 61.5% +81 Elo. Terrible result for the new Nullmove R I suspect that causes it, even if it is an unlucky result partly, it can't be a good engine, at this time control at least and probably also not at longer times.

The 42nd iteration was probably okay though:

42/80 1241:17+1.49 1...Kxg7 2.Qh5 Rg8 3.Bxe6 Nxe6 4.Nd5 Qd8
5.f4 Kh8 6.Qxf7 Nxf4 7.Bxf4 Rg7
8.Qe6 Ne5 9.Nxe7 Qxe7 10.Qxd6 Qxd6
11.Rxd6 Re8 12.Rd4 a5 13.a3 Nc6
14.Rd6 Rxe4 (199.832.176.132) 2683


43/80 1379:42+1.41-- 1...Kxg7 2.Qh5 Rg8 3.Bxe6 Nxe6 4.Nd5 Qd8
5.f4 Kh8 (222.367.846.257) 2686

43/80 1494:11+1.33-- 1...Kxg7 2.Qh5 (241.328.826.134) 2691

best move: Kg8xg7 time: 1494:16.380 min n/s: 2.691.862 nodes: 241.328.826.134
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
noctiferus
Posts: 364
Joined: Sun Oct 04, 2009 1:27 pm
Location: Italy

Re: New Vitruvius?

Post by noctiferus »

Sorry for the delay, I got a flu!
I'll summarize the analysis and audio of youtube ASAP