Spanish Serpent

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

Moderator: Ras

User avatar
Eelco de Groot
Posts: 4724
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Spanish Serpent

Post by Eelco de Groot »

Just wanted to post that the changed Rainbow Serpent now wants to play Ruy Lopez :) at high depth (as opposed to the French of version 001, which is also nice to see I find personally). There is no big relevance to playing strength or anything, because the engine lacks the capacity to really judge opening moves. And with more threads probably it will play something different the next time I try. But it is just nice to see at least a little bit familiar and established opening and not, heaven forbid, the Scandinavian :(.

Also wanted to say that the sources at least will be put up as a separate Stockfish branch tomorrow I think, so if someone would want to compile them they can do so easily and then I'll try to keep it synchronized with Stockfish for a while, if it can be semiautomatic. It is just a tool: I did not test for strength but it is more for second opinions with a free engine. Most of the time it will be just like Stockfish. But thanks to some code from Jörg Oster at least this SF version will also detect three times repetitions instead of two, and some other code from Jörg just be a little bit better at detecting Zugzwang although I think that is only a very limited effect, mainly because Jörg's code has a deeper verification search and Rainbow Serpent too.

Most Windows people will think they can't compile this themselves though and after Stockfish moved to C++11 I agree it has become, not hard, but a little bit harder because older versions of MinGW don't suffice for C++11. But first let me put up my GitHub branch (and maybe I can borrow a 64 bit Windows somehow and a then install a C++11 compiler for it :) Some day I'll get around to that, but not tomorrow)

[d]rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq -

Engine: Rainbow Serpent Sf20150607_002 MOD MP (Q6700, 4 threads, 32 bit, 512 MB)
by Tord Romstad, Marco Costalba and Joona Kiiski


31/53 9:57 +0.16 1.d4 d5 2.c4 {c4 after d4 d5 used to be very hard for a computer} e6 3.Nf3 Be7 4.Bf4 Nf6
5.e3 O-O 6.Nc3 Nbd7 7.cxd5 Nxd5
8.Nxd5 exd5 9.Bd3 c5 10.O-O c4
11.Bc2 h6 12.b3 b5 13.Qb1 c3 14.b4 (1.440.195.244) 2408
.
.
.
32/53 12:36 +0.23 1.d4 d5 2.c4 e6 3.Nf3 Be7 4.Nc3 Nf6
5.Bf4 {Is this playable?} O-O 6.e3 Nbd7 7.a3 c5 8.cxd5 Nxd5
9.Nxd5 exd5 10.dxc5 Nxc5 11.Be2 Bf6
12.Rc1 Ne4 13.Nd4 Qa5+ 14.Kf1 (1.819.554.659) 2404

33/53 15:35 +0.17-- 1.d4 d5 (2.257.951.088) 2414


33/53 22:08 +0.12 1.e4 e6 {French Tarrasch again} 2.d4 d5 3.Nd2 c5 4.Ngf3 Nf6
5.exd5 exd5 6.Bb5+ Bd7 7.Bxd7+ Nbxd7
8.O-O Be7 9.dxc5 Nxc5 10.Nb3 Nce4
11.Nfd4 O-O 12.Nf5 Re8 13.Nxe7+ Qxe7
14.Be3 (3.225.104.617) 2428
.
.
.
34/53 29:14 +0.12 1.e4 e6 2.d4 d5 3.Nd2 c5 4.Ngf3 Nf6
5.exd5 exd5 6.Bb5+ Bd7 7.Bxd7+ Nbxd7
8.O-O Be7 9.dxc5 Nxc5 10.Nb3 Nce4
11.Nfd4 O-O 12.Nf5 Re8 13.Nxe7+ Rxe7
14.Be3 (4.261.922.190) 2429
.
.
.
35/55 54:43 +0.22 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.d3 Nf6
5.O-O a6 6.c3 Ba7 7.d4 d6 8.dxe5 Nxe5
9.Nxe5 dxe5 10.Qxd8+ Kxd8 11.Nd2 Ke7
12.Nf3 Nxe4 13.Nxe5 f6 14.Nf3 (7.822.945.053) 2382
.
.
.
36/57 77:48 +0.15 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.d3 Nf6
5.O-O d6 6.c3 a5 7.h3 O-O 8.Re1 Re8
9.Na3 Be6 10.Bxe6 Rxe6 11.Nb5 Bb6
12.Bg5 h6 13.Bxf6 Rxf6 14.d4 (11.191.934.272) 2397
.
.
.

37/61 192:39 +0.14 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6
5.O-O Be7 6.Bxc6 dxc6 7.d3 Bd6
8.Nbd2 Be6 9.b3 O-O 10.Nc4 Bxc4
11.bxc4 Re8 12.Be3 Nd7 13.Rb1 b5
14.Qd2 (27.049.548.438) 2339

38/61 213:35 +0.17 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6
5.O-O Be7 6.Nc3 b5 7.Bb3 O-O 8.Re1 d6
9.Nd5 Bg4 10.c3 Na5 11.Nxe7+ Qxe7
12.Bc2 c5 13.d3 Nc6 14.h3 (29.976.259.032) 2339


best move: e2-e4 time: 242:23.365 min n/s: 2.339.040 nodes: 34.130.909.718
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
User avatar
Eelco de Groot
Posts: 4724
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: Spanish Serpent

Post by Eelco de Groot »

Hello All,

Sources of the Rainbow Serpent dev. version are now available as branch 'Serpent' from the development version of Stockfish. No new ideas really, and as I said no strenghth testing but my hope is that it is a bit better documented now on GitHub.

https://github.com/Kingdefender/Stockfish/tree/Serpent

There's a button 'Download ZIP' that will give you the sources for external compilation. I don't have any 64 bit compiles available, 32 bit is no problem.

I hope you have some fun with it,
Eelco
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
jpqy
Posts: 556
Joined: Thu Apr 24, 2008 9:31 am
Location: Belgium

Re: Spanish Serpent

Post by jpqy »

Thanks Eelco :!:

Here you have a x64 compile : https://www.dropbox.com/s/8ige1nhx8r9pr ... 64.7z?dl=0

JP.
User avatar
Eelco de Groot
Posts: 4724
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: Spanish Serpent

Post by Eelco de Groot »

Thanks for making a 64 bit compile Jean-Paul!

Maybe I can add the logo that Kyodai once made on ImmortalChess

Image
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