3200 elo engines vs Nolot test position

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

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
hammerklavier
Posts: 102
Joined: Mon Nov 28, 2011 9:17 am

3200 elo engines vs Nolot test position

Post by hammerklavier » Sat Oct 11, 2014 4:32 am

[d]r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w KQkq - 0 1

The old Test Nolot, #3 Position

Stockfish 5 (3285 ccrl) can´t find the correct move. Nxg5

Phenom x4 9750 2.4Ghz

  • 34/44 10:28 1.141.190.354 1.815.190 +0,59 h2h3 Bg4xf3 g2xf3 Ng5f7 Rh1g1 g7g6 Bc1e3 Be7g5 f3f4 Bg5h4 Qd1e2 Qd8d7 000 000 Ng3f1 Nf7h6 Nf1d2 Kc8b8 Nd2f3 Bh4e7 Be3f2 Rd8f8 Kc1b1 Nh6f5 h3h4 a7a6 Rg1g2 Nf5g7 Rg2g4 Ng7f5 Rd1h1 Qd7e8 Rg4g2 Nf5g7 Qe2d2 Qe8f7 Nf3g5 Be7xg5 f4xg5 Ng7f5
Rybka 3 Dynamic found the correct move in 20 seconds.

gladius
Posts: 538
Joined: Tue Dec 12, 2006 9:10 am

Re: 3200 elo engines vs Nolot test position

Post by gladius » Sat Oct 11, 2014 4:46 am

Current dev version finds it at depth 32, took 115 seconds running on 3 threads.

info depth 32 seldepth 41 score cp 60 lowerbound nodes 589822272 nps 5123765 time 115115 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 d1g4 e1g1 f7g8 g7f5 e7f8 h6f8 c8f8

hammerklavier
Posts: 102
Joined: Mon Nov 28, 2011 9:17 am

Re: 3200 elo engines vs Nolot test position

Post by hammerklavier » Sat Oct 11, 2014 4:50 am

Fantastic! i will try with the dev version.
What hardware you use?

zullil
Posts: 5842
Joined: Mon Jan 08, 2007 11:31 pm
Location: PA USA
Full name: Louis Zulli

Re: 3200 elo engines vs Nolot test position

Post by zullil » Sat Oct 11, 2014 10:39 am

hammerklavier wrote:[d]r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w KQkq - 0 1

The old Test Nolot, #3 Position

Stockfish 5 (3285 ccrl) can´t find the correct move. Nxg5
Depth 31 in less than 30 seconds with latest Stockfish (16 threads):

Code: Select all

info depth 31 currmove h2h3 currmovenumber 1
info depth 31 currmove f3g5 currmovenumber 2
info depth 31 seldepth 40 score cp 64 lowerbound nodes 371974662 nps 13163982 time 28257 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 c6e5
info depth 31 currmove f3g5 currmovenumber 1
info depth 31 seldepth 40 score cp 70 lowerbound nodes 380898412 nps 13217378 time 28818 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 c6e5 e1g1 f7g8 a1d1 e5d3 d1d3
info depth 31 currmove f3g5 currmovenumber 1
info depth 31 seldepth 40 score cp 79 lowerbound nodes 394487034 nps 13281050 time 29703 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 d1g4 e1g1 f7g8 g3f5 c8f8 f1f4 g4f5
info depth 31 currmove f3g5 currmovenumber 1
info depth 31 seldepth 42 score cp 91 lowerbound nodes 417782507 nps 13446924 time 31069 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 d1g4 e1g1 f7g8 g3f5 c8f8 f1f4 g4f5 g7f5 f8c8
info depth 31 currmove f3g5 currmovenumber 1
info depth 31 seldepth 42 score cp 106 lowerbound nodes 443999990 nps 13587538 time 32677 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 d1g4 e1g1 f7g8 g3f5 a7a6 f1f4 g4f5
info depth 31 currmove f3g5 currmovenumber 1
info depth 31 seldepth 42 score cp 128 lowerbound nodes 485437609 nps 13877179 time 34981 multipv 1 pv f3g5 g4d1 g5e6 d8c8 e6g7 e8f7 c1h6 d1g4 e1g1 f7g8 g3f5 a7a6 f1f4 g4f5 d3f5
info depth 31 currmove f3g5 currmovenumber 1
info depth 31 currmove c1g5 currmovenumber 2
info depth 31 currmove d3c4 currmovenumber 29
info depth 31 currmove d3h7 currmovenumber 44
info depth 31 seldepth 58 score cp 147 nodes 771394642 nps 14545833 time 53032 multipv 1 pv f3g5 g4d1 g5e6 d8b8 e6g7 e8d8 e1d1 b7b5 h1f1 d8c8 g3f5 e7f8 g7e6 c8d7 e6f4 c6e7 f5e7 f8e7 f4d5 b8g8 d3b5 d7d8 d5f4 e7g5 f1f2 g5f4 c1f4 g8g6 d1e1 a8b8 b5c4 h8f8 e1f1 f8f4 f2f4 b8b2 f4f2 b2f2 f1f2 g6c2 c4e2 c2c3

rabbits23
Posts: 144
Joined: Fri Sep 12, 2014 2:57 am
Location: Randwick Australia

Re: 3200 elo engines vs Nolot test position

Post by rabbits23 » Mon Oct 13, 2014 6:32 am

Whew; that's one complicatd position.
Allan

Henk
Posts: 5883
Joined: Mon May 27, 2013 8:31 am

Re: 3200 elo engines vs Nolot test position

Post by Henk » Mon Oct 13, 2014 7:13 am

There are no engines that passed 3200 ELO or even 3100. Or is that an inflated ELO rating.

Uri Blass
Posts: 8677
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: 3200 elo engines vs Nolot test position

Post by Uri Blass » Mon Oct 13, 2014 7:28 am

Henk wrote:There are no engines that passed 3200 ELO or even 3100. Or is that an inflated ELO rating.
How do you know?

BBauer
Posts: 658
Joined: Wed Mar 08, 2006 7:58 pm

Re: 3200 elo engines vs Nolot test position

Post by BBauer » Mon Oct 13, 2014 7:41 am

Here my results computed on a small labtop, weak AMD A6 proc, 2GHz, 2Gb hash.

Code: Select all

FEN: r2qk2r/ppp1b1pp/2n1p3/3pP1n1/3P2b1/2PB1NN1/PP4PP/R1BQK2R w KQkq - 0 1
Stockfish_14101221_x64_modern:
 28/41	01:50	 191.082k	1.727k	+0,55	h2-h3 Lg4xf3 g2xf3 Sg5-f7 Dd1-e2 g7-g6 Th1-g1 Le7-g5 f3-f4 Lg5-h4 Lc1-d2 Sf7-h6 O-O-O Dd8-d7 Sg3-f1 O-O-O Kc1-b1 Th8-f8 Sf1-e3 Kc8-b8 Td1-f1 a7-a6 b2-b3 Sc6-e7 Tg1-g4 Sh6xg4 De2xg4 g6-g5 f4xg5 Tf8xf1+ Se3xf1 Lh4-f2 Ld3xh7 Se7-f5 g5-g6
 29/41	02:07	 220.015k	1.725k	+0,58	h2-h3 Lg4xf3 g2xf3 Sg5-f7 Dd1-e2 g7-g6 Lc1-d2 Le7-g5 f3-f4 Lg5-h4 Th1-g1 Lh4xg3+ Tg1xg3 Dd8-h4 De2-f3 O-O-O O-O-O Sf7-h6 f4-f5 Sh6xf5 Ld3xf5 e6xf5 Ld2-g5 Dh4-e4 Lg5xd8 Th8xd8
 30/41+	02:32	 265.421k	1.735k	+0,64	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-f8 Lc1-h6 Sc6xd4
 30/41+	02:38	 274.382k	1.734k	+0,70	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-f8 Lc1-h6 Sc6xd4 Ta1xd1
 30/41+	02:42	 282.575k	1.737k	+0,79	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-f8 Lc1-h6 Ld1-a4 b2-b3 Sc6xe5 d4xe5 La4-e8 Sg7-f5+ Kf8-f7 O-O
 30/41+	02:53	 301.220k	1.739k	+0,91	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-f8 Lc1-h6 Ld1-g4 O-O+ Kf8-g8 Sg3-f5 Le7-f8 Tf1-f4 Lg4xf5
 30/41+	03:06	 325.925k	1.745k	+1,06	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-f8 Lc1-h6 Ld1-g4 O-O+ Kf8-g8 Sg3-f5 Le7-f8 Tf1-f4 Lg4xf5 Ld3xf5 Sc6-d8
 30/41+	03:23	 355.629k	1.746k	+1,28	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-f8 Lc1-h6 Ld1-g4 O-O+ Kf8-g8 Sg3-f5 Le7-f8 Tf1-f4 Lg4xf5 Ld3xf5 Sc6-d8
 30/53	05:36	 564.112k	1.679k	+1,34	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-d8 Ke1xd1 b7-b5 Th1-f1 Kd8-c8 Sg3-f5 Le7-f8 Sg7-e6 Kc8-d7 Se6-f4 Sc6-e7 Sf5xe7 Lf8xe7 Sf4xd5 Db8-g8 Ld3xb5+ Kd7-d8 Sd5xe7 Kd8xe7 g2-g3 Dg8-g6 b2-b3 a7-a5 Lc1-a3+ Ke7-d8 Lb5-e2 Dg6-e4 Ta1-c1 De4-g2 La3-c5 Dg2xh2
 31/53+	05:53	 592.332k	1.675k	+1,40	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-d8 Ke1xd1 b7-b5 Th1-f1 Kd8-c8 Sg3-f5 Le7-f8 Sg7-e6 Kc8-d7 Sf5-g7 Lf8xg7 Se6xg7 Kd7-c8 Sg7-e6 Db8-b6 Tf1-f6 Sc6xe5 d4xe5
 31/53+	06:05	 611.691k	1.674k	+1,46	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-d8 Ke1xd1 Le7-f8 Sg7-e6+ Kd8-e8 Sg3-h5 Lf8-e7 Ld3-f5 Sc6-d8 Se6-g7+ Ke8-f8 Th1-f1 Kf8-g8 Kd1-c2 c7-c5
 31/53	06:29	 646.439k	1.660k	+1,54	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-d7 Ld3-f5+ Kd7-d8 Ke1xd1 Sc6-a5 Kd1-c2 b7-b5 Sg3-h5 a7-a6 Sh5-f4 c7-c6 Sf4-e6+ Kd8-c8 Se6-c5+ Kc8-c7 Sg7-e6+ Kc7-c8 Th1-f1 Le7xc5 Se6xc5+ Kc8-c7 Lf5-e6 Db8-e8 Tf1-f7+ Kc7-b8 Lc1-d2 Ta8-a7 Ta1-f1 Sa5-c4 Sc5-d7+ Ta7xd7 Le6xd7 De8-g8 Tf7-f2 Kb8-c7 Ld7-f5 Sc4xd2 Kc2xd2
 32/53-	08:07	 799.115k	1.638k	+1,48	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-d8 Ke1xd1
 32/53	08:21	 819.594k	1.635k	+1,51	Sf3xg5 Lg4xd1 Sg5xe6 Dd8-b8 Se6xg7+ Ke8-d8 Ke1xd1
Nxg5 is found in 2:32 min on depth 30+.
Kind regards
Bernhard

Henk
Posts: 5883
Joined: Mon May 27, 2013 8:31 am

Re: 3200 elo engines vs Nolot test position

Post by Henk » Mon Oct 13, 2014 8:47 am

Uri Blass wrote:
Henk wrote:There are no engines that passed 3200 ELO or even 3100. Or is that an inflated ELO rating.
How do you know?
The one and only CEGT rating list.

ernest
Posts: 1875
Joined: Wed Mar 08, 2006 7:30 pm

Re: 3200 elo engines vs Nolot test position

Post by ernest » Mon Oct 13, 2014 7:32 pm

hammerklavier wrote:The old Test Nolot, #3 Position
Stockfish 5 (3285 ccrl) can´t find the correct move. Nxg5
You are a poor victim of multithread non-reproducibility!...

Repeat your test a few times and come back...

(on my 1st try, my SF5 found Nxg5 very quickly, at depth 32)

Post Reply