SF-McBrain 4.0 Release

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

Moderators: hgm, Rebel, chrisw

User avatar
Ovyron
Posts: 4556
Joined: Tue Jul 03, 2007 4:30 am

Re: SF-McBrain 4.0 Release

Post by Ovyron »

MikeB wrote:The next release of SF-McB will have Oster's patch - it's already in the they dev version.
Ah, nice, thanks! My problem with the Dev version is that my own compiles are slower than the official Stockfish 9 release, while my compiles of McBrain are faster than your releases, so it seems any fork I'd like to test would be better used on the McBrain branch.
MikeB wrote:I'm glad you posed this question , since with Oster's code, I had re-write the tactical to get it to work right. Fairly simple rewrite - added two short lines of code ( so that the output looks like mutiipv is set to 1) .
Any chance to add an option so that the Tactical setting still shows the PVs and scores of the other lines it analyses? It seems the Tactical setting is the only way to mix MultiPV and "go searchmoves" in the Shredder GUI.

Supposing a Tactical=2 setting and MultiPV=4 will miss the best move of a position (Tactical=3 or MultiPV=8 would find it, but it's too costly to use everywhere), after I have refuted all 4 main moves and want a 5th alternative, there's no way to use MultiPV=4+"go searchmoves" to exclude the 4 moves I already refuted, because the GUI will go back to SinglePV after using "go searchmoves." Tactical=2+"go searchmoves" to exclude those moves does what I want, but McBrain isn't showing me the score of the other 3 moves it analyses.
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: SF-McBrain 4.0 Release

Post by MikeB »

Ovyron wrote: Ah, nice, thanks! My problem with the Dev version is that my own compiles are slower than the official Stockfish 9 release, while my compiles of McBrain are faster than your releases, so it seems any fork I'd like to test would be better used on the McBrain branch.
My releases are not quite as fast any more using clang, but clang offers better compatibility with other macs and complies much more quickly than gcc. Plus gcc PGO is now broken on my machine since High Sierra was installed. I tested the official stockfish vs my self compile pf SF 9 and the difference was 1%, which I can live with.

Ovyron wrote: Any chance to add an option so that the Tactical setting still shows the PVs and scores of the other lines it analyses? It seems the Tactical setting is the only way to mix MultiPV and "go searchmoves" in the Shredder GUI.

Supposing a Tactical=2 setting and MultiPV=4 will miss the best move of a position (Tactical=3 or MultiPV=8 would find it, but it's too costly to use everywhere), after I have refuted all 4 main moves and want a 5th alternative, there's no way to use MultiPV=4+"go searchmoves" to exclude the 4 moves I already refuted, because the GUI will go back to SinglePV after using "go searchmoves." Tactical=2+"go searchmoves" to exclude those moves does what I want, but McBrain isn't showing me the score of the other 3 moves it analyses.
Can you show me from the terminal how you are using MultiPV=4+"go searchmoves" and what the expected output is? - not familiar with that technique. If you can, show both SF and McBrain and the output you get.
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: SF-McBrain 4.0 Release

Post by MikeB »

decided to let it run to see how long it takes with normal search

[d]r4r1k/ppq4p/2n1bppP/2P1p3/2B2n2/1N3Q2/PBP2PP1/3RR1K1 w - - 0 1

takes about 15 minutes with normal search with current dev-SF-McB

Code: Select all

dep	score	nodes	time	(not shown:  tbhits	knps	seldep) 
 40	+5.65 	27.0G	22:17.87	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Nd2 Rf5 Nf3 Qe8 Rd5 Rd8 Rxe5 Qxe5 Bxe5+ Rxe5 Nxe5 Rd2 Bb3 Re2 Nf7+ Kg8 Nd6+ Kf8 Nxb7 Re8 Kh2 Ke7 Nd6 Rf8 Kh3 Kd7 Bd5 a6 Kg4 Ke7 Bb7 f3 g3 Rb8 Kxf3 a5 Kf4 Ke6 Ke3 a4 Kd4 Kd7 Kd5 Kc7 Bc6 Rf8 f 
 39	+5.50 	24.9G	20:37.74	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Nd2 Qe8 Nf3 Rf5 Rd5 Rd8 Rxe5 Qxe5 Bxe5+ Rxe5 Nxe5 Rd2 Bb3 Re2 Nf7+ Kg8 Nd6+ Kf8 Nxb7 Re8 Kh2 Ke7 Nd6 Rf8 Kh3 Kd7 Bd5 Ke7 Kg4 a5 Bb7 Kd8 Bf3 Kc7 Ne4 Rf5 Ng5 Kc8 Nxh7 Rh5 Kxf4 Rxh6 Nf8 Kc7 Ke5 
 38	+5.09 	22.5G	18:45.70	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Nd2 Qe8 Nf3 Rf5 Rd5 Rd8 Rxe5 Qxe5 Bxe5+ Rxe5 Nxe5 Rd2 Bb3 Re2 Nf7+ Kg8 Nd6+ Kf8 Nxb7 Re4 c6 Ke7 Kh2 f3 gxf3 Rh4+ Kg3 g5 Bd5 Rxh6 Kg4 Rh4+ Kf5 Rh6 Kxg5 Rh2 Be4 Rxf2 Kf4 Rh2 Bf5 Rh4+ Ke5 Rc4 B 
 38	+5.00!	22.0G	18:20.79	Qxf4! 
 38	+4.86!	21.4G	17:49.41	Qxf4! 
 38	+4.71!	21.0G	17:32.60	Qxf4! 
 38	+4.57!	20.7G	17:18.48	Qxf4! 
 38	+4.42!	20.4G	17:01.52	Qxf4! 
 38	+4.27!	20.1G	16:49.40	Qxf4! 
 38	+4.13!	19.9G	16:40.32	Qxf4! 
 38	+3.98!	19.7G	16:30.41	Qxf4! 
 38	+3.84!	19.6G	16:23.99	Qxf4! 
 38	+3.69!	19.5G	16:21.03	Qxf4! 
 38	+3.55!	19.4G	16:16.06	Qxf4! 
 38	+3.40!	19.3G	16:09.31	Qxf4! 
 38	+3.25!	19.1G	15:59.90	Qxf4! 
 38	+3.11!	19.0G	15:56.35	Qxf4! 
 38	+2.96!	18.9G	15:52.92	Qxf4! 
 38	+2.82!	18.9G	15:50.36	Qxf4! 
 38	+2.67!	18.8G	15:46.72	Qxf4! 
 38	+2.52!	18.7G	15:42.89	Qxf4! 
 38	+2.38!	18.6G	15:38.64	Qxf4! 
 38	+2.23!	18.5G	15:35.74	Qxf4! 
 38	+2.09!	18.5G	15:33.42	Qxf4! 
 38	+1.94!	18.4G	15:29.82	Qxf4! 
 38	+1.80!	18.4G	15:27.20	Qxf4! 
 38	+1.65!	18.3G	15:24.81	Qxf4! 
 38	+1.50!	18.3G	15:22.89	Qxf4! 
 38	+1.36!	18.3G	15:21.50	Qxf4! 
 38	+1.21!	18.2G	15:17.33	Qxf4! 
 38	+1.07!	18.1G	15:16.64	Qxf4! 
 38	+0.97?	17.9G	15:03.06	Nd2 Bxc4? 
 
this one is about the power of the bishops QxN and RxN and then the nice deflection move with Rd7...
User avatar
Ovyron
Posts: 4556
Joined: Tue Jul 03, 2007 4:30 am

Re: SF-McBrain 4.0 Release

Post by Ovyron »

MikeB wrote:Can you show me from the terminal how you are using MultiPV=4+"go searchmoves" and what the expected output is?
From the terminal it works as expected, it's the Shredder GUI that is setting the engine back to SinglePV whenever you use "go searchmoves". I was using the Tactical setting as a workaround back when I didn't know it was actually searching the PVs and scores but not showing them.

The reason MultiPV is superior to the Tactical setting is that for positions where only 2 or 3 moves are playable, MultiPV will show, say, that the 4th move is really bad, so the user can adjust and use MultiPV lines down so that McBrain only looks at the playable moves, while Tactical=2 will waste time analyzing the 4th bad move, and Tactical=3 will waste a lot of time analyzing the 4th, 5th, 6th, 7th, and 8th move, and the analysis of the top 3 will suffer, timewise.

If a move is forced in a position, with MultiPV you can see that the rest of the moves are really bad and make the move instantly. With the Tactical setting you have no clue and would let it analyze in full.

But the Tactical setting provides an "Always On" option for MultiPV, with an option to show the rest of the moves this could be very useful.
Vinvin
Posts: 5228
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: SF-McBrain 4.0 Release

Post by Vinvin »

MikeB wrote:One more position

[d]r4r1k/ppq4p/2n1bppP/2P1p3/2B2n2/1N3Q2/PBP2PP1/3RR1K1 w - - 0 1
I very like this one !
SF9, 6*4GHz, 3 runs : 6m29s, 19m41s and not found in 40 min.

Code: Select all

Stockfish_9_x64_popcnt:
...
 35/63	05:43	 3 468 752k	10 101k	+1,17	1.Nd2 Bxc4 2.Nxc4 Ne6 3.Qd5 Qe7 4.Nd6 Rab8 5.Qc4 Na5 6.Qe4 Nc6 7.Qd5 Nf4 8.Qc4 Qd7 9.Bc1 g5 10.Qe4 Nd8 11.Bb2 Ng6 12.g3 Nf7 13.Nxf7+ Qxf7 14.Rd6 Qxa2 15.Bc3 Qa3 16.Rd3 Qa2 17.Rd7 Rf7 18.Red1 Nf8 19.Rxf7 Qxf7 20.Rd6 Ne6 21.Qf3 Rd8 22.Bxe5
 36/50-	06:04	 3 681 396k	10 110k	+1,10	1.Nd2 Bxc4
 36/50-	06:06	 3 701 508k	10 112k	+1,02	1.Nd2 Bxc4
 36/50+	06:11	 3 755 547k	10 114k	+1,10	1.Nd2
 36/58+	06:29	 3 940 398k	10 129k	+1,25	1.Qxf4
 36/63+	06:59	 4 272 043k	10 174k	+4,12	1.Qxf4

Stockfish_9_x64_popcnt:
...
 40/64	16:11	 9 630 580k	9 909k	+1,04	1.Nd2 Bxc4 2.Nxc4 Rab8 3.Qg4 Qe7 4.Rd7 Qe6 5.Qxe6 Nxe6 6.Nd6 Kg8 7.Rd1 Rfd8 8.Ne4 Rxd7 9.Rxd7 Rf8 10.Rxb7 f5 11.Nd6 Rb8 12.Rxb8+ Nxb8 13.Bxe5 Nc6 14.Bf6 Nxc5 15.c3 Nd7 16.Bg5 Nde5 17.Kf1 Nf7 18.Nxf7 Kxf7 19.Ke2 Ne5 20.Bf4 Ke6 21.f3 a5 22.Bxe5 Kxe5 23.Kd3 g5 24.c4 f4
 41/55	18:25	 10 787 915k	9 762k	+1,06	1.Nd2 Bxc4 2.Nxc4 Rab8 3.Qg4 Qe7 4.Rd7 Qe6 5.Qxe6 Nxe6 6.Nd6 Kg8 7.Rd1 Rfd8 8.Ne4 Rxd7 9.Rxd7 Rf8 10.Rxb7 f5 11.Nd6 Rb8 12.Rxb8+ Nxb8 13.Bxe5 Nc6 14.Bf6 Nxc5 15.c3 Nd7 16.Bg5 Nde5 17.Kf1 Nf7 18.Nxf7 Kxf7 19.Ke2 Ne5 20.Bf4 Ke6 21.f3 Kd5 22.g3 Nd7 23.Be3 a5 24.Bd4 Ne5 25.Bxe5 Kxe5 26.Ke3 g5
 42/58+	19:41	 11 570 909k	9 790k	+1,13	1.Qxf4
 42/58+	19:44	 11 597 512k	9 792k	+1,21	1.Qxf4
 42/58+	21:10	 12 522 808k	9 860k	+4,34	1.Qxf4
 42/70	22:46	 13 602 149k	9 952k	+5,32	1.Qxf4 exf4 2.Rxe6 Ne5 3.Rxe5 fxe5 4.Rd7 Qb8 5.Na5 Rf5 6.Nxb7 Qc8 7.Re7 Qc6 8.Rxe5 Qf6 9.Nd6 Qxe5 10.Bxe5+ Rxe5 11.Nf7+ Kg8 12.Nxe5+ Kf8 13.Nd3 Re8 14.Bd5 Re2 15.Bb3 f3 16.gxf3 Re8 17.Kg2 Ke7 18.f4 Rd8 19.Kf3 Re8 20.Bc4 Rc8 21.Ke4 Rf8 22.Bd5 a5 23.Bc4 Kd7 24.Ne5+ Kc7 25.Nf7 Kc6 26.Ng5 Re8+ 27.Kd4 Rd8+ 28.Ke5 Rd2 29.Nxh7 Rxc2 30.Bd5+ Kxc5
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: SF-McBrain 4.0 Release

Post by MikeB »

Vinvin wrote:
MikeB wrote:One more position

[d]r4r1k/ppq4p/2n1bppP/2P1p3/2B2n2/1N3Q2/PBP2PP1/3RR1K1 w - - 0 1
I very like this one !
SF9, 6*4GHz, 3 runs : 6m29s, 19m41s and not found in 40 min.

Code: Select all

Stockfish_9_x64_popcnt:
...
 35/63	05:43	 3 468 752k	10 101k	+1,17	1.Nd2 Bxc4 2.Nxc4 Ne6 3.Qd5 Qe7 4.Nd6 Rab8 5.Qc4 Na5 6.Qe4 Nc6 7.Qd5 Nf4 8.Qc4 Qd7 9.Bc1 g5 10.Qe4 Nd8 11.Bb2 Ng6 12.g3 Nf7 13.Nxf7+ Qxf7 14.Rd6 Qxa2 15.Bc3 Qa3 16.Rd3 Qa2 17.Rd7 Rf7 18.Red1 Nf8 19.Rxf7 Qxf7 20.Rd6 Ne6 21.Qf3 Rd8 22.Bxe5
 36/50-	06:04	 3 681 396k	10 110k	+1,10	1.Nd2 Bxc4
 36/50-	06:06	 3 701 508k	10 112k	+1,02	1.Nd2 Bxc4
 36/50+	06:11	 3 755 547k	10 114k	+1,10	1.Nd2
 36/58+	06:29	 3 940 398k	10 129k	+1,25	1.Qxf4
 36/63+	06:59	 4 272 043k	10 174k	+4,12	1.Qxf4

Stockfish_9_x64_popcnt:
...
 40/64	16:11	 9 630 580k	9 909k	+1,04	1.Nd2 Bxc4 2.Nxc4 Rab8 3.Qg4 Qe7 4.Rd7 Qe6 5.Qxe6 Nxe6 6.Nd6 Kg8 7.Rd1 Rfd8 8.Ne4 Rxd7 9.Rxd7 Rf8 10.Rxb7 f5 11.Nd6 Rb8 12.Rxb8+ Nxb8 13.Bxe5 Nc6 14.Bf6 Nxc5 15.c3 Nd7 16.Bg5 Nde5 17.Kf1 Nf7 18.Nxf7 Kxf7 19.Ke2 Ne5 20.Bf4 Ke6 21.f3 a5 22.Bxe5 Kxe5 23.Kd3 g5 24.c4 f4
 41/55	18:25	 10 787 915k	9 762k	+1,06	1.Nd2 Bxc4 2.Nxc4 Rab8 3.Qg4 Qe7 4.Rd7 Qe6 5.Qxe6 Nxe6 6.Nd6 Kg8 7.Rd1 Rfd8 8.Ne4 Rxd7 9.Rxd7 Rf8 10.Rxb7 f5 11.Nd6 Rb8 12.Rxb8+ Nxb8 13.Bxe5 Nc6 14.Bf6 Nxc5 15.c3 Nd7 16.Bg5 Nde5 17.Kf1 Nf7 18.Nxf7 Kxf7 19.Ke2 Ne5 20.Bf4 Ke6 21.f3 Kd5 22.g3 Nd7 23.Be3 a5 24.Bd4 Ne5 25.Bxe5 Kxe5 26.Ke3 g5
 42/58+	19:41	 11 570 909k	9 790k	+1,13	1.Qxf4
 42/58+	19:44	 11 597 512k	9 792k	+1,21	1.Qxf4
 42/58+	21:10	 12 522 808k	9 860k	+4,34	1.Qxf4
 42/70	22:46	 13 602 149k	9 952k	+5,32	1.Qxf4 exf4 2.Rxe6 Ne5 3.Rxe5 fxe5 4.Rd7 Qb8 5.Na5 Rf5 6.Nxb7 Qc8 7.Re7 Qc6 8.Rxe5 Qf6 9.Nd6 Qxe5 10.Bxe5+ Rxe5 11.Nf7+ Kg8 12.Nxe5+ Kf8 13.Nd3 Re8 14.Bd5 Re2 15.Bb3 f3 16.gxf3 Re8 17.Kg2 Ke7 18.f4 Rd8 19.Kf3 Re8 20.Bc4 Rc8 21.Ke4 Rf8 22.Bd5 a5 23.Bc4 Kd7 24.Ne5+ Kc7 25.Nf7 Kc6 26.Ng5 Re8+ 27.Kd4 Rd8+ 28.Ke5 Rd2 29.Nxh7 Rxc2 30.Bd5+ Kxc5
final release version of SF-McB-v9...18 threads, 12 core Mac Pro, 8192M hash

Code: Select all

dep	score	nodes	time	(not shown:  tbhits	knps	seldep)
 37	+5.90 	10.1G	8:46.86	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Rd5 Re8 Nd4 a5 Nf3 b5 cxb6 Qxb6 Bb3 Qxb3 cxb3 Ra7 Nxe5 Kg8 Bd4 Rc7 Rxa5 Rd8 Kh2 Rb7 Kh3 Rf8 Bc5 Re8 Bd6 f3 Nxf3 Rf7 Kg4 Rf5 Rxf5 gxf5+ Kxf5 Re2 a4 Rxf2 a5 Rb2 a6 Rxb3 
 36	+5.67 	8.67G	7:30.11	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Rd5 Re8 Nd4 a5 Nf3 b5 cxb6 Qxb6 Bb3 Qf6 Nxe5 Re6 Nf7+ Kg8 Bxf6 Kxf7 Bg5 Re1+ Kh2 a4 Bc4 Kf8 Rd4 Ra5 Bxf4 Ke8 Bg8 Re7 Kg3 Raa7 Bg5 Red7 Re4+ Kf8 Bc4 Ra5 Kh4 Rc5 Be6 Rd6 c4 Rxe6 Rxe6 Rxc4+ Kg3 
 35	+5.32 	7.34G	6:22.00	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Rd5 Re8 Nd4 a5 Nf3 b5 cxb6 Qxb6 Bb3 Qxb3 cxb3 Ra7 Nxe5 Kg8 Bd4 Rc7 Rxa5 Rd8 Nf3 Re7 Rc5 Red7 Be5 Rf7 a4 Rf5 Kf1 Re8 a5 g5 Rc7 Rexe5 Nxe5 Rxe5 Rg7+ Kh8 a6 
 34	+5.06 	6.75G	5:52.72	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Rd5 Re8 Nd4 a5 Nf3 b5 cxb6 Qxb6 Bb3 Qxb3 cxb3 Ra7 Nxe5 Kg8 Bd4 Rb7 Rxa5 Rbe7 Ng4 Kf7 Kh2 Re6 Ne5+ Kg8 Nf3 R6e7 Be5 Rf7 Kh3 Rf5 a4 Rh5+ Kg4 Rxh6 Bxf4 
 33	+4.78 	6.41G	5:35.96	Qxf4 exf4 Rxe6 Ne5 Rxe5 fxe5 Rd7 Qb8 Rd5 b5 Rxe5 bxc4 Re7+ Rf6 Bxf6+ Kg8 Rg7+ Kf8 Be7+ Ke8 Bd6 Qd8 Nd2 Rc8 Ne4 Qh4 Be7 Qxe7 Nd6+ Qxd6 cxd6 Rd8 Rxh7 Rxd6 Rxa7 Rd1+ Kh2 Rd5 h7 Rh5+ Kg1 Kf8 a4 Rh6 a5 g5 h8=Q+ Rxh8 Ra8+ Kg7 Rxh8 Kxh 
 33	+4.80!	6.38G	5:34.40	Qxf4! 
 33	+4.68!	6.28G	5:29.64	Qxf4! 
 33	+4.55!	6.19G	5:25.27	Qxf4! 
 33	+4.43!	6.12G	5:21.44	Qxf4! 
 33	+4.30!	6.03G	5:17.09	Qxf4! 
 33	+4.18!	5.96G	5:13.39	Qxf4! 
 33	+4.05!	5.90G	5:10.57	Qxf4! 
 33	+3.93!	5.86G	5:08.37	Qxf4! 
 33	+3.80!	5.81G	5:05.66	Qxf4! 
 33	+3.68!	5.76G	5:03.21	Qxf4! 
 33	+3.55!	5.72G	5:01.55	Qxf4! 
 33	+3.43!	5.72G	5:01.22	Qxf4! 
 33	+3.30!	5.62G	4:56.11	Qxf4! 
 33	+3.18!	5.59G	4:54.61	Qxf4! 
 33	+3.05!	5.56G	4:53.38	Qxf4! 
 33	+2.93!	5.54G	4:52.12	Qxf4! 
 33	+2.80!	5.52G	4:51.04	Qxf4! 
 33	+2.68!	5.49G	4:49.66	Qxf4! 
 33	+2.55!	5.47G	4:48.41	Qxf4! 
 33	+2.43!	5.45G	4:47.64	Qxf4! 
 33	+2.30!	5.44G	4:47.10	Qxf4! 
 33	+2.18!	5.43G	4:46.40	Qxf4! 
 33	+2.05!	5.40G	4:45.02	Qxf4! 
 33	+1.93!	5.39G	4:44.36	Qxf4! 
 33	+1.80!	5.36G	4:42.91	Qxf4! 
 33	+1.68!	5.34G	4:41.92	Qxf4! 
 33	+1.55!	5.32G	4:40.95	Qxf4! 
 33	+1.43!	5.31G	4:40.25	Qxf4! 
 33	+1.30!	5.29G	4:39.43	Qxf4! 
 33	+1.18!	5.29G	4:39.33	Qxf4! 
 33	+1.05!	5.29G	4:39.20	Qxf4! 
 32	+0.93 	5.06G	4:27.46	Nd2 Bxc4 Nxc4 Qe7 Nd6 Ne6 Qd5 Rab8 Bc1 Nc7 Qc4 Qe6 Qa4 Ne8 Ne4 g5 Qb3 Qf5 Bb2 Na5 Qc3 Nc6 Qc4 Na5 Qd5 Nc6 Qd7 Qg6 Qh3 g4 Qh4 Nc7 Rd6 Ne8 Rd7 Qf5 Rd2 Nc7 Nd6 Qe6 
 32	+0.89?	4.94G	4:21.33	Nd2 Bxc4? 
 32	+1.09!	3.60G	3:10.57	Nd2! 
 31	+0.96 	2.51G	2:13.85	Nd2 Bxc4 Nxc4 Rae8 Nd6 Rd8 Qe4 Rd7 Ba3 Ne6 c3 Re7 Qf3 Ng5 Qd5 Nf7 Bc1 Qd7 Be3 Ncd8 Qb3 Qc6 Nb5 g5 Nxa7 Qe6 Nb5 Nxh6 Bc1 Qxb3 axb3 Kg7 b4 Ne6 Nd6 Rd7 Be3 
 31	+0.95!	1.92G	1:42.72	Nd2! 
 31	+0.85?	1.91G	1:42.03	Nd2 Bxc4? 
 30	+0.92 	1.86G	1:39.48	Nd2 Bxc4 Nxc4 Rae8 Nd6 Re7 Ba3 Ne6 c3 Ng5 Qd5 Nf7 Bc1 Qd7 Be3 Ncd8 Rb1 Nxd6 cxd6 Re6 Red1 b6 c4 Nf7 c5 f5 g3 Kg8 Kg2 Rd8 Qb3 bxc5 Bxc5 Nxh6 
 29	+0.89 	1.47G	1:18.48	Nd2 Bxc4 Nxc4 Rad8 Nd6 Rd7 Bc1 Ne6 Qd5 Re7 c3 Ncd8 Ne4 Nc6 Be3 f5 Nd6 f4 Bc1 Ned8 Bb2 Re6 c4 Qe7 f3 g5 Bc3 Rff6 Rb1 b6 Ne4 Rg6 cxb6 Rxh6 bxa7 Qxa7+ c5 
 29
User avatar
Marek Soszynski
Posts: 582
Joined: Wed May 10, 2006 7:28 pm
Location: Birmingham, England

Re: SF-McBrain 4.0 Release

Post by Marek Soszynski »

MikeB wrote:One more position

[d]r4r1k/ppq4p/2n1bppP/2P1p3/2B2n2/1N3Q2/PBP2PP1/3RR1K1 w - - 0 1
Where is this from? Can you post the game? Thanks.
Marek Soszynski
Paloma
Posts: 1167
Joined: Thu Dec 25, 2008 9:07 pm
Full name: Herbert L

Re: SF-McBrain 4.0 Release

Post by Paloma »

yanquis1972
Posts: 1766
Joined: Wed Jun 03, 2009 12:14 am

Re: SF-McBrain 4.0 Release

Post by yanquis1972 »

very cool michael, unfortunately i'm getting a 404 on the link in the original post.
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: SF-McBrain 4.0 Release

Post by MikeB »

yanquis1972 wrote:very cool michael, unfortunately i'm getting a 404 on the link in the original post.
Correct, version is 4.0 is no longer available. A newer version will be posted shortly. This was not intentional, but it s what it is. Sorry about that.