Test Position, white to play and draw

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

Moderator: Ras

Hobacle
Posts: 244
Joined: Thu Mar 04, 2021 3:57 pm
Full name: Nott Q Suree

Re: Test position, white to play and draw.

Post by Hobacle »

peter wrote: ↑Sat Aug 06, 2022 9:55 am
Hobacle wrote: ↑Sat Aug 06, 2022 8:08 am [fen]r5b1/7P/p4R2/2B4p/rp6/4N1P1/pk6/3K4 w - - 0 1[/fen]
Not working anymore, or do you have anything better for White then this?

r5b1/7P/p4R2/2B4p/rp6/4N1P1/pk6/3K4 w - - 0 1

Analysis by Crystallite 1.1a-bmi2:

1.Bd4+ Kb3 2.h8Q a1R+ 3.Bxa1 Rxa1+ 4.Kd2 Rd8+ 5.Nd5 Rxd5+ 6.Ke3 Be6 7.Kf2 Ra2+ 8.Kf3 Rd1 9.Kf4 Bc4 10.Kg5 Rd3 11.Qb8 Rg2 12.Kh6 Rdxg3 13.Qd6 Rd3 14.Qe5 Rc3 15.Rb6 Ra2 16.Qxh5 a5 17.Kg7 a4 18.Kf6 Rf2+ 19.Ke7 Re2+ 20.Kf6 Rce3 21.Qh1 Re1 22.Qh2 Re6+ 23.Rxe6 Rxe6+ 24.Kg7 a3 25.Qd2 a2 26.Qd1+ Ka3 27.Qc1+ Ka4 28.Qd1+ b3 29.Qd7+ Kb4 30.Qd2+ Kb5 31.Qg5+ Ka4 32.Qc1 Re7+ 33.Kh6 Re4 34.Kg7 Be2 35.Qc6+ Ka3 36.Qc1+ Kb4 37.Qd2+ Ka4 38.Qb2 Rc4 39.Kh6 Bf3 40.Kg7 Kb4 41.Kf6 Rc6+ 42.Kg7 Rc3 43.Qa1 Bd5 44.Qb2 Bc4 45.Kf8 Rc2 46.Qe5 a1Q
Black is clearly winning: -+ (-6.35 ++) Depth: 40/91 00:11:37 28813MN, tb=187071955

Output after some Forward (with the one and another one wrongly drawing line yet too)- Backward,
regards
Why don't 1.Rf2+ Kb3/Kb1 2.Rb2+ work?
Last edited by Hobacle on Sat Aug 06, 2022 10:24 am, edited 1 time in total.
Top
peter
Posts: 3410
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Test position, white to play and draw.

Post by peter »

Hobacle wrote: ↑Sat Aug 06, 2022 10:16 am
peter wrote: ↑Sat Aug 06, 2022 9:55 am
Hobacle wrote: ↑Sat Aug 06, 2022 8:08 am [fen]r5b1/7P/p4R2/2B4p/rp6/4N1P1/pk6/3K4 w - - 0 1[/fen]
Not working anymore, or do you have anything better for White then this?

r5b1/7P/p4R2/2B4p/rp6/4N1P1/pk6/3K4 w - - 0 1

Analysis by Crystallite 1.1a-bmi2:

1.Bd4+ Kb3 2.h8Q a1R+ 3.Bxa1 Rxa1+ 4.Kd2 Rd8+ 5.Nd5 Rxd5+ 6.Ke3 Be6 7.Kf2 Ra2+ 8.Kf3 Rd1 9.Kf4 Bc4 10.Kg5 Rd3 11.Qb8 Rg2 12.Kh6 Rdxg3 13.Qd6 Rd3 14.Qe5 Rc3 15.Rb6 Ra2 16.Qxh5 a5 17.Kg7 a4 18.Kf6 Rf2+ 19.Ke7 Re2+ 20.Kf6 Rce3 21.Qh1 Re1 22.Qh2 Re6+ 23.Rxe6 Rxe6+ 24.Kg7 a3 25.Qd2 a2 26.Qd1+ Ka3 27.Qc1+ Ka4 28.Qd1+ b3 29.Qd7+ Kb4 30.Qd2+ Kb5 31.Qg5+ Ka4 32.Qc1 Re7+ 33.Kh6 Re4 34.Kg7 Be2 35.Qc6+ Ka3 36.Qc1+ Kb4 37.Qd2+ Ka4 38.Qb2 Rc4 39.Kh6 Bf3 40.Kg7 Kb4 41.Kf6 Rc6+ 42.Kg7 Rc3 43.Qa1 Bd5 44.Qb2 Bc4 45.Kf8 Rc2 46.Qe5 a1Q
Black is clearly winning: -+ (-6.35 ++) Depth: 40/91 00:11:37 28813MN, tb=187071955

Output after some Forward (with the one and another one wrongly drawing line yet too)- Backward,
regards
Why doesn't 1.Rf2+ work? 1.Rf2+ Kb3/Kb1 2.Rb2+ Kxb2 then we get into the familiar position.
Stupid me
:)
Peter.
Top
Vinvin
Posts: 5297
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: Test Position, white to play and draw

Post by Vinvin »

Not easy to find an engine who solves this position, but I finally found Crystal 3.2.
5 runs on my AMD 5950x, 16 cores@4Ghz
Timing to find the mainline and a 0.00 eval : 2m10s, 2m52s, 3m21s, 1m6s, 1m16s. Average : 2m9s

The 5 outputs :

Code: Select all

FEN: r5b1/7P/p6p/2B5/rp1R4/4N1P1/pk6/3K4 w - - 1 1

Crystal 3.2 x64 avx2:
Found 513 tablebases
NNUE evaluation using nn-190f102a22c3.nnue enabled
...
 20/37	01:10	 1.836.185k	26.042k	-11,12	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd6 Rh8 6.Rd7 Kb2 7.Kd3 a5 8.Rb7 Ra1 9.g4 Rc1 10.Ke4 Re1 11.Kf3 a4 12.Rxb4 Rxh7 13.Nf5 Rc1 14.Rb5 Rc3+ 15.Kf4 a3
 21/37-	01:20	 2.090.364k	26.064k	-11,20	1.Rd2+ Kb1
 21/37-	01:23	 2.188.134k	26.108k	-11,24	1.Rd2+ Kb1
 21/37-	01:40	 2.626.650k	26.135k	-11,39	1.Rd2+ Kb1
 21/62	01:57	 3.067.722k	26.140k	-11,36	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bc4 10.Bf6 Bf7 11.Be5 Bb3 12.Bf6 a5 13.Bg7 Bc4 14.Bf6 Be6 15.Be5 Bf7 16.Bf6 Bc4 17.Be5 R8d7 18.Bh8 Bg8 19.Bf6 a4 20.Bh8 Rd8
 22/53-	02:04	 3.247.392k	26.131k	-11,60	1.Rd2+ Kb1
 22/53+	02:04	 3.248.675k	26.132k	-11,56	1.Rd2+
 22/53+	02:10	 3.390.661k	25.940k	-0,70	1.Rd2+
 22/53	02:10	 3.395.340k	25.929k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d5 10.Bf6
 23/29	02:11	 3.399.815k	25.925k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d5 10.Bf6
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/38	01:40	 2.618.754k	26.085k	-11,36	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Rb7 a5 7.Kf3 Kb2 8.g4 Ra1 9.Ke4 Re1 10.Kf4 a4 11.Rxb4 Rxh7 12.Re4 Rc1 13.g5 Rf7+ 14.Kg3 a3 15.Ng2 a2 16.g6 Rg7 17.Re2+ Rc2
 21/33-	01:57	 3.061.498k	25.985k	-11,56	1.Rd2+ Kb1
 21/33-	02:17	 3.564.220k	25.919k	-11,60	1.Rd2+ Kb1
 21/33-	02:42	 4.212.586k	25.858k	-11,66	1.Rd2+ Kb1
 21/35+	02:45	 4.281.378k	25.859k	-11,11	1.Rd2+
 21/60+	02:52	 4.443.703k	25.770k	-0,50	1.Rd2+
 21/60	02:52	 4.447.956k	25.765k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 h5 10.Bg7 Bf7 11.Bf6 Bc4 12.Bg7
 22/51	02:53	 4.462.455k	25.749k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 h5 10.Bg7 Bf7 11.Bf6 Bc4 12.Bg7

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/41	02:24	 3.805.553k	26.337k	-11,50	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd6 Rh8 6.Rxh6 Ba2 7.Nc2+ Kb2 8.Kd2 a5 9.Ne3 Bb1 10.Nc4+ Kb3 11.Rc6 Bxh7 12.Rc7 Rd8+ 13.Ke3 Rd3+ 14.Kf4 Rd4+ 15.Ke5 Rxc4 16.Rxh7 Ra1 17.Rb7 a4 18.g4 Rxg4 19.Rh7
 21/37-	02:55	 4.596.125k	26.254k	-11,96	1.Rd2+ Kb1
 21/37-	03:12	 5.044.175k	26.269k	-12,00	1.Rd2+ Kb1
 21/37+	03:12	 5.044.813k	26.269k	-11,96	1.Rd2+
 21/37+	03:12	 5.045.062k	26.269k	-11,74	1.Rd2+
 21/47	03:12	 5.048.655k	26.270k	-11,59	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bb3 10.Bf6 R8d5 11.Bg7 Ba4 12.Bf6 Bd7 13.Bh8 Bc8 14.Bf6 a5 15.Bg7 Bf5 16.Bh8 Be4 17.Bf6 Bh7 18.Bh8 a4 19.Bg7 Bg6 20.Bf6 a3
 22/25-	03:15	 5.142.510k	26.275k	-11,30	1.Rd2+ Kb1
 22/35+	03:15	 5.142.651k	26.275k	-11,26	1.Rd2+
 22/35+	03:19	 5.237.857k	26.247k	-1,65	1.Rd2+
 22/69	03:21	 5.293.837k	26.222k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Bf6
 23/61	03:23	 5.339.590k	26.199k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Bf6

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/66	00:52	 1.346.540k	25.727k	-10,73	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Be5 Bf7 11.Bg7 R8d6 12.Bh8 Bb3 13.Be5 Bc4 14.Bh8 Rd7 15.Bf6 Be6 16.Bh8 R7d5 17.Bf6 Bf7 18.Bh8 Bg8 19.Bf6 Bh7 20.Bh8 h5 21.Bf6 Bg8 22.Bh8 Rd7 23.Bf6 Rd8 24.Bh8 Bf7 25.Be5 Bc4 26.Bf6 Bb3 27.Be5 a5 28.Bg7 Bg8 29.Be5 Be6 30.Bg7 R8d6 31.Bf6 h4
 21/44-	00:52	 1.359.735k	25.733k	-10,73	1.Rd2+ Kb1
 21/53+	00:52	 1.359.876k	25.733k	-10,69	1.Rd2+
 21/53	00:52	 1.359.962k	25.733k	-10,69	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Be5 Be6 11.Bg7 R8d5 12.Bh8 Bg8 13.Bg7 a5 14.Bh8 h5 15.Bf6 Rd8 16.Be5 Bc4 17.Bf6 Bd3 18.Be5 a4 19.Bf6 h4 20.gxh4
 22/50-	01:00	 1.564.621k	25.894k	-10,73	1.Rd2+ Kb1
 22/50+	01:00	 1.564.627k	25.894k	-10,61	1.Rd2+
 22/53+	01:04	 1.668.407k	25.745k	-1,60	1.Rd2+
 22/65	01:06	 1.701.472k	25.684k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 23/49-	01:06	 1.702.202k	25.686k	-0,03	1.Rd2+ Kb1
 23/49+	01:06	 1.707.782k	25.705k	 0,00	1.Rd2+
 23/49	01:06	 1.707.793k	25.705k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 24/39	01:06	 1.708.574k	25.706k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/38	00:43	 1.152.548k	26.626k	-10,88	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Rb7 Kb2 7.Kd3 a5 8.g4 Ra1 9.Kd4 Re1 10.Nf5 Bc2 11.Kc5 b3 12.Nd4 Rxh7 13.Nxc2 Re4
 21/39-	00:52	 1.389.640k	26.550k	-11,10	1.Rd2+ Kb1
 21/39-	00:57	 1.514.268k	26.496k	-11,14	1.Rd2+ Kb1
 21/39-	01:06	 1.762.630k	26.463k	-11,20	1.Rd2+ Kb1
 21/39	01:08	 1.805.127k	26.446k	-11,19	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Kd3 Kb2 7.g4 a5 8.Rb7 Ra1 9.Kd4 Rc1 10.Ke5 Re1 11.Kf4 a4 12.Ng2 Rf1+ 13.Kg3 Rf3+ 14.Kh2 Rc3 15.Nh4 a3 16.Ng6 Rxh7 17.Rxh7 Bc2 18.Rb7 b3 19.Nh8
 22/33+	01:16	 2.009.717k	26.179k	-0,21	1.Rd2+
 22/44+	01:16	 2.009.754k	26.180k	-0,12	1.Rd2+
 22/44+	01:16	 2.009.772k	26.179k	 0,00	1.Rd2+
 22/44	01:16	 2.009.799k	26.179k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bf7 10.Bg7 Bc4 11.Bf6
 23/45	01:16	 2.009.885k	26.179k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bf7
Top
acase
Posts: 994
Joined: Wed Mar 15, 2006 1:14 am
Location: Columbus, Ohio USA
Full name: Andrew R Case

Re: Test Position, white to play and draw

Post by acase »

Vinvin wrote: ↑Sun Aug 07, 2022 3:13 am Not easy to find an engine who solves this position, but I finally found Crystal 3.2.
5 runs on my AMD 5950x, 16 cores@4Ghz
Timing to find the mainline and a 0.00 eval : 2m10s, 2m52s, 3m21s, 1m6s, 1m16s. Average : 2m9s

The 5 outputs :

Code: Select all

FEN: r5b1/7P/p6p/2B5/rp1R4/4N1P1/pk6/3K4 w - - 1 1

Crystal 3.2 x64 avx2:
Found 513 tablebases
NNUE evaluation using nn-190f102a22c3.nnue enabled
...
 20/37	01:10	 1.836.185k	26.042k	-11,12	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd6 Rh8 6.Rd7 Kb2 7.Kd3 a5 8.Rb7 Ra1 9.g4 Rc1 10.Ke4 Re1 11.Kf3 a4 12.Rxb4 Rxh7 13.Nf5 Rc1 14.Rb5 Rc3+ 15.Kf4 a3
 21/37-	01:20	 2.090.364k	26.064k	-11,20	1.Rd2+ Kb1
 21/37-	01:23	 2.188.134k	26.108k	-11,24	1.Rd2+ Kb1
 21/37-	01:40	 2.626.650k	26.135k	-11,39	1.Rd2+ Kb1
 21/62	01:57	 3.067.722k	26.140k	-11,36	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bc4 10.Bf6 Bf7 11.Be5 Bb3 12.Bf6 a5 13.Bg7 Bc4 14.Bf6 Be6 15.Be5 Bf7 16.Bf6 Bc4 17.Be5 R8d7 18.Bh8 Bg8 19.Bf6 a4 20.Bh8 Rd8
 22/53-	02:04	 3.247.392k	26.131k	-11,60	1.Rd2+ Kb1
 22/53+	02:04	 3.248.675k	26.132k	-11,56	1.Rd2+
 22/53+	02:10	 3.390.661k	25.940k	-0,70	1.Rd2+
 22/53	02:10	 3.395.340k	25.929k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d5 10.Bf6
 23/29	02:11	 3.399.815k	25.925k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d5 10.Bf6
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/38	01:40	 2.618.754k	26.085k	-11,36	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Rb7 a5 7.Kf3 Kb2 8.g4 Ra1 9.Ke4 Re1 10.Kf4 a4 11.Rxb4 Rxh7 12.Re4 Rc1 13.g5 Rf7+ 14.Kg3 a3 15.Ng2 a2 16.g6 Rg7 17.Re2+ Rc2
 21/33-	01:57	 3.061.498k	25.985k	-11,56	1.Rd2+ Kb1
 21/33-	02:17	 3.564.220k	25.919k	-11,60	1.Rd2+ Kb1
 21/33-	02:42	 4.212.586k	25.858k	-11,66	1.Rd2+ Kb1
 21/35+	02:45	 4.281.378k	25.859k	-11,11	1.Rd2+
 21/60+	02:52	 4.443.703k	25.770k	-0,50	1.Rd2+
 21/60	02:52	 4.447.956k	25.765k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 h5 10.Bg7 Bf7 11.Bf6 Bc4 12.Bg7
 22/51	02:53	 4.462.455k	25.749k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 h5 10.Bg7 Bf7 11.Bf6 Bc4 12.Bg7

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/41	02:24	 3.805.553k	26.337k	-11,50	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd6 Rh8 6.Rxh6 Ba2 7.Nc2+ Kb2 8.Kd2 a5 9.Ne3 Bb1 10.Nc4+ Kb3 11.Rc6 Bxh7 12.Rc7 Rd8+ 13.Ke3 Rd3+ 14.Kf4 Rd4+ 15.Ke5 Rxc4 16.Rxh7 Ra1 17.Rb7 a4 18.g4 Rxg4 19.Rh7
 21/37-	02:55	 4.596.125k	26.254k	-11,96	1.Rd2+ Kb1
 21/37-	03:12	 5.044.175k	26.269k	-12,00	1.Rd2+ Kb1
 21/37+	03:12	 5.044.813k	26.269k	-11,96	1.Rd2+
 21/37+	03:12	 5.045.062k	26.269k	-11,74	1.Rd2+
 21/47	03:12	 5.048.655k	26.270k	-11,59	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bb3 10.Bf6 R8d5 11.Bg7 Ba4 12.Bf6 Bd7 13.Bh8 Bc8 14.Bf6 a5 15.Bg7 Bf5 16.Bh8 Be4 17.Bf6 Bh7 18.Bh8 a4 19.Bg7 Bg6 20.Bf6 a3
 22/25-	03:15	 5.142.510k	26.275k	-11,30	1.Rd2+ Kb1
 22/35+	03:15	 5.142.651k	26.275k	-11,26	1.Rd2+
 22/35+	03:19	 5.237.857k	26.247k	-1,65	1.Rd2+
 22/69	03:21	 5.293.837k	26.222k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Bf6
 23/61	03:23	 5.339.590k	26.199k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Bf6

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/66	00:52	 1.346.540k	25.727k	-10,73	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Be5 Bf7 11.Bg7 R8d6 12.Bh8 Bb3 13.Be5 Bc4 14.Bh8 Rd7 15.Bf6 Be6 16.Bh8 R7d5 17.Bf6 Bf7 18.Bh8 Bg8 19.Bf6 Bh7 20.Bh8 h5 21.Bf6 Bg8 22.Bh8 Rd7 23.Bf6 Rd8 24.Bh8 Bf7 25.Be5 Bc4 26.Bf6 Bb3 27.Be5 a5 28.Bg7 Bg8 29.Be5 Be6 30.Bg7 R8d6 31.Bf6 h4
 21/44-	00:52	 1.359.735k	25.733k	-10,73	1.Rd2+ Kb1
 21/53+	00:52	 1.359.876k	25.733k	-10,69	1.Rd2+
 21/53	00:52	 1.359.962k	25.733k	-10,69	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Be5 Be6 11.Bg7 R8d5 12.Bh8 Bg8 13.Bg7 a5 14.Bh8 h5 15.Bf6 Rd8 16.Be5 Bc4 17.Bf6 Bd3 18.Be5 a4 19.Bf6 h4 20.gxh4
 22/50-	01:00	 1.564.621k	25.894k	-10,73	1.Rd2+ Kb1
 22/50+	01:00	 1.564.627k	25.894k	-10,61	1.Rd2+
 22/53+	01:04	 1.668.407k	25.745k	-1,60	1.Rd2+
 22/65	01:06	 1.701.472k	25.684k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 23/49-	01:06	 1.702.202k	25.686k	-0,03	1.Rd2+ Kb1
 23/49+	01:06	 1.707.782k	25.705k	 0,00	1.Rd2+
 23/49	01:06	 1.707.793k	25.705k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 24/39	01:06	 1.708.574k	25.706k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/38	00:43	 1.152.548k	26.626k	-10,88	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Rb7 Kb2 7.Kd3 a5 8.g4 Ra1 9.Kd4 Re1 10.Nf5 Bc2 11.Kc5 b3 12.Nd4 Rxh7 13.Nxc2 Re4
 21/39-	00:52	 1.389.640k	26.550k	-11,10	1.Rd2+ Kb1
 21/39-	00:57	 1.514.268k	26.496k	-11,14	1.Rd2+ Kb1
 21/39-	01:06	 1.762.630k	26.463k	-11,20	1.Rd2+ Kb1
 21/39	01:08	 1.805.127k	26.446k	-11,19	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Kd3 Kb2 7.g4 a5 8.Rb7 Ra1 9.Kd4 Rc1 10.Ke5 Re1 11.Kf4 a4 12.Ng2 Rf1+ 13.Kg3 Rf3+ 14.Kh2 Rc3 15.Nh4 a3 16.Ng6 Rxh7 17.Rxh7 Bc2 18.Rb7 b3 19.Nh8
 22/33+	01:16	 2.009.717k	26.179k	-0,21	1.Rd2+
 22/44+	01:16	 2.009.754k	26.180k	-0,12	1.Rd2+
 22/44+	01:16	 2.009.772k	26.179k	 0,00	1.Rd2+
 22/44	01:16	 2.009.799k	26.179k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bf7 10.Bg7 Bc4 11.Bf6
 23/45	01:16	 2.009.885k	26.179k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bf7
Hello Vincent,

Did you give this one a try? I had posted it earlier.

White to play and draw, the move you are looking for is 1.b6!!

[fen]nq6/pPpB4/2K5/1P2BN2/3Pk2p/5p2/8/6bN w - - 0 1[/fen]
Top
Hobacle
Posts: 244
Joined: Thu Mar 04, 2021 3:57 pm
Full name: Nott Q Suree

Re: Test Position, white to play and draw

Post by Hobacle »

Vinvin wrote: ↑Sun Aug 07, 2022 3:13 am Not easy to find an engine who solves this position, but I finally found Crystal 3.2.
5 runs on my AMD 5950x, 16 cores@4Ghz
Timing to find the mainline and a 0.00 eval : 2m10s, 2m52s, 3m21s, 1m6s, 1m16s. Average : 2m9s

The 5 outputs :

Code: Select all

FEN: r5b1/7P/p6p/2B5/rp1R4/4N1P1/pk6/3K4 w - - 1 1

Crystal 3.2 x64 avx2:
Found 513 tablebases
NNUE evaluation using nn-190f102a22c3.nnue enabled
...
 20/37	01:10	 1.836.185k	26.042k	-11,12	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd6 Rh8 6.Rd7 Kb2 7.Kd3 a5 8.Rb7 Ra1 9.g4 Rc1 10.Ke4 Re1 11.Kf3 a4 12.Rxb4 Rxh7 13.Nf5 Rc1 14.Rb5 Rc3+ 15.Kf4 a3
 21/37-	01:20	 2.090.364k	26.064k	-11,20	1.Rd2+ Kb1
 21/37-	01:23	 2.188.134k	26.108k	-11,24	1.Rd2+ Kb1
 21/37-	01:40	 2.626.650k	26.135k	-11,39	1.Rd2+ Kb1
 21/62	01:57	 3.067.722k	26.140k	-11,36	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bc4 10.Bf6 Bf7 11.Be5 Bb3 12.Bf6 a5 13.Bg7 Bc4 14.Bf6 Be6 15.Be5 Bf7 16.Bf6 Bc4 17.Be5 R8d7 18.Bh8 Bg8 19.Bf6 a4 20.Bh8 Rd8
 22/53-	02:04	 3.247.392k	26.131k	-11,60	1.Rd2+ Kb1
 22/53+	02:04	 3.248.675k	26.132k	-11,56	1.Rd2+
 22/53+	02:10	 3.390.661k	25.940k	-0,70	1.Rd2+
 22/53	02:10	 3.395.340k	25.929k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d5 10.Bf6
 23/29	02:11	 3.399.815k	25.925k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d5 10.Bf6
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/38	01:40	 2.618.754k	26.085k	-11,36	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Rb7 a5 7.Kf3 Kb2 8.g4 Ra1 9.Ke4 Re1 10.Kf4 a4 11.Rxb4 Rxh7 12.Re4 Rc1 13.g5 Rf7+ 14.Kg3 a3 15.Ng2 a2 16.g6 Rg7 17.Re2+ Rc2
 21/33-	01:57	 3.061.498k	25.985k	-11,56	1.Rd2+ Kb1
 21/33-	02:17	 3.564.220k	25.919k	-11,60	1.Rd2+ Kb1
 21/33-	02:42	 4.212.586k	25.858k	-11,66	1.Rd2+ Kb1
 21/35+	02:45	 4.281.378k	25.859k	-11,11	1.Rd2+
 21/60+	02:52	 4.443.703k	25.770k	-0,50	1.Rd2+
 21/60	02:52	 4.447.956k	25.765k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 h5 10.Bg7 Bf7 11.Bf6 Bc4 12.Bg7
 22/51	02:53	 4.462.455k	25.749k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 h5 10.Bg7 Bf7 11.Bf6 Bc4 12.Bg7

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/41	02:24	 3.805.553k	26.337k	-11,50	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd6 Rh8 6.Rxh6 Ba2 7.Nc2+ Kb2 8.Kd2 a5 9.Ne3 Bb1 10.Nc4+ Kb3 11.Rc6 Bxh7 12.Rc7 Rd8+ 13.Ke3 Rd3+ 14.Kf4 Rd4+ 15.Ke5 Rxc4 16.Rxh7 Ra1 17.Rb7 a4 18.g4 Rxg4 19.Rh7
 21/37-	02:55	 4.596.125k	26.254k	-11,96	1.Rd2+ Kb1
 21/37-	03:12	 5.044.175k	26.269k	-12,00	1.Rd2+ Kb1
 21/37+	03:12	 5.044.813k	26.269k	-11,96	1.Rd2+
 21/37+	03:12	 5.045.062k	26.269k	-11,74	1.Rd2+
 21/47	03:12	 5.048.655k	26.270k	-11,59	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bb3 10.Bf6 R8d5 11.Bg7 Ba4 12.Bf6 Bd7 13.Bh8 Bc8 14.Bf6 a5 15.Bg7 Bf5 16.Bh8 Be4 17.Bf6 Bh7 18.Bh8 a4 19.Bg7 Bg6 20.Bf6 a3
 22/25-	03:15	 5.142.510k	26.275k	-11,30	1.Rd2+ Kb1
 22/35+	03:15	 5.142.651k	26.275k	-11,26	1.Rd2+
 22/35+	03:19	 5.237.857k	26.247k	-1,65	1.Rd2+
 22/69	03:21	 5.293.837k	26.222k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Bf6
 23/61	03:23	 5.339.590k	26.199k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Bf6

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/66	00:52	 1.346.540k	25.727k	-10,73	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Be5 Bf7 11.Bg7 R8d6 12.Bh8 Bb3 13.Be5 Bc4 14.Bh8 Rd7 15.Bf6 Be6 16.Bh8 R7d5 17.Bf6 Bf7 18.Bh8 Bg8 19.Bf6 Bh7 20.Bh8 h5 21.Bf6 Bg8 22.Bh8 Rd7 23.Bf6 Rd8 24.Bh8 Bf7 25.Be5 Bc4 26.Bf6 Bb3 27.Be5 a5 28.Bg7 Bg8 29.Be5 Be6 30.Bg7 R8d6 31.Bf6 h4
 21/44-	00:52	 1.359.735k	25.733k	-10,73	1.Rd2+ Kb1
 21/53+	00:52	 1.359.876k	25.733k	-10,69	1.Rd2+
 21/53	00:52	 1.359.962k	25.733k	-10,69	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 Bb3 10.Be5 Be6 11.Bg7 R8d5 12.Bh8 Bg8 13.Bg7 a5 14.Bh8 h5 15.Bf6 Rd8 16.Be5 Bc4 17.Bf6 Bd3 18.Be5 a4 19.Bf6 h4 20.gxh4
 22/50-	01:00	 1.564.621k	25.894k	-10,73	1.Rd2+ Kb1
 22/50+	01:00	 1.564.627k	25.894k	-10,61	1.Rd2+
 22/53+	01:04	 1.668.407k	25.745k	-1,60	1.Rd2+
 22/65	01:06	 1.701.472k	25.684k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 23/49-	01:06	 1.702.202k	25.686k	-0,03	1.Rd2+ Kb1
 23/49+	01:06	 1.707.782k	25.705k	 0,00	1.Rd2+
 23/49	01:06	 1.707.793k	25.705k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 24/39	01:06	 1.708.574k	25.706k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bg7 R8d6 10.Bh8 Bf7 11.Bg7
 
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
...
 20/38	00:43	 1.152.548k	26.626k	-10,88	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Rb7 Kb2 7.Kd3 a5 8.g4 Ra1 9.Kd4 Re1 10.Nf5 Bc2 11.Kc5 b3 12.Nd4 Rxh7 13.Nxc2 Re4
 21/39-	00:52	 1.389.640k	26.550k	-11,10	1.Rd2+ Kb1
 21/39-	00:57	 1.514.268k	26.496k	-11,14	1.Rd2+ Kb1
 21/39-	01:06	 1.762.630k	26.463k	-11,20	1.Rd2+ Kb1
 21/39	01:08	 1.805.127k	26.446k	-11,19	1.Rd2+ Kb1 2.Bd4 Bb3+ 3.Ke2 a1Q 4.Bxa1 Kxa1 5.Rd7 Rh8 6.Kd3 Kb2 7.g4 a5 8.Rb7 Ra1 9.Kd4 Rc1 10.Ke5 Re1 11.Kf4 a4 12.Ng2 Rf1+ 13.Kg3 Rf3+ 14.Kh2 Rc3 15.Nh4 a3 16.Ng6 Rxh7 17.Rxh7 Bc2 18.Rb7 b3 19.Nh8
 22/33+	01:16	 2.009.717k	26.179k	-0,21	1.Rd2+
 22/44+	01:16	 2.009.754k	26.180k	-0,12	1.Rd2+
 22/44+	01:16	 2.009.772k	26.179k	 0,00	1.Rd2+
 22/44	01:16	 2.009.799k	26.179k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bf7 10.Bg7 Bc4 11.Bf6
 23/45	01:16	 2.009.885k	26.179k	 0,00	1.Rd2+ Kb1 2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Be5 Bf7
Hello, I think I have another version of this composition, with a harder beginning move:
Hobacle wrote: ↑Sat Aug 06, 2022 8:08 am [fen]r5b1/7P/p4R2/2B4p/rp6/4N1P1/pk6/3K4 w - - 0 1[/fen]
1.Rf2+!
Top
peter
Posts: 3410
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Test Position, white to play and draw

Post by peter »

Hobacle wrote: ↑Sun Aug 07, 2022 12:24 pm Hello, I think I have another version of this composition, with a harder beginning move:
Hobacle wrote: ↑Sat Aug 06, 2022 8:08 am [fen]r5b1/7P/p4R2/2B4p/rp6/4N1P1/pk6/3K4 w - - 0 1[/fen]
1.Rf2+!
Haha, you think all the others will fool themselves like I did?
:)
You gave this one already, but it's indeed even a little bit harder for the engines getting mislead in this variation by 1.Bd4+? also. Now to get right continuations soon even Crystallite likes to have MultiPV=2 to start with.

In original position first one move is found almost normally from most of the engines I tried, just the continuation with the rook- sac is something else and the 0.00- eval from starting postion I got from Crystallite only so far.

So to make it more usable as for a testposition for engines I'd rather take the one after the first move from White and Black, when the sac is to be found at once.
Not as beautiful as for a study maybe, but as for the engines at least solvable, which is a criterion for test positions at least too, there's no use in studies (of which there are still many as well as unsolvalbe with reasonable hardware- time) to test some engines with, if not a single one solves it. (Or even worse, only a single one is able to, which is often enough the case as for Sting and cyclic zugzwang- studies e.g.)

Just my two cents regards
Peter.
Top
Hobacle
Posts: 244
Joined: Thu Mar 04, 2021 3:57 pm
Full name: Nott Q Suree

Re: Test Position, white to play and draw

Post by Hobacle »

peter wrote: ↑Sun Aug 07, 2022 12:48 pm Haha, you think all the others will fool themselves like I did?
:)
You gave this one already, but it's indeed even a little bit harder for the engines getting mislead in this variation by 1.Bd4+? also. Now to get right continuations soon even Crystallite likes to have MultiPV=2 to start with.

In original position first one move is found almost normally from most of the engines I tried, just the continuation with the rook- sac is something else and the 0.00- eval from starting postion I got from Crystallite only so far.

So to make it more usable as for a testposition for engines I'd rather take the one after the first move from White and Black, when the sac is to be found at once.
Not as beautiful as for a study maybe, but as for the engines at least solvable, which is a criterion for test positions at least too, there's no use in studies (of which there are still many as well as unsolvalbe with reasonable hardware- time) to test some engines with, if not a single one solves it. (Or even worse, only a single one is able to, which is often enough the case as for Sting and cyclic zugzwang- studies e.g.)
I knew that. Not to fool others, I just found a position with a harder beginning move with one piece misplaced from the original study than the one OP posted and wanted to share it with others.
Top
peter
Posts: 3410
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Test Position, white to play and draw

Post by peter »

Hobacle wrote: ↑Sun Aug 07, 2022 1:45 pm I knew that. Not to fool others, I just found a position with a harder beginning move with one piece misplaced from the original study than the one OP posted and wanted to share it.
Don't get me wrong, dear Hobacle, as for my personal pov I just wanted to put the spot on engine- testing according to the thread- titel.
As for this one, the engine- side, I estimated the maybe really somewhat more difficult position of yours as for the "seduction" 1.Bd4+?, maybe that makes it really even harder for humans too, especially if they mix up engine- output of mate- score for the side of White's and the one of Black"s as it happened to me. Cardinal mistake of course was to look at engines' output as for evals more then as for the moves anyhow
:)
As for a test position original one isn't so good if you don't judge evals at all neither, (what you normally don't do in suites with criterion solved or not only) because many engines would "solve" it by accident, choosing the right first move but for wrong reasons as for the follow- up moves.

That's why I'd rather take your variation or (even better to me) the position after first move from White and from Black, because that makes it not much easier but solvable for engines and only if the rook- sac, which is the point of the position anyhow, is really "found" too.

E.g. LC0 on 3070ti with the net playing in TCEC latest:


[fen]r5b1/7P/p7/2B4p/rp6/4N1P1/p4R2/1k1K4 w - - 0 1[/fen]



Analysis by Lc0 v0.30.0-dev+git.75c6ada:

2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6
-/+ (-1.01) Depth: 7/13 00:00:04 0kN
...
2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Nc4 a1Q 5.Bd4 Bxc4 6.Qh7+ Ka2+ 7.Bxa1 Bb3+ 8.Ke2 Re8+ 9.Kf2
-/+ (-1.50) Depth: 8/16 00:00:04 1kN
2.h8Q a1Q 3.Bd4 Rd8 4.Rd2 Rxd4 5.Qxd4 Qxd4 6.Rxd4 a5 7.Nf1 Kb2 8.Ne3 Kc3
-+ (-2.41) Depth: 8/16 00:00:04 1kN
2.h8Q a1Q 3.Bd4 Rd8 4.Rd2 Rxd4 5.Qxd4 Qxd4 6.Rxd4 a5 7.Nf1 Kb2 8.Ne3 Kc3
-+ (-2.65) Depth: 8/17 00:00:04 2kN
2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Nc4 a1Q 5.Bd4 Bxc4 6.Qh7+ Ka2+ 7.Bxa1 Bb3+ 8.Ke1 Re8+ 9.Kf2 Kxa1
-+ (-2.18) Depth: 8/17 00:00:04 2kN
...
2.Rb2+ Kxb2 3.h8Q+ Kb1 4.Qa1+ Kxa1 5.Kc1 b3 6.Nc2+ bxc2 7.Be7 Rd4 8.Bf6 Rad8 9.Bh8 Bc4 10.Bg7 R8d6 11.Be5 a5 12.Bg7 a4 13.Bh8 a3 14.Bg7 h4 15.gxh4 Bd3 16.Bh8 Be4 17.h5 Bh7 18.h6 Bf5 19.Be5 Bh7 20.Bg7 R6d5
-+ (-2.28) Depth: 22/46 00:00:40 814kN, tb=198

Still wrong eval, but right solution as for the chosen move "for correct reasons" regarding some follow- up moves,
regards
Peter.
Top
Hobacle
Posts: 244
Joined: Thu Mar 04, 2021 3:57 pm
Full name: Nott Q Suree

Re: Test Position, white to play and draw

Post by Hobacle »

peter wrote: ↑Sun Aug 07, 2022 1:59 pm Don't get me wrong, dear Hobacle, as for my personal pov I just wanted to put the spot on engine- testing according to the thread- titel.
As for this one, the engine- side, I estimated the maybe really somewhat more difficult position of yours as for the "seduction" 1.Bd4+?, maybe that makes it really even harder for humans too, especially if they mix up engine- output of mate- score for the side of White's and the one of Black"s as it happened to me. Cardinal mistake of course was to look at engines's output as for evals more then as for the moves anyhow
:)
As for a test position original one isn't so good if you don't judge evals at all neither, (what you normally don't do in suites with criterion solved or not only) because many engines would "solve" it by accident, choosing the right first move but for wrong reasons as for the follow- up moves.

That's why I'd rather take your variation or (even better to me) the position after first move from White and from Black, because that makes it not much easier but solvable for engines and only if the rook- sac, which is the point of the position anyhow, is really "found" too,
regards
I agree your point, since this is a thread about "test position".
Top
acase
Posts: 994
Joined: Wed Mar 15, 2006 1:14 am
Location: Columbus, Ohio USA
Full name: Andrew R Case

Re: Test Position, white to play and draw

Post by acase »

Why not complicate it some more by adding Queens into the mix and giving the white rook a possible checking move at the e3 square to confuse the engines even more? :lol: :twisted:

Try this out:
[fen]r5b1/7P/p7/2B4p/rp1R4/k3N1P1/pq2Q3/3K4 w - - 0 1[/fen]

Obviously 1.QxQ+ is the first move, also note that 1.Rd3+ allows black to block it and deliver a cross check to white with the Bishop on g8 by playing 1...Bb3+
Top

Return to “Computer Chess Club: General Topics”