7 steps to heaven. 7 testpositions of variable difficulty!

[lech]

With the removal of the wrong WAC.230 ...b2 position and the addition of this, as i understand computers are only be capable of solving the first 2(the easy ones) of the 7. That's unfortunate and i hope to see someday a program to solve 2-3 of these 5 tough ones....

My testing Sting SF 3 (bred on Stockfish 2.1.1 - modified null move search) gets it in 50 minutes (2 threads 1056M hash).
[d]1b6/3kB3/8/1KpPp2p/4Pp1P/5P2/8/8 w - - 0 1 bm Bf8;

Code: Select all

info depth 55 seldepth 63 multipv 1 score cp 177 nodes 1295351015 nps 1205485 ti
me 1074547 pv e7c5 b8c7 c5b6 c7b8 b5c4 b8d6 b6f2 d6c7 c4b4 d7d6 f2c5 d6d7 c5g1 c
7d6 b4c4 d6e7 g1f2 e7d6 f2c5 d6c7 d5d6 c7a5 c5b4 a5b6 c4d5 b6d4 b4d2 d4b2 d5c5 b
2d4 c5d5
info depth 56
info currmove e7c5 currmovenumber 1
info nodes 1693686623 nps 1191569 time 1421391
info currmove b5c5 currmovenumber 2
info currmove d5d6 currmovenumber 3
info nodes 1751065480 nps 1192722 time 1468125
info currmove b5c4 currmovenumber 4
info currmove e7d8 currmovenumber 5
info currmove e7g5 currmovenumber 6
info nodes 1752196588 nps 1192578 time 1469250
info currmove e7f6 currmovenumber 7
info nodes 1764634282 nps 1191854 time 1480578
info currmove b5b6 currmovenumber 8
info currmove e7d6 currmovenumber 9
info currmove e7f8 currmovenumber 10
info depth 56 seldepth 0 multipv 1 score cp 185 lowerbound nodes 3627347184 nps
1189172 time 3050313 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8h4 g7e5 h4f2 c5c4 h5h4 c4d3 f
2e3 d3e2 h4h3 e2f1 d7d8 e5d6
info nodes 3627347184 nps 1189172 time 3050313
info currmove e7f8 currmovenumber 1

Critter-1.6a (8 threads, 8GB hash) gets this in about 5 minutes:

Code: Select all

33/78+ 05:01    3267M  10827K  +1.28 Bf8 Bc7 Kxc5 Bd8 Bg7 Bxh4 Bxe5 Bf2+ Kc4 h4 Kd3 Be3 Ke2 h3 Kf1 Kc8 Bc3 Bb6 e5 Kd7 Bd2 Be3 e6+ Kd6 Bb4+ Kc7 Ba3 Bd4 Bf8 Bh8 Kg1 Kd8 Kh2
33/78+ 07:15    4712M  10825K  +1.41 Bf8 Bc7 Kxc5 Bd8 Bg7 Bxh4 Bxe5 Bf2+ Kc4 h4 Kd3 Be3 Ke2 h3 Kf1 Kc8 Bc3 Bb6 e5 Kd7 Bd2 Be3 e6+ Kd6 Bb4+ Kc7 Ba3 Kd8 d6 Ke8 d7+ Kd8 Bd6 Bd4 Bxf4 Bf6 Kf2 Be7 Bh2 Bh4+ Kg1 Ke7 Bc7 Kxe6 d8=Q Bxd8 Bxd8 Kf5 Bh4 Kf4 Kf2

Louis, I think that many engines can find 1.Bf8 move.
Time is only question of nps (hardware).

[lech]

Avoid 1.Bxc7? that loses as also all other moves that lose too and play 1.c6! that draws.
[d]8/p1pp3p/8/2P5/5B2/p2K4/7P/1k6 w - - 0 1

1.Bxh7! wins! Classic Bxh7 sacrifice, but here the win is more difficult than the classic ones.
[d]r1bq1rk1/1p1nbppp/p3p3/2ppP3/5B1P/2PBP3/PP1N1PP1/R2QK2R w KQ - 0 1

The two positions are dedicated for engines (or for Peter) ?

For engines of course.
The Bxh7+ sacrifice is particularly interesting since it is a normal position that could happen in a game and it shows the ability of the engine to attack(calculate correctly the attack).

George, as I see in Sting analysis both are very easy for engines and rather are far to be some problem (step to heaven).

[Aser Huerga]

Avoid 1.Bxc7? that loses as also all other moves that lose too and play 1.c6! that draws.
[d]8/p1pp3p/8/2P5/5B2/p2K4/7P/1k6 w - - 0 1

1.Bxh7! wins! Classic Bxh7 sacrifice, but here the win is more difficult than the classic ones.
[d]r1bq1rk1/1p1nbppp/p3p3/2ppP3/5B1P/2PBP3/PP1N1PP1/R2QK2R w KQ - 0 1

The two positions are dedicated for engines (or for Peter) ?

For engines of course.
The Bxh7+ sacrifice is particularly interesting since it is a normal position that could happen in a game and it shows the ability of the engine to attack(calculate correctly the attack).

George, as I see in Sting analysis both are very easy for engines and rather are far to be some problem (step to heaven).

The first position (c6!) is not easy for most engines: try Houdini, Critter, Hiarcs, Komodo ...

[lech]

Avoid 1.Bxc7? that loses as also all other moves that lose too and play 1.c6! that draws.
[d]8/p1pp3p/8/2P5/5B2/p2K4/7P/1k6 w - - 0 1

1.Bxh7! wins! Classic Bxh7 sacrifice, but here the win is more difficult than the classic ones.
[d]r1bq1rk1/1p1nbppp/p3p3/2ppP3/5B1P/2PBP3/PP1N1PP1/R2QK2R w KQ - 0 1

The two positions are dedicated for engines (or for Peter) ?

For engines of course.
The Bxh7+ sacrifice is particularly interesting since it is a normal position that could happen in a game and it shows the ability of the engine to attack(calculate correctly the attack).

George, as I see in Sting analysis both are very easy for engines and rather are far to be some problem (step to heaven).

The first position (c6!) is not easy for most engines: try Houdini, Critter, Hiarcs, Komodo ...

Try Stockfish!
Sting SF 3 solves it very quickly! If I remember well, less than 1 minute (2 threads).
Probably Stockfish do it immediatelly.

[Aser Huerga]

The first position (c6!) is not easy for most engines: try Houdini, Critter, Hiarcs, Komodo ...

Try Stockfish!
Sting SF 3 solves it very quickly! If I remember well, less than 1 minute (2 threads).
Probably Stockfish do it immediatelly.

Yes, Stockfish finds it quickly, then comes Rybka, but all others are blind, even playing c6 manually Houdini takes some minutes to see the draw.
Good test position!

[lech]

Avoid 1.Bxc7? that loses as also all other moves that lose too and play 1.c6! that draws.
[d]8/p1pp3p/8/2P5/5B2/p2K4/7P/1k6 w - - 0 1
1.Bxh7! wins! Classic Bxh7 sacrifice, but here the win is more difficult than the classic ones.
[d]r1bq1rk1/1p1nbppp/p3p3/2ppP3/5B1P/2PBP3/PP1N1PP1/R2QK2R w KQ - 0 1

Stockfish-222-32-ja solves 1st position in 9 sec (Sting-sf-3 in 17 sec.).
And the second Stockfish in 952 sec./depth 27 (Sting 338 sec./ depth 26), Critter 1.6a in 19 minutes 44 seconds/ depth 23.

Sting-sf-3 (2 threads; 1056M hash)
[d]r1bq1rk1/1p1nbppp/p3p3/2ppP3/5B1P/2PBP3/PP1N1PP1/R2QK2R w KQ - 0 1 bm Bxh7;

Code: Select all

info depth 25 seldepth 41 multipv 1 score cp 56 nodes 173336092 nps 556177 time
311656 pv d2f3 h7h6 f3g5 d8b6 d3h7 g8h8 d1c2 c5c4 h7g8 f7f5 g8e6 e7g5 e6d7 g5f4
d7c8 f4e3 c8d7 e3c5 e5e6 d5d4 c3d4 c5b4 e1f1 b6d4 a1d1 d4e4 c2e4 f5e4 f1e2 b4e7
h4h5 h8g8 d7a4 a8d8 a4c2 d8d1 h1d1
info depth 26
info currmove d2f3 currmovenumber 1
info nodes 178464407 nps 556450 time 320719
info currmove d1h5 currmovenumber 2
info nodes 179933945 nps 556371 time 323406
info currmove h4h5 currmovenumber 3
info nodes 180808337 nps 556360 time 324984
info currmove f4g3 currmovenumber 4
info currmove d3c2 currmovenumber 5
info currmove d1g4 currmovenumber 6
info nodes 182736654 nps 556487 time 328375
info currmove d1c2 currmovenumber 7
info nodes 183326352 nps 556429 time 329469
info currmove e1f1 currmovenumber 8
info currmove d3e4 currmovenumber 9
info currmove d3g6 currmovenumber 10
info currmove f4h6 currmovenumber 11
info currmove e1g1 currmovenumber 12
info currmove d3e2 currmovenumber 13
info currmove d2b1 currmovenumber 14
info currmove c3c4 currmovenumber 15
info nodes 184732871 nps 556320 time 332062
info currmove g2g4 currmovenumber 16
info currmove d1f3 currmovenumber 17
info nodes 184901612 nps 556279 time 332390
info currmove e1e2 currmovenumber 18
info currmove d3c4 currmovenumber 19
info currmove d3b5 currmovenumber 20
info currmove g2g3 currmovenumber 21
info currmove d3f5 currmovenumber 22
info currmove h1h3 currmovenumber 23
info currmove f4g5 currmovenumber 24
info currmove d2f1 currmovenumber 25
info currmove a1b1 currmovenumber 26
info currmove d2e4 currmovenumber 27
info currmove f4h2 currmovenumber 28
info currmove d3f1 currmovenumber 29
info currmove b2b3 currmovenumber 30
info currmove b2b4 currmovenumber 31
info currmove d2b3 currmovenumber 32
info currmove d1b3 currmovenumber 33
info currmove a2a4 currmovenumber 34
info currmove d3b1 currmovenumber 35
info currmove d1b1 currmovenumber 36
info currmove d2c4 currmovenumber 37
info currmove e3e4 currmovenumber 38
info currmove f2f3 currmovenumber 39
info currmove h1h2 currmovenumber 40
info currmove h1f1 currmovenumber 41
info currmove a1c1 currmovenumber 42
info currmove d1c1 currmovenumber 43
info currmove h1g1 currmovenumber 44
info currmove d1e2 currmovenumber 45
info currmove a2a3 currmovenumber 46
info currmove d1a4 currmovenumber 47
info currmove d3h7 currmovenumber 48
info depth 26 seldepth 0 multipv 1 score cp 68 lowerbound nodes 188225347 nps 55
6751 time 338078 pv d3h7 g8h7 d1h5 h7g8 d2f3 f7f6 f3g5 f6g5 h4g5 f8f5 h5h7 g8f7
g5g6 f7e8 h7g7 e7f8 g7h8 d8b6 g6g7 b6b2 e1g1
info nodes 188225347 nps 556751 time 338078
info currmove d3h7 currmovenumber 1
info depth 26 seldepth 0 multipv 1 score cp 80 lowerbound nodes 189270546 nps 55
7164 time 339703 pv d3h7 g8h7 d1h5 h7g8 d2f3 f7f6 f3g5 f6g5 h4g5 f8f5 h5h7 g8f7
g5g6 f7e8 h7g7 e7f8 g7h8 d8b6 g6g7 b6b2 e1g1
info nodes 189270546 nps 557164 time 339703
info currmove d3h7 currmovenumber 1
info depth 26 seldepth 0 multipv 1 score cp 98 lowerbound nodes 191983802 nps 55
8092 time 344000 pv d3h7 g8h7 d1h5 h7g8 d2f3 f7f6 f3g5 f6g5 h4g5 f8f5 h5h7 g8f7
g5g6 f7e8 h7g7 e7f8 g7h8 d8b6 g6g7 b6b2 e1g1
info nodes 191983802 nps 558092 time 344000
info currmove d3h7 currmovenumber 1
info depth 26 seldepth 0 multipv 1 score cp 126 lowerbound nodes 197370131 nps 5
59394 time 352828 pv d3h7 g8h7 d1h5 h7g8 d2f3 f7f6 f3g5 f6g5 h4g5 f8f5 h5h7 g8f7
 g5g6 f7e8 h7g7 e7f8 g7h8 d8b6 g6g7 b6b2 e1g1
info nodes 197370131 nps 559369 time 352844
info currmove d3h7 currmovenumber 1
info depth 26 seldepth 0 multipv 1 score cp 167 lowerbound nodes 208122808 nps 5
61522 time 370640 pv d3h7 g8h7 d1h5 h7g8 d2f3 f7f6 f3g5 f6g5 h4g5 f8f5 h5h7 g8f7
 g5g6 f7e8 h7g7 e7f8 g7h8 d8b6 g6g7 b6b2 e1g1
info nodes 208122808 nps 561498 time 370656
info currmove d3h7 currmovenumber 1
info depth 26 seldepth 0 multipv 1 score cp 228 lowerbound nodes 216328457 nps 5
63400 time 383969 pv d3h7 g8h7 d1h5 h7g8 d2f3 f7f6 f3g5 f6g5 h4g5 f8f5 h5h7 g8f7
 g5g6 f7e8 h7g7 e7f8 g7h8 d8b6 g6g7 b6b2 e1g1
info nodes 216328457 nps 563378 time 383984
info currmove d3h7 currmovenumber 1

[ernest]

The first position (c6!) is not easy for most engines: try Houdini, Critter, Hiarcs, Komodo ...

Well Houdini is by far not as fast as Stockfish, but does get it in 2.6 minutes on my old DualCore:

8/p1pp3p/8/2P5/5B2/p2K4/7P/1k6 w - - 0 1
Analysis by Houdini 3 w32 Nlm:
..............
1.Bxc7 a2 2.Be5 a1Q 3.Bxa1 Kxa1 4.Kc4 Kb2 5.Kb4 h5 6.Kc4 Ka3 7.Kc3 a5 8.Kc4 Ka4 9.h3 h4 10.Kc3 Kb5 11.Kd4 Kb4 12.c6 dxc6 13.Kd3 a4 14.Kc2 c5 15.Kb2 a3+ 16.Kb1 Kb3 17.Ka1 c4 18.Kb1 c3
-+ (-4.09) Depth: 27/48 00:00:27 101mN, tb=516
1.Bxc7 a2 2.Be5 a1Q 3.Bxa1 Kxa1 4.Kc4 Kb2 5.Kb4 h5 6.Kc4 Ka3 7.Kc3 a5 8.Kc4 Ka4 9.h3 h4 10.Kc3 Kb5 11.Kd4 Kb4 12.Kd3 Kxc5 13.Kc2 d5 14.Kb3 d4 15.Kb2 a4 16.Kc2 Kc4 17.Kd2 a3 18.Kc2 d3+ 19.Kc1 Kc3
-+ (-4.23) Depth: 28/50 00:00:39 144mN, tb=868
1.Bxc7 a2
-+ (-4.41 --) Depth: 29/50 00:00:42 152mN, tb=1114
1.Bxc7 a2
-+ (-4.76 --) Depth: 29/50 00:00:44 161mN, tb=1258
1.Bxc7 a2
-+ (-5.70 --) Depth: 29/50 00:00:48 173mN, tb=1695
1.Bxc7 a2 2.Be5 a1Q 3.Bxa1 Kxa1 4.Kc4 Ka2 5.Kb4 Kb2 6.Kc4 Ka3 7.Kb5 Kb3 8.h3 Kc3 9.h4 h6 10.Ka6 Kd4 11.Kb5 h5 12.Ka4 Kc4 13.c6 dxc6 14.Ka3 c5 15.Kb2 a5 16.Kb1 a4 17.Kb2 a3+ 18.Kxa3 Kc3 19.Ka2 c4 20.Kb1 Kd2
-+ (-7.02) Depth: 29/54 00:01:15 281mN, tb=2239
1.c6
-+ (-4.06 !) Depth: 29/57 00:02:34 590mN, tb=2773
1.c6
-+ (-3.71 !) Depth: 29/57 00:02:43 627mN, tb=2974
1.c6
-+ (-2.77 !) Depth: 29/57 00:02:46 637mN, tb=2987
1.c6
= (-0.03 !) Depth: 29/57 00:03:03 697mN, tb=3058
1.c6 dxc6 2.Be5 a2 3.Ba1 Kxa1 4.Kc2 a5 5.h3 h6 6.Kc1 h5 7.h4 c5 8.Kc2 c4 9.Kc1 c3 10.Kc2 c5 11.Kc1 c2 12.Kxc2 a4 13.Kc1 c4 14.Kc2 a3 15.Kc1 c3 16.Kc2
= (0.00) Depth: 29/57 00:03:17 747mN, tb=3086
1.c6 dxc6 2.Be5 a2 3.Ba1 Kxa1 4.Kc2 a5 5.h3 h6 6.Kc1 h5 7.h4 c5 8.Kc2 c4 9.Kc1 c3 10.Kc2 c5 11.Kc1 c2 12.Kxc2 a4 13.Kc1 c4 14.Kc2 a3 15.Kc1 c3 16.Kc2
= (0.00) Depth: 29/57 00:03:30 789mN, tb=3118
1.c6 dxc6 2.Be5 a2 3.Ba1 Kxa1 4.Kc2 a5 5.h3 h6 6.Kc1 h5 7.h4 c5 8.Kc2 c4 9.Kc1 c3 10.Kc2 c5 11.Kc1 c2 12.Kxc2 a4 13.Kc1 c4 14.Kc2 a3 15.Kc1 c3 16.Kc2
= (0.00) Depth: 29/57 00:03:46 843mN, tb=3173
1.c6 dxc6 2.Be5 a2 3.Ba1 Kxa1 4.Kc2 a5 5.h3 h6 6.Kc1 h5 7.h4 c5 8.Kc2 c4 9.Kc1 c3 10.Kc2 c5 11.Kc1 c2 12.Kxc2 a4 13.Kc1 c4 14.Kc2 a3 15.Kc1 c3 16.Kc2
= (0.00) Depth: 29/57 00:04:13 935mN, tb=3289

[lech]

With the removal of the wrong WAC.230 ...b2 position and the addition of this, as i understand computers are only be capable of solving the first 2(the easy ones) of the 7. That's unfortunate and i hope to see someday a program to solve 2-3 of these 5 tough ones....

My testing Sting SF 3 (bred on Stockfish 2.1.1 - modified null move search) gets it in 50 minutes (2 threads 1056M hash).
[d]1b6/3kB3/8/1KpPp2p/4Pp1P/5P2/8/8 w - - 0 1 bm Bf8;

Code: Select all

info depth 55 seldepth 63 multipv 1 score cp 177 nodes 1295351015 nps 1205485 ti
me 1074547 pv e7c5 b8c7 c5b6 c7b8 b5c4 b8d6 b6f2 d6c7 c4b4 d7d6 f2c5 d6d7 c5g1 c
7d6 b4c4 d6e7 g1f2 e7d6 f2c5 d6c7 d5d6 c7a5 c5b4 a5b6 c4d5 b6d4 b4d2 d4b2 d5c5 b
2d4 c5d5
info depth 56
info currmove e7c5 currmovenumber 1
info nodes 1693686623 nps 1191569 time 1421391
info currmove b5c5 currmovenumber 2
info currmove d5d6 currmovenumber 3
info nodes 1751065480 nps 1192722 time 1468125
info currmove b5c4 currmovenumber 4
info currmove e7d8 currmovenumber 5
info currmove e7g5 currmovenumber 6
info nodes 1752196588 nps 1192578 time 1469250
info currmove e7f6 currmovenumber 7
info nodes 1764634282 nps 1191854 time 1480578
info currmove b5b6 currmovenumber 8
info currmove e7d6 currmovenumber 9
info currmove e7f8 currmovenumber 10
info depth 56 seldepth 0 multipv 1 score cp 185 lowerbound nodes 3627347184 nps
1189172 time 3050313 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8h4 g7e5 h4f2 c5c4 h5h4 c4d3 f
2e3 d3e2 h4h3 e2f1 d7d8 e5d6
info nodes 3627347184 nps 1189172 time 3050313
info currmove e7f8 currmovenumber 1

Stockfish gets this in less than 4 minutes (8 threads, 8GB hash):

Code: Select all

info depth 52 seldepth 80 score cp 161 lowerbound nodes 1568059696 nps 7002392 time 223932 multipv 1 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8h4 g7e5 h4g3 e5d4 h5h4 e4e5 h4h3 e5e6 d7e7 c5c6 h3h2 d4c5 e7f6 e6e7 h2h1q e7e8q h1f3 e8e6 f6g7 c5d4 g7f8 e6c8 f8f7 c8f5 f7e7 d4c5

I modified code and now Sting gets this in 3 minutes on my slow dualcore!

Sting-sf-3, 2 threads, 1056M hash
[d]1b6/3kB3/8/1KpPp2p/4Pp1P/5P2/8/8 w - - 0 1 bm Bf8;

Code: Select all

info depth 45 seldepth 50 multipv 1 score cp 193 nodes 86482011 nps 1043133 time
 82906 pv e7c5 b8c7 c5b6 c7d6 b6a5 d6e7 a5e1 e7d6 e1c3 d6c7 b5c5 c7d6 c5c4 d6b8
c3b2 b8c7 c4c5 c7d6 c5b5 d6b8 b5a6 b8c7 a6b7 c7d6 b2a1 d6c7 a1c3 c7d6 c3b2 d6c7
b2c3
info depth 46
info currmove e7c5 currmovenumber 1
info nodes 134053116 nps 1028211 time 130375
info currmove b5c5 currmovenumber 2
info currmove e7g5 currmovenumber 3
info currmove b5c4 currmovenumber 4
info currmove b5b6 currmovenumber 5
info currmove e7f6 currmovenumber 6
info currmove d5d6 currmovenumber 7
info nodes 136853365 nps 1031151 time 132719
info currmove e7d8 currmovenumber 8
info currmove e7d6 currmovenumber 9
info currmove e7f8 currmovenumber 10
info depth 46 seldepth 0 multipv 1 score cp 202 lowerbound nodes 192485195 nps 1
010007 time 190578 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8e7 d5d6 e7d6 c5d5 d6e7 g7e5 e7d
8 e5f4 d8h4 f4d6 h4e7 d6e5 e7h4 e5b8 h4e7 b8f4 h5h4 f4h2 h4h3 e4e5 e7d8 f3f4
info nodes 192485195 nps 1010007 time 190578
info currmove e7f8 currmovenumber 1
info depth 46 seldepth 0 multipv 1 score cp 210 lowerbound nodes 206897608 nps 1
009177 time 205016 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8e7 d5d6 e7d6 c5d5 d6e7 g7e5 e7d
8 e5f4 d8h4 e4e5 h4f2 f4h2 h5h4 d5e4 f2g3 h2g3 h4g3 e4e3 d7e6 f3f4 e6d5 e3f3
info nodes 206897608 nps 1009177 time 205016
info currmove e7f8 currmovenumber 1
info depth 46 seldepth 0 multipv 1 score cp 222 lowerbound nodes 216350104 nps 1
008479 time 214531 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8h4 g7e5 h4f2 c5c4 h5h4 c4d3 f2e
3 d3e2 d7c8 e5d6 c8d7 e4e5 e3d4 e2f1 d4c3 f1g2 c3e1 g2h3
info nodes 216350104 nps 1008479 time 214531
info currmove e7f8 currmovenumber 1
info depth 46 seldepth 0 multipv 1 score cp 240 lowerbound nodes 230698676 nps 1
008038 time 228859 pv e7f8 b8c7 b5c5 c7d8 f8g7 d8e7 d5d6 e7d6 c5d5 d6e7 g7e5 e7h
4 e5f4 h4f2 d5e5 h5h4 e5f5 h4h3 f5g4 f2e3 f4h2 d7e6 g4h3
info nodes 230698676 nps 1008038 time 228859

[peter]

Hi again, George!

May I at first ask you once again for the origin of the study?
You see, I just asked you twice already.


Here's my so far fastest way to get a more or less automatic solution by an engine with good support of GUI and human input.
The feature of "deep position analysis" of Fritz GUI brings up 1.Rxh1 with 3 variants in setup of parameters by itself too, but it takes lot of time to get it into .pgn- output and much more time to get an eval of the two main alternative variants 1.Rc4 and 1.Rxb4 not much lower than that of the best move.

The evals of these stay less bad for White for a very long time given the starting position only, as well with Fritz GUI's position analysis as with chessbase 13's "deep analysis" or "cloud analysis" with two engines. (This night e.g. cloud analysis with Stockfish and Houdini failed to bring up 1.Rxh1 as best move once again.
Maybe someone being well skilled with IdEA could try to get the solution this way, I'm not an expert of this tool.

Here's Fritz13's output of deep position analysis with H3 on 12 cores with 8G hash after about 2 hours, with given 1.Rxh1 as a single half move, just to bring it quickly into comparison to the evals of the alternative variants Houdini computes by itself from the very start.




[Event "?"]
[Site "?"]
[Date "2013.03.21"]
[Round "?"]
[White "Neue Partie"]
[Black "?"]
[Result "?"]
[SetUp "1"]
[FEN "1k6/3p4/1B6/4Pp1p/1p5R/1p4p1/pP3n2/K6n w - -"]

1. Rxh1 (1. Rxb4 Ka8 2. Rc4 g2 3. Rc8+ Kb7 4. Rg8 Ng4
5. Bd4 f4 6. e6 dxe6 7. Rg6 e5 8. Rb6+ Kc7 $19 {+15.41/25})
(1. Rc4 g2 2. Rc1 Nh3 3. e6 dxe6 4. Rc4 f4 5. Ba7+ Kb7
6. Rxb4+ Kc7 7. Rb5 Ng3 8. Rc5+ Kd6 $19 {+10.69/25})
1... Nxh1 2. e6 dxe6 3. Bc7+ Kb7 4. Bxg3 f4 5. Bf2 e5
6. Bh4 Kc6 7. Bg3 Kc5 8. Bf2+ Kb5 $19 {+10.05/39}

Time per move was 135 sec., there were 3 variants to be computed at first half move for white side.

Of course one can say that's very much help of the user, just to show the best move to the GUI and to the engine.
That's right, but I don't talk about some kind of positions' test to get rating lists or Elo or something like this, of which we have blind alleys enough for testing and developing engines already.


As for my personal POV the task of further development of engines isn't Elo anymore, it's the numeric eval- output.
If such isn't to be got counting for sure as for absolute numeric height, it should at least show the right relationship, the right gradient as for comparison to other moves in analysis and in backward analysis over the variants.

Good GUI- features and good human input for analysis of complicated positions (are there any others to be analysed? ) are to be used and tested together with engines maybe, to get more measurable results for further development of engines and GUIs nowadays.

[Aser Huerga]

The first position (c6!) is not easy for most engines: try Houdini, Critter, Hiarcs, Komodo ...

Well Houdini is by far not as fast as Stockfish, but does get it in 2.6 minutes on my old DualCore

Try Houdini 3 in 1 core (to avoid smp randomness), it will take a lot of time.