Smartpruning at 1.0 will abort early without changing result. Default at 1.33 has a pretty low chance of changing move, and saves up enough time to be worth it in games. If you want it to spend maximum time it can set to 0.0. For testing tactics suites I would set it to probably 0.9 or so just to be 100% sure it will solve the correct number, while also getting a nice time savings.MMarc wrote: ↑Wed May 06, 2020 10:03 pmIf I understand it ok, disabling SmartPruningFactor will let continue the search for positions in which the engine wouldn't have changed its best move anyway, so that it will yield the same test result. To hope for a better result, one would have to add variability to the search. There used to be a Dirichlet noise parameter in older Lc0 binaries, but it seems to be gone in the v.25.1.cc2150dx wrote: ↑Wed May 06, 2020 8:05 pm Thanks for the info
However I found this: SmartPruningFactor - Do not spend time on the moves which cannot become bestmove given the remaining time to search.
When no other move can overtake the current best, the search stops, saving the time. Values greater than
1 stop less promising moves from being considered even earlier. Values less than 1 causes hopeless moves
to still have some attention. When set to 0, smart pruning is deactivated.
Tough Tactical Test 1
Moderators: hgm, Rebel, chrisw
-
- Posts: 99
- Joined: Sat Mar 10, 2018 6:16 am
Re: Tough Tactical Test 1
-
- Posts: 12540
- Joined: Wed Mar 08, 2006 8:57 pm
- Location: Redmond, WA USA
Re: Tough Tactical Test 1
Houdini Tactical
Code: Select all
Analysis of C:\Users\dannc\Downloads\ttt1.epd
Analyzing engine: Houdini Tactical
5/11/2020 6:00:27 PM Level: 60 Seconds
Source : List of EPD/PGN files
C:\Users\dannc\Downloads\ttt1.epd
Colours : White moves=True, Black moves=True
Direction : forward
Games (PGN) : From 1 to 999999
Moves : From 1 to 250
Positions (EPD) : All positions
Use Engine(s) : List
Houdini Tactical
Use seconds per move : 60 Seconds per move
Analysis Lines : Minimum search depth=2
Activate abort analysis : False
1) .. g5-g4 Re2xg2
2) Re1-e4 Nf4xd5
3) Ne1-d3, Ne1-f3 Ne1-d3 * 1 Second
4) Rd1-b1 g5-g6
5) Bg7-e5 Bg7-e5 * 24 Seconds
6) .. Bf8-e7 Kc6-b7
7) g5-g6 g5-g6 * 2 Seconds
8) Bd3xe4 Bd3xe4 * 19 Seconds
9) Nf3xd4 Qc3xa3
10) Bf3xh5 Bf3xh5 * 4 Seconds
11) .. Qc5-d4 Ne2-f4
12) .. Rb2-b3 Rb2-b3 * 4 Seconds
13) Nh4-f3 Nh4-f3 * 8 Seconds
14) Rh3-h5 Rh3-h5 * 0 Seconds
15) Nd4-f5 Nd4-f5 * 23 Seconds
16) Bc3xe5 Ba4-b3
17) f2-f4 f2-f4 * 7 Seconds
18) Bc2xh7 Nf3-g5
19) .. Nf6-h5 Rb3xc3
20) d5-d6 d5-d6 * 3 Seconds
21) Bg2-h3 g3-g4
22) Rc4xc5 Rc4xc5 * 6 Seconds
23) .. Nh4xg2 Nh4xg2 * 15 Seconds
24) f5-f6 f5-f6 * 41 Seconds
25) .. h7-h5 h7-h5 * 40 Seconds
26) e5xf6 e5xf6 * 11 Seconds
27) Rf1-f3 Be5xg7
28) Rf1-f5 Rf1-f5 * 15 Seconds
29) Nf3-h2 Nf3-h2 * 28 Seconds
30) Rc6-d6 Rc6-d6 * 37 Seconds
31) Ng3-f5 Ng3-f5 * 44 Seconds
32) .. c7-c6 c7-c6 * 7 Seconds
33) .. Qf6-g5 Bd7xb5
34) Rd2-d3 Rd2-d3 * 20 Seconds
35) .. Qc7-c8 c5-c4
36) .. g4-g3 g4-g3 * 50 Seconds
37) Ng5-e6 Ng5-e6 * 4 Seconds
38) Ne5-g6 Qd4-b2
39) Rf1-f3 f5xg6
40) .. Rd8-d4 Rd8-d4 * 2 Seconds
41) c4xd5 c4xd5 * 55 Seconds
42) .. Rh8xh2 Rh8xh2 * 2 Seconds
43) Qd2-f4 Qd2-f4 * 13 Seconds
44) Qf1-g1 Qf1-g1 * 13 Seconds
45) Rb1-e1 Rb1-e1 * 8 Seconds
46) .. Qh6-h3 Qh6-h3 * 25 Seconds
47) Qe4-e5, Qe4-c6, Qe4-b7, Kg2-g1 f6-f7
48) .. e4-e3 Re7xe6
49) Rb6xc6 a4-a5
50) .. Ng6-h4 Ng6-h4 * 13 Seconds
51) Be2-c4 Be2-c4 * 33 Seconds
52) .. Rg7-h7 Rg7-h7 * 7 Seconds
53) Qf2xd2 Qf2xd2 * 25 Seconds
54) .. Ra7-g7 Ra7-g7 * 19 Seconds
55) .. Rc3xc7 Rc3xc7 * 4 Seconds
56) .. Kf6xg6 Kf6-e5
57) .. Ra4xe4 Ra4xe4 * 3 Seconds
58) .. Be5xd4 Be5xd4 * 12 Seconds
59) Rb2-g2 Rb2-g2 * 40 Seconds
60) .. Bb5-c6 Bb5-c6 * 6 Seconds
61) .. Nc1xe2 Nc1xb3
62) .. Ra2xe2 Ra2xe2 * 43 Seconds
63) .. Ng3-e2 Ng3-e2 * 12 Seconds
64) Qa7xd7 Qa7xa6
65) Qb2xb5 Qb2xb5 * 40 Seconds
66) Rf5xf7 h2xg3
67) .. Qe7-e4 Qe7-e4 * 9 Seconds
68) .. Bg6-e4 d3-d2
69) Ra8-a7 Ra8-d8
70) Bh6-g7 Bh6-g7 * 19 Seconds
71) Bb2-c1 f6xg7
72) .. Ba8xe4 Ng7xh5
73) .. f4-f3 Qe6xd5
74) .. Rd7xd6 Rd7xd6 * 42 Seconds
75) Kc4xd4 Nc8-b6
76) .. Bc8xh3 Ne7-g6
77) Bd3xh7 Qd1-h5
78) Kg1-h2 Kg1-h2 * 5 Seconds
79) Nb3-c5 Nb3-c5 * 6 Seconds
80) .. Bc8xh3 Bc8xh3 * 23 Seconds
81) Bg6-h7 Bg6-h7 * 37 Seconds
82) Bd3xh7 Bd3xh7 * 1 Second
83) Rh1xh7 Rh1xh7 * 1 Second
84) f4-f5 f4-f5 * 4 Seconds
85) .. a4-a3 Bd8-a5
86) .. Bd6xh2 Bd6xh2 * 3 Seconds
87) Rg1-g4 g5-g6
88) .. Be7-d6 Be7-d6 * 4 Seconds
89) .. Nc6xd4 Nc6xd4 * 2 Seconds
90) e5-e6 e5-e6 * 5 Seconds
91) Bc2xg6 Bc2xg6 * 22 Seconds
92) .. h5xg4 Nf6xg4
93) .. Ng3-f1 Ng3-f1 * 7 Seconds
94) .. Ng5xf3 Ra8-e8
95) Qf3xf8 Qf3xc6
96) Qe2-h5 Qe2-h5 * 12 Seconds
97) .. Ra8-g8 Ra8-g8 * 29 Seconds
98) .. Nd6xe4 Nd6xe4 * 33 Seconds
99) Bf4xh6 Bf4xh6 * 7 Seconds
100) Bf7xg6 Bf7xg6 * 27 Seconds
65 of 100 matching moves
5/11/2020 7:41:59 PM, Total time: 1:41:32 AM Rated time: 53:06 = 3186 Seconds
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
-
- Posts: 1167
- Joined: Thu Dec 25, 2008 9:07 pm
- Full name: Herbert L
Re: Tough Tactical Test 1
So Houdidit-6.03-Tactical has more solutions and is faster.
Houdidit6.03:
70 of 100 matching moves
Rated time: 43:21 = 2601 Seconds
Houdini (6.02?)
65 of 100 matching moves
Rated time: 53:06 = 3186 Seconds
Houdidit6.03:
70 of 100 matching moves
Rated time: 43:21 = 2601 Seconds
Houdini (6.02?)
65 of 100 matching moves
Rated time: 53:06 = 3186 Seconds
-
- Posts: 5228
- Joined: Thu Mar 09, 2006 9:40 am
- Full name: Vincent Lejeune
-
- Posts: 6442
- Joined: Tue Jan 09, 2007 12:31 am
- Location: PA USA
- Full name: Louis Zulli
Re: Tough Tactical Test 1
-
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: Tough Tactical Test 1
and better node counting ...
-
- Posts: 1167
- Joined: Thu Dec 25, 2008 9:07 pm
- Full name: Herbert L
Re: Tough Tactical Test 1
or the position of the sun ...
-
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: Tough Tactical Test 1
[d]3rr1bk/6np/2p1n3/1pPpNpPp/pP1QpP1P/P2PP3/2K5/B7 w - - 0 1
BD 13.1
a lot better than before ....
it got to zero on the first run ..fairly quickly
BD 13.1
a lot better than before ....
it got to zero on the first run ..fairly quickly
Code: Select all
34 -1.30! 5.78G 1:28.60 Ng6+!
34 -1.34! 2.77G 0:45.06 Ng6+!
34 -1.37! 2.69G 0:43.79 Ng6+!
33 -1.40! 2.61G 0:42.73 Ng6+!
33 -1.42? 2.58G 0:42.10 Ng6+ hxg6?
32 -1.40! 2.52G 0:41.26 Ng6+!
32 -1.42? 2.51G 0:41.07 Ng6+ hxg6?
31 -1.40 2.50G 0:40.82 Ng6+ hxg6 Qxg7+ Nxg7 Bxg7+ Kh7 d4 Rc8 Kb1 Be6 Kc2 Ra8 Kd1 Ra7 Kc2 Bc8 Kd1 Kg8 Kd2 Ra8 Ke1 Kf7 Be5 Ke6 Ke2 Ra7 Bb8 Ra6 Be5 Ra8 Kd1 Ra7 Kd2 Ba6 Ke2 Rh7 Bd6 Rf7 Ke1 Kd7 Kd1 Kc8 Ke2 Bb7 Kd2 Ba8 Bf8 Ra7 Kd1
31 -1.46! 2.29G 0:37.63 Ng6+!
31 -2.07! 2.11G 0:35.04 Ng6+!
31 -2.54! 1.92G 0:32.20 Ng6+!
31 -2.91! 1.79G 0:30.45 Ng6+!
31 -3.20! 1.68G 0:28.88 Ng6+!
31 -3.41! 1.52G 0:26.65 Ng6+!
31 -3.57! 1.37G 0:24.49 Ng6+!
31 -3.69! 1.20G 0:22.07 Ng6+!
31 -3.83! 736.1M 0:14.56 Qb2!
31 -3.85! 714.0M 0:14.11 Qb2!
30 -3.86 705.2M 0:13.93 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rc8 Ba1 Kf7 Kd2 Ne8 Nxf5 Rxc6 Bh8 Nd6 Nd4 Rc7 Be5 Rd7 Ke3 Kg6 Bf6 Nf5+ Ke4 Nxd4 Bxd4 Rxd4+ Kxd4 Kf5 Ke3 Kg4 Ke4 Kxh4 Kf3 Bd5+ Ke3 Kg4 f5
30 -3.78! 605.8M 0:11.90 Qb2!
30 -3.82! 605.6M 0:11.89 Qb2!
29 -3.73 589.9M 0:11.58 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rc8 Bc3 Kf7 Kb2 Bd3 Kc1 Ne8 Kd2 Bc4 Nxf5 Rxc6 Be5 Rc8 Ke3 Rd8 Bd4 Ke6 Ng3 Nd6 Bb6 Rd7 f5+ Nxf5+ Kf4 Rd3 Nxh5
29 -3.71? 529.0M 0:10.34 Qb2 d4?
29 -3.67? 475.0M 0:09.24 Qb2 d4?
29 -3.64? 473.1M 0:09.21 Qb2 d4?
28 -3.60 467.1M 0:09.09 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kd1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rc8 Kd2 Bd5 Nxb5 Bxc6 Nd4 Be4 Ke1 Ne8 Ne6 Nd6 Nc5 Nc4 Be5 Nxa3 Kf2 Bd5 Bf6 Nb5 Nxa4 Rc2+ Ke3 Rg2 Nb6 Be4 g6 hxg6 Nd5 Rg3+ Ke2
28 -3.66! 465.8M 0:09.06 Qb2!
28 -3.69? 462.5M 0:08.99 Qb2 d4?
28 -3.65? 345.6M 0:06.74 Qb2 d4?
28 -3.62? 242.8M 0:04.96 Qb2 d4?
27 -3.45! 234.6M 0:04.80 Qb2!
27 -3.53! 234.4M 0:04.79 Qb2!
27 -3.56? 218.3M 0:04.48 Qb2 d4?
27 -3.52? 170.4M 0:03.49 Qb2 d4?
27 -3.50? 153.5M 0:03.15 Qb2 d4?
26 -3.43 120.8M 0:02.49 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rc8 Bc3 Bd3 Kd2 Be4 Nxb5 Bxc6 Nd4 Be4 b5 Ne8 Ne6 Nd6 b6 Nc4+ Ke1 Nxa3 Nc7 Nc4
25 -3.37 85.6M 0:01.76 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rd8 Bc3 Rc8 Kd2 Bd5 Nxb5 Bxc6 Nd4 Be4 Nb5 Ne8 Be5 Rc2+ Ke1 Bd3 g6 Bxb5 gxh7+ Kxh7
24 -3.34 66.1M 0:01.36 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 Bc3 Re4 c6 Ne8 Nxf5 Rxf4 Ng3 Rf3 Ne4 Kf7 Kd2 Bd5 c7
23 -3.21 58.3M 0:01.21 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rc8 Bc3 Kf7 Kd2 Ne8 Nxf5 Rxc6 Be5 Rc8 Bd4 Ke6 Ne3 Nd6 f5+ Nxf5 Nxf5 Kxf5 Bc3
22 -3.03 42.4M 0:00.87 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Nxd4 Rxb2 Bxb2 Kg8 c6 Rc8 Bc3 Kf7 Kd2 Ne8 Nxf5 Rxc6 Bd4 Re6 g6+ hxg6 Ng3 Bb3 Be5 Kf8
21 -3.02 33.7M 0:00.69 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kd1 Rxb2 Bxb2 Rc8 Nxd4 Kg8 c6 Kf7 Kd2 Kf8 Ke3 Bd5 g6 hxg6 Nxb5 Bxc6 Nd4 Re8+ Kd3 Rd8
20 -3.15 29.5M 0:00.61 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kd1 Rxb2 Bxb2 Rc8 Nxd4 Kg8 Kc1 Re8 Kd2 Kf8 c6 Kg8 Nf3 Re2+ Kc1 Ne6 Nd4 Re1+ Kc2 Nxd4+ Bxd4 Bb3+ Kc3 Rc1+ Kd3
19 -3.07 25.8M 0:00.54 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Qxe2 Rxe2 Bxd4 Kg8 Be5 Ne6 Ne7+ Kf8 Nxf5 Bd3 Nh6 Nd8 Bd6+ Ke8 f5 Kd7 Ng8 Bxf5 Kd1 Ra2 Nf6+ Kc6
18 -3.01 18.0M 0:00.38 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Qxe2 Rxe2 Bxd4 Kg8 Be5 Ne6 Ne7+ Kf8 Nxf5 Bd3 Nh6 Bg6 Bd6+ Ke8 f5 Nd4 fxg6
17 -3.06 9.91M 0:00.21 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rde8 Qxe2 Rxe2 Bxd4 Kg8 Bxg7 Kxg7 Nd4 Re4 Nxf5+ Kg6 Nd6 Rxf4 c6 Be6
16 -2.96 5.89M 0:00.12 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kc2 Re2+ Kd1 Rxb2 Bxb2 Rc8 Nxd4 Kg8 Kd2 Re8 Kc2 Rd8 Kc1 Ne8 Ba1 Nc7 Bc3 Kf8 c6 Rc8 g6 hxg6
15 -2.95 4.63M 0:00.10 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kd1 Rxb2 Bxb2 Rc8 Nxd4 Kg8 Kc1 Ra8 Bc3 Re8 g6 hxg6 Kb2 Rd8 Nc2
14 -2.79 4.59M 0:00.10 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kd1 Rxb2 Bxb2 Rc8 Nxd4 Kg8 Kc1 Re8 Kd1 Kf8 Kd2 Rd8 Kc1 Ne8 Nc6 Rd3 Ne5 Rh3 g6 hxg6 c6
13 -2.54 1.71M 0:00.04 Qb2 d4 Nxc6 exd3+ Kd1 dxe3 Nxd8 Nd4 Qxd4 Bb3+ Ke1 d2+ Kf1 d1=Q+ Qxd1 Bxd1 Nf7+ Kg8 Nh6+ Kf8 Bxg7+ Kxg7 Nxf5+ Kf8 Nd6 Rd8 f5 Bg4 f6 Rd7 Ke1
12 -2.57 1.29M 0:00.03 Qb2 d4 Nxc6 exd3+ Kd1 dxe3 Nxd8 Nd4 Qxd4 Bb3+ Ke1 d2+ Kf1 d1=Q+ Qxd1 Bxd1 Nf7+ Kg8 Nh6+ Kf8 Bxg7+ Kxg7 Nxf5+ Kf8 Nd6 Re7 Nxb5 Bb3
11 -2.46 939673 0:00.02 Qb2 d4 Nxc6 exd3+ Kxd3 Nxf4+ exf4 Bc4+ Kd2 Re2+ Kc1 Rxb2 Kxb2 Ra8 Kc1 Re8 Bxd4
10 -2.23 754209 0:00.02 Qb2 d4 Nxc6 exd3+ Kd1 dxe3 Nxd8 e2+ Ke1 Rxd8 Kd2 Nd4 Qxd4 Rxd4 Bxd4
9 -2.23 392509 0:00.01 Qb2 d4 Nxc6 exd3+ Kd1 Nxf4 Nxd8 Bb3+ Ke1 Rxe3+ Kf1 Rf3+ Kg1 Ne2+ Kg2
8 -2.25 173995 0:00.00 Qc3 d4 Qd2 Rd5 dxe4 dxe3 Qxe3 fxe4
7 -2.27 91730 0:00.00 Qb2 d4 Nxc6 Rd5 Nxd4 Rxd4 exd4
6 -1.64 53347 0:00.00 Qb2 d4 Nxc6 Rd5 Kd1 dxe3
5 -1.82 43074 0:00.00 Qb2 d4 Nxc6 dxe3 Nxd8 exd3+ Kb1 Rxd8
4 -2.02 23258 0:00.00 Qb2 d4 Nxc6 Rd7
3 -2.02 17934 0:00.00 Qb2 d4 Kc1
2 -1.62 9045 0:00.00 Qc3 d4
1 NNUE evaluation using C:\cluster.mfb\Popcnt-LP\eval\nn-811243cea2cf.nnue enabled
1 Found 510 tablebases
1 -1.44 6650 0:00.00 Qc3
-
- Posts: 12540
- Joined: Wed Mar 08, 2006 8:57 pm
- Location: Redmond, WA USA
Re: Tough Tactical Test 1
I really appreciate your efforts, Mike.
When I am validating a test set, some Mike variants are always on the prowl. And I would be a flaming idiot if it were not so.
When I am validating a test set, some Mike variants are always on the prowl. And I would be a flaming idiot if it were not so.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
-
- Posts: 416
- Joined: Thu Aug 02, 2012 7:48 pm
- Location: Germany
Re: Tough Tactical Test 1
i did a quick test around TTT1 with current SF.dev (Stockfish_22060708_x64_avx2) and it fared pretty well imo:
it solved 88 out of 100 with these conditions:
Ryzen 3600 [6 cores, 12 threads]; 1024 mb hash; no Tb's; 15 sec per position
...these are the 12 positions that were not solved within 15 sec,
a 10 min search with 8192 mb hash ended this way (hashes were always cleared, solving-time in minutes added in the end, sorted by time-consumption):
3q1rnk/3b1p1p/2p2p2/r2Np2Q/1p2P3/pB1PPRN1/2P3PP/6K1 w - - 0 1 bm Nf5 0:24
2b2rk1/3nbppp/1q2p3/n2pP3/p1pP1PN1/prP2N1P/2BB2PK/1R2QR2 w - - 0 1 bm Bxh7+ 0:42
r2q1rk1/2n1bpp1/b2p4/3Pp1PQ/1p6/p3B2P/PPPN1P2/2KR2R1 w - - 0 1 bm Rg4 0:46
b4rk1/6np/3p1pp1/1r2p1PP/1p2P3/1PqN1P2/2P3RQ/1KBR4 b - - 0 1 bm Bxe4 1:23
r4qrk/P1N5/1pPp3b/1B2p3/R3Pn1p/5Pn1/4PBPK/Q3N1R1 b - - 0 1 bm Nf1+ 1:31
1k6/6Bq/b6p/2p2p2/1pPp3r/pP1Pb3/P1K3Q1/6Rr w - - 0 1 bm Be5+ 1:32
k4r1r/Pp6/bP1bq1p1/3N2p1/R1pPppB1/4P1PP/5PK1/3QBR2 b - - 0 1 bm ...f3+ 3:13
1b2r1r1/5p1k/3pb1pB/2pnP1Q1/q2P1N2/p1P5/1p3PP1/1R2R1K1 w - - 0 1 bm Re4 5:06
1qrr1b2/6p1/ppkn1P2/3pP3/1PP5/2BP1pP1/BQ3P2/1RRN1n1K b - - 0 1 bm .. Bf8-e7 5:38
3rr1bk/6np/2p1n3/1pPpNpPp/pP1QpP1P/P2PP3/2K5/B7 w - - 0 1 bm Ng6+ not solved within 10 min... typical stupidity and probably much dependant on the NN getting used !?
...while examining i excluded 2 positions from getting judged (for myself only) because i consider them as "not a good testposition"...theses are:
r4r1k/1p4pp/1p1p3q/p2Np1n1/P3Ppb1/2PP1N2/1P1RKPPP/R2Q4 b - - 0 1 bm Nxf3 ...[Bxf3 does transpose !?}
5rqk/4R2p/5Pp1/1p4P1/4Q3/2p3P1/6K1/8 w - - 0 1 bm Qe5 [Kg1 and other moves are completely winning as well]
if im wromg about this plz dont shy to mention
regards
it solved 88 out of 100 with these conditions:
Ryzen 3600 [6 cores, 12 threads]; 1024 mb hash; no Tb's; 15 sec per position
...these are the 12 positions that were not solved within 15 sec,
a 10 min search with 8192 mb hash ended this way (hashes were always cleared, solving-time in minutes added in the end, sorted by time-consumption):
3q1rnk/3b1p1p/2p2p2/r2Np2Q/1p2P3/pB1PPRN1/2P3PP/6K1 w - - 0 1 bm Nf5 0:24
2b2rk1/3nbppp/1q2p3/n2pP3/p1pP1PN1/prP2N1P/2BB2PK/1R2QR2 w - - 0 1 bm Bxh7+ 0:42
r2q1rk1/2n1bpp1/b2p4/3Pp1PQ/1p6/p3B2P/PPPN1P2/2KR2R1 w - - 0 1 bm Rg4 0:46
b4rk1/6np/3p1pp1/1r2p1PP/1p2P3/1PqN1P2/2P3RQ/1KBR4 b - - 0 1 bm Bxe4 1:23
r4qrk/P1N5/1pPp3b/1B2p3/R3Pn1p/5Pn1/4PBPK/Q3N1R1 b - - 0 1 bm Nf1+ 1:31
1k6/6Bq/b6p/2p2p2/1pPp3r/pP1Pb3/P1K3Q1/6Rr w - - 0 1 bm Be5+ 1:32
k4r1r/Pp6/bP1bq1p1/3N2p1/R1pPppB1/4P1PP/5PK1/3QBR2 b - - 0 1 bm ...f3+ 3:13
1b2r1r1/5p1k/3pb1pB/2pnP1Q1/q2P1N2/p1P5/1p3PP1/1R2R1K1 w - - 0 1 bm Re4 5:06
1qrr1b2/6p1/ppkn1P2/3pP3/1PP5/2BP1pP1/BQ3P2/1RRN1n1K b - - 0 1 bm .. Bf8-e7 5:38
3rr1bk/6np/2p1n3/1pPpNpPp/pP1QpP1P/P2PP3/2K5/B7 w - - 0 1 bm Ng6+ not solved within 10 min... typical stupidity and probably much dependant on the NN getting used !?
...while examining i excluded 2 positions from getting judged (for myself only) because i consider them as "not a good testposition"...theses are:
r4r1k/1p4pp/1p1p3q/p2Np1n1/P3Ppb1/2PP1N2/1P1RKPPP/R2Q4 b - - 0 1 bm Nxf3 ...[Bxf3 does transpose !?}
5rqk/4R2p/5Pp1/1p4P1/4Q3/2p3P1/6K1/8 w - - 0 1 bm Qe5 [Kg1 and other moves are completely winning as well]
if im wromg about this plz dont shy to mention
regards
Wahrheiten sind Illusionen von denen wir aber vergessen haben dass sie welche sind.