How powerful is the Queen compared to 2 Rooks ?

[Dann Corbit]

You are correct there are too many board configurations, but in the majority of them you can give white with the Queen to Komodo versus either GM Nakamura or GM MLV or any other top GM with the two Rooks and I will bet money on Komodo.

I will bet on Komodo (on good hardware) against Nakamura or even Carlsen with either configuration.

The best programs on strong hardware are several hundred Elo above the best humans.

Two rooks verses a queen is a very minor difference, in comparison.

[Chessqueen]

You are correct there are too many board configurations, but in the majority of them you can give white with the Queen to Komodo versus either GM Nakamura or GM MLV or any other top GM with the two Rooks and I will bet money on Komodo.

I will bet on Komodo (on good hardware) against Nakamura or even Carlsen with either configuration.

The best programs on strong hardware are several hundred Elo above the best humans.

Two rooks verses a queen is a very minor difference, in comparison.

I was giving my point of view in the statement that Larry made of a Queen versus two Rooks with 5 pawns each, I know that there are millions of configurations but of course that you do NOT want to give a disadvantage position to White with the Queen such as this one, we are talking about positions that are equal from the start.

[D]8/4r1kp/p5p1/1p3p2/1P2rP1P/P6P/3Q2K1/8 w - - 0 1

Stockfish_10_x64:
1/2 00:00 97 32k -1.40 Qd2-d6
2/2 00:00 143 48k -0.90 Qd2-d6 h7-h6 Qd6xa6
3/4 00:00 364 91k -1.53 Qd2-b2+ Kg7-h6 Qb2-c1 Re7-d7
4/5 00:00 799 160k -1.38 Qd2-d6 Re7-e6 Qd6-c5 Re4-e3
5/6 00:00 2k 204k -1.67 Qd2-b2+ Kg7-f7 Qb2-h8 h7-h5 Qh8-h6 Re4-e3
6/10 00:00 5k 259k -1.79 Kg2-f3 Re7-e6 Qd2-c1 Re6-d6 Qc1-b1 Kg7-h6
7/9 00:00 9k 314k -1.69 Qd2-d6 Re4-e2+ Kg2-f1 Re2-e3 Qd6xa6 Re7-e4 Qa6-b7+ Kg7-h6 Qb7xb5
8/11 00:00 12k 340k -2.18 Qd2-d6 Re7-e6 Qd6-c7+ Kg7-h6 Qc7-f7 Re4-e2+ Kg2-f1 Re6-e3 Qf7-f6 Re2-e1+ Kf1-f2
9/14 00:00 22k 309k -2.37 Qd2-d6 Re4-e2+ Kg2-f1 Re2-e6 Qd6-c5 Kg7-h6 Kf1-g2 Kh6-h5 Kg2-f3 Re6-e4
10/13 00:00 29k 331k -2.02 Qd2-d6 Re4-e2+ Kg2-g1 Re2-e1+ Kg1-f2 Re7-e2+ Kf2-f3 Re2-e3+ Kf3-f2 Re1-e2+ Kf2-f1 Kg7-h6 Qd6xa6
11/18 00:00 61k 379k -2.13 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e4 Qf6xa6 Re1-e3+ Kf3-g2 Re4xf4 h4-h5 Kh6xh5
12/22 00:00 91k 391k -1.93 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e4 Qf6xa6 Re1-e3+ Kf3-g2 Re4xf4 h4-h5 Rf4-h4 Qa6xb5 Re3xa3 Qb5-b7 Ra3-a2+ Kg2-f1 Rh4xh3 h5xg6
13/25 00:00 147k 430k -2.26 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Qg5-h4 Re6-e2+ Kg2-f1 Re2-e3 Kf1-f2 Rd3xa3 h5xg6 Re3xh3
14/31 00:00 200k 456k -1.77 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e6 Qf6-g5+ Kh6-g7 h4-h5 Re6-e2+ Kf2-f3 Re2-e3+ Kf3-f2 Re1-e2+ Kf2-f1 Re2-c2 h3-h4 Rc2-c3 h5xg6 h7xg6
15/27 00:00 237k 464k -1.94 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 Qg5-d8 Re1-e2 Qd8-d3 Kg7-h6 Qd3xe2 Re6xe2 Kf3xe2 Kh6-h5 Ke2-f1
16/39 00:01 531k 522k -2.33 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Qg5-h4 Re6-e3 Qh4-g5 Rd3xa3 h5-h6+ Kg7-f7 Qg5-d8 Re3-g3+ Kg2-f2 Ra3-f3+ Kf2-e2 Rg3xh3
17/38 00:01 677k 537k -1.87 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-a2 Qf6-g5+ Kh6-g7 h4-h5 Ra2xa3+ Kf3-f2 Re1-e8 h5xg6 h7xg6 h3-h4 Ra3-a2+ Kf2-f1 Ra2-h2 Qg5-g3 Rh2-h1+ Kf1-g2 Rh1-c1
18/40 00:02 1,196k 544k -1.33 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 h7-h6 Qg5-d8 g6xh5 Qd8-d5 Kg7-g6 Qd5-d8 Re3-e2+ Kg2-f1 Re6-e8 Qd8-b6+ Re2-e6 Qb6-g1+ Kg6-f7
19/38 00:03 1,692k 557k -1.21 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-g1 Kg7-f7 Qg5-h6 Kf7-g8 Qh6-g5 Kg8-g7 h3-h4 Re6-e3 Kg1-f2 Rd3xa3 h5-h6+ Kg7-f7 Qg5-d8
20/34- 00:03 1,781k 557k -1.30 Qd2-b2+ Kg7-h6
20/34- 00:03 1,787k 557k -1.40 Qd2-b2+ Kg7-h6
20/40- 00:03 1,952k 555k -1.75 Qd2-b2+ Kg7-h6
20/43+ 00:03 2,051k 556k -1.58 Qd2-b2+
20/44 00:03 2,208k 557k -1.53 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Re6-e3 Qg2-b7+ Kg7-h6 Qb7xa6 Rh3-f3+ Kf2-g2 Re3xa3 Qa6xb5 Rf3-b3 Qb5-e2 Rb3xb4 Qe2-d1 Ra3-a4 Qd1-e2 Rb4-e4 Qe2-d2
21/46- 00:04 2,422k 559k -1.62 Qd2-b2+ Kg7-h6
21/46 00:04 2,632k 560k -1.64 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-f2 Re3xh3 h5xg6 h7xg6 Qg5-g2 Re6-e3 Qg2-b7+ Kg7-h6 Qb7xa6 Rh3-f3+ Kf2-g2 Re3xa3 Qa6xb5 Rf3-g3+ Kg2-f1 Ra3-f3+ Kf1-e2 Rf3xf4 Qb5-d5 Rf4-e4+ Ke2-f2 Rg3-g4 b4-b5 Re4-b4
22/43- 00:04 2,736k 561k -1.74 Qd2-b2+ Kg7-h6
22/43- 00:04 2,800k 562k -1.83 Qd2-b2+ Kg7-h6
22/57- 00:06 3,665k 528k -2.45 Qd2-b2+ Kg7-h6
22/57 00:07 3,844k 531k -2.43 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-g1 Rd3xh3 Qg5-g2 Rh3xh5 Qg2-b7+ Kg7-h6 Qb7-c8 Re6-e1+ Kg1-f2 Rh5-h1 Qc8xa6 Re1-f1+ Kf2-g3 Rf1-g1+ Kg3-f2 Rg1-g4 Qa6xb5 Rh1-h2+ Kf2-f3 Rh2-h3+ Kf3-e2 Rg4xf4 Qb5-c5 Rf4-e4+ Ke2-f2
23/47- 00:08 4,331k 537k -2.53 Qd2-b2+ Kg7-h6
23/47- 00:08 4,526k 540k -2.62 Qd2-b2+ Kg7-h6
23/47- 00:08 4,774k 541k -2.77 Qd2-b2+ Kg7-h6
23/50+ 00:09 5,138k 544k -2.65 Qd2-b2+
23/50+ 00:09 5,174k 544k -2.37 Qd2-b2+
23/50 00:09 5,269k 544k -2.42 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-f2 Re3xh3 h5xg6 h7xg6 Qg5-d8 Re6-e4 Qd8-a5 Re4xf4+ Kf2-g2 Rh3-h4 Qa5xa6 Rh4-g4+ Kg2-h3 Rf4-f3+ Kh3-h2 Rf3-f2+ Kh2-h1 Rf2-f1+ Kh1-h2 f5-f4 Qa6-b7+ Kg7-h6 Qb7xb5 Rf1-f2+ Kh2-h1 Rg4-g5
24/55+ 00:12 6,656k 542k -2.32 Qd2-b2+
24/55- 00:13 7,195k 534k -2.51 Qd2-b2+ Kg7-h6
24/55- 00:14 7,536k 537k -2.66 Qd2-b2+ Kg7-h6
24/55 00:14 7,834k 539k -2.56 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Qg5-h4 h7-h6 Qh4-f2 Re6-e3 Qf2-b2+ Rd3-c3 h3-h4 Kg7-h7 h5xg6+ Kh7xg6 Qb2-a1 Re3-d3 Kg2-f1 Kg6-h5 a3-a4 Kh5-g4 a4xb5 a6xb5 Kf1-e2 Kg4xf4
25/42- 00:15 8,642k 544k -2.65 Qd2-b2+ Kg7-h6
25/42- 00:16 8,941k 547k -2.75 Qd2-b2+ Kg7-h6
25/42- 00:16 9,343k 550k -2.89 Qd2-b2+ Kg7-h6
25/42 00:18 10,142k 550k -2.82 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-b8 Re6-e8 Qb8-a7 Re3-e2+ Kf2-g1
26/49 00:21 11,748k 546k -2.89 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-b8 Re6-e8 Qb8-a7 Re3-e2+ Kf2-g1 Re8-e6 Qa7-d4+ Kf6-f7 Qd4-d7+ Re6-e7 Qd7-d5+ Kf7-g7 Qd5-d4+ Kg7-h7 Qd4-f6
27/57 00:22 12,448k 550k -2.80 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-c8 Re3-e2+ Kf2-g1 Re2-e1+ Kg1-f2 Kf6-f7 Qc8-b7+ Kf7-g8 Kf2-g3 Re1-e4 Qb7-d7 Kg8-f8 Kg3-h3 Re6-e7 Qd7-c8+ Kf8-g7 Qc8xa6 Re4xf4
28/64+ 00:23 12,763k 551k -2.71 Qd2-b2+
28/64+ 00:23 12,829k 552k -2.61 Qd2-b2+
28/64- 00:24 13,608k 555k -2.90 Qd2-b2+ Kg7-h6
28/64 00:27 15,346k 552k -2.91 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-c8 Re3-e2+ Kf2-g1 Re2-e1+ Kg1-f2 Kf6-f7 Qc8-b7+ Kf7-g8 Kf2-g3 Re1-e4 Qb7-d7 Re6-e7 Qd7-c8+ Kg8-h7 Qc8xa6 Re4xb4 Qa6-a1 Re7-e4 Qa1-h1+ Kh7-g7
29/51 00:30 16,721k 556k -2.91 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-c8 Re3-e2+ Kf2-g1 Re2-e1+ Kg1-f2 Kf6-f7 Qc8-b7+ Kf7-g8 Kf2-g3 Re1-e4 Qb7-d7 Re6-e7 Qd7-c8+ Kg8-h7 Qc8xa6 Re4xb4 Qa6-a1
30/57+ 00:31 17,376k 558k -2.82 Qd2-b2+
30/57+ 00:31 17,552k 557k -2.72 Qd2-b2+
30/57 00:32 17,814k 554k -2.79 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-g8 Qb7-a8+ Kg8-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Re6-e7 Qd7-d5+ Re3-e6 Kf2-g3 Kf7-g7 Kg3-h4 Kg7-h6 Qd5-d8 Re7-e8
31/44+ 00:33 18,273k 551k -2.70 Qd2-b2+
31/49- 00:35 19,512k 552k -2.89 Qd2-b2+ Kg7-h6
31/49+ 00:40 22,386k 554k -2.75 Qd2-b2+
31/50+ 00:43 23,593k 545k -2.54 Qd2-b2+
31/56 00:49 26,920k 544k -2.89 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-g8 Qb7-a8+ Kg8-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Re6-e7 Qd7-d5+ Re3-e6 Kf2-g3 Kf7-g7 Kg3-h4 Kg7-h6 Qd5-d8 Re7-e8 Qd8-g5+ Kh6-g7
32/47+ 00:50 27,375k 542k -2.79 Qd2-b2+
32/47+ 00:50 27,454k 542k -2.70 Qd2-b2+
32/47 00:51 27,752k 543k -2.81 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-g8 Qb7-a8+ Kg8-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Re6-e7 Qd7-d5+ Re3-e6 Kf2-g3 Kf7-g7 Kg3-h4 Kg7-h6 Qd5-d8 Re7-e8 Qd8-g5+ Kh6-h7 Qg5-g1 Kh7-g7 Kh4-g5
33/48+ 00:52 28,317k 544k -2.71 Qd2-b2+
33/51+ 00:52 28,552k 544k -2.62 Qd2-b2+
33/51+ 00:54 30,073k 547k -2.47 Qd2-b2+

Stockfish_10_x64:
1/2 00:00 97 32k -1.40 Qd2-d6
2/2 00:00 143 48k -0.90 Qd2-d6 h7-h6 Qd6xa6
3/4 00:00 364 91k -1.53 Qd2-b2+ Kg7-h6 Qb2-c1 Re7-d7
4/5 00:00 799 160k -1.38 Qd2-d6 Re7-e6 Qd6-c5 Re4-e3
5/6 00:00 2k 234k -1.67 Qd2-b2+ Kg7-f7 Qb2-h8 h7-h5 Qh8-h6 Re4-e3
6/10 00:00 5k 273k -1.79 Kg2-f3 Re7-e6 Qd2-c1 Re6-d6 Qc1-b1 Kg7-h6
7/9 00:00 9k 326k -1.69 Qd2-d6 Re4-e2+ Kg2-f1 Re2-e3 Qd6xa6 Re7-e4 Qa6-b7+ Kg7-h6 Qb7xb5
8/11 00:00 12k 340k -2.18 Qd2-d6 Re7-e6 Qd6-c7+ Kg7-h6 Qc7-f7 Re4-e2+ Kg2-f1 Re6-e3 Qf7-f6 Re2-e1+ Kf1-f2
9/14 00:00 22k 309k -2.37 Qd2-d6 Re4-e2+ Kg2-f1 Re2-e6 Qd6-c5 Kg7-h6 Kf1-g2 Kh6-h5 Kg2-f3 Re6-e4
10/13 00:00 29k 316k -2.02 Qd2-d6 Re4-e2+ Kg2-g1 Re2-e1+ Kg1-f2 Re7-e2+ Kf2-f3 Re2-e3+ Kf3-f2 Re1-e2+ Kf2-f1 Kg7-h6 Qd6xa6
11/18 00:00 61k 381k -2.13 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e4 Qf6xa6 Re1-e3+ Kf3-g2 Re4xf4 h4-h5 Kh6xh5
12/22 00:00 91k 393k -1.93 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e4 Qf6xa6 Re1-e3+ Kf3-g2 Re4xf4 h4-h5 Rf4-h4 Qa6xb5 Re3xa3 Qb5-b7 Ra3-a2+ Kg2-f1 Rh4xh3 h5xg6
13/25 00:00 147k 430k -2.26 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Qg5-h4 Re6-e2+ Kg2-f1 Re2-e3 Kf1-f2 Rd3xa3 h5xg6 Re3xh3
14/31 00:00 200k 456k -1.77 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e6 Qf6-g5+ Kh6-g7 h4-h5 Re6-e2+ Kf2-f3 Re2-e3+ Kf3-f2 Re1-e2+ Kf2-f1 Re2-c2 h3-h4 Rc2-c3 h5xg6 h7xg6
15/27 00:00 237k 467k -1.94 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 Qg5-d8 Re1-e2 Qd8-d3 Kg7-h6 Qd3xe2 Re6xe2 Kf3xe2 Kh6-h5 Ke2-f1
16/39 00:01 531k 522k -2.33 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Qg5-h4 Re6-e3 Qh4-g5 Rd3xa3 h5-h6+ Kg7-f7 Qg5-d8 Re3-g3+ Kg2-f2 Ra3-f3+ Kf2-e2 Rg3xh3
17/38 00:01 677k 535k -1.87 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-a2 Qf6-g5+ Kh6-g7 h4-h5 Ra2xa3+ Kf3-f2 Re1-e8 h5xg6 h7xg6 h3-h4 Ra3-a2+ Kf2-f1 Ra2-h2 Qg5-g3 Rh2-h1+ Kf1-g2 Rh1-c1
18/40 00:02 1,196k 545k -1.33 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 h7-h6 Qg5-d8 g6xh5 Qd8-d5 Kg7-g6 Qd5-d8 Re3-e2+ Kg2-f1 Re6-e8 Qd8-b6+ Re2-e6 Qb6-g1+ Kg6-f7
19/38 00:03 1,692k 559k -1.21 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-g1 Kg7-f7 Qg5-h6 Kf7-g8 Qh6-g5 Kg8-g7 h3-h4 Re6-e3 Kg1-f2 Rd3xa3 h5-h6+ Kg7-f7 Qg5-d8
20/34- 00:03 1,781k 559k -1.30 Qd2-b2+ Kg7-h6
20/34- 00:03 1,787k 559k -1.40 Qd2-b2+ Kg7-h6
20/40- 00:03 1,952k 557k -1.75 Qd2-b2+ Kg7-h6
20/43+ 00:03 2,051k 558k -1.58 Qd2-b2+
20/44 00:03 2,208k 559k -1.53 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Re6-e3 Qg2-b7+ Kg7-h6 Qb7xa6 Rh3-f3+ Kf2-g2 Re3xa3 Qa6xb5 Rf3-b3 Qb5-e2 Rb3xb4 Qe2-d1 Ra3-a4 Qd1-e2 Rb4-e4 Qe2-d2
21/46- 00:04 2,422k 562k -1.62 Qd2-b2+ Kg7-h6
21/46 00:04 2,632k 563k -1.64 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-f2 Re3xh3 h5xg6 h7xg6 Qg5-g2 Re6-e3 Qg2-b7+ Kg7-h6 Qb7xa6 Rh3-f3+ Kf2-g2 Re3xa3 Qa6xb5 Rf3-g3+ Kg2-f1 Ra3-f3+ Kf1-e2 Rf3xf4 Qb5-d5 Rf4-e4+ Ke2-f2 Rg3-g4 b4-b5 Re4-b4
22/43- 00:04 2,736k 564k -1.74 Qd2-b2+ Kg7-h6
22/43- 00:04 2,800k 564k -1.83 Qd2-b2+ Kg7-h6
22/57- 00:06 3,665k 530k -2.45 Qd2-b2+ Kg7-h6
22/57 00:07 3,844k 533k -2.43 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-g1 Rd3xh3 Qg5-g2 Rh3xh5 Qg2-b7+ Kg7-h6 Qb7-c8 Re6-e1+ Kg1-f2 Rh5-h1 Qc8xa6 Re1-f1+ Kf2-g3 Rf1-g1+ Kg3-f2 Rg1-g4 Qa6xb5 Rh1-h2+ Kf2-f3 Rh2-h3+ Kf3-e2 Rg4xf4 Qb5-c5 Rf4-e4+ Ke2-f2
23/47- 00:08 4,331k 539k -2.53 Qd2-b2+ Kg7-h6
23/47- 00:08 4,526k 542k -2.62 Qd2-b2+ Kg7-h6
23/47- 00:08 4,774k 544k -2.77 Qd2-b2+ Kg7-h6
23/50+ 00:09 5,138k 546k -2.65 Qd2-b2+
23/50+ 00:09 5,174k 546k -2.37 Qd2-b2+
23/50 00:09 5,269k 547k -2.42 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-f2 Re3xh3 h5xg6 h7xg6 Qg5-d8 Re6-e4 Qd8-a5 Re4xf4+ Kf2-g2 Rh3-h4 Qa5xa6 Rh4-g4+ Kg2-h3 Rf4-f3+ Kh3-h2 Rf3-f2+ Kh2-h1 Rf2-f1+ Kh1-h2 f5-f4 Qa6-b7+ Kg7-h6 Qb7xb5 Rf1-f2+ Kh2-h1 Rg4-g5
24/55+ 00:12 6,656k 544k -2.32 Qd2-b2+
24/55- 00:13 7,195k 536k -2.51 Qd2-b2+ Kg7-h6
24/55- 00:13 7,536k 538k -2.66 Qd2-b2+ Kg7-h6
24/55 00:14 7,834k 541k -2.56 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Qg5-h4 h7-h6 Qh4-f2 Re6-e3 Qf2-b2+ Rd3-c3 h3-h4 Kg7-h7 h5xg6+ Kh7xg6 Qb2-a1 Re3-d3 Kg2-f1 Kg6-h5 a3-a4 Kh5-g4 a4xb5 a6xb5 Kf1-e2 Kg4xf4
25/42- 00:15 8,642k 545k -2.65 Qd2-b2+ Kg7-h6
25/42- 00:16 8,941k 548k -2.75 Qd2-b2+ Kg7-h6
25/42- 00:16 9,343k 551k -2.89 Qd2-b2+ Kg7-h6
25/42 00:18 10,142k 552k -2.82 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-b8 Re6-e8 Qb8-a7 Re3-e2+ Kf2-g1
26/49 00:21 11,748k 547k -2.89 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-b8 Re6-e8 Qb8-a7 Re3-e2+ Kf2-g1 Re8-e6 Qa7-d4+ Kf6-f7 Qd4-d7+ Re6-e7 Qd7-d5+ Kf7-g7 Qd5-d4+ Kg7-h7 Qd4-f6
27/57 00:22 12,448k 551k -2.80 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-c8 Re3-e2+ Kf2-g1 Re2-e1+ Kg1-f2 Kf6-f7 Qc8-b7+ Kf7-g8 Kf2-g3 Re1-e4 Qb7-d7 Kg8-f8 Kg3-h3 Re6-e7 Qd7-c8+ Kf8-g7 Qc8xa6 Re4xf4
28/64+ 00:23 12,763k 553k -2.71 Qd2-b2+
28/64+ 00:23 12,829k 553k -2.61 Qd2-b2+
28/64- 00:24 13,608k 556k -2.90 Qd2-b2+ Kg7-h6
28/64 00:27 15,346k 553k -2.91 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-c8 Re3-e2+ Kf2-g1 Re2-e1+ Kg1-f2 Kf6-f7 Qc8-b7+ Kf7-g8 Kf2-g3 Re1-e4 Qb7-d7 Re6-e7 Qd7-c8+ Kg8-h7 Qc8xa6 Re4xb4 Qa6-a1 Re7-e4 Qa1-h1+ Kh7-g7
29/51 00:30 16,721k 557k -2.91 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-f6 Qb7-a8 Kf6-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Kf7-g8 Qd7-c8+ Kg8-g7 Qc8-b7+ Kg7-f6 Qb7-c8 Re3-e2+ Kf2-g1 Re2-e1+ Kg1-f2 Kf6-f7 Qc8-b7+ Kf7-g8 Kf2-g3 Re1-e4 Qb7-d7 Re6-e7 Qd7-c8+ Kg8-h7 Qc8xa6 Re4xb4 Qa6-a1
30/57+ 00:31 17,376k 559k -2.82 Qd2-b2+
30/57+ 00:31 17,552k 558k -2.72 Qd2-b2+
30/57 00:32 17,814k 555k -2.79 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-g8 Qb7-a8+ Kg8-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Re6-e7 Qd7-d5+ Re3-e6 Kf2-g3 Kf7-g7 Kg3-h4 Kg7-h6 Qd5-d8 Re7-e8
31/44+ 00:33 18,273k 552k -2.70 Qd2-b2+
31/49- 00:35 19,512k 553k -2.89 Qd2-b2+ Kg7-h6
31/49+ 00:40 22,386k 554k -2.75 Qd2-b2+
31/50+ 00:43 23,593k 546k -2.54 Qd2-b2+
31/56 00:49 26,920k 544k -2.89 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-g8 Qb7-a8+ Kg8-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Re6-e7 Qd7-d5+ Re3-e6 Kf2-g3 Kf7-g7 Kg3-h4 Kg7-h6 Qd5-d8 Re7-e8 Qd8-g5+ Kh6-g7
32/47+ 00:50 27,375k 542k -2.79 Qd2-b2+
32/47+ 00:50 27,454k 543k -2.70 Qd2-b2+
32/47 00:51 27,752k 543k -2.81 Qd2-b2+ Kg7-h6 Qb2-f6 Re4-e2+ Kg2-f1 Re2-e1+ Kf1-f2 Re7-e2+ Kf2-f3 Re2-e6 Qf6-g5+ Kh6-g7 h4-h5 Re1-e3+ Kf3-g2 Re3-d3 Kg2-f2 Rd3xh3 h5xg6 h7xg6 Qg5-g2 Rh3xa3 Qg2-b7+ Kg7-g8 Qb7-a8+ Kg8-f7 Qa8-d5 Ra3-e3 Qd5-d7+ Re6-e7 Qd7-d5+ Re3-e6 Kf2-g3 Kf7-g7 Kg3-h4 Kg7-h6 Qd5-d8 Re7-e8 Qd8-g5+ Kh6-h7 Qg5-g1 Kh7-g7 Kh4-g5
33/48+ 00:52 28,317k 544k -2.71 Qd2-b2+
33/51+ 00:52 28,552k 545k -2.62 Qd2-b2+
33/51+ 00:54 30,073k 548k -2.47 Qd2-b2+

[Dann Corbit]

I was giving my point of view in the statement that Larry made of a Queen versus two Rooks with 5 pawns each, I know that there are millions of configurations but of course that you do NOT want to give a disadvantage position to White with the Queen such as this one, we are talking about positions that are equal from the start.

Of course, connected rooks on an open file get large positional bonus value.
And part of the problem with random board configurations is that almost all of them are not even because of positional score.
So it is a very difficult problem to solve exactly.
In fact, I think it is safe to say:
1. Nobody on earth knows the exact values of the chessmen to 4 significant digits.
2. The chessmen change in value in connection with their configuration on the board
3. The chessmen change in value in connection with the number and types of other chessmen on the board
4. The chessmen change in value in connection with tactical formations (e.g. Alekhine's gun, e.g. a knight with a pawn protector close to the enemy king).

I think we can make calculations to get an approximation to the value of chessmen in different situations (e.g. opening, midgame, endgame...) but it takes a lot of data to get a good approximation.

I think your posts are interesting and thoughtful. But I think you tend to draw very strong assumptions as a result of a single game played out. Of course, those assumptions might be right. But I do not think less than 800 games adequately demonstrates a point such as "How much is a knight and a bishop worth, compared to a bishop pair?" or "Is a queen worth the same as two rooks?"

Which is not to say that your experiments are not interesting. They are interesting. In fact, I do the same sort of experiments myself.

[Chessqueen]

I was giving my point of view in the statement that Larry made of a Queen versus two Rooks with 5 pawns each, I know that there are millions of configurations but of course that you do NOT want to give a disadvantage position to White with the Queen such as this one, we are talking about positions that are equal from the start.

Of course, connected rooks on an open file get large positional bonus value.
And part of the problem with random board configurations is that almost all of them are not even because of positional score.
So it is a very difficult problem to solve exactly.
In fact, I think it is safe to say:
1. Nobody on earth knows the exact values of the chessmen to 4 significant digits.
2. The chessmen change in value in connection with their configuration on the board
3. The chessmen change in value in connection with the number and types of other chessmen on the board
4. The chessmen change in value in connection with tactical formations (e.g. Alekhine's gun, e.g. a knight with a pawn protector close to the enemy king).

I think we can make calculations to get an approximation to the value of chessmen in different situations (e.g. opening, midgame, endgame...) but it takes a lot of data to get a good approximation.

I think your posts are interesting and thoughtful. But I think you tend to draw very strong assumptions as a result of a single game played out. Of course, those assumptions might be right. But I do not think less than 800 games adequately demonstrates a point such as "How much is a knight and a bishop worth, compared to a bishop pair?" or "Is a queen worth the same as two rooks?"

Which is not to say that your experiments are not interesting. They are interesting. In fact, I do the same sort of experiments myself.

[Chessqueen]

I was giving my point of view in the statement that Larry made of a Queen versus two Rooks with 5 pawns each, I know that there are millions of configurations but of course that you do NOT want to give a disadvantage position to White with the Queen such as this one, we are talking about positions that are equal from the start.

Of course, connected rooks on an open file get large positional bonus value.
And part of the problem with random board configurations is that almost all of them are not even because of positional score.
So it is a very difficult problem to solve exactly.
In fact, I think it is safe to say:
1. Nobody on earth knows the exact values of the chessmen to 4 significant digits.
2. The chessmen change in value in connection with their configuration on the board
3. The chessmen change in value in connection with the number and types of other chessmen on the board
4. The chessmen change in value in connection with tactical formations (e.g. Alekhine's gun, e.g. a knight with a pawn protector close to the enemy king).

I think we can make calculations to get an approximation to the value of chessmen in different situations (e.g. opening, midgame, endgame...) but it takes a lot of data to get a good approximation.

I think your posts are interesting and thoughtful. But I think you tend to draw very strong assumptions as a result of a single game played out. Of course, those assumptions might be right. But I do not think less than 800 games adequately demonstrates a point such as "How much is a knight and a bishop worth, compared to a bishop pair?" or "Is a queen worth the same as two rooks?"

Which is not to say that your experiments are not interesting. They are interesting. In fact, I do the same sort of experiments myself.

I believe that the Alpha Zero team has done a lot of tests with different endgame pieces + pawns configurations. not necessarily thousand of games, but meaningful configurations that helped them to determine a more exact value of each pieces, in which computer has proven different and more exact value than what we humans has given each pieces for so long.

[Dann Corbit]

I believe that the Alpha Zero team has done a lot of tests with different endgame pieces + pawns configurations. not necessarily thousand of games, but meaningful configurations that helped them to determine a more exact value of each pieces, in which computer has proven different and more exact value than what we humans has given each pieces for so long.

AlphaZero searches 60 thousand positions per second and it trained for 9 hours.
60,000 Nodes / second * 3600 seconds / hour * 9 hours = 1,944,000,000 nodes. About 2 billion chess positions were considered by the neural network.

[Chessqueen]

I believe that the Alpha Zero team has done a lot of tests with different endgame pieces + pawns configurations. not necessarily thousand of games, but meaningful configurations that helped them to determine a more exact value of each pieces, in which computer has proven different and more exact value than what we humans has given each pieces for so long.

Alpha Zero searches 60 thousand positions per second and it trained for 9 hours.
60,000 Nodes / second * 3600 seconds / hour * 9 hours = 1,944,000,000 nodes. About 2 billion chess positions were considered by the neural network.

That is Amazing! How close is Komodo or SF in developing a Neural Network version? This calculation power is what give Alpha Zero the capability to see so deep and to restrict the opponent moves and handicap its options like in this game:
https://www.youtube.com/watch?v=jS26Ct34YrQ

In order to get an Idea oh how many times more nodes can Alpha Zero calculate than lets say Komodo or SF10 on the fastest hardware or computer that SF10 can find, can you give us an Idea of how many Nodes / per second can SF10 calculate on the fastest computer that SF can get hook on?

[Dann Corbit]

I believe that the Alpha Zero team has done a lot of tests with different endgame pieces + pawns configurations. not necessarily thousand of games, but meaningful configurations that helped them to determine a more exact value of each pieces, in which computer has proven different and more exact value than what we humans has given each pieces for so long.

Alpha Zero searches 60 thousand positions per second and it trained for 9 hours.
60,000 Nodes / second * 3600 seconds / hour * 9 hours = 1,944,000,000 nodes. About 2 billion chess positions were considered by the neural network.

That is Amazing! How close is Komodo or SF in developing a Neural Network version? This calculation power is what give Alpha Zero the capability to see so deep and to restrict the opponent moves and handicap its options like in this game:
https://www.youtube.com/watch?v=jS26Ct34YrQ

It is really a different paradigm.

However, the Komodo team has taken a step in that direction by building the MCTS version of Komodo.

That is because the neural net tools like LC0 use MCTS (Monte Carlo Tree Search).

What that means is that they simply sample some prospective move sequences and choose the ones that have a good result.
Monte Carlo is a famous gambling place, which is where the MC part of the name comes from.

Now, the neural net evaluation is nothing at all like the hand-tuned evaluation (except that they both decide if a given chess position is good or bad).
But what you will end up with by adding a neural net search is another LC0.
I guess you could make one version stronger than another by spending more training effort, but I guess what that would really do is split the training work force and you would end up with less excellence for any of the teams than you would get with a single entrant.

[Chessqueen]

I believe that the Alpha Zero team has done a lot of tests with different endgame pieces + pawns configurations. not necessarily thousand of games, but meaningful configurations that helped them to determine a more exact value of each pieces, in which computer has proven different and more exact value than what we humans has given each pieces for so long.

Alpha Zero searches 60 thousand positions per second and it trained for 9 hours.
60,000 Nodes / second * 3600 seconds / hour * 9 hours = 1,944,000,000 nodes. About 2 billion chess positions were considered by the neural network.

That is Amazing! How close is Komodo or SF in developing a Neural Network version? This calculation power is what give Alpha Zero the capability to see so deep and to restrict the opponent moves and handicap its options like in this game:
https://www.youtube.com/watch?v=jS26Ct34YrQ

It is really a different paradigm.

However, the Komodo team has taken a step in that direction by building the MCTS version of Komodo.

That is because the neural net tools like LC0 use MCTS (Monte Carlo Tree Search).

What that means is that they simply sample some prospective move sequences and choose the ones that have a good result.
Monte Carlo is a famous gambling place, which is where the MC part of the name comes from.

Now, the neural net evaluation is nothing at all like the hand-tuned evaluation (except that they both decide if a given chess position is good or bad).
But what you will end up with by adding a neural net search is another LC0.
I guess you could make one version stronger than another by spending more training effort, but I guess what that would really do is split the training work force and you would end up with less excellence for any of the teams than you would get with a single entrant.

In order to get an Idea oh how many more nodes can Alpha Zero calculate than lets say Komodo or SF10 on the fastest hardware or computer that SF10 can find, can you give us an Idea of how many Nodes / per second can Komodo or SF10 calculate on the fastest computer that Komodo can get hook on?

[Dann Corbit]

On a cluster system, a souped up version of Stockfish can analyze 4.8 billion NPS (see the top entry in this list):
http://www.ipmanchess.yolasite.com/amd- ... -bench.php

However, there is no way to know the Elo of such a system because it does not scale in a linear fashion.

I guess an interesting way to look at it is:
Super GM's think in tens of positions in a second.
LC0/LeelaZero think in tens of thousands of positions per second.
Stockfish can think in tens/hundreds/thousands of millions of positions per second.

I think what this shows us is that nodes per second is a very poor measure of strength.
I think it also shows us that the humans have the best evaluation ability, the computer NN approach is second best, and the alpha-beta searchers are third.
The speed of the NN searchers more than make up for the evaluation ability and the speed of the alpha-beta searchers being extreme is why they can compete