White 287 vs Black 847
White starts from setup 287 (NRKNRQBB), Black from setup 847 (RKNRQNBB). In the encyclopedia since 2026-10-15: neither side starts better.
|cp| ≤ 5 at 400M · 800M · 1.6B · 4B nodes, two independent engines agree
How to play with different setups
White sets up the pieces by position No. 287, Black by position No. 847, mirrored on their own side; the pawns stand as usual.
Moves as usual: White starts; everything else follows the Chess960 rules.
The two setups
White: setup #287 N R K N R Q B B
King on the c-file, rooks on b, e, queen on the f-file; one bishop starts in a corner; the knights start away from the centre. The king already stands on a castled square, so one of the castlings moves only the rook. In ordinary symmetric play setup 287 gives White +0.07 (Morozov Balance Index, 400M nodes).
Black: setup #847 R K N R Q N B B
King on the b-file, rooks on a, d, queen on the e-file; one bishop starts in a corner; the knights start on central files. Castling follows the Chess960 rule: king and rook end on the classical squares. In ordinary symmetric play setup 847 gives White +0.30 (Morozov Balance Index, 400M nodes).
Why the pair is balanced
At the deepest rungs (400M, 800M, 1.6B nodes) the evaluation is exactly 0.00 each time, reaching search depth 62.
Admission rule: |cp| ≤ 5 at four budgets up to 4 billion nodes, no one-sided drift along the ladder, and confirmation by two independent engines — Berserk and Lc0 — each below its own median on symmetric positions. The same corridor is required at every deeper budget (8 G, 16 G) at which the pair has been evaluated.
The shallow rungs agreed early: from 100 thousand to 128 million nodes the evaluation never left ±13 cp.
For comparison, in symmetric play setup 287 gives White +0.07 and setup 847 gives White +0.30: with both sides on the same rank the first move is worth something. In this pair Black's own setup absorbs that edge completely.
White is not forced into a single path: the best first move is g4, and 3 first moves are within 10 cp of it.
Along the engine's main line (30 moves) the evaluation stayed within ±18 cp the whole way: the balance holds as the pieces come out, not only on move one.
Engine self-play agrees: 10 games, White won 1, drawn 9, Black won 0 (White scored 55.0%).
Verification ladder
Nine independent searches of the starting position, each with a larger node budget. Evaluation in centipawns from White's point of view.
| Stage | Nodes | Depth | Eval, cp | Main line |
|---|---|---|---|---|
| S0 | 100 000 | 13 | -2 | 1. f4 f5 2. Nb3 Bxb3 3. axb3 d5 4. g4 g6 |
| S1 | 500 000 | 23 | 0 | 1. f4 f5 2. Nb3 g6 3. g4 fxg4 4. Bxb7 Kxb7 |
| S2 | 2 000 000 | 27 | 0 | 1. f4 f5 2. Nb3 g6 3. g4 fxg4 4. h3 gxh3 |
| S3 | 8 000 000 | 30 | +13 | 1. f4 f5 2. Nb3 g6 3. g4 d5 4. gxf5 gxf5 |
| S4 | 32 000 000 | 34 | 0 | 1. f4 f5 2. Nb3 a5 3. g4 fxg4 4. Nc5 d5 |
| S5 | 128 000 000 | 48 | 0 | 1. f4 f5 2. Nb3 a5 3. g4 fxg4 4. Nc3 c6 |
| S6 | 400 000 000 | 55 | 0 | 1. f4 f5 2. Nb3 a5 3. g4 fxg4 4. Nc5 d5 |
| S7 | 800 000 000 | 59 | 0 | 1. f4 f5 2. Nb3 a5 3. g4 fxg4 4. Nc5 d5 |
| S8 | 1 600 000 000 | 62 | 0 | 1. f4 f5 2. Nb3 a5 3. g4 fxg4 4. Nc5 d5 |
Deep verification
The search ladder reaches 1.6 billion nodes; admission needs a 4-billion rung and two further engines on top of that. Beyond even that, every pair was re-checked at budgets up to ten times larger and in thousands of self-play games. This section shows what each check returned for this pair.
| Nodes | Eval, cp |
|---|---|
| 4 000 000 000 | 0 |
| 8 000 000 000 | not checked |
| 16 000 000 000 | not checked |
This pair stays within the ±5 cp corridor up to 4 000 000 000 nodes.
Independent engines, run on the same position:
- Berserk 14: -9
- Lc0 0.32.1: +4
Independent confirmations: 3 of 3. An engine counts as confirming when its reading of the pair is closer to equality than its own reading of a typical symmetric Chess960 position (+10 for Berserk, +18 for Lc0). Where only one engine confirms, the pair is balanced according to that engine alone, and this page says so.
Own weight of each setup, measured against the classical array: White +133, Black +129, difference +4. A balanced pair needs White's setup to be weaker by about the price of the first move.
Extended self-play: 10 games, 1 White wins, 9 draws, 0 Black wins, White scored 55.0%.
Best first moves
The 3 best first moves for White, searched together at 400 000 000 nodes.
| # | Eval, cp | Depth | Main line |
|---|---|---|---|
| 1 | 0 | 53 | 1. g4 f5 2. Nb3 fxg4 3. f4 a5 4. Nc5 d5 5. e4 Nd7 |
| 2 | 0 | 53 | 1. Nb3 f5 2. f4 a5 3. g4 fxg4 4. Nc5 d5 5. e4 Nd7 |
| 3 | 0 | 53 | 1. f4 f5 2. Nb3 a5 3. g4 fxg4 4. Nc5 d5 5. e4 Nd7 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 18 cp, mean 3.6 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | f4 | +13 | f5 | 0 |
| 2 | b3 | -14 | g6 | -11 |
| 3 | g3 | -6 | d5 | -10 |
| 4 | c3 | -11 | Nd7 | -11 |
| 5 | Bd4 | -8 | e5 | -3 |
| 6 | fxe5 | 0 | Nxe5 | 0 |
| 7 | Nc2 | -9 | a5 | -10 |
| 8 | Nb2 | -8 | Ne7 | -14 |
| 9 | Nd3 | -18 | N7c6 | -7 |
| 10 | Nxe5 | -6 | Nxe5 | -14 |
| 11 | h3 | -4 | Bf7 | -4 |
| 12 | Qf2 | 0 | Qe6 | -4 |
| 13 | O-O | -3 | O-O | -5 |
| 14 | Ne1 | -9 | Nd7 | 0 |
| 15 | Bxh8 | 0 | Kxh8 | 0 |
| 16 | Qd4+ | -4 | Kg8 | 0 |
| 17 | g4 | 0 | fxg4 | 0 |
| 18 | Rxf7 | 0 | Rxf7 | 0 |
| 19 | Bxd5 | 0 | Qf5 | 0 |
| 20 | Bxf7+ | 0 | Kxf7 | 0 |
| 21 | e4 | 0 | Qe6 | 0 |
| 22 | hxg4 | 0 | Ne5 | 0 |
| 23 | Qd5 | 0 | Qxd5 | 0 |
| 24 | exd5 | 0 | g5 | 0 |
| 25 | d4 | 0 | Nxg4 | 0 |
| 26 | Nf3 | 0 | Kf6 | +5 |
| 27 | Re1 | +2 | Rd8 | 0 |
| 28 | Re6+ | 0 | Kf5 | +2 |
| 29 | Re7 | +3 | Rxd5 | 0 |
| 30 | Rxh7 | 0 | Ke4 | 0 |
Engine self-play
10 games at 500 000 nodes per move: White won 1, drawn 9, Black won 0 — White scored 55.0%. Average length 40.0 moves.
Sample game 1 (1-0)
1. Nb3 g6 2. g4 a5 3. f4 d6 4. e4 a4 5. Nd4 e5 6. Ne2 f5 7. exf5 e4 8. Ndc3 d5 9. O-O-O Ra5 10. a3 gxf5 11. gxf5 Qf7 12. d3 exd3 13. Rxd3 Qxf5 14. Red1 c6 15. Bd4 Bxd4 16. Nxd4 Qf6 17. f5 Nd6 18. Qf4 Kc8 19. Bg2 Nd7 20. Re3 Nc5 21. Rh3 Nc4 22. Bf1 Qe5 23. Qh4 Nd6 24. Nf3 Qe3+ 25. Nd2 Qe5 26. Re1 Nce4 27. Ndxe4 dxe4 28. Nxe4 Nxf5 29. Qf2 Nd4 30. Rd3 Kb8 31. h4 Ka8 32. Qd2 b6 33. Nc3 Qf6 34. Rd1 Bc4 35. Rxd4 Rxd4 36. Qxd4 Qxd4 37. Rxd4 Bxf1 38. Rxa4 Rxa4 39. Nxa4 Kb7 40. Kd2 c5 41. Ke3 Kc6 42. Nc3 Bc4 43. Kf4 Bf7 44. Ke5 Bg6 45. Ne4 Kb5 46. Nd6+ Ka4 47. Nc4 Kb5 48. Ne3 Kc6 49. c3 Bb1 50. a4 Bg6 51. Nd5 Bc2 52. Nf6 Bd3 53. Ke6 Bc2 54. Kf7 b5 55. a5 b4 56. cxb4 cxb4 57. a6 Kb6 58. Nd5+ Ka7 59. Nxb4 Be4 60. Kg7 h5 61. Kh6 Bf3 62. Kg5 Bd1 63. Nd5 Kxa6 64. Nf6 Ka5 65. Nxh5 Bc2 66. Nf6 Kb6 67. h5 Kc7 68. Nd5+ Kd6 69. Nf4 Bh7 70. Ng6 Bg8 71. Kf6 Kc5 72. h6 Bd5 73. h7 Be4 74. h8=Q Bb7 75. Ne5 Ba6 76. Qa8
Sample game 2 (1/2-1/2)
1. f4 f5 2. Nb3 g5 3. g3 a5 4. Bd4 Bxd4 5. Nxd4 gxf4 6. gxf4 d5 7. Nc3 e6 8. O-O-O Ra6 9. d3 Qe7 10. Kb1 Bf7 11. Ka1 c5 12. Nf3 Nd6 13. Nb1 a4 14. e4 d4 15. Ne5 Be8 16. Nd2 Ng6 17. Rb1 Rc8 18. a3 Qh4 19. Nec4 Nxc4 20. Nxc4 Qxf4 21. exf5 Qxf5 22. Qg1 b5 23. Ne5 Ka7 24. Nxg6 Qxg6 25. Qf2 Qf7 26. Qh4 Qc7 27. Re2 Rb6 28. Rbe1 Bd7 29. Rg1 Qd6 30. Rf2 b4 31. Rf7 bxa3 32. Rgg7 axb2+ 33. Kb1 Rc7 34. Qd8 Ka6 35. Rxd7 Rxd7 36. Rxd7 Qb8 37. Qxb8 Rxb8 38. Rd6+ Ka5 39. Rd7 Ka6 40. Rd6+ Ka5
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 287 / 847 in the app's encyclopedia, then choose how to play.
- Local game — two players on one device, with a clock.
- With a mentor — a clearly labelled training bot with hints, at your level.
- Challenge a friend — an online game by link; both players get the same pair, colours swap on rematch.
Play this pair against the computer, right here Play a friend
FEN (Shredder, castling rights by rook files): rknrqnbb/pppppppp/8/8/8/8/PPPPPPPP/NRKNRQBB w BEad - 0 1
Mirror twin: White 576 vs Black 112 — a different position (castling squares are fixed), not evaluated by the search.