White 36 vs Black 817
White starts from setup 36 (NBBNQRKR), Black from setup 817 (BRKBNQRN). In the encyclopedia since 2026-10-03: neither side starts better.
|cp| ≤ 5 at 400M · 800M · 1.6B · 4B nodes, two independent engines agree
The two setups
White: setup #36 N B B N Q R K R
King on the g-file, rooks on f, h, queen on the e-file; no 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 36 gives White +0.26 (Morozov Balance Index, 400M nodes).
Black: setup #817 B R K B N Q R N
King on the c-file, rooks on b, g, 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 817 gives White +0.31 (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 57.
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 shallow rungs wandered between -11 and +17 cp before settling — a reminder that quick engine checks overstate imbalances, and why the ladder goes deep.
For comparison, in symmetric play setup 36 gives White +0.26 and setup 817 gives White +0.31: 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 c4, and 3 first moves are within 10 cp of it.
Along the engine's main line (30 moves) the largest swing was 77 cp (around move 10), ending at 0 cp — the position stays playable for both sides.
Engine self-play agrees: 10 games, White won 0, drawn 10, Black won 0 (White scored 50.0%).
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 | 12 | +17 | 1. c4 f5 2. d4 g5 3. e4 b5 4. cxb5 Nd6 |
| S1 | 500 000 | 18 | +1 | 1. c4 f5 2. h4 e6 3. d4 g5 4. h5 b6 |
| S2 | 2 000 000 | 22 | -11 | 1. c4 f5 2. e4 e6 3. d4 g5 4. Nb3 fxe4 |
| S3 | 8 000 000 | 27 | -2 | 1. c4 f5 2. d4 e6 3. h4 b6 4. e4 g5 |
| S4 | 32 000 000 | 35 | 0 | 1. c4 f5 2. d4 g5 3. Nb3 e6 4. e4 Ng6 |
| S5 | 128 000 000 | 39 | -5 | 1. c4 f5 2. d4 e6 3. Nb3 b6 4. f3 g5 |
| S6 | 400 000 000 | 49 | 0 | 1. c4 f5 2. d4 g5 3. Nb3 Ng6 4. e4 Nd6 |
| S7 | 800 000 000 | 55 | 0 | 1. c4 f5 2. d4 g5 3. Nb3 Ng6 4. e4 Nd6 |
| S8 | 1 600 000 000 | 57 | 0 | 1. c4 f5 2. d4 g5 3. Nb3 Ng6 4. e4 e6 |
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 | not checked |
| 8 000 000 000 | not checked |
| 16 000 000 000 | not checked |
This pair held exactly 0.00 up to 1 600 000 000 nodes.
Independent engines, run on the same position:
- Berserk 14: 0
- Lc0 0.32.1: -6
Independent confirmations: 2 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 +74, Black +72, difference +2. A balanced pair needs White's setup to be weaker by about the price of the first move.
Extended self-play: 10 games, 0 White wins, 10 draws, 0 Black wins, White scored 50.0%.
Best first moves
The 3 best first moves for White, searched together at 400 000 000 nodes.
| # | Eval, cp | Depth | Main line |
|---|---|---|---|
| 1 | 0 | 42 | 1. c4 f5 2. d4 g5 3. Nb3 b6 4. e4 e6 5. exf5 exf5 |
| 2 | -9 | 42 | 1. Nb3 g5 2. c4 Ng6 3. d4 e6 4. e4 f5 5. f3 b6 |
| 3 | -4 | 41 | 1. d4 g5 2. e4 Ng6 3. Nb3 Nd6 4. e5 Nc4 5. c3 b5 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 77 cp, mean 25.6 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | c4 | -13 | f5 | -19 |
| 2 | d4 | -10 | g5 | -2 |
| 3 | e4 | 0 | e5 | -15 |
| 4 | d5 | -16 | f4 | -27 |
| 5 | h3 | -31 | h5 | -37 |
| 6 | Nc3 | -35 | g4 | -44 |
| 7 | Nb3 | -45 | Nf7 | -37 |
| 8 | h4 | -41 | g3 | -31 |
| 9 | Nb5 | -52 | b6 | -64 |
| 10 | Nxa7+ | -77 | Kb7 | -61 |
| 11 | Nb5 | -61 | c6 | -60 |
| 12 | Nc3 | -48 | gxf2+ | -61 |
| 13 | Rxf2 | -61 | Nf6 | -64 |
| 14 | Rh3 | -67 | Ka7 | -53 |
| 15 | Bd3 | -55 | Bb7 | -51 |
| 16 | Kh1 | -51 | Rc8 | -48 |
| 17 | a4 | -44 | Kb8 | -36 |
| 18 | Bd2 | -33 | cxd5 | -36 |
| 19 | Nxd5 | -25 | Bxd5 | -17 |
| 20 | cxd5 | -8 | Nd6 | -1 |
| 21 | Bb4 | 0 | Ng4 | 0 |
| 22 | Re2 | 0 | Be7 | 0 |
| 23 | Ba6 | 0 | Nb7 | 0 |
| 24 | Bc3 | 0 | Qf6 | 0 |
| 25 | a5 | 0 | Rg6 | 0 |
| 26 | Bxb7 | 0 | Kxb7 | 0 |
| 27 | axb6 | 0 | Qxb6 | 0 |
| 28 | Na5+ | 0 | Kb8 | 0 |
| 29 | b3 | 0 | Rxc3 | 0 |
| 30 | Rxc3 | 0 | Bb4 | 0 |
Engine self-play
10 games at 500 000 nodes per move: White won 0, drawn 10, Black won 0 — White scored 50.0%. Average length 43.5 moves.
Sample game 1 (1/2-1/2)
1. c4 f5 2. e4 e6 3. Nc3 b6 4. d4 g5 5. Nb3 Ng6 6. f3 Nd6 7. c5 bxc5 8. dxc5 Nf7 9. exf5 exf5 10. Qe2 Nh4 11. h3 Qe8 12. Qxe8 Rxe8 13. Rd1 Ne5 14. Nd4 Be7 15. Nxf5 Bxc5+ 16. Kh2 Bb4 17. Nxh4 Bxc3 18. Nf5 Bxb2 19. Bxg5 Bxf3 20. Rdf1 Nf7 21. Ne7+ Kb7 22. Bh4 Be5+ 23. Kg1 Bd4+ 24. Kh2 Be5+ 25. Kg1 Bd4+ 26. Kh2
Sample game 2 (1/2-1/2)
1. Nb3 f5 2. d4 b6 3. c4 g5 4. e4 e6 5. f3 Ng6 6. exf5 exf5 7. Rf2 a5 8. O-O a4 9. Na1 Bf6 10. Nc2 Nd6 11. b3 f4 12. Ba3 Nh4 13. d5 O-O-O 14. Bxd6 Qxd6 15. Nc3 Qf8 16. Ne4 axb3 17. axb3 Be5 18. Re2 Qg7 19. Na3 Nf5 20. Nb5 Rde8 21. Ra2 Bb7 22. c5 g4 23. c6 dxc6 24. dxc6 Bd4+ 25. Kh1 Bxc6 26. Na7+ Kd7 27. Nxc6 Ng3+ 28. Nxg3 Rxe1 29. Rxe1 fxg3 30. Ne7 Qxe7 31. Bf5+ Kd6 32. Rxe7 Kxe7 33. hxg3 gxf3 34. gxf3 Rxg3 35. Be4 h5 36. Rg2 h4 37. b4 c5 38. bxc5 bxc5 39. Rc2 Ke6 40. Re2 Be5 41. Rc2 Rh3+ 42. Kg1 Ke7 43. Rc4 Rg3+ 44. Kh1 Rh3+ 45. Kg1 Kd6 46. Ba8 Bd4+ 47. Kg2 Rg3+ 48. Kh1 Ke5 49. Bb7 Rh3+ 50. Kg2 Rg3+ 51. Kh2 Bg1+ 52. Kh1 Bd4 53. Kh2 Kf5 54. Rc1 Kf6 55. Bc6 Ke5 56. Rc4 Kf6 57. Ra4 Bg1+ 58. Kh1 Bd4 59. Kh2 Kg5 60. Bb7 Bg1+ 61. Kh1 h3 62. Ra8 h2 63. Rf8 Kh6 64. Rf4 Kh5 65. Bd5 Rg5 66. Bc4 Rg7 67. Bf7+ Kg5 68. Rg4+ Kf6 69. Rxg7 Kxg7 70. Bc4 Kf6 71. Kg2 Ke5 72. Bb3 Kd4
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 36 / 817 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): brkbnqrn/pppppppp/8/8/8/8/PPPPPPPP/NBBNQRKR w FHbg - 0 1
Mirror twin (the same position mirrored): White 926 vs Black 139.