White 453 vs Black 521
White starts from setup 453 (RNBBNKQR), Black from setup 521 (RNQBBKNR). In the encyclopedia since 2026-11-01: 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. 453, Black by position No. 521, 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 #453 R N B B N K Q R
King on the f-file, rooks on a, h, queen on the g-file; no 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 453 gives White +0.22 (Morozov Balance Index, 400M nodes).
Black: setup #521 R N Q B B K N R
King on the f-file, rooks on a, h, queen on the c-file; no 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 521 gives White +0.09 (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 55.
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 wandered between -10 and +35 cp before settling — a reminder that quick engine checks overstate imbalances, and why the ladder goes deep.
For comparison, in symmetric play setup 453 gives White +0.22 and setup 521 gives White +0.09: 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 d3, and 2 first moves are within 10 cp of it.
Along the engine's main line (30 moves) the evaluation stayed within ±24 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 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 | 15 | +35 | 1. c4 h5 2. Nc3 c6 3. h3 e5 4. e4 d6 |
| S1 | 500 000 | 18 | -1 | 1. Nc3 e6 2. h4 Ne7 3. h5 O-O 4. g4 d5 |
| S2 | 2 000 000 | 23 | +4 | 1. e4 e6 2. d3 Ne7 3. h4 O-O 4. h5 h6 |
| S3 | 8 000 000 | 28 | +17 | 1. e4 d5 2. d3 e6 3. c4 Ne7 4. h4 c5 |
| S4 | 32 000 000 | 36 | -4 | 1. d3 d5 2. e4 e6 3. f4 Ne7 4. Be3 O-O |
| S5 | 128 000 000 | 40 | -10 | 1. Nc3 e6 2. e4 Ne7 3. h4 O-O 4. h5 c5 |
| S6 | 400 000 000 | 45 | 0 | 1. Nc3 e6 2. e4 Ne7 3. h4 c5 4. h5 d5 |
| S7 | 800 000 000 | 49 | 0 | 1. e4 d5 2. d3 e6 3. f4 c5 4. c4 Ne7 |
| S8 | 1 600 000 000 | 55 | 0 | 1. e4 d5 2. d3 e6 3. f4 c5 4. c4 Ne7 |
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: -5
- 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 +28, Black +34, difference -6. 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 | 43 | 1. d3 d5 2. e4 e6 3. f4 Ne7 4. c4 O-O 5. Qf2 f5 |
| 2 | 0 | 43 | 1. e4 d5 2. d3 e6 3. f4 Ne7 4. c4 O-O 5. Qf2 c5 |
| 3 | -18 | 42 | 1. Nc3 e6 2. d3 c5 3. e4 Ne7 4. f4 d5 5. Qf2 Nbc6 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 24 cp, mean 6.0 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | e4 | +9 | e6 | -11 |
| 2 | d3 | -18 | d5 | -18 |
| 3 | Nc3 | -17 | Ne7 | -18 |
| 4 | h4 | -2 | c5 | -4 |
| 5 | h5 | +6 | h6 | 0 |
| 6 | Bf4 | +1 | Bc7 | -12 |
| 7 | Qh2 | -16 | Bxf4 | -11 |
| 8 | Qxf4 | -3 | d4 | +7 |
| 9 | Nb1 | +18 | b5 | +12 |
| 10 | Nf3 | +24 | Qd8 | +10 |
| 11 | Rh3 | +15 | Nbc6 | -10 |
| 12 | a4 | +2 | bxa4 | -5 |
| 13 | Nbd2 | -3 | O-O | -8 |
| 14 | Rg3 | -14 | Kh8 | -20 |
| 15 | Nc4 | -6 | Nc8 | 0 |
| 16 | Rxa4 | 0 | Nb6 | 0 |
| 17 | Ra6 | 0 | Nxc4 | 0 |
| 18 | dxc4 | -2 | Qe7 | -4 |
| 19 | Ne5 | -2 | Nb4 | -4 |
| 20 | Qg4 | -8 | Rg8 | -13 |
| 21 | Ra1 | -8 | a5 | 0 |
| 22 | Rga3 | -12 | Nc6 | -8 |
| 23 | Nd3 | 0 | e5 | -1 |
| 24 | Qh3 | 0 | Bd7 | 0 |
| 25 | Bg4 | 0 | Bxg4 | 0 |
| 26 | Qxg4 | 0 | Rgb8 | 0 |
| 27 | Qf5 | 0 | Qc7 | 0 |
| 28 | g4 | 0 | f6 | 0 |
| 29 | Qf3 | 0 | Ra7 | 0 |
| 30 | Ra4 | 0 | Rba8 | 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 41.2 moves.
Sample game 1 (1/2-1/2)
1. e4 c5 2. d3 f6 3. g4 e6 4. c4 Ne7 5. Qg2 Nbc6 6. O-O O-O 7. f4 d5 8. cxd5 exd5 9. f5 Bc7 10. Nc3 Kh8 11. Kh1 dxe4 12. dxe4 Nd4 13. Be3 Bc6 14. Nd3 b6 15. Bxd4 cxd4 16. Ne2 g5 17. Nxd4 Rd8 18. Nxc6 Nxc6 19. Nf2 Rd2 20. Rc1 Ne5 21. Bb3 Nd3 22. e5 Rxf2 23. Rxf2 Nxc1 24. Rc2 Nxb3 25. Rxc7 Qd8 26. Qb7 Qd1+ 27. Kg2 Qxg4+ 28. Kh1 Qd1+ 29. Kg2 Qe2+ 30. Kg1 Qe1+ 31. Kg2 Qd2+ 32. Kh1 Qd1+ 33. Kg2 Qd2+ 34. Kg3 Qd3+ 35. Kg2
Sample game 2 (1/2-1/2)
1. d3 c5 2. e4 Nc6 3. g4 e6 4. h4 Nge7 5. h5 O-O 6. g5 d5 7. Qg2 Bc7 8. a4 Qd8 9. Nd2 Kh8 10. h6 g6 11. c3 f5 12. gxf6 Ng8 13. Nb3 Qd6 14. Be3 b6 15. Nd2 Nxf6 16. Bg5 Nd8 17. Rh3 Nc6 18. Kg1 Bd8 19. f4 Qc7 20. Qg3 dxe4 21. dxe4 Nh5 22. Bxh5 gxh5 23. Ng2 Rg8 24. Qh4 Bxg5 25. fxg5 Qe7 26. Rg3 Bg6 27. Rf1 Raf8 28. Rf6 Ne5 29. Qf4 Ng4 30. e5 Qd8 31. Rxg4 hxg4 32. Nc4 Qd1+ 33. Kf2 Bf5 34. Nd6 a6 35. Kg3 b5 36. a5 b4 37. cxb4 cxb4 38. Qf2 Qb3+ 39. Ne3 Qd3 40. Rf7 Qd4 41. Qf4 Qxb2 42. g6 hxg6 43. h7 g5 44. hxg8=R+ Rxg8 45. Qf2 Qxe5+ 46. Kg2 Bg6 47. Re7 Rg7 48. Ndc4 Qf4 49. Rxe6 Be4+ 50. Kg1 g3 51. Qxf4 gxf4 52. Rxe4 fxe3 53. Nxe3 Rb7 54. Nd5 Rb5 55. Nxb4 Rxa5 56. Re7 Kg8 57. Kg2 Rg5 58. Ra7 Rg7 59. Rxa6 Rg4 60. Nd5 Kf7 61. Ne3 Rg5 62. Ra4 Ke6 63. Ng4 Kd6 64. Kxg3 Kc6 65. Kf4 Kb5 66. Ra1 Rg6 67. Ne3 Re6 68. Rc1 Rc6 69. Rb1+ Kc5 70. Rc1+ Kb5 71. Rb1+ Kc5 72. Rc1+
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 453 / 521 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): rnqbbknr/pppppppp/8/8/8/8/PPPPPPPP/RNBBNKQR w AHah - 0 1
Mirror twin: White 698 vs Black 537 — a different position (castling squares are fixed), not evaluated by the search.