White 520 vs Black 284
White starts from setup 520 (RBNQBKNR), Black from setup 284 (NBRKNRBQ). 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. 520, Black by position No. 284, 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 #520 R B N Q B K N R
King on the f-file, rooks on a, h, queen on the d-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 520 gives White +0.19 (Morozov Balance Index, 400M nodes).
Black: setup #284 N B R K N R B Q
King on the d-file, rooks on c, f, queen on the h-file; no bishop starts in a corner; the knights start away from the centre. Castling follows the Chess960 rule: king and rook end on the classical squares. In ordinary symmetric play setup 284 gives White +0.19 (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 52.
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 -28 and +6 cp before settling — a reminder that quick engine checks overstate imbalances, and why the ladder goes deep.
For comparison, in symmetric play setup 520 gives White +0.19 and setup 284 gives White +0.19: 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 Nf3, and 2 first moves are within 10 cp of it.
Along the engine's main line (30 moves) the evaluation stayed within ±30 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 8, Black won 2 (White scored 40.0%).
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 | +6 | 1. Nf3 c6 2. d4 f5 3. Bd2 O-O-O 4. O-O Bd5 |
| S1 | 500 000 | 17 | +2 | 1. Nf3 f5 2. d4 g5 3. Nxg5 Qg7 4. Nf3 c5 |
| S2 | 2 000 000 | 19 | -17 | 1. c3 f5 2. Nf3 c5 3. d4 cxd4 4. cxd4 g5 |
| S3 | 8 000 000 | 27 | -28 | 1. d4 f5 2. h4 g6 3. a4 c5 4. c3 c4 |
| S4 | 32 000 000 | 33 | -24 | 1. c3 f5 2. Nf3 c5 3. O-O e5 4. d3 O-O-O |
| S5 | 128 000 000 | 36 | -10 | 1. d4 f5 2. Nf3 c6 3. a4 O-O-O 4. Bd2 Nd6 |
| S6 | 400 000 000 | 43 | 0 | 1. d4 f5 2. Nf3 c6 3. O-O O-O-O 4. e3 g6 |
| S7 | 800 000 000 | 51 | 0 | 1. d4 f5 2. Nf3 c6 3. e3 O-O-O 4. O-O g6 |
| S8 | 1 600 000 000 | 52 | 0 | 1. d4 f5 2. Nf3 c6 3. e3 O-O-O 4. O-O g6 |
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: 0
- 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 +48, Black +33, difference +15. 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, 8 draws, 2 Black wins, White scored 40.0%.
Best first moves
The 3 best first moves for White, searched together at 400 000 000 nodes.
| # | Eval, cp | Depth | Main line |
|---|---|---|---|
| 1 | +7 | 40 | 1. Nf3 f5 2. d4 c6 3. c3 g6 4. O-O O-O-O 5. Nd3 d6 |
| 2 | +2 | 40 | 1. d4 f5 2. Nf3 c6 3. c3 g6 4. O-O O-O-O 5. Nd3 d6 |
| 3 | -13 | 39 | 1. c3 f5 2. Nf3 c5 3. d4 g5 4. dxc5 Rxc5 5. Qd4 Qxd4 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 30 cp, mean 11.7 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | d4 | -25 | f5 | -28 |
| 2 | Nf3 | +2 | c6 | +6 |
| 3 | a4 | -6 | O-O-O | -24 |
| 4 | Ba2 | -13 | Bxa2 | -6 |
| 5 | Nxa2 | -4 | Qg8 | -30 |
| 6 | Bb4 | -14 | d6 | -12 |
| 7 | Qc1 | -15 | a6 | -19 |
| 8 | c4 | -15 | Ba7 | -19 |
| 9 | Qc2 | -16 | g5 | -18 |
| 10 | e3 | -20 | f4 | -17 |
| 11 | O-O-O | -19 | Nb6 | -22 |
| 12 | Nd2 | -22 | Nc7 | -24 |
| 13 | Rhf1 | -16 | Kb8 | -17 |
| 14 | Kb1 | -20 | h5 | -19 |
| 15 | Nc1 | -13 | Ka8 | -14 |
| 16 | g3 | -10 | fxe3 | -9 |
| 17 | fxe3 | -8 | Ne6 | -13 |
| 18 | Ba5 | -14 | Rc8 | -15 |
| 19 | Ne2 | -10 | Nd7 | -17 |
| 20 | h3 | -13 | Bb8 | -16 |
| 21 | Bc3 | -15 | Nc7 | -21 |
| 22 | Qd3 | -13 | Rxf1 | -5 |
| 23 | Rxf1 | -4 | e5 | 0 |
| 24 | dxe5 | -11 | Nxe5 | -7 |
| 25 | Bxe5 | -6 | dxe5 | 0 |
| 26 | Qf5 | 0 | Ba7 | 0 |
| 27 | Ne4 | 0 | Bxe3 | 0 |
| 28 | a5 | 0 | Rd8 | 0 |
| 29 | Qf7 | 0 | Qxf7 | 0 |
| 30 | Rxf7 | 0 | Ne6 | 0 |
Engine self-play
10 games at 500 000 nodes per move: White won 0, drawn 8, Black won 2 — White scored 40.0%. Average length 47.4 moves.
Sample game 1 (1/2-1/2)
1. d4 g5 2. c3 f5 3. Nf3 Qg7 4. Bd2 h6 5. O-O c6 6. Bd3 Nb6 7. b3 O-O-O 8. a4 d5 9. a5 Nd7 10. a6 b6 11. Qc2 Qf6 12. g3 Bc7 13. c4 e5 14. dxe5 Bxe5 15. Rb1 Nc7 16. cxd5 Bxd5 17. b4 f4 18. Nb3 Rde8 19. Bc4 fxg3 20. fxg3 Bxc4 21. Qxc4 Qg6 22. Rbc1 Rf6 23. Nbd4 Bxd4+ 24. Nxd4 Rxf1+ 25. Kxf1 c5 26. bxc5 bxc5 27. Nf3 Qd6 28. Qd3 Qc6 29. h3 Re6 30. Be3 Nxa6 31. Bf2 Nc7 32. Kg1 Qe4 33. Qc3 Nb5 34. Qb3 a6 35. Bxc5 Nxc5 36. Rxc5+ Kb7 37. Rc2 Qe3+ 38. Qxe3 Rxe3 39. Kf2 Re6 40. Nd2 Rf6+ 41. Ke3 Rc6 42. Ra2 Rc3+ 43. Kf2 Kb6 44. Ra4 Na7 45. Rb4+ Nb5 46. Ra4 Nd6 47. Rd4 Nb7 48. Rb4+ Kc7 49. Ne4 Rc6
Sample game 2 (0-1)
1. Nf3 g5 2. c3 Qg7 3. e4 f5 4. d4 fxe4 5. Bxe4 Nd6 6. Bc2 Nc4 7. Qe2 c5 8. O-O Nab6 9. b3 Na3 10. Be4 d5 11. Bd3 Nd7 12. dxc5 e5 13. Bd2 Nxc5 14. Bxg5+ Kd7 15. Bh4 a6 16. Qe3 e4 17. Nd4 exd3 18. Nxd3 Nxd3 19. Qh3+ Kc7 20. Qxd3 Kb6 21. f4 Ka7 22. Rae1 Bd6 23. f5 Qh6 24. Bg3 Bxg3 25. hxg3 Qb6 26. Kh2 Nb5 27. Nxb5+ axb5 28. Rf4 Kb8 29. Qd4 Qf6 30. a4 bxa4 31. Qxa4 Qc6 32. Qa5 Qxc3 33. Rb4 Qc7 34. Qa6 Rxf5 35. Rxb7+ Qxb7 36. Qd6+ Rc7 37. Re8+ Ka7 38. Qa3+ Kb6 39. Qb4+ Kc6 40. Qa4+ Qb5 41. Qa8+ Kd6 42. g4 Bf7 43. Rd8+ Ke6 44. gxf5+ Kxf5 45. Rc8 Qc5 46. Rxc7 Qxc7+ 47. Kg1 d4 48. Qa1 Qe5 49. Qa7 Kg6 50. Qb6+ Kg5 51. Qd8+ Kg4 52. Qc8+ Be6 53. Qb7 Bd5 54. Qc8+ Qe6 55. Qc1 Qe3+ 56. Qxe3 dxe3 57. Kf1 Kf4 58. b4
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 520 / 284 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): nbrknrbq/pppppppp/8/8/8/8/PPPPPPPP/RBNQBKNR w AHcf - 0 1
Mirror twin: White 541 vs Black 588 — a different position (castling squares are fixed), not evaluated by the search.