White 183 vs Black 899
White starts from setup 183 (NRBNKRQB), Black from setup 899 (BRKQRNNB). In the encyclopedia since 2026-10-21: neither side starts better.
|cp| ≤ 5 at 400M · 800M · 1.6B · 4B nodes, two independent engines agree
The two setups
White: setup #183 N R B N K R Q B
King on the e-file, rooks on b, f, queen on the g-file; one 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 183 gives White +0.17 (Morozov Balance Index, 400M nodes).
Black: setup #899 B R K Q R N N B
King on the c-file, rooks on b, e, queen on the d-file; one bishop starts in a corner; the knights start on central files. The king already stands on a castled square, so one of the castlings moves only the rook. In ordinary symmetric play setup 899 gives White +0.27 (Morozov Balance Index, 400M nodes).
Why the pair is balanced
At the deepest rungs (400M, 800M, 1.6B nodes) the evaluation never left ±3 centipawns of equality, reaching search depth 56.
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 -2 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 183 gives White +0.17 and setup 899 gives White +0.27: 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 f4, and 3 first moves are within 10 cp of it.
Along the engine's main line (30 moves) the evaluation stayed within ±22 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 | 14 | +35 | 1. g3 e5 2. f4 exf4 3. Rxf4 g5 4. Rf2 d5 |
| S1 | 500 000 | 17 | -2 | 1. f4 b6 2. g3 Bxh1 3. Qxh1 g6 4. Qd5 Nf6 |
| S2 | 2 000 000 | 24 | +10 | 1. g4 e5 2. d3 g6 3. f4 b6 4. Bxa8 Rxa8 |
| S3 | 8 000 000 | 27 | +15 | 1. g3 e5 2. Nc3 g6 3. d3 b6 4. Bxa8 Rxa8 |
| S4 | 32 000 000 | 35 | +12 | 1. g3 e5 2. Nc3 g6 3. d3 Ne6 4. Nb3 b6 |
| S5 | 128 000 000 | 38 | 0 | 1. g3 e5 2. Nc3 g6 3. d3 Ne6 4. Nb3 b6 |
| S6 | 400 000 000 | 46 | -2 | 1. g3 e5 2. Nc3 g6 3. d3 Ne6 4. Nb3 f5 |
| S7 | 800 000 000 | 49 | +3 | 1. f4 b6 2. d3 g6 3. Bd2 f5 4. Nf2 Nf6 |
| S8 | 1 600 000 000 | 56 | -3 | 1. f4 b6 2. d3 g6 3. Bd2 f5 4. Nf2 e5 |
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: -8
- Lc0 0.32.1: +8
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 +56, Black +82, difference -26. 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 | -1 | 43 | 1. f4 b6 2. d3 g6 3. Bd2 f5 4. Nf2 e5 5. fxe5 Qe7 |
| 2 | -9 | 43 | 1. d3 b6 2. f4 g6 3. Bd2 f5 4. Nf2 e5 5. fxe5 Qe7 |
| 3 | -5 | 42 | 1. g3 e5 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 22 cp, mean 4.4 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | g3 | -9 | e5 | -1 |
| 2 | Nc3 | -1 | g6 | +2 |
| 3 | d3 | +7 | Ne6 | +7 |
| 4 | f4 | -5 | exf4 | -17 |
| 5 | Bxf4 | -14 | Bxc3+ | -11 |
| 6 | bxc3 | -7 | b6 | -13 |
| 7 | Bxa8 | -13 | Rxa8 | -13 |
| 8 | Bd2 | -16 | f5 | -17 |
| 9 | O-O-O | -18 | Ne7 | -22 |
| 10 | g4 | -15 | a5 | -14 |
| 11 | gxf5 | -4 | gxf5 | -4 |
| 12 | Nb3 | 0 | Kb7 | -9 |
| 13 | a4 | 0 | d5 | -2 |
| 14 | e4 | 0 | fxe4 | 0 |
| 15 | dxe4 | 0 | Qd6 | -5 |
| 16 | exd5 | -9 | Qxd5 | 0 |
| 17 | Qg4 | -3 | Nc5 | -4 |
| 18 | Bg5 | 0 | Nxb3+ | 0 |
| 19 | Kb2 | 0 | Qg8 | 0 |
| 20 | cxb3 | 0 | Nc6 | 0 |
| 21 | Rf6 | 0 | Rad8 | 0 |
| 22 | Rxd8 | 0 | Nxd8 | 0 |
| 23 | h4 | 0 | Re6 | 0 |
| 24 | Qf3+ | 0 | Nc6 | 0 |
| 25 | Rxe6 | 0 | Qxe6 | 0 |
| 26 | h5 | 0 | Ka7 | 0 |
| 27 | h6 | 0 | Nb8 | 0 |
| 28 | Qd3 | 0 | Qf7 | 0 |
| 29 | Qe4 | 0 | Na6 | 0 |
| 30 | Be3 | 0 | Nc5 | 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 35.4 moves.
Sample game 1 (1/2-1/2)
1. f4 b6 2. c3 g6 3. Ne3 f5 4. d3 e5 5. fxe5 Bxe5 6. Nac2 Qe7 7. g3 Bxh1 8. Qxh1 Ne6 9. Bd2 b5 10. Nd5 Qf7 11. g4 c6 12. gxf5 gxf5 13. Nde3 f4 14. O-O-O Nf6 15. Nf5 Kb7 16. Kb1 Qh5 17. e4 fxe3 18. Ncxe3 Qxh2 19. Qxh2 Bxh2 20. Ng7 Nxg7 21. Rxf6 Re7 22. Rh6 Bf4 23. Rxh7 Rbe8 24. Nd5 cxd5 25. Bxf4 Ne6 26. Rh4 Rf7 27. Bh2 Rg8 28. c4 Rg2 29. Rh8 Nc7 30. cxb5 Rff2 31. Be5 Re2 32. Bf6 Rgf2 33. Bd4 Rg2 34. Bf6 Nxb5 35. a4 Ref2 36. Be5 d6 37. Rh7+ Kb8 38. Bh8 d4 39. axb5 Rxb2+ 40. Ka1 Ra2+
Sample game 2 (1/2-1/2)
1. c4 g6 2. g3 e5 3. b3 f5 4. Bb2 Nf6 5. Nc2 b6 6. Nc3 Ne6 7. d3 Bxh1 8. Qxh1 Qe7 9. O-O-O O-O 10. Kb1 c6 11. Ba3 c5 12. Bb2 d6 13. e3 Qb7 14. Qxb7 Rxb7 15. f3 Bg7 16. d4 Rd7 17. dxc5 dxc5 18. Rxd7 Nxd7 19. Rd1 Nf6 20. Nb5 a6 21. Nc3 Rd8 22. Rxd8+ Nxd8 23. e4 Nc6 24. Kc1 Nd4 25. Ne1 fxe4 26. Nxe4 Nxe4 27. fxe4 Kf7 28. Kd1 Ke6 29. Bc1 Bf6 30. Bd2 Kd7 31. Bc1 Kd6 32. Nd3 Nf3 33. h3 Ng1 34. Nf2 Nf3 35. Ke2 Nd4+ 36. Kf1 Nf3 37. Ke2 Nd4+ 38. Kd1 Nc6 39. Nd3 Bd8 40. Be3 Kc7 41. Ke2 Kd6 42. Bd2 Nd4+
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 183 / 899 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): brkqrnnb/pppppppp/8/8/8/8/PPPPPPPP/NRBNKRQB w BFbe - 0 1
Mirror twin: White 770 vs Black 51 — a different position (castling squares are fixed), not evaluated by the search.