White 248 vs Black 647
White starts from setup 248 (NBRKBQNR), Black from setup 647 (RNBKRQNB). 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. 248, Black by position No. 647, 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 #248 N B R K B Q N R
King on the d-file, rooks on c, h, queen on the f-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 248 gives White +0.32 (Morozov Balance Index, 400M nodes).
Black: setup #647 R N B K R Q N B
King on the d-file, rooks on a, e, queen on the f-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 647 gives White +0.15 (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 59.
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 -18 and -3 cp before settling — a reminder that quick engine checks overstate imbalances, and why the ladder goes deep.
For comparison, in symmetric play setup 248 gives White +0.32 and setup 647 gives White +0.15: 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 ±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 | -16 | 1. e4 g6 2. O-O-O f5 3. exf5 gxf5 4. f4 a5 |
| S1 | 500 000 | 21 | -3 | 1. e4 g6 2. O-O-O f5 3. exf5 gxf5 4. f4 d6 |
| S2 | 2 000 000 | 25 | -11 | 1. e4 d6 2. d4 e5 3. dxe5 Rxe5 4. c3 f5 |
| S3 | 8 000 000 | 33 | -14 | 1. e4 d6 2. d4 e5 3. dxe5 Rxe5 4. c3 f5 |
| S4 | 32 000 000 | 39 | -18 | 1. e4 d6 2. d4 e5 3. dxe5 Rxe5 4. c3 f5 |
| S5 | 128 000 000 | 40 | -6 | 1. f4 g6 2. c3 f5 3. Nf3 d6 4. Bh4 Nf6 |
| S6 | 400 000 000 | 45 | 0 | 1. e4 d6 2. d4 g6 3. O-O-O e5 4. dxe5 Bxe5 |
| S7 | 800 000 000 | 52 | 0 | 1. f4 g6 2. c3 f5 3. d4 Nf6 4. Bh4 b6 |
| S8 | 1 600 000 000 | 59 | 0 | 1. f4 g6 2. c3 f5 3. d4 Nf6 4. Bh4 b6 |
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: -2
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 +47, Black +67, difference -20. 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 | +6 | 46 | 1. f4 g6 2. c3 f5 3. d4 Nf6 4. Bh4 a5 5. Nf3 Ne4 |
| 2 | 0 | 46 | 1. c3 g6 2. f4 f5 3. d4 Nf6 4. Bh4 a5 5. Nf3 Ne4 |
| 3 | 0 | 46 | 1. d4 g6 2. c3 f5 3. f4 Nf6 4. Bh4 a5 5. Nf3 Ne4 |
Main-line play-out
30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 24 cp, mean 5.6 cp, final 0 cp.
| # | White | cp | Black | cp |
|---|---|---|---|---|
| 1 | e4 | -15 | d6 | -12 |
| 2 | d4 | -19 | e5 | -13 |
| 3 | d5 | -1 | f5 | +18 |
| 4 | c4 | +21 | g6 | +18 |
| 5 | Ne2 | +15 | Nf6 | +6 |
| 6 | Ng3 | +15 | a5 | +21 |
| 7 | f3 | -6 | a4 | 0 |
| 8 | exf5 | 0 | gxf5 | -17 |
| 9 | Bxf5 | 0 | Bxf5 | -24 |
| 10 | Nxf5 | -24 | Nxd5 | -17 |
| 11 | cxd5 | -16 | Qxf5 | -18 |
| 12 | Bh4+ | -10 | Kc8 | -10 |
| 13 | Qc4 | -6 | Na6 | 0 |
| 14 | O-O | -3 | Nc5 | -6 |
| 15 | Bf2 | -4 | b6 | 0 |
| 16 | Nc2 | 0 | Qg6 | 0 |
| 17 | Rcd1 | 0 | e4 | 0 |
| 18 | f4 | 0 | Bxb2 | 0 |
| 19 | f5 | 0 | Qf7 | 0 |
| 20 | Bxc5 | 0 | dxc5 | 0 |
| 21 | Qb5 | 0 | a3 | 0 |
| 22 | Qc6 | 0 | Ra5 | 0 |
| 23 | f6 | 0 | Rf8 | 0 |
| 24 | d6 | 0 | Qd7 | 0 |
| 25 | Qxe4 | 0 | Ra4 | 0 |
| 26 | Qd5 | 0 | Rxf6 | 0 |
| 27 | Rxf6 | 0 | Bxf6 | 0 |
| 28 | Qf3 | 0 | Bg7 | 0 |
| 29 | dxc7 | 0 | Bd4+ | 0 |
| 30 | Kh1 | 0 | Kxc7 | 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 52.7 moves.
Sample game 1 (1/2-1/2)
1. e4 g6 2. c3 d6 3. O-O-O f5 4. exf5 Bxf5 5. Bxf5 gxf5 6. f4 Nd7 7. Nc2 O-O-O 8. Kb1 Kb8 9. Nf3 e6 10. Bh4 Bf6 11. g3 a6 12. Ne3 Nc5 13. d3 e5 14. fxe5 dxe5 15. Ng5 Bxg5 16. Bxg5 Ne7 17. Bxe7 Rxe7 18. Nxf5 Red7 19. d4 h5 20. Qf3 Rf7 21. dxc5 Rxd1+ 22. Rxd1 Rxf5 23. Qd3 Ka7 24. Kc2 Rf2+ 25. Rd2 Qxc5 26. b4 Rxd2+ 27. Kxd2 Qf2+ 28. Qe2 Qf5 29. Qc4 Kb8 30. Qg8+ Ka7 31. Qc4 Kb8 32. a4 Qb1 33. Qg8+ Ka7 34. Qc4 Qb2+ 35. Ke3 Qxh2 36. Qc5+ Kb8 37. Qf8+ Ka7 38. Qc5+ Kb8 39. Qxe5 b6 40. a5 Qg1+
Sample game 2 (1/2-1/2)
1. e4 g6 2. c3 d6 3. O-O-O f5 4. exf5 Bxf5 5. Bxf5 gxf5 6. f4 Nd7 7. Nc2 O-O-O 8. Kb1 Kb8 9. Nf3 e6 10. Bh4 Bf6 11. g3 a6 12. Ne3 Nc5 13. d3 e5 14. fxe5 dxe5 15. Ng5 Bxg5 16. Bxg5 Ne7 17. Bxe7 Rxe7 18. Nxf5 Red7 19. d4 h5 20. Qf3 Rf7 21. dxc5 Rxd1+ 22. Rxd1 Rxf5 23. Qd3 Ka7 24. Kc2 Rf2+ 25. Rd2 Qxc5 26. b4 Rxd2+ 27. Kxd2 Qf2+ 28. Qe2 Qf5 29. Qc4 Kb8 30. Qg8+ Ka7 31. Qc4 Kb8 32. a4 Qb1 33. Qg8+ Ka7 34. Qc4 Qb2+ 35. Ke3 Qxh2 36. Qc5+ Kb8 37. Qf8+ Ka7 38. Qc5+ Kb8 39. Qxe5 b6 40. a5 Qg1+
Play it in the CH960 app
Open this page on a phone with the CH960 app installed, or find pair 248 / 647 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): rnbkrqnb/pppppppp/8/8/8/8/PPPPPPPP/NBRKBQNR w CHae - 0 1
Mirror twin: White 621 vs Black 210 — a different position (castling squares are fixed), not evaluated by the search.