♚Open in the app

White 458 vs Black 335

White starts from setup 458 (RNNKBBQR), Black from setup 335 (NRQKRNBB). In the encyclopedia since 2026-10-09: neither side starts better.

|cp| ≤ 5 at 400M · 800M · 1.6B · 4B nodes, two independent engines agree

The two setups

White: setup #458 R N N K B B Q R

King on the d-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 458 gives White +0.30 (Morozov Balance Index, 400M nodes).

Black: setup #335 N R Q K R N B B

King on the d-file, rooks on b, e, queen on the c-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 335 gives White +0.37 (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 same corridor is required at every deeper budget (8 G, 16 G) at which the pair has been evaluated.

The shallow rungs wandered between -1 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 458 gives White +0.30 and setup 335 gives White +0.37: 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 f3, and 2 first moves are within 10 cp of it.

Along the engine's main line (28 moves) the largest swing was 32 cp (around move 1), 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 11 -1 1. Nc3 g6 2. g4 b5 3. e4 b4 4. Na4 d6
S1 500 000 19 0 1. Nc3 g6 2. f4 b5 3. e4 b4 4. Na4 Nb6
S2 2 000 000 24 +35 1. d4 g6 2. f3 f5 3. Nb3 d6 4. Nc3 c6
S3 8 000 000 26 +19 1. d4 g6 2. f3 f5 3. e4 d6 4. Nc3 fxe4
S4 32 000 000 30 +7 1. d4 g6 2. f3 d6 3. a4 c5 4. Bc3 f5
S5 128 000 000 38 0 1. d4 g6 2. f3 f5 3. Nd3 Bc4 4. e3 b5
S6 400 000 000 49 0 1. d4 g6 2. f3 f5 3. a4 c5 4. Bc3 cxd4
S7 800 000 000 53 0 1. d4 g6 2. f3 f5 3. e4 d6 4. Nd2 Nb6
S8 1 600 000 000 57 0 1. d4 g6 2. f3 f5 3. e4 d6 4. Nd2 Nb6

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: +14

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 +43, Black 0, difference +43. 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 +3 39 1. f3 g6 2. d4 f5 3. a4 c5 4. Bc3 cxd4 5. Bxd4 Bxd4
2 0 39 1. d4 g6 2. f3 f5 3. a4 c5 4. Bc3 cxd4 5. Bxd4 Bxd4
3 -18 38 1. Nc3 c5 2. e4 g6 3. f4 f5 4. e5 Ne6 5. Qe3 Nb6

Main-line play-out

28 moves, 5 000 000 nodes per half-move. Largest |evaluation| 32 cp, mean 4.0 cp, final 0 cp.

# White cp Black cp
1 d4 +32 g6 +15
2 f3 0 f5 -7
3 e4 -14 d6 0
4 Nc3 -2 fxe4 -10
5 fxe4 -3 c5 -9
6 dxc5 -7 Qxc5 0
7 Qxc5 -8 dxc5 -8
8 Nd3 -14 c4 -8
9 Nb4 -11 O-O-O+ -3
10 Kc1 0 Bd4 -1
11 Be2 0 Nd7 -5
12 a4 -10 a5 -4
13 Nbd5 +3 e6 0
14 Nf4 +2 Nc5 +11
15 Kb1 +7 Bxc3 0
16 Bxc3 -3 e5 0
17 Nh3 +1 b6 0
18 Nf2 +4 Nc7 +5
19 Ra3 +6 N7a6 +8
20 Bf3 0 h5 0
21 Nd1 +8 Nb4 0
22 Ne3 0 Kb8 0
23 Re1 0 Ka7 0
24 Kc1 0 Rf8 0
25 Bxe5 0 Nb3+ 0
26 Kb1 0 Nd2+ 0
27 Kc1 0 Nb3+ 0
28 Kb1 0 Nd2+ 0
29 Kc1 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 32.4 moves.

Sample game 1 (1/2-1/2)
1. a4 g6 2. Nc3 f5 3. g4 f4 4. e3 Ne6 5. exf4 Nxf4 6. Qg3 Rf8 7. N1e2 Nxe2 8. Bxe2 d6 9. f4 a6 10. Rf1 Nb6 11. a5 Nd5 12. f5 Qd7 13. Qd3 Nb4 14. Qe4 Nc6 15. fxg6 Rxf1 16. Bxf1 hxg6 17. Qxg6 O-O-O 18. h3 d5 19. Bg2 Be6 20. Bg3 Rg8 21. Qh6 Bg7 22. Qh4 Rh8 23. Qg5 Rg8 24. Qh4 Rh8 25. Qg5 Rg8
Sample game 2 (1/2-1/2)
1. Nc3 c5 2. f4 d6 3. d4 g5 4. dxc5 Qxc5 5. e3 gxf4 6. Nd3 Qb6 7. exf4 f5 8. g4 Qxg1 9. Rxg1 fxg4 10. Rxg4 Be6 11. Rg5 Bf6 12. Rg3 Bh4 13. Rg7 Bf6 14. Rg3 Bh4 15. Re3 Bxe1 16. Nxe1 Bd7 17. O-O-O Ng6 18. Ng2 Nc7 19. Bd3 Rf8 20. Rde1 Bf5 21. Bxf5 Rxf5 22. Rh3 Rf7 23. Rf1 O-O-O 24. f5 Ne5 25. Rh4 Rg8 26. Ne3 Nd7 27. Ne4 Ne8 28. Nd5 Nc7 29. Ne3 Ne8 30. Nd5 Nc7

Play it in the CH960 app

Open this page on a phone with the CH960 app installed, or find pair 458 / 335 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): nrqkrnbb/pppppppp/8/8/8/8/PPPPPPPP/RNNKBBQR w AHbe - 0 1

Mirror twin: White 693 vs Black 336 — a different position (castling squares are fixed), not evaluated by the search.

All pairs of the encyclopedia · All asymmetric setups