♚Open in the app

White 505 vs Black 641

White starts from setup 505 (RQNBBKNR), Black from setup 641 (BRNBKRQN). In the encyclopedia since 2026-10-21: 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. 505, Black by position No. 641, mirrored on their own side; the pawns stand as usual.

Black · No. 641
White · No. 505

Moves as usual: White starts; everything else follows the Chess960 rules.

Printable card (A4, landscape)

The two setups

White: setup #505 R Q N B B K N R

King on the f-file, rooks on a, h, queen on the b-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 505 gives White +0.26 (Morozov Balance Index, 400M nodes).

Black: setup #641 B R N B K R Q N

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 641 gives White +0.16 (Morozov Balance Index, 400M nodes).

Why the pair is balanced

At the deepest rungs (400M, 800M, 1.6B nodes) the evaluation never left ±5 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 agreed early: from 100 thousand to 128 million nodes the evaluation never left ±13 cp.

For comparison, in symmetric play setup 505 gives White +0.26 and setup 641 gives White +0.16: with both sides on the same rank the first move is worth something. In this pair Black's own setup absorbs that edge completely.

The engine's first choice for White is h4; the alternatives are slightly weaker but keep the game level.

Along the engine's main line (30 moves) the largest swing was 34 cp (around move 16), ending at 0 cp — the position stays playable for both sides.

Engine self-play agrees: 10 games, White won 1, drawn 9, Black won 0 (White scored 55.0%).

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 -3 1. c4 f5 2. d4 Qxc4 3. Nf3 e6 4. Bb3 Qc6
S1 500 000 19 0 1. h4 c5 2. c3 f5 3. d4 b5 4. Bc2 Qd5
S2 2 000 000 24 +13 1. h4 c5 2. c3 f5 3. d4 b5 4. Bc2 Qd5
S3 8 000 000 27 -1 1. h4 b6 2. e3 e5 3. Nce2 Nd6 4. c4 Nxc4
S4 32 000 000 33 -5 1. h4 b6 2. e3 e5 3. Nce2 f5 4. c3 c5
S5 128 000 000 43 0 1. h4 b6 2. e3 f5 3. d4 c5 4. Nd3 Nd6
S6 400 000 000 51 0 1. h4 b6 2. e3 f5 3. d4 c5 4. Nd3 Nd6
S7 800 000 000 50 0 1. h4 b6 2. e3 f5 3. d4 c5 4. Nd3 Nd6
S8 1 600 000 000 56 -5 1. h4 b6 2. e3 e5 3. d4 Ng6 4. h5 Nh4

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: +3

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 +24, Black +49, difference -25. A balanced pair needs White's setup to be weaker by about the price of the first move.

Extended self-play: 10 games, 1 White wins, 9 draws, 0 Black wins, White scored 55.0%.

Best first moves

The 3 best first moves for White, searched together at 400 000 000 nodes.

# Eval, cp Depth Main line
1 0 41 1. h4 b6 2. e3 e5 3. d4 Ng6 4. h5 Nh4 5. Rh2 g5
2 -33 40 1. d4 g5 2. c4 b6 3. Nd3 f5 4. Bb3 e6 5. Ba4 Bf6
3 -36 40 1. c4 f5 2. Bb3 g5 3. d4 b6 4. Nd3 e6 5. h4 Bf6

Main-line play-out

30 moves, 5 000 000 nodes per half-move. Largest |evaluation| 34 cp, mean 13.3 cp, final 0 cp.

# White cp Black cp
1 h4 +2 b6 -17
2 Nf3 0 c5 -18
3 d4 -17 f5 -16
4 c3 -11 Bc7 -23
5 Nd3 +6 c4 +22
6 Nb4 +22 Bb7 +17
7 a4 +20 Nd6 +20
8 b3 +23 Nhf7 +15
9 bxc4 +18 Nxc4 +13
10 Bb3 +20 Nfd6 +13
11 Ne5 +14 Rc8 +12
12 h5 +16 a5 +15
13 Qa2 +8 Bb8 +12
14 Nbd3 +10 Bd5 +23
15 Rb1 +25 g5 +27
16 hxg6 +34 hxg6 +27
17 f3 +26 g5 +26
18 Bf2 +18 Be6 +15
19 g3 +34 Ne4 +32
20 Bxc4 +15 Bxc4 +10
21 Nxc4 +13 Nxc3 +9
22 Nd6+ +6 Bxd6 +10
23 Qxg8 +10 Rxg8 +12
24 Rxb6 +1 Kf7 0
25 Rh7+ 0 Rg7 0
26 Rxg7+ 0 Kxg7 +6
27 Ra6 +7 Bb4 +5
28 Ra7 +3 Kf6 0
29 Rxd7 +2 Nxa4 +1
30 Ra7 0 Nc3 0

Engine self-play

10 games at 500 000 nodes per move: White won 1, drawn 9, Black won 0 — White scored 55.0%. Average length 44.5 moves.

Sample game 1 (1/2-1/2)
1. h4 f5 2. Nf3 b6 3. d4 c5 4. c4 Qxc4 5. Bb3 Qb5 6. Qc2 Bxf3 7. gxf3 cxd4 8. Nd3 e6 9. Rg1 g6 10. Ba4 Qd5 11. Nb4 Qb7 12. Nc6 b5 13. Nxd8 Kxd8 14. Bb4 Rg8 15. Bb3 Ne7 16. Qd2 Nc6 17. Bd6 Nf7 18. Bxb8 Qxb8 19. a4 Ke7 20. axb5 Qxb5 21. Ba4 Qb4 22. Ke1 Nfe5 23. Bxc6 Nxc6 24. Qxb4+ Nxb4 25. Kd2 a6 26. Ra3 Rb8 27. Rb3 a5 28. Ra1 Rb5 29. f4 Ke8 30. Raa3 Ke7 31. Ra1 h6 32. Ra4 Kf6 33. Rxa5 Rxa5 34. Rxb4 Ke7 35. Rxd4 Ra1 36. b4 Rh1 37. b5 Rb1 38. Ke3 Rb3+ 39. Rd3 Rxb5 40. Kf3 Rb4 41. Rd1 e5 42. fxe5 Rxh4 43. Rd6 Rg4
Sample game 2 (1/2-1/2)
1. h4 c5 2. Nf3 b6 3. c3 f5 4. d4 Bc7 5. Nb3 Qd5 6. c4 Qxc4 7. dxc5 Qd5 8. Qc1 Nf7 9. Nbd4 bxc5 10. Bb3 Rxb3 11. axb3 Bb6 12. Ba5 Bb7 13. Nb5 Qxb3 14. Nc7+ Bxc7 15. Bxc7 d6 16. Ba5 O-O 17. Qb1 e6 18. h5 f4 19. Qa2 Qxa2 20. Rxa2 a6 21. Nd2 Na7 22. Ra1 Rc8 23. Rg1 Nb5 24. g3 fxg3 25. Rxg3 e5 26. f3 Nd4 27. Bc3 Nf5 28. Rg1 d5 29. Kf2 d4 30. Ba5 N7d6 31. b3 Ne3 32. Rac1 Bd5 33. Bb6 c4 34. bxc4 Nexc4 35. Nxc4 Nxc4 36. Ba7 Ra8 37. Bc5 Rc8 38. Bb4 Rc6 39. Rb1 Kf7 40. Be1 Rh6 41. Rg5 Kf6 42. f4 g6 43. Ra1 Bb7 44. Ba5 Nxa5 45. Rxa5 exf4 46. Rg4 Rxh5 47. Rxf4+ Rf5 48. Raxf5+ gxf5 49. Rxd4 Ke5

Play it in the CH960 app

Open this page on a phone with the CH960 app installed, or find pair 505 / 641 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): brnbkrqn/pppppppp/8/8/8/8/PPPPPPPP/RQNBBKNR w AHbf - 0 1

Mirror twin: White 553 vs Black 219 — a different position (castling squares are fixed), not evaluated by the search.

All pairs of the encyclopedia · All asymmetric setups