Dataset Viewer
Auto-converted to Parquet Duplicate
generatedAt
string
engine
string
initMs
int64
cubesPerDepth
int64
rows
list
2026-09-13T14:19:30.127Z
min2phase (Kociemba two-phase)
78
200
[ { "scrambleDepth": 10, "cubes": 200, "meanSolutionMoves": 12.4, "medianSolutionMoves": 10, "maxSolutionMoves": 21, "meanSolveMs": 0.1 }, { "scrambleDepth": 15, "cubes": 200, "meanSolutionMoves": 19.2, "medianSolutionMoves": 20, "maxSolutionMoves": 21, "meanSolveMs...

Rubik's Cube Solver Benchmark — min2phase Solution Length

Solution length and solve time for the LK Forge Rubik's Cube Solver, which uses the min2phase (Kociemba two-phase) algorithm. 200 seeded random scrambles per depth, solving the exact shipped module.

Headline numbers (200 cubes each)

Scramble depth (random turns) Mean solution moves Median Max Mean solve time
10 12.4 10 21 0.10 ms
15 19.2 20 21 0.29 ms
20 20.4 21 21 0.50 ms
25 20.5 21 21 0.56 ms
30 20.5 21 21 0.47 ms
40 20.5 21 21 0.46 ms

Scrambling more does not make the cube harder to solve. Past ~20 random turns the cube is already effectively maximally scrambled, so the solution length plateaus at ~20.5 moves (this build caps solutions at 21 moves in the half-turn metric — consistent with God's Number = 20). Every solve completes in well under a millisecond in plain JavaScript.

Files

  • data/rubik-benchmark.json — per scramble depth: mean/median/max solution moves and mean solve time; plus one-time pruning-table initMs.

Method

  • Engine: the shipped ES modules (solver.js → vendored min2phase.js, scramble.js, constants.js), imported and run unmodified.
  • Scrambles: seeded randomScramble(depth) applied to the solved cube, then solved from the resulting facelet string.
  • Solution moves counted in the half-turn metric (HTM), as returned by the solver.

Reproduce

node scripts/benchmark-rubik.mjs

Citation

LK Forge (2026). Rubik's Cube Solver Benchmark — min2phase Solution Length. https://lkforge.com/tools/rubik/

Everything runs client-side, no login, no tracking.

Downloads last month
15