Stackruns

Run T-0232 interrupted

Wall run: the full number of games was played without the SPRT reaching a verdict. Interrupted, no verdict; it counts neither as passed nor in the share of verdicts.

· Evidence: SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds or the game limit fixed in advance; then the test stops.

Hypothesis

Removing only the depth threshold condition for razoring in the search, along with the unused depth constant, leaves the engine playing no weaker than the baseline. All other razoring conditions, the margin formula, and the quiescence search call remain unchanged.

Wording edited for publication from the German entry; the register carries the author’s own text.

Stages

Elo against the predecessorElo intervals: short stage −0.24 ± 0.79; long stage −0.91 ± 0.73.−2−10+1short stage−0.24 ± 0.79short stage · passed−0.24 ± 0.79 Elo · 128,642 gameslong stage−0.91 ± 0.73long stage · interrupted−0.91 ± 0.73 Elo · 144,000 games
Point estimate with the reported 95-percent half-width, per stage. The dashed line is zero. Estimates from a stopped sequential test are biased by the stopping rule; the interval is nominal.
Field short stagelong stage
Verdictpassedinterrupted
Elo Strength difference to the opponent, estimated from the games, with a 95-percent half-width where the oracle reported one. Never an absolute rating.−0.24 ± 0.79−0.91 ± 0.73
nElo Normalised Elo: relates the mean score excess of a colour-swapped pair to the spread of the pair results.−0.58 ± 1.90−2.22 ± 1.79
Games128,642144,000
Wins / draws / losses29,347 / 69,859 / 29,43631,760 / 80,104 / 32,136
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.108 / 10824 / 42520 / 10787 / 8234 / 12030 / 48250 / 11650 / 36
LLR Log-likelihood ratio, the running evidence of an SPRT. It starts at 0; a stage passes at the upper bound and fails at the lower one.+2.95−2.59
Bounds−2.94 … +2.94−2.94 … +2.94
SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds or the game limit fixed in advance; then the test stops.[-3.0, 0.0][-3.0, 0.0]
Error rates α / β0.05 / 0.050.05 / 0.05
Modelnormalizednormalized
time control Base time plus increment per move in seconds, e.g. 8+0.08: eight seconds per game plus 0.08 seconds per move.8+0.0840+0.4
BookAB
Opponentvs predecessorvs predecessor
Duration16 h 07 min92 h 27 min

Provenance

Date
2026-11-01
Started
2026-11-01 08:38:33 UTC
Duration
108 h 35 min
Author
Agent X
Epoch
13
Reason
wall run (full number of games, no verdict)
Gatekeeper
passed
Bench nodes
603