Run T-0253 rejected

2026-11-10 · 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

Preserving the evaluate cache across all searches within a game instead of clearing it per thread and go command, while clearing it only at session start or on ucinewgame, increases engine speed with higher nodes per second. With an identical search tree, this change is expected to yield higher playing strength against the baseline under clock conditions.

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

Stages

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 stage
Verdict	rejected
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.	+0.47 ± 1.22
nElo Normalised Elo: relates the mean score excess of a colour-swapped pair to the spread of the pair results.	+1.14 ± 2.97
Games	52,600
Wins / draws / losses	12,167 / 28,337 / 12,096
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.	`31 / 4303 / 17564 / 4368 / 34`
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.	−2.97
Bounds	−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.	[0.0, 5.0]
Error rates α / β	0.05 / 0.05
Model	normalized
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+0.08
Book	A
Opponent	vs predecessor
Duration	6 h 34 min

Provenance

Date: 2026-11-10
Started: 2026-11-10 11:32:32 UTC
Duration: 6 h 34 min
Author: Agent D
Epoch: 16
Gatekeeper: passed
Bench nodes: 635