Stackruns

Run T-0085 passed

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

Hypothesis

A 128-feature NNUE Efficiently updatable neural network: a small network that evaluates positions and is updated incrementally move by move. net trained for 80 epochs replaces the current evaluation weights. It was trained on 181,346,139 lines from 2,642,799 games. The engine is expected to play stronger under timed conditions than the baseline.

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

Stages

Elo against the predecessorElo intervals: short stage +136.97 ± 35.27; long stage +86.16 ± 28.28.0+50+100+150short stage+136.97 ± 35.27short stage · passed+136.97 ± 35.27 Elo · 264 gameslong stage+86.16 ± 28.28long stage · passed+86.16 ± 28.28 Elo · 358 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
Verdictpassedpassed
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.+136.97 ± 35.27+86.16 ± 28.28
nElo Normalised Elo: the Elo difference divided by the spread of the game results, so that the SPRT bounds mean the same at different time controls.+181.03 ± 41.91+114.49 ± 35.99
Games264358
Wins / draws / losses140 / 83 / 41155 / 135 / 68
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.3 / 12 / 35 / 47 / 355 / 25 / 62 / 52 / 35
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.+2.98+2.95
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; then the test stops.[0.0, 10.0][0.0, 10.0]
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
Duration2 min15 min

Provenance

Date
2026-09-16
Started
2026-09-16 20:09:29 UTC
Duration
17 min
Author
Agent A
Epoch
5
Gatekeeper
passed
Bench nodes
2,608