Stackruns

Run T-0124 passed

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

Hypothesis

An N384 network with width 384 and a paired AVX2 head, trained over 8 epochs with a learning rate smoothly decaying to zero per step, increases playing strength compared to the baseline using a P256 network when combined with batched accumulator updates that apply all NNUE Efficiently updatable neural network: a small network that evaluates positions and is updated incrementally move by move. changes of a move in a single pass while preserving the search tree.

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

Stages

Elo against the predecessorElo intervals: short stage +54.64 ± 15.01; long stage +58.45 ± 14.72.0+20+40+60+80short stage+54.64 ± 15.01short stage · passed+54.64 ± 15.01 Elo · 936 gameslong stage+58.45 ± 14.72long stage · passed+58.45 ± 14.72 Elo · 840 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.+54.64 ± 15.01+58.45 ± 14.72
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.+82.44 ± 22.26+95.12 ± 23.50
Games936840
Wins / draws / losses333 / 416 / 187300 / 380 / 160
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.9 / 79 / 183 / 151 / 465 / 64 / 167 / 154 / 30
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+2.97
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, 5.0][0.0, 5.0]
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
Duration7 min35 min

Provenance

Date
2026-09-24
Started
2026-09-24 12:59:09 UTC
Duration
42 min
Author
Agent X
Epoch
6
Gatekeeper
passed
Bench nodes
2,904