Records
Null cell: uniform setups against uniform setups, greedy in both seats
Registration 2511aae8-dff0-4534-919e-abe3e434d69b (study setups/structured-prior), run 2026-09-19 on AMD Ryzen AI Max (Strix Halo), 32 threads, 125 GB unified memory, Radeon 8060S via ROCm, Linux 6.18. Server commit f64c784, research commit 79af9a4.
Command:
cd server && cargo run --release -p armies-arena --bin matchup -- --red 'greedy-uniform-a=greedy:{"setup":"uniform"}' --blue 'greedy-uniform-b=greedy:{"setup":"uniform"}' --games 400 --seed 0 --rules training --threads 8 --moves --out ../research/runs/setup-prior/null.json --record-dir ../research/runs/setup-prior/records-null --tag 2511aae8-dff0-4534-919e-abe3e434d69b
cd research && python3 scripts/paired_contrast.py runs/setup-prior/null.json| Player | Wins | Draws | Losses | Effective | 95% Wilson | Mean plies | ms/move |
|---|---|---|---|---|---|---|---|
| greedy-uniform-a | 185 | 17 | 198 | 0.484 | 0.435 to 0.533 | 152 | 0.002 |
| greedy-uniform-b | 198 | 17 | 185 | 0.516 | 0.467 to 0.565 | 152 | 0.002 |
| ending: flag_captured | 380 | ||||||
| ending: no_moves | 3 | ||||||
| ending: quiet_limit | 17 |
Per-seed paired contrast for contestant a (score as Red plus score as Blue minus one): mean minus 0.033, 95% normal interval minus 0.143 to 0.078 over 200 seeds (68 positive, 61 split, 71 negative). The interval covers zero, as the registration requires. Contestant a won 89 games as Red and 96 as Blue. Draw share 4.2%. Wall time 0.1 s on 8 threads; this run shared the machine with other agents, which does not change deterministic scores. Twenty-nine of the 400 games (7.2%) ended within 20 plies, split 14 to 15 between the two identical contestants: the early-ending baseline of uniform against uniform.
Provenance
Archived JSON: runs/setup-prior/null.json (SHA-256 bcf520c46ff1f8722f69c0d94fc7eade97797bee63c8ab5f27ab8a4325af24fd), 13.9 MB with every game's move list and per-ply detail, so it is not copied next to this record. Games planned 400, attempted 400, completed 400, interrupted 0. No deviation from the registration.