{"meta":{"board":"Board 1 · run 2","date":"2026-09-27","seeds":"0–9","n_per_game":10,"job":"b1_lx_01_v2","suite":"v3.1","cap_s":60,"levels":6,"tick_ms":250,"stale_ms":1000,"expire_ms":1000,"void_lag_ms":100,"voids":0,"source_sha256":"2173f762133957b8693a49078db7dcae32f79f6337f71435db6095147ab89d76","generated_by":"site/arena/build/make_v3.py"},"games":[{"id":"snake","name":"Snake Rush","what":"Keep a growing snake alive. The clock tightens three screws at once: a hunger timer that forces it to the food, wall cells that appear more and more often, and a speed that rises from 0.5 to 2 cells per tick.","controls":"One control (direction), four options.","anchors":{"naive":[13.75,2],"heuristic":[27.75,3],"planner":[49.88,5],"seeds":48},"wall":{"t":10.0,"note":"10.0 s equals the opening hunger limit (40 ticks at 4 Hz).","cause":null},"clip":{"src":"media/v3-SnakeRush-burst.mp4","poster":"media/v3-SnakeRush-burst.jpg","seed_index":1,"players":["T-dual Burst","Jev","SemIf-4B"]}},{"id":"rotorwash_ramp","name":"Rotorwash-Ramp","what":"Fly a helicopter with a sling load through a cave. A collapse front chases from behind and accelerates, the roof comes down, the storm grows and engine power fades.","controls":"Three simultaneous controls: collective, cyclic, winch. 240 Hz float physics, platform-pinned.","anchors":{"naive":[11.14,2],"heuristic":[24.49,3],"planner":[45.83,5],"seeds":32},"wall":null,"clip":{"src":"media/v3-RotorwashRamp-burst.mp4","poster":"media/v3-RotorwashRamp-burst.jpg","seed_index":2,"players":["T-dual Burst","Jev","SemIf-4B"]}},{"id":"booster_gauntlet","name":"Booster Gauntlet","what":"Land a rocket booster on a drone ship, again and again. Each touchdown re-drops it; thrust-to-weight rises and touchdown limits tighten with the clock. Each hop must land within 22 s of its drop.","controls":"Multi-control timing: ignition, throttle, divert, legs. The less-seen probe (no-hover hoverslam, relights).","anchors":{"naive":[12.53,2],"heuristic":[24.63,3],"planner":[43.83,5],"seeds":32},"wall":{"t":22.0,"note":"22.0 s equals the first hop's landing deadline (a hop must land within 22 s of its drop).","cause":"In Board 1 every round that ended at exactly 22.0 s (25 of 60, including all 10 of T-dual Burst's) was a hop timeout: the booster was still flying under control but had not landed its first hop. None of them was a crash."},"clip":{"src":"media/v3-BoosterGauntlet-burst.mp4","poster":"media/v3-BoosterGauntlet-burst.jpg","seed_index":0,"players":["T-dual Burst","Jev","SemIf-4B"]}}],"players":[{"id":"t_dual/chord","label":"T-dual Burst","sub":"27B · Nagi-ENORMOUS Burst (T-dual weights) · Burst readout: one forward pass sets every control","kind":"nagi","composite":0.7933,"ci95":[0.72,0.8533],"inflation_p50":2.2693,"in_domain":{"snake":["cl:snake"],"rotorwash_ramp":["cl:hover","ma:crane","ma:hover2"],"booster_gauntlet":["cl:hover","ma:hover2"]},"serving":{"gpu":"H100","weights_format":"merged-bf16","kernels":"fused","concurrency_mode":"batched"},"games":{"snake":{"n":10,"n_void":0,"censored":0,"win":0.63,"median_t":10.0,"rmst":10.0,"median_level":1.0,"ladder":0.008,"km":[[0.0,1.0],[4.5,0.9],[6.5,0.8],[9.5,0.7],[10.0,0.3],[10.5,0.2],[13.0,0.1],[16.0,0.0]],"alive_at":{"10":0.3,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":226.7,"fresh":0.65,"stale":0,"expiry":0,"t_fail":{"0":6.5,"1":10.0,"2":9.5,"3":4.5,"4":10.0,"5":10.5,"6":13.0,"7":10.0,"8":10.0,"9":16.0},"at_wall":4},"rotorwash_ramp":{"n":10,"n_void":0,"censored":0,"win":0.9,"median_t":26.038,"rmst":24.23,"median_level":3.0,"ladder":0.457,"km":[[0.0,1.0],[12.57,0.9],[13.06,0.8],[21.28,0.7],[24.75,0.6],[25.67,0.5],[26.4,0.4],[28.37,0.3],[28.87,0.2],[29.75,0.1],[31.56,0.0]],"alive_at":{"10":1.0,"20":0.8,"30":0.1,"40":0.0,"50":0.0,"60":0.0},"p50_ms":241.8,"fresh":0.659,"stale":0,"expiry":0,"t_fail":{"0":24.754,"1":21.283,"2":25.671,"3":26.404,"4":28.871,"5":31.558,"6":13.062,"7":12.575,"8":28.371,"9":29.75},"at_wall":null},"booster_gauntlet":{"n":10,"n_void":0,"censored":0,"win":0.85,"median_t":22.0,"rmst":22.0,"median_level":3.0,"ladder":0.392,"km":[[0.0,1.0],[22.0,0.0]],"alive_at":{"10":1.0,"20":1.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":249.5,"fresh":0.659,"stale":0,"expiry":0,"t_fail":{"0":22.0,"1":22.0,"2":22.0,"3":22.0,"4":22.0,"5":22.0,"6":22.0,"7":22.0,"8":22.0,"9":22.0},"at_wall":10}},"rank":1},{"id":"enormous_cl/kcall","label":"ENORMOUS-CL","sub":"27B · Nagi-ENORMOUS with closed-loop control training · one call per control","kind":"nagi","composite":0.6733,"ci95":[0.5733,0.76],"inflation_p50":2.8599,"in_domain":{"snake":["cl:snake"],"rotorwash_ramp":["cl:hover","ma:crane","ma:hover2"],"booster_gauntlet":["cl:hover","ma:hover2"]},"serving":{"gpu":"H100","weights_format":"merged-bf16","kernels":"fused","concurrency_mode":"batched"},"games":{"snake":{"n":10,"n_void":0,"censored":0,"win":0.74,"median_t":11.0,"rmst":11.65,"median_level":2.0,"ladder":0.037,"km":[[0.0,1.0],[3.5,0.9],[7.0,0.8],[10.0,0.7],[11.0,0.4],[12.5,0.2],[16.0,0.1],[22.0,0.0]],"alive_at":{"10":0.7,"20":0.1,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":208.5,"fresh":0.815,"stale":0,"expiry":0,"t_fail":{"0":12.5,"1":12.5,"2":3.5,"3":11.0,"4":7.0,"5":11.0,"6":11.0,"7":22.0,"8":10.0,"9":16.0},"at_wall":1},"rotorwash_ramp":{"n":10,"n_void":0,"censored":0,"win":0.58,"median_t":17.698,"rmst":17.439,"median_level":2.0,"ladder":0.234,"km":[[0.0,1.0],[11.94,0.9],[12.72,0.8],[14.04,0.7],[16.51,0.6],[17.4,0.5],[18.0,0.4],[18.39,0.3],[19.49,0.2],[19.69,0.1],[26.21,0.0]],"alive_at":{"10":1.0,"20":0.1,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":599.4,"fresh":0.341,"stale":0,"expiry":0,"t_fail":{"0":19.492,"1":14.042,"2":12.721,"3":11.938,"4":16.512,"5":18.392,"6":19.692,"7":26.208,"8":17.996,"9":17.4},"at_wall":null},"booster_gauntlet":{"n":10,"n_void":0,"censored":0,"win":0.7,"median_t":22.0,"rmst":19.835,"median_level":3.0,"ladder":0.302,"km":[[0.0,1.0],[14.8,0.9],[15.3,0.8],[17.25,0.7],[19.0,0.6],[22.0,0.0]],"alive_at":{"10":1.0,"20":0.6,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":701.6,"fresh":0.297,"stale":0,"expiry":0,"t_fail":{"0":22.0,"1":22.0,"2":15.3,"3":19.0,"4":22.0,"5":22.0,"6":17.25,"7":22.0,"8":14.8,"9":22.0},"at_wall":6}},"rank":2},{"id":"t_dual/kcall","label":"T-dual K-call","sub":"27B · same weights as T-dual Burst · one model call per control","kind":"nagi","composite":0.66,"ci95":[0.5733,0.7433],"inflation_p50":2.7992,"in_domain":{"snake":["cl:snake"],"rotorwash_ramp":["cl:hover","ma:crane","ma:hover2"],"booster_gauntlet":["cl:hover","ma:hover2"]},"serving":{"gpu":"H100","weights_format":"merged-bf16","kernels":"fused","concurrency_mode":"batched"},"games":{"snake":{"n":10,"n_void":0,"censored":0,"win":0.61,"median_t":10.0,"rmst":9.95,"median_level":1.0,"ladder":0.006,"km":[[0.0,1.0],[5.5,0.9],[6.5,0.8],[9.5,0.7],[10.0,0.2],[12.5,0.1],[15.5,0.0]],"alive_at":{"10":0.2,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":211.4,"fresh":0.756,"stale":0,"expiry":0,"t_fail":{"0":6.5,"1":12.5,"2":10.0,"3":10.0,"4":10.0,"5":9.5,"6":5.5,"7":10.0,"8":10.0,"9":15.5},"at_wall":5},"rotorwash_ramp":{"n":10,"n_void":0,"censored":0,"win":0.54,"median_t":18.267,"rmst":18.343,"median_level":2.5,"ladder":0.267,"km":[[0.0,1.0],[11.12,0.9],[11.19,0.8],[14.22,0.7],[15.05,0.6],[15.37,0.5],[21.17,0.4],[22.19,0.3],[23.25,0.2],[23.28,0.1],[26.6,0.0]],"alive_at":{"10":1.0,"20":0.5,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":609.9,"fresh":0.33,"stale":0,"expiry":0,"t_fail":{"0":15.367,"1":21.167,"2":23.254,"3":23.283,"4":22.192,"5":15.046,"6":26.596,"7":11.117,"8":14.225,"9":11.188},"at_wall":null},"booster_gauntlet":{"n":10,"n_void":0,"censored":0,"win":0.83,"median_t":22.0,"rmst":21.85,"median_level":3.0,"ladder":0.385,"km":[[0.0,1.0],[20.5,0.9],[22.0,0.0]],"alive_at":{"10":1.0,"20":1.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":752.9,"fresh":0.277,"stale":0,"expiry":0,"t_fail":{"0":22.0,"1":22.0,"2":22.0,"3":22.0,"4":22.0,"5":22.0,"6":20.5,"7":22.0,"8":22.0,"9":22.0},"at_wall":9}},"rank":3},{"id":"enormous_old/kcall","label":"ENORMOUS-old","sub":"27B · the released Nagi-ENORMOUS, no in-domain training · one call per control","kind":"nagi","composite":0.41,"ci95":[0.3433,0.47],"inflation_p50":2.133,"in_domain":{"snake":[],"rotorwash_ramp":[],"booster_gauntlet":[]},"serving":{"gpu":"H100","weights_format":"merged-bf16","kernels":"fused","concurrency_mode":"batched"},"games":{"snake":{"n":10,"n_void":0,"censored":0,"win":0.53,"median_t":6.5,"rmst":8.65,"median_level":1.0,"ladder":0.014,"km":[[0.0,1.0],[5.0,0.9],[5.5,0.7],[6.0,0.6],[6.5,0.4],[10.0,0.2],[15.5,0.1],[16.0,0.0]],"alive_at":{"10":0.2,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":243.9,"fresh":0.656,"stale":0,"expiry":0,"t_fail":{"0":6.5,"1":15.5,"2":5.5,"3":6.0,"4":10.0,"5":5.5,"6":5.0,"7":10.0,"8":6.5,"9":16.0},"at_wall":2},"rotorwash_ramp":{"n":10,"n_void":0,"censored":0,"win":0.52,"median_t":15.587,"rmst":15.988,"median_level":2.0,"ladder":0.182,"km":[[0.0,1.0],[12.23,0.9],[12.87,0.8],[13.04,0.7],[13.99,0.6],[15.2,0.5],[15.98,0.4],[16.53,0.3],[16.8,0.2],[21.28,0.1],[21.96,0.0]],"alive_at":{"10":1.0,"20":0.2,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":726.8,"fresh":0.295,"stale":0,"expiry":0,"t_fail":{"0":12.867,"1":15.196,"2":16.529,"3":13.992,"4":13.042,"5":21.962,"6":12.229,"7":15.979,"8":21.279,"9":16.804},"at_wall":null},"booster_gauntlet":{"n":10,"n_void":0,"censored":0,"win":0.18,"median_t":8.525,"rmst":8.685,"median_level":1.0,"ladder":0.0,"km":[[0.0,1.0],[8.15,0.9],[8.35,0.7],[8.45,0.6],[8.5,0.5],[8.55,0.4],[8.75,0.3],[8.85,0.2],[8.95,0.1],[9.95,0.0]],"alive_at":{"10":0.0,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":840.8,"fresh":0.197,"stale":11,"expiry":8,"t_fail":{"0":8.75,"1":8.95,"2":8.15,"3":8.85,"4":8.35,"5":8.35,"6":8.5,"7":8.45,"8":9.95,"9":8.55},"at_wall":0}},"rank":4},{"id":"jev/vendor_single","label":"Jev","sub":"Remote API, one request per decision · network time included · 1.1 s client timeout","kind":"external","composite":0.3867,"ci95":[0.2933,0.4933],"inflation_p50":0.9861,"in_domain":{"snake":null,"rotorwash_ramp":null,"booster_gauntlet":null},"serving":{"gpu":null,"weights_format":null,"kernels":null,"concurrency_mode":"remote"},"games":{"snake":{"n":10,"n_void":0,"censored":0,"win":0.3,"median_t":6.0,"rmst":8.9,"median_level":1.0,"ladder":0.058,"km":[[0.0,1.0],[2.0,0.9],[2.5,0.8],[3.5,0.7],[4.5,0.6],[5.0,0.5],[7.0,0.3],[14.0,0.2],[19.5,0.1],[24.0,0.0]],"alive_at":{"10":0.3,"20":0.1,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":216.8,"fresh":0.851,"stale":0,"expiry":0,"t_fail":{"0":7.0,"1":7.0,"2":24.0,"3":3.5,"4":2.0,"5":5.0,"6":4.5,"7":19.5,"8":2.5,"9":14.0},"at_wall":0},"rotorwash_ramp":{"n":10,"n_void":0,"censored":0,"win":0.46,"median_t":15.081,"rmst":16.651,"median_level":2.0,"ladder":0.202,"km":[[0.0,1.0],[9.39,0.9],[10.65,0.8],[12.52,0.7],[13.65,0.6],[14.45,0.5],[15.71,0.4],[17.18,0.3],[19.12,0.2],[20.08,0.1],[33.76,0.0]],"alive_at":{"10":0.9,"20":0.2,"30":0.1,"40":0.0,"50":0.0,"60":0.0},"p50_ms":212.9,"fresh":0.905,"stale":0,"expiry":0,"t_fail":{"0":20.079,"1":13.65,"2":10.646,"3":19.121,"4":12.521,"5":15.712,"6":9.388,"7":33.762,"8":14.45,"9":17.183},"at_wall":null},"booster_gauntlet":{"n":10,"n_void":0,"censored":0,"win":0.4,"median_t":8.975,"rmst":10.555,"median_level":1.0,"ladder":0.029,"km":[[0.0,1.0],[8.2,0.9],[8.45,0.8],[8.6,0.7],[8.9,0.6],[8.95,0.5],[9.0,0.4],[9.05,0.3],[12.3,0.2],[15.05,0.1],[17.05,0.0]],"alive_at":{"10":0.3,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":215.5,"fresh":0.878,"stale":0,"expiry":0,"t_fail":{"0":8.95,"1":9.0,"2":8.2,"3":8.9,"4":8.45,"5":17.05,"6":12.3,"7":9.05,"8":15.05,"9":8.6},"at_wall":0}},"rank":5},{"id":"semif_4b/kcall","label":"SemIf-4B","sub":"OpenJev / SemIf Qwen3.5-4B · self-hosted on the same H100 path · one call per control","kind":"external","composite":0.0767,"ci95":[0.04,0.1167],"inflation_p50":1.8835,"in_domain":{"snake":null,"rotorwash_ramp":null,"booster_gauntlet":null},"serving":{"gpu":"H100","weights_format":"bf16","kernels":"sdpa","concurrency_mode":"batched"},"games":{"snake":{"n":10,"n_void":0,"censored":0,"win":0.19,"median_t":4.5,"rmst":5.2,"median_level":1.0,"ladder":0.0,"km":[[0.0,1.0],[4.5,0.4],[5.5,0.3],[6.0,0.2],[6.5,0.1],[7.0,0.0]],"alive_at":{"10":0.0,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":101.5,"fresh":0.952,"stale":0,"expiry":0,"t_fail":{"0":6.0,"1":4.5,"2":4.5,"3":4.5,"4":7.0,"5":5.5,"6":6.5,"7":4.5,"8":4.5,"9":4.5},"at_wall":0},"rotorwash_ramp":{"n":10,"n_void":0,"censored":0,"win":0.0,"median_t":3.75,"rmst":3.838,"median_level":1.0,"ladder":0.0,"km":[[0.0,1.0],[3.54,0.9],[3.62,0.8],[3.66,0.7],[3.72,0.6],[3.74,0.5],[3.76,0.4],[3.85,0.3],[4.15,0.2],[4.17,0.0]],"alive_at":{"10":0.0,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":347.1,"fresh":0.481,"stale":0,"expiry":0,"t_fail":{"0":3.721,"1":3.538,"2":4.171,"3":4.171,"4":3.658,"5":3.758,"6":3.742,"7":4.146,"8":3.854,"9":3.617},"at_wall":null},"booster_gauntlet":{"n":10,"n_void":0,"censored":0,"win":0.04,"median_t":4.65,"rmst":5.14,"median_level":1.0,"ladder":0.0,"km":[[0.0,1.0],[4.4,0.9],[4.6,0.8],[4.65,0.4],[4.7,0.2],[4.85,0.1],[9.55,0.0]],"alive_at":{"10":0.0,"20":0.0,"30":0.0,"40":0.0,"50":0.0,"60":0.0},"p50_ms":501.5,"fresh":0.343,"stale":0,"expiry":0,"t_fail":{"0":9.55,"1":4.4,"2":4.65,"3":4.7,"4":4.6,"5":4.65,"6":4.65,"7":4.85,"8":4.7,"9":4.65},"at_wall":0}},"rank":6}],"unranked":[{"label":"CLM-8B","reason":"Not ranked. It failed its own conformance gate P3 before play (its README example scored outside tolerance), so its scores cannot be mapped 1:1 to the option sets (preregistered rule D15). The champion rule was therefore computed against the two external competitors that played."}],"champion":{"champion":"T-dual Burst","externals":["Jev","SemIf-4B"],"rule":"Mean over the 3 games of the paired per-seed win rate against the external competitors only; ties: sum of RMST ≤ 60 s, then mean level at failure.","table":[{"label":"T-dual Burst","per_game":{"snake":0.825,"rotorwash_ramp":0.95,"booster_gauntlet":1.0},"composite":0.925,"rmst_sum":56.23,"mean_level":2.4,"rank":1},{"label":"ENORMOUS-CL","per_game":{"snake":0.875,"rotorwash_ramp":0.85,"booster_gauntlet":0.95},"composite":0.8917,"rmst_sum":48.92,"mean_level":2.17,"rank":2},{"label":"T-dual K-call","per_game":{"snake":0.8,"rotorwash_ramp":0.75,"booster_gauntlet":1.0},"composite":0.85,"rmst_sum":50.14,"mean_level":2.23,"rank":3},{"label":"ENORMOUS-old","per_game":{"snake":0.775,"rotorwash_ramp":0.8,"booster_gauntlet":0.45},"composite":0.675,"rmst_sum":33.32,"mean_level":1.47,"rank":4}]},"pairwise":{"snake":{"t_dual/chord":{"t_dual/kcall":0.5,"enormous_cl/kcall":0.4,"enormous_old/kcall":0.6,"jev/vendor_single":0.7,"semif_4b/kcall":0.95},"t_dual/kcall":{"t_dual/chord":0.5,"enormous_cl/kcall":0.3,"enormous_old/kcall":0.65,"jev/vendor_single":0.7,"semif_4b/kcall":0.9},"enormous_cl/kcall":{"t_dual/chord":0.6,"t_dual/kcall":0.7,"enormous_old/kcall":0.65,"jev/vendor_single":0.9,"semif_4b/kcall":0.85},"enormous_old/kcall":{"t_dual/chord":0.4,"t_dual/kcall":0.35,"enormous_cl/kcall":0.35,"jev/vendor_single":0.7,"semif_4b/kcall":0.85},"jev/vendor_single":{"t_dual/chord":0.3,"t_dual/kcall":0.3,"enormous_cl/kcall":0.1,"enormous_old/kcall":0.3,"semif_4b/kcall":0.5},"semif_4b/kcall":{"t_dual/chord":0.05,"t_dual/kcall":0.1,"enormous_cl/kcall":0.15,"enormous_old/kcall":0.15,"jev/vendor_single":0.5}},"rotorwash_ramp":{"t_dual/chord":{"t_dual/kcall":0.9,"enormous_cl/kcall":0.8,"enormous_old/kcall":0.9,"jev/vendor_single":0.9,"semif_4b/kcall":1.0},"t_dual/kcall":{"t_dual/chord":0.1,"enormous_cl/kcall":0.5,"enormous_old/kcall":0.6,"jev/vendor_single":0.5,"semif_4b/kcall":1.0},"enormous_cl/kcall":{"t_dual/chord":0.2,"t_dual/kcall":0.5,"enormous_old/kcall":0.5,"jev/vendor_single":0.7,"semif_4b/kcall":1.0},"enormous_old/kcall":{"t_dual/chord":0.1,"t_dual/kcall":0.4,"enormous_cl/kcall":0.5,"jev/vendor_single":0.6,"semif_4b/kcall":1.0},"jev/vendor_single":{"t_dual/chord":0.1,"t_dual/kcall":0.5,"enormous_cl/kcall":0.3,"enormous_old/kcall":0.4,"semif_4b/kcall":1.0},"semif_4b/kcall":{"t_dual/chord":0.0,"t_dual/kcall":0.0,"enormous_cl/kcall":0.0,"enormous_old/kcall":0.0,"jev/vendor_single":0.0}},"booster_gauntlet":{"t_dual/chord":{"t_dual/kcall":0.55,"enormous_cl/kcall":0.7,"enormous_old/kcall":1.0,"jev/vendor_single":1.0,"semif_4b/kcall":1.0},"t_dual/kcall":{"t_dual/chord":0.45,"enormous_cl/kcall":0.7,"enormous_old/kcall":1.0,"jev/vendor_single":1.0,"semif_4b/kcall":1.0},"enormous_cl/kcall":{"t_dual/chord":0.3,"t_dual/kcall":0.3,"enormous_old/kcall":1.0,"jev/vendor_single":0.9,"semif_4b/kcall":1.0},"enormous_old/kcall":{"t_dual/chord":0.0,"t_dual/kcall":0.0,"enormous_cl/kcall":0.0,"jev/vendor_single":0.0,"semif_4b/kcall":0.9},"jev/vendor_single":{"t_dual/chord":0.0,"t_dual/kcall":0.0,"enormous_cl/kcall":0.1,"enormous_old/kcall":1.0,"semif_4b/kcall":0.9},"semif_4b/kcall":{"t_dual/chord":0.0,"t_dual/kcall":0.0,"enormous_cl/kcall":0.0,"enormous_old/kcall":0.1,"jev/vendor_single":0.1}}},"out_order":{"snake":{"0":[[6.0,"SemIf-4B"],[6.5,"ENORMOUS-old"],[6.5,"T-dual Burst"],[6.5,"T-dual K-call"],[7.0,"Jev"],[12.5,"ENORMOUS-CL"]],"1":[[4.5,"SemIf-4B"],[7.0,"Jev"],[10.0,"T-dual Burst"],[12.5,"ENORMOUS-CL"],[12.5,"T-dual K-call"],[15.5,"ENORMOUS-old"]],"2":[[3.5,"ENORMOUS-CL"],[4.5,"SemIf-4B"],[5.5,"ENORMOUS-old"],[9.5,"T-dual Burst"],[10.0,"T-dual K-call"],[24.0,"Jev"]],"3":[[3.5,"Jev"],[4.5,"SemIf-4B"],[4.5,"T-dual Burst"],[6.0,"ENORMOUS-old"],[10.0,"T-dual K-call"],[11.0,"ENORMOUS-CL"]],"4":[[2.0,"Jev"],[7.0,"ENORMOUS-CL"],[7.0,"SemIf-4B"],[10.0,"ENORMOUS-old"],[10.0,"T-dual Burst"],[10.0,"T-dual K-call"]],"5":[[5.0,"Jev"],[5.5,"ENORMOUS-old"],[5.5,"SemIf-4B"],[9.5,"T-dual K-call"],[10.5,"T-dual Burst"],[11.0,"ENORMOUS-CL"]],"6":[[4.5,"Jev"],[5.0,"ENORMOUS-old"],[5.5,"T-dual K-call"],[6.5,"SemIf-4B"],[11.0,"ENORMOUS-CL"],[13.0,"T-dual Burst"]],"7":[[4.5,"SemIf-4B"],[10.0,"ENORMOUS-old"],[10.0,"T-dual Burst"],[10.0,"T-dual K-call"],[19.5,"Jev"],[22.0,"ENORMOUS-CL"]],"8":[[2.5,"Jev"],[4.5,"SemIf-4B"],[6.5,"ENORMOUS-old"],[10.0,"ENORMOUS-CL"],[10.0,"T-dual Burst"],[10.0,"T-dual K-call"]],"9":[[4.5,"SemIf-4B"],[14.0,"Jev"],[15.5,"T-dual K-call"],[16.0,"ENORMOUS-CL"],[16.0,"ENORMOUS-old"],[16.0,"T-dual Burst"]]},"rotorwash_ramp":{"0":[[3.721,"SemIf-4B"],[12.867,"ENORMOUS-old"],[15.367,"T-dual K-call"],[19.492,"ENORMOUS-CL"],[20.079,"Jev"],[24.754,"T-dual Burst"]],"1":[[3.538,"SemIf-4B"],[13.65,"Jev"],[14.042,"ENORMOUS-CL"],[15.196,"ENORMOUS-old"],[21.167,"T-dual K-call"],[21.283,"T-dual Burst"]],"2":[[4.171,"SemIf-4B"],[10.646,"Jev"],[12.721,"ENORMOUS-CL"],[16.529,"ENORMOUS-old"],[23.254,"T-dual K-call"],[25.671,"T-dual Burst"]],"3":[[4.171,"SemIf-4B"],[11.938,"ENORMOUS-CL"],[13.992,"ENORMOUS-old"],[19.121,"Jev"],[23.283,"T-dual K-call"],[26.404,"T-dual Burst"]],"4":[[3.658,"SemIf-4B"],[12.521,"Jev"],[13.042,"ENORMOUS-old"],[16.512,"ENORMOUS-CL"],[22.192,"T-dual K-call"],[28.871,"T-dual Burst"]],"5":[[3.758,"SemIf-4B"],[15.046,"T-dual K-call"],[15.712,"Jev"],[18.392,"ENORMOUS-CL"],[21.962,"ENORMOUS-old"],[31.558,"T-dual Burst"]],"6":[[3.742,"SemIf-4B"],[9.388,"Jev"],[12.229,"ENORMOUS-old"],[13.062,"T-dual Burst"],[19.692,"ENORMOUS-CL"],[26.596,"T-dual K-call"]],"7":[[4.146,"SemIf-4B"],[11.117,"T-dual K-call"],[12.575,"T-dual Burst"],[15.979,"ENORMOUS-old"],[26.208,"ENORMOUS-CL"],[33.762,"Jev"]],"8":[[3.854,"SemIf-4B"],[14.225,"T-dual K-call"],[14.45,"Jev"],[17.996,"ENORMOUS-CL"],[21.279,"ENORMOUS-old"],[28.371,"T-dual Burst"]],"9":[[3.617,"SemIf-4B"],[11.188,"T-dual K-call"],[16.804,"ENORMOUS-old"],[17.183,"Jev"],[17.4,"ENORMOUS-CL"],[29.75,"T-dual Burst"]]},"booster_gauntlet":{"0":[[8.75,"ENORMOUS-old"],[8.95,"Jev"],[9.55,"SemIf-4B"],[22.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"1":[[4.4,"SemIf-4B"],[8.95,"ENORMOUS-old"],[9.0,"Jev"],[22.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"2":[[4.65,"SemIf-4B"],[8.15,"ENORMOUS-old"],[8.2,"Jev"],[15.3,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"3":[[4.7,"SemIf-4B"],[8.85,"ENORMOUS-old"],[8.9,"Jev"],[19.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"4":[[4.6,"SemIf-4B"],[8.35,"ENORMOUS-old"],[8.45,"Jev"],[22.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"5":[[4.65,"SemIf-4B"],[8.35,"ENORMOUS-old"],[17.05,"Jev"],[22.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"6":[[4.65,"SemIf-4B"],[8.5,"ENORMOUS-old"],[12.3,"Jev"],[17.25,"ENORMOUS-CL"],[20.5,"T-dual K-call"],[22.0,"T-dual Burst"]],"7":[[4.85,"SemIf-4B"],[8.45,"ENORMOUS-old"],[9.05,"Jev"],[22.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"8":[[4.7,"SemIf-4B"],[9.95,"ENORMOUS-old"],[14.8,"ENORMOUS-CL"],[15.05,"Jev"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]],"9":[[4.65,"SemIf-4B"],[8.55,"ENORMOUS-old"],[8.6,"Jev"],[22.0,"ENORMOUS-CL"],[22.0,"T-dual Burst"],[22.0,"T-dual K-call"]]}},"confirmation":{"title":"Burst vs Jev on fresh seeds","player":"Nagi-ENORMOUS Burst","same_as":"T-dual Burst","same_note":"The released name of T-dual Burst: the same weights and byte-identical prompts (variant A of the preregistration), served on the same pinned H100 path.","opponent":"Jev","seeds":"10–33","n_per_game":{"snake":24,"rotorwash_ramp":24,"booster_gauntlet":24},"estimand":"Θ = mean over the 3 games of the paired per-seed win rate of Burst against Jev (fail later = 1, tie = ½); 95% t interval over the 24 seed indices (t with 23 df on the per-index game mean).","theta":0.6736111111111112,"lb975":0.577429612891048,"ub975":0.7697926093311743,"se":0.04649464354837324,"df":23,"method":"t_23 on c_i (no dropped index)","gates":{"V_rt":{"passed":true,"replay_exact":true,"voids":0,"rule":"records replay exactly, every cell valid, seeds 0–9 equal Board 1, identical serving path"},"R_RT":{"passed":true,"H0":"Theta <= 0.35","lb975":0.577429612891048,"rule":"lower 97.5% bound > 0.35, i.e. not clearly worse than Jev"},"C1":{"passed":true,"lb975":0.577429612891048,"signflip_p":0.001809981900180998,"rule":"lower 97.5% bound > 0.50 and sign-flip p ≤ 0.025"}},"per_game":{"snake":{"W":0.4583333333333333,"n":24,"lb":0.2526263177861815,"signflip_p":0.7355126448735513,"holm_alpha":0.025,"passed":false,"rmst":9.333333333333334,"rmst_ci95":[7.1454563244616836,11.521210342204984],"rmst_jev":10.666666666666666,"rmst_jev_ci95":[7.929440746042799,13.403892587290533],"at_wall":{"burst":2,"jev":1},"cause":null},"rotorwash_ramp":{"W":0.5833333333333334,"n":24,"lb":0.3368341926480352,"signflip_p":0.2699273007269927,"holm_alpha":0.0125,"passed":false,"rmst":23.075666666666667,"rmst_ci95":[20.679660722339516,25.471672610993817],"rmst_jev":21.495083333333337,"rmst_jev_ci95":[17.304650223802625,25.68551644286405],"at_wall":null,"cause":null},"booster_gauntlet":{"W":0.9791666666666666,"n":24,"lb":0.9253746417019976,"signflip_p":9.99990000099999e-06,"holm_alpha":0.008333333333333333,"passed":true,"rmst":22.0,"rmst_ci95":null,"rmst_jev":12.260416666666666,"rmst_jev_ci95":[10.640742073405688,13.880091259927644],"at_wall":{"burst":24,"jev":1},"cause":"All 24 of Burst's rounds were hop timeouts at 22.0 s (still flying, first hop not landed). Jev crashed before 22.0 s in 23 of 24 rounds (landing without legs, torn legs or breakup) and timed out in 1."}},"c2_rule":"tested only because C1 passed: Holm over the 3 games at one-sided 0.025, t and sign-flip both at the Holm level","screen_0_9":{"vs_jev":{"theta":0.8666666666666667,"lb975":0.6999394105374099,"ub975":1.0333939227959235},"retest_vs_board1":{"theta":0.43333333333333335,"lb975":0.26285892942457956,"ub975":0.6038077372420871},"label":"screen, n = 10, not confirmatory"},"decision":"RELEASE","receipts":{"release_decision.json":"e820a29e6596d8dee6d7919bc2d420933fdb6cff9977eb254e81a5da607a47d9","results.json":"cb4875716db62396419e2675aa6f72183140ce1984869f0edfb2e351196407e4"}}}