alex
a545f33cc7
board: lab expanded to five Windows guests via ZFS clones - the single-VM constraint was shaping the work
2026-09-08 16:37:08 -04:00
alex
ed73b9fe7a
board: lane AI4 - ModCount decomposed with zero residual, all EIPs named; reference-game corrections
2026-09-08 16:31:02 -04:00
alex
c0c685d07f
board: lane G3 growth is clamp-decided not capacity-gated; a live-verified module was wrong on widened float literals
2026-09-08 15:57:33 -04:00
alex
3e5117bc05
board: VM140 holder = lane W3
2026-09-08 15:49:45 -04:00
alex
c810c552e5
standalone report from merged main: default 209 -> 131 closed 78; note the with-inputs 127 distinction
2026-09-08 15:49:04 -04:00
alex
1b5553b0aa
board: lane EV event posting; two events.md corrections; the withheld-roster control measurement
2026-09-08 15:46:01 -04:00
alex
c997c6ebce
board: lane W2 - multiplayer verified end to end, ModCount confirmed, rcex closed, Player.Status writer found
2026-09-08 15:34:45 -04:00
alex
2b7758123f
rules 21-22: lanes never touch the shared worktree; union-resolve is not concatenation. VM released
2026-09-08 15:34:12 -04:00
alex
3fe74bc6a5
standalone report from merged main: 209 -> 131 leaves, closed 78 regressed 0
2026-09-08 15:10:17 -04:00
alex
449daeb621
roadmap: item 6 premise refuted - no writer exists, it needs a predicate not a watchpoint
2026-09-08 15:07:02 -04:00
alex
7574623f73
board: lane T2 treaty stamps close 26; roadmap item 6 refuted - no writer exists; tail ranked and unactionable
2026-09-08 15:06:46 -04:00
alex
f259655026
dashboard refresh
2026-09-08 15:03:10 -04:00
alex
a4cc50043d
board: lane B6 - no build orders exist at load, so shpt[0] is Rung B; build queue read; stack-built fptr tables are a third blind spot
2026-09-08 14:58:45 -04:00
alex
214e00186d
roadmap: item 1 done (E1+C3); new item 1b S11 civilian growth, 1c ship repair cost
2026-09-08 14:53:33 -04:00
alex
247285ac28
board: lane C3 output turn path; unspent construction cascades to money; S11 civilian growth is next
2026-09-08 14:53:17 -04:00
alex
6669c4ab1e
dashboard refresh
2026-09-08 14:48:59 -04:00
alex
3b4142b566
board: note the self-inflicted build-dir race that mimicked a flaky test
2026-09-08 14:48:58 -04:00
alex
ce4603a380
board: lane AI2 - task selection is a stable sort on a constant priority table; order-method mapping complete; throttle settled
2026-09-08 14:46:12 -04:00
alex
8411c7e9c7
board: VM140 holder = lane W2
2026-09-08 14:14:49 -04:00
alex
35ccaf5d94
dashboard refresh
2026-09-08 13:56:49 -04:00
alex
4549f5eb5c
board: lane H - oracle reproduced, instrument perturbation found, P2 held so Rung B stands
2026-09-08 13:56:44 -04:00
alex
132d41c44e
roadmap: restore item 2, mark item 4 done, refresh current numbers
2026-09-08 13:53:39 -04:00
alex
3f014a42eb
board+roadmap: lane E1 income 25/25; correct roadmap item 1 - the turn path uses ComputeOutput, not ComputeMaxIncome
2026-09-08 13:53:20 -04:00
alex
08da5a1a06
standalone report regenerated from main: 209 -> 158 leaves, closed 51 regressed 0, 16/44 phases
2026-09-08 13:46:17 -04:00
alex
6e10def1b5
dashboard refresh after visibility
2026-09-08 13:43:31 -04:00
alex
069d7e649b
board: lane E3 visibility - 128 leaves across five pairs, 0 regressed; the gate is AFlags not VFlags
2026-09-08 13:42:36 -04:00
alex
d617285b0c
board: lane AI1 - the AI is a StrategyClient running after the autosave; orders go through the TurnCommands we already own; P2 routed to the VM
2026-09-08 13:39:38 -04:00
alex
19afa1f78a
dashboard refresh
2026-09-08 13:28:52 -04:00
alex
f6a914beee
board: lane E2 census wired; ship construction identified as a new blocker by a falsified prediction
2026-09-08 13:28:52 -04:00
alex
c4f1c2407a
roadmap: rewrite around the engine as the product; milestone split into Rung A (modulo command stream) and Rung B (needs AI); 13 stale board rows marked superseded
2026-09-08 13:04:13 -04:00
alex
7f4d5c5b89
dashboard refresh
2026-09-08 12:52:21 -04:00
alex
90fba303ac
board: lanes A2 and D2 - alliance rule, ModCount rule, ship designs do have a serializer
2026-09-08 12:52:20 -04:00
alex
ab79e16843
board: multiplayer revival - direct IP joins without GameSpy; identity constants; ranked plan
2026-09-08 12:51:39 -04:00
alex
1754cc20ac
board: lane P2 nav classifier; type 2 is the Liir drive, a naming error not a gap
2026-09-08 12:45:27 -04:00
alex
52581efbfb
board: VM140 holder = lane H
2026-09-08 12:15:33 -04:00
alex
cfe07fef80
dashboard refresh
2026-09-08 12:15:23 -04:00
alex
f328b82b67
board: lane N output term closed and live-verified; blocker moves to the income tail; VM released
2026-09-08 12:13:40 -04:00
alex
58e3d85f3f
board: lane V2 vtable inversion; 5,045 functions with no direct callers; tail draw surface is unobserved not absent
2026-09-08 12:06:22 -04:00
alex
7645399f97
dashboard refresh
2026-09-08 12:04:41 -04:00
alex
ef458bb3c8
board: lane B5 retreat pipeline; OID allocation closed; predicted original bug
2026-09-08 12:02:32 -04:00
alex
0253941d16
board: lane Y generator gap measured and attributed; BnkEl inversion oracle; divisor defect understated
2026-09-08 11:42:25 -04:00
alex
95b2bc81cf
board: VM140 holder = lane N
2026-09-08 11:01:06 -04:00
alex
c3ec32be01
dashboard refresh
2026-09-08 11:00:57 -04:00
alex
a77b1c3427
board: RNG ledger closed with zero residual; trade-raid generation behind a virtual edge; VM released
2026-09-08 11:00:56 -04:00
alex
707da6f848
dashboard refresh
2026-09-08 10:39:40 -04:00
alex
f8f5581204
board: standalone scaffold runs on all 11 saves; 5 leaves closed, 0 regressed; ranked byte-match blockers
2026-09-08 10:39:39 -04:00
alex
c504729341
lane S2: the standalone scaffold, and the measured distance to the byte-match
...
tools/standalone_report.py drives sots-engine's sots_turn over each
consecutive-turn save pair and diffs the result against the game's own
post-turn save with state_checksum.py, which localises to named leaves and
proves its own coverage by re-serialisation.
turn1-state -> turn2-state baseline 209 diverging, after 204, closed 5
turn2-state -> turn3-state baseline 108 diverging, after 103, closed 5
regressed 0 on both
`regressed` is reported next to `closed` and never netted off. It earned its
place immediately: committing the phase-31 player-status restore turned two
agreeing leaves into disagreeing ones, because the phase writes 1 and the file
carries 4.
The stable-system stand-in feeding the colony pass is a labelled hypothesis and
it survived a changed workload -- the same 3 ntdev leaves closed on both pairs,
six agreements, zero disagreements.
Two things deliberately NOT implemented: the TShn/ltis counters (18 leaves, a
`+1` would close them, but "+1 across one observed turn" is a hypothesis, not a
reading), and the RNG state write-back (an advanced-but-incomplete generator is
wrong in a different way from an untouched one).
dashboard.py gains section 6, reading verify/results/standalone/status.json:
phases modelled/committed per driver, baseline vs after, closed vs regressed,
the subsystem breakdown of what still differs, and the RNG gap. Sections 6-8
renumbered to 7-9; the delta footer tracks the two new counts.
DASHBOARD_README.md documents every number.
findings/control-flow/standalone-scaffold.md has the ranked blocker list.
2026-09-08 10:35:56 -04:00
alex
932f366597
dashboard refresh
2026-09-08 10:07:19 -04:00
alex
c15a45d0c6
board: lane I complete inlined-draw inventory; corrects lane J's 14 to 11; residual stated honestly
2026-09-08 10:07:18 -04:00
alex
1d7c6ef63e
board+rules: lane J combat resolver; inlined draws invisible to call-graph sweeps; Ghidra sizes untrustworthy
2026-09-08 09:38:55 -04:00
