fabi/Sylpheed
1
0
Fork 0
Code Issues 9 Pull Requests Actions Packages Projects Releases Wiki Activity
Files
642341b12b3adb269d552aec1df24dd09a8cb129
Sylpheed/docs/re/data
History
sylph-decoder 642341b12b re: rest_plateau() picks the wrong plateau -- and it is the whole residual 
rest_plateau() selects the LONGEST run of identical adjacent poses, which need
not be the run covering the screen's settle instant. rest_vs_settle left a 21.9 %
disagreement that I recorded as ambiguous by construction. It is not.

  CONTROL  exactly one plateau, covering the settle instant:
           3 072 / 3 072 agree (100.0 %)
  TEST     more than one plateau, at least one covering:
           1 622 elements, agree on 586 (36.1 %)
           of the 1 036 disagreements, rest() landed on a run NOT covering the
           settle instant: 1 036 -- all of them, no exceptions

Both poses are genuinely held in these cases -- they are plateau cases, not
transients -- so this is rest() returning a pose the screen has ALREADY LEFT by
the time it settles.

This corrects my own METHOD entry of two iterations ago, which said a candidate
cannot be adjudicated against the incumbent it replaces. Too strong. The bare
comparison cannot; the comparison plus a structural property that independently
says which side is wrong in each disagreement can. What I lacked was not an
oracle but a discriminator.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Wuu56cE8vJGTBtn1ppsk8v
2026-08-30 12:38:18 +00:00
..
a-press-ab-legs.txt
re: the A-press A/B is run -- signed-in profile, no swallow, menu opens
2026-08-30 08:05:43 +00:00
a-press-fault-log-extract.txt
re: the A-press fault is SOLVED -- Xenia swallows input, the guest pump is unbounded
2026-08-30 06:44:49 +00:00
adv-stream-assignment.txt
re: run 'grep the corpus for the claim' on this corpus -- four still standing
2026-08-30 10:35:27 +00:00
aiparams-census.txt
re: every AIParams object mapped to its stage, zero residual
2026-08-27 22:12:58 +00:00
aiparams-s02.txt
re: decode the stage table set — phases, routes, sub-objectives, AI parameters
2026-08-24 11:15:34 +00:00
archive-naming.txt
re: the Enumerate declaration tables close DefTables (130 -> 1413 of 1425 IDXD)
2026-08-27 19:59:20 +00:00
arsenal-chain.txt
re: the 59-of-131 arsenal question is closed -- an item names a hardpoint
2026-08-27 12:49:22 +00:00
attract-frame-match.txt
re: the boot intro and the attract loop are the same video, and the
2026-08-28 18:29:39 +00:00
b-on-main-menu.txt
re: B on the main menu goes to the TITLE -- measured, filling an empty evidence cell
2026-08-30 11:06:18 +00:00
bgm-wave-census.txt
re: a music bank is two stems that play together -- not intro+loop, and
2026-08-28 18:13:10 +00:00
boot-settle-run1.tsv
re: settle_time measured -- the title is drawn at 2 s, not at rest.t s 4.18
2026-08-29 15:21:49 +00:00
boot-splash-gap-draws.csv
re: the boot splash black gap is ~9 units, measured in draws not luminance
2026-08-29 20:36:15 +00:00
boot-timeline-2026-08-29.tsv
re: I cannot measure this emulator's clock -- and that answers the port's 2%
2026-08-29 13:16:02 +00:00
collisionset.txt
re: confirm the cmesh<->model link via GameResourceID; world unit is oracle-blocked
2026-08-26 09:41:46 +00:00
config-keys.txt
re: the config sections are real disc data -- 64/65, with a clean 0/54 control
2026-08-27 11:54:36 +00:00
config-records.txt
re: decode the config records -- render/input constants, and a correction
2026-08-27 12:02:13 +00:00
cutscene-message-table.txt
re: the sound-cue table, and cutscene dialogue's speaker/portrait/timing
2026-08-26 03:24:15 +00:00
datasheet-schema.txt
re: the disc ships the unit datasheet's own SCHEMA (and 文字列 was not a placeholder)
2026-08-28 08:22:36 +00:00
doc-link-audit.txt
docs: the UI decode's own evidence images were unreachable -- 11 links repaired
2026-08-29 02:34:16 +00:00
eff-bit-alpha-test.txt
re(ui): premultiplied alpha refuted for bit 0x02; parking the field
2026-08-29 03:23:40 +00:00
eff-bit-census.txt
re(ui): the eff-name implication for bit 0x02 is refuted disc-wide
2026-08-29 03:20:24 +00:00
effect-homes.txt
re: the EF_IDX_ prefix - ptc_pack has 727 names, and the map reaches 128 of 137
2026-08-28 10:01:53 +00:00
eightax-detail-test.txt
re(ui): settle 8AX vs ptbase statically -- the game draws the full-res one
2026-08-29 02:40:53 +00:00
focus-record-alpha-census.txt
re: 210 focus records pulse, not 2 -- and 8 of them fail in a way that looks right
2026-08-29 19:45:17 +00:00
focus-ring-period-corr.npy
re: the main menu's focus ring spins continuously -- period 2.18 s, measured
2026-08-29 11:50:32 +00:00
forced-backdrop-ink-thresholds.txt
re: reconcile the two ink counts -- we agree at >1 and not at >0
2026-08-30 07:07:21 +00:00
forced-backdrop-key-source.txt
re: not ONE of the 80 forced instances has a file-read key
2026-08-30 07:11:12 +00:00
forced-backdrop-necessity.txt
re: forced_backdrop DECIDES 62 of its 80 instances, not 6
2026-08-30 06:53:16 +00:00
forced-backdrop-pixel-cost.txt
re: the forced-backdrop rule's pixel cost -- 38 screens go black without it
2026-08-30 07:04:05 +00:00
gamepart-class-ids.txt
re: what the game actually reads at boot -- config.ini, and which
2026-08-28 18:49:19 +00:00
gamma-call-evidence.txt
re(ui): the game does query the display gamma -- measured, with its control
2026-08-29 02:51:32 +00:00
generic-record-partition.txt
re: correction -- Generic is a per-FILE header record, not a table
2026-08-27 12:29:30 +00:00
hangar-loadouts.txt
re: who selects a loadout -- the pak entry does; the Arsenal pak is stage-scoped
2026-08-27 13:09:07 +00:00
heap-ledger-gpu-writeback-probe.txt
re: the allocation failure is NOT the freeze -- two patched builds say so
2026-08-26 15:22:32 +00:00
heap-ledger-stage01.txt
re: the heap ledger refutes my own leak hypothesis
2026-08-26 12:22:40 +00:00
hud-config.txt
re: the objective/tutorial marker records; S24-S29 are CHALLENGE, not story
2026-08-27 15:46:31 +00:00
idxd-tag-hash.txt
re: recover the IDXD record-key / field-tag hash (8643/8643)
2026-08-25 10:38:18 +00:00
idxd-unnamed-field-keys.txt
re: measure the nameless IDXD field keys — 504 is 504 entries, not 504 names
2026-08-26 03:36:23 +00:00
intro-audio-channel-census.txt
re: the boot intro's audio output -- five live channels, not a stereo mix
2026-08-30 08:13:50 +00:00
intro-audio-decomposition.txt
re: ask #4 answered -- the intro is a 5.1 WMAPro bed at 0.600 plus three streams
2026-08-30 08:19:03 +00:00
isl-builtins-26-28-29-sites.txt
re(data): every call site of built-ins 26, 28, 29 and 101, with craft types
2026-08-26 03:43:31 +00:00
isl-command-table.txt
re: recover the ISL interpreter's command table — 57 opcodes, committed
2026-08-26 01:44:49 +00:00
isl-phase-guards-all.txt
isl: built-in 104 is request_next -- named from the game's own log string
2026-08-27 09:31:01 +00:00
isl-schedule-all.txt
re: the trailing data table is a per-phase TIMELINE of scheduled routines
2026-08-27 07:07:35 +00:00
isl-stage02-conditions.txt
isl: rename built-ins 8/9/93 to stopwatch_start/_elapsed/_stop
2026-08-27 09:21:04 +00:00
isl-stage02-dialogue.txt
re: caption pages are utterances, and the dialogue tool was truncating 356 names
2026-08-26 03:11:38 +00:00
isl-stage02-phase-ends.txt
isl: rename built-ins 8/9/93 to stopwatch_start/_elapsed/_stop
2026-08-27 09:21:04 +00:00
isl-stage02-phase-guards.txt
isl: rename built-ins 8/9/93 to stopwatch_start/_elapsed/_stop
2026-08-27 09:21:04 +00:00
isl-stage02-schedule.txt
re: the trailing data table is a per-phase TIMELINE of scheduled routines
2026-08-27 07:07:35 +00:00
isl-stage02.txt
isl: rename built-ins 8/9/93 to stopwatch_start/_elapsed/_stop
2026-08-27 09:21:04 +00:00
isl-timers.txt
isl: rename built-ins 8/9/93 to stopwatch_start/_elapsed/_stop
2026-08-27 09:21:04 +00:00
jp-title-at-rest.txt
re: the Japanese title at rest -- the capture MISSION has wanted since 2026-08-29
2026-08-30 11:37:35 +00:00
kf-angle-census.txt
re(ui): decode keyframe +12 as screen-plane rotation in degrees
2026-08-28 22:42:00 +00:00
kf-record-census.txt
formats: a keyframe's time comes before its pose, and none of them was missing
2026-08-29 14:01:39 +00:00
kf-unknown-4-8-census.txt
re: the keyframe block s +4 and +8 are angles used almost always as a 180 flip
2026-08-29 18:30:29 +00:00
main-game-unnamed.txt
re: the GP_MAIN_GAME unnamed block -- 333 of 337 already owned, 4 genuinely new
2026-08-27 20:17:10 +00:00
mcol-verify.txt
re: MCOL solved -- a closed triangle collision mesh in a uniform grid
2026-08-26 09:12:36 +00:00
menu-bgm-loop-measured.txt
re: the menu BGM loops at 61.93 s, and the game never reaches the fade
2026-08-30 09:34:42 +00:00
menu-bgm-loop-start.txt
re: the menu loop starts at 9.44 s -- measured, and the fix was scheduling
2026-08-30 10:13:14 +00:00
menu-bgm-wrap-timing.txt
re: the menu loop WATCHED -- three wraps, 61.81 s, and my placement is refuted
2026-08-30 10:03:52 +00:00
menu-bgm-xma-loop-fields.txt
re: the loop IS a runtime XMA field -- and reading it contradicts my audio measurement
2026-08-30 09:52:13 +00:00
miscbin-names.txt
re: name-resolve all 40 MiscBin entries, and demonstrate the REGN<->MCOL pairing
2026-08-26 09:21:00 +00:00
mission-scoring.txt
re: all 24 settings objects named -- stage\StageParameter_S<NN>.tbl
2026-08-27 19:05:28 +00:00
movie-audio-layouts.csv
re(media): the disc ships movies in TWO audio profiles, and 28 of them are 5.1
2026-08-29 08:57:56 +00:00
name-block-bases.txt
re: the base-solver's confidence axis was inverted
2026-08-28 00:42:46 +00:00
name-lookup-callsites.txt
re: the lookup argument is decoded at 46 of 48 sites -- and it refutes my
2026-08-28 19:52:02 +00:00
nav-autorepeat-and-settled-b.txt
re: three empty evidence cells filled in one run -- auto-repeat, B on a settled title, plate redraw
2026-08-30 11:18:18 +00:00
order-crosscheck-reach.txt
docs: withdraw yesterday's "paint order is a sequence" -- wrong source
2026-08-29 05:19:44 +00:00
paint-key-census.txt
re: retract the paint-key census and redo it over all 21 184 sprites
2026-08-26 10:06:23 +00:00
paint-order-audit.txt
re(ui): measure the paint-order hedge -- exact on 4 of 5, and bound the rest
2026-08-29 02:22:01 +00:00
paint-order-tie-pixel-cost.txt
re: the paint-order tie-break costs the port zero pixels, not 24 pairs
2026-08-29 14:47:19 +00:00
paint-order-ties-gp_title.txt
re: the paint-order tie-break costs one pixel, on one screen we do not ship
2026-08-29 19:35:29 +00:00
palogo-eff-plateau-vs-fallback.txt
re: the port's two extra elements are PLATEAU cases -- refutation succeeds, and its point gets bigger
2026-08-30 11:54:43 +00:00
phase-watch-s02-full.txt
re: bounded scan fixes the freeze; mission-over branch confirmed on the oracle
2026-08-25 16:07:26 +00:00
phase-watch-s02.txt
re: state 1 = not yet deployed, and arrivals are directly observed
2026-08-25 15:01:24 +00:00
plate-pulse-timeseries.txt
re: the PRESS A plate PULSES -- measured, and it keeps an authored entry
2026-08-30 07:37:49 +00:00
plate-timing-run1.tsv
re: the PRESS-A plate arrives 2.13 s after the title settles, measured twice
2026-08-29 12:15:22 +00:00
plate-timing-run2.tsv
re: the PRESS-A plate arrives 2.13 s after the title settles, measured twice
2026-08-29 12:15:22 +00:00
plateau-census.txt
re(ui): size the rest() guess disc-wide, and refute my own proposed fix
2026-08-28 23:23:55 +00:00
plateau-choice.txt
re: rest_plateau() picks the wrong plateau -- and it is the whole residual
2026-08-30 12:38:18 +00:00
plateauless-endstate.txt
re(ui): the structural case for last -- 2 293 of 2 305, checked disc-wide
2026-08-29 04:34:21 +00:00
present-rate-controls-2026-08-29.json
re: I cannot measure this emulator's clock -- and that answers the port's 2%
2026-08-29 13:16:02 +00:00
preset-messages.txt
re: the player-facing chatter partition; correcting the wingman claim
2026-08-27 17:42:28 +00:00
prt-parts.txt
re: the variant reading is refuted; HudResource is the in-game screen config
2026-08-27 14:42:20 +00:00
ptloop-leaf-sweep-positions.txt
re: ask #2 -- t=357.7 came from a GPU draw capture, and a sweep cannot date a frame
2026-08-30 07:15:45 +00:00
ratc-child-name-audit.txt
re(ui): a RATC child's name is stated, not inferred -- and it was hiding every menu background
2026-08-29 07:26:56 +00:00
ratc-name-fix-render-effect.txt
re(ui): a RATC child's name is stated, not inferred -- and it was hiding every menu background
2026-08-29 07:26:56 +00:00
ratc-tan-frame-sequence.txt
re(ui): the 60 nameless RATC children are frames, not children -- .tan decoded
2026-08-29 07:39:09 +00:00
record-loop-length-census.txt
re: a nested record's +0x08 is its loop length -- the plate's period is 120, not 105
2026-08-29 19:40:23 +00:00
rest-fallback-audit.txt
re: my own 1697 audited -- and the first correction failed its control
2026-08-30 12:13:46 +00:00
rest-fallback-census.txt
re: the rest() fallback -- its example dissolved, the question got bigger
2026-08-30 11:50:07 +00:00
rest-rule-blast-radius.txt
re(ui): quantify what changing the rest rule would do disc-wide
2026-08-29 04:24:40 +00:00
rest-vs-settle.txt
re: propose settle-instant posing -- and my control cannot validate it
2026-08-30 12:22:44 +00:00
rotation-top-level-examples.txt
re(ui): rotation is not nested-only, and the pivot-anchored scale is measured
2026-08-28 22:46:44 +00:00
rotation-toplevel-census.txt
re(ui): finish the top-level rotation census; record the JP-capture blocker
2026-08-28 23:46:54 +00:00
script-manifest.txt
re: correction -- the dialogue table was already in the corpus, and one negative was false
2026-08-27 13:45:35 +00:00
se-cue-runtime-offsets.txt
re: confirm, move and back -- every cue the five screens need is now
2026-08-28 19:38:09 +00:00
se-ui-cues.txt
re: the menu's sound events are named on the disc; the binding to them is
2026-08-28 18:33:58 +00:00
se-wave-riff-decode.txt
media: expose se_wave_riff -- the menu's SE cues, assembled where the format lives
2026-08-29 08:41:42 +00:00
settle-vs-rest-against-captures.txt
re: settle_time() itself beats rest() against the game -- on the one screen that adjudicates
2026-08-30 12:30:07 +00:00
slow-xpr-files.txt
docs: the suite is heavy, not hung -- correcting my own "cannot terminate"
2026-08-29 04:02:34 +00:00
splash-phase-timeline.txt
re(ui): deliver the measured splash sequence the port has to author
2026-08-29 05:00:54 +00:00
splash-ramp-check.txt
re(ui): Q1 -- the interpolation law holds, the group timeline does not
2026-08-28 22:56:17 +00:00
stage-dialogue-with-durations.txt
re: the seek table's sample total is not the wave length -- durations corrected
2026-08-26 06:17:07 +00:00
stage-numbering.txt
re: the stage numbering is consistent everywhere; the manifest still declares the cut S17
2026-08-27 13:27:19 +00:00
stage-s02-tables.txt
re: decode the stage table set — phases, routes, sub-objectives, AI parameters
2026-08-24 11:15:34 +00:00
stage-tutorial-records.txt
re: S18-S23 stage records were never missing; only S17 is absent
2026-08-25 09:34:58 +00:00
tbm-submenu-attempt.txt
re: the .tbm submenu is not reached -- and the second tap was never delivered
2026-08-30 10:30:49 +00:00
tbm-submenu-reached.txt
re: a submenu is REACHED -- and correlation cannot identify it
2026-08-30 10:49:48 +00:00
tie-break-live-over-time-gp_title.txt
re: the paint-order tie-break costs one pixel, on one screen we do not ship
2026-08-29 19:35:29 +00:00
tie-break-pixel-cost-gp_title.txt
re: the paint-order tie-break costs one pixel, on one screen we do not ship
2026-08-29 19:35:29 +00:00
title-glow-alpha-per-frame.csv
re: the title's build-in measured in the guest's draw stream -- the flashes are real
2026-08-29 19:56:34 +00:00
title-state-machine.txt
re: the ten states are ONE PHASE of five -- and phase 0 is the splash
2026-08-28 20:05:27 +00:00
turret-coverarea.txt
re: CoverArea is a 6-bit MOUNT mask; the three other turret leftovers closed
2026-08-27 15:35:09 +00:00
ui-scale-census-with-leaves.txt
re: 125 percent is not the only odd scale, and ptlogo_eff2 is a pop not a steady state
2026-08-29 18:16:20 +00:00
unit-datasheet-static.txt
re: the static unit datasheet and the AI flight model
2026-08-27 12:20:03 +00:00
unit-substructures.txt
re: the destructible-subsystem model -- a unit's sub-records
2026-08-27 12:39:50 +00:00
unitgroup-s02.txt
re: decode the UnitGroup squadron roster, and correct yesterday's container reading
2026-08-24 11:07:33 +00:00
voice-bank-manifest.txt
re: the seek table's sample total is not the wave length -- durations corrected
2026-08-26 06:17:07 +00:00
voice-region-cap-sweep.txt
formats: drop the 1.5 MB cap that truncated 17 voice regions' first stream
2026-08-30 08:55:47 +00:00
voice-region-chunk-census.txt
re: count the voice-region population properly -- 25 three-chunk, and 17 of them were broken
2026-08-30 09:46:08 +00:00
voice-region-start-audit.txt
re: resolve_movie_voice_region starts INSIDE the first stream, 8 of 10 multichannel regions
2026-08-30 08:41:22 +00:00
voice-region-start-clip.txt
formats: drop the 1.5 MB cap that truncated 17 voice regions' first stream
2026-08-30 08:55:47 +00:00
voice-stream-cue-map.txt
re: a long voice cue is three presentations of one take, and our guard clips the first
2026-08-29 15:09:40 +00:00
voice-three-stream-sizes.txt
re: the dual-mono explanation for the extra stream bytes does not generalise
2026-08-29 15:37:46 +00:00
voice-three-streams-runtime.txt
re: the three streams of a voice cue are decoded CONCURRENTLY, not alternatives
2026-08-29 15:48:35 +00:00
waitobj-healthy-stability-run2.txt
re: run 3 did not freeze -- a counter-example to my own inducer hypothesis
2026-08-25 11:56:00 +00:00
waitobj-healthy-stability.txt
re: control run withdraws the T68/T69 signature too -- nothing survives
2026-08-25 11:07:37 +00:00
waitobj-s02-inducer.txt
re: second freeze at ~96s with the inducer on; T74/T75 signature withdrawn
2026-08-25 10:10:14 +00:00
waitobj-s02.txt
re: the frozen wait-object capture, and screen_id was never a freeze test
2026-08-25 09:02:48 +00:00
weapon-shell-static.txt
re: the static weapon datasheet -- Weapon, Shell, AssortMissileParam
2026-08-27 12:11:03 +00:00
zero-scale-census.txt
re(ui): scale 0 means collapsed, not "unset" -- stop drawing it full size
2026-08-28 23:38:37 +00:00