A resolve-check on HANDOFF's own rows. Q8 read "SE audio is undecodable
from the disc -- no XACT container exists anywhere". menu-audio-cues.md
retracted exactly that ("### Retracting 'cannot be extracted'") and
locates three cues in Static.slb that decode to PCM: d-pad move 0x1ec0
(4 packets), (B) back 0x0ec0 (2), (A) confirm 0x5d6c0 (6), all mono
48 kHz. The retraction landed in docs/re/ and the page the port reads
kept the superseded text -- the fourth time in this corpus.
Writing the rule down has not worked, so there is a tool now.
handoff_lint.py flags every HANDOFF line making a strong negative claim
that links a doc containing retraction language. First run: found the Q8
row, plus one benign false positive (Q3 links a doc whose retraction is
about a sprite count, not about the tie-break -- checked, and HANDOFF
repeats none of the retracted figures). The lint also caught its own bug
first: it reported existing docs as missing because it joined a guessed
repo root, so it now resolves links relative to the file as markdown does.
Separately, EXTRAS's paint-order risk narrows twice more. Of its 15 tied
pairs only 2 overlap, and of those, ptloop01 x ptloop02 are loop*
animations compose skips by default -- so exactly ONE tie can be drawn:
ptframe3 x ptframe4, overlapping 102x132 px. Against live-extras.png that
contested region correlates +0.9622, better than the whole frame (+0.9440)
and inside the range of regions where order cannot matter (+0.8502 /
+0.9903). Consistent with our order, not proof: correlation cannot see a
swap between locally similar art.
15 -> 2 -> 1 -> consistent is now the whole paint-order risk on the five
screens, and HANDOFF says so.
Q3 was delivered as "decoded: a u16 layer key at +0x0A". The audit last
iteration showed the key does not fully order a screen -- elements
sharing a key are tied, and on the title that tie-break decides two total
occlusions. This searches for what breaks the tie, and closes it as a
negative with reach.
The game paints the five tied ptlogo_back2eff glows in the order
eff1, eff2, eff5, eff3, eff4. Three static structures were searched:
1. The declaration table. Entries 14-18 are byte-identical apart from the
pivot, which is only half the sprite's own size.
2. The T8aD headers. All five carry identical +0x04 (0x8832) and
identical +0x08/+0x0A (32899, the key itself), differing only in
position and tile count. Searched exhaustively -- every offset
0x00-0x7f, u8/u16/u32, ascending and descending:
fields sorting to the MEASURED order: 0
fields sorting to the DECLARATION order (control): 64
The control is the point: 64 fields can be found that reproduce a
known ordering, so the scan finds ordering fields when they exist. It
finds none for the order the game uses.
3. The RATC child order -- a genuinely different permutation on other
screens -- gives eff1..eff5 here, declaration order again.
All three static orderings give eff1..eff5; the game gives eff1,2,5,3,4.
That agrees with ui-screen-runtime's conclusion from the other direction:
the game builds a reordered child list at load time and paints that.
Q3 now reads honestly: the layer key is decoded and orders 4 of the 5
measured bundles exactly; the tie-break within a key is undecodable, and
a consumer must use a measured order or accept declaration index as an
arbitrary stand-in. The port's exposure remains 2 overlapping tied pairs
on EXTRAS.
METHOD: an exhaustive field search needs a positive control, or "found
nothing" is worthless.
Cross-checked the instrument built last iteration against an independent
grabber while both watched the same screen, and it fails.
A single long-lived ffmpeg x11grab stream degrades and then freezes:
862 frames in 540.1 s = 1.60 fps (it starts at 3.98)
t=450/480/510/540 s: surface mean 5.21, identical every time
At that same moment `import` read surface mean 125.65, and a freshly
started ffmpeg stream read 122.43 -- agreeing with import to 3%. So the
acquisition was broken, not the analysis: the stream replayed a stale
frame while the screen was 24x brighter.
That withdraws last iteration's headline. "2391 frames over 600 s from
t=0, max glyph 0" cannot distinguish "the title never appeared" from "the
stream froze early and repeated one frame 2391 times". Its 3.98 fps was
measured over the first 20 s, before the degradation. Sample count is not
coverage unless the samples are known independent.
Fixed: the stream is now torn down and restarted every 30 s. Startup is
~0.3 s, cheap against the title's window, and it guarantees live frames.
Separately, the cache hypothesis was tested and is SUPPORTED. cache,
cache0, cache1, cache_host moved aside (to /tmp/xenia-cache-aside, not
deleted) and the surface renders again: import reads mean 54.8 and 68.6
with 100% non-black warm content, against 0.07 and 0.08% non-black in the
black run; 773 of 862 probe frames had >2% non-black. One run each side
and many kill -9s before the black one, so it is supported, not proven --
the old caches are kept for reproduction.
Still no title, but that number now comes from a stalling probe and
establishes nothing either way.
METHOD: validating a probe on static images tests its analysis, not its
acquisition -- cross-check against an independent grabber during a run.
The named experiment was to attach the fast probe at t=0 so the boot
title could not be missed. Done, default config, English:
2391 frames in 600.4 s = 3.98 fps; max glyph 0; hits 0
Ten minutes sampled four times a second FROM LAUNCH, no green-(A) glyph.
So "the plate only shows in an early boot window I keep missing" is mine
and refuted -- the third explanation refuted in three iterations.
Then the check that should have come first. Splitting the raw root grab
into bands:
y 0- 44 (GTK menu bar) 100.00% non-black mean 210.50
y 45-719 (game surface) 0.08% non-black mean 0.07
The game is rendering black, reproducibly across back-to-back samples,
while the guest is alive and polling input (XamInputGetKeystrokeEx past
1201 calls) and MEM-WATCH keeps reporting. The crop and every pixel
oracle were correct; there was nothing on the surface to detect.
What this does NOT do is retroactively explain the earlier failures, and
claiming so would be the fourth over-reach in a row. Those runs had
content: run 2 sampled mean 33.1, run 3's classifier measured real
frame-to-frame rmse, the gamma_type=0 run measured mean 122.8. The
failure mode CHANGED over the session; black is the newest and worst.
Hypothesis for the regression, untested: canary's shader/pipeline cache
is 47 MB and was last written 23:49 on Aug 28, during the failed runs,
and this session has kill -9'd the emulator repeatedly. The test is to
move cache* aside and boot again -- one line and one run, not done.
Two METHOD lines: ask whether the screen is drawing anything before
explaining why a feature of it is missing; and a newly found fault does
not retroactively explain older failures.
Last iteration I blamed four failed runs on the probe sampling every
~41 s, slower than the title screen lasts, and withdrew three earlier
conclusions on that basis. Building the fix tested the claim and killed
it.
The speedup is real and control-verified. One long-lived ffmpeg x11grab
stream, raw RGB, glyph counted in numpy -- no per-sample process startup,
no PNG encode, no convert -crop:
wrapper `screenshot` 3.98 s per sample (emulator running)
import -window root -> PPM 1.20 s
long-lived x11grab stream 0.29 s 13.7x
The counter is byte-identical to is_title.py: 753 on the committed title
capture, 327 on the main menu.
Pointed at a running game it says the opposite of what I expected:
332 frames in 85.3 s = 3.89 fps; max glyph 0
1674 frames in 420.0 s = 3.99 fps; max glyph 0
1674 consecutive samples over seven unbroken minutes, four per second,
zero green-(A) pixels. Sampling rate was a real defect that happened not
to be the cause.
So "neither locale reaches the interactive title without a pad press" --
withdrawn last iteration for want of evidence -- is reinstated, now as a
dense measurement, with its reach stated: a MID-RUN window only, silent
about the boot title.
Leading hypothesis, unconfirmed: the PRESS (A) plate appears only in the
boot title window and the attract loop's title carries none, which is
exactly what title_states_capture.sh was written to test. The experiment
is to start the fast probe from t=0 rather than attach to a run already
in progress.
METHOD: fixing the instrument is how you test the explanation that blamed
it -- a plausible mechanism is a hypothesis, and the fix is its
experiment, not its proof.
Four consecutive runs failed to reach the interactive title, across two
locales, two launch paths and two display-gamma settings. I attributed it
in turn to a stale oracle, to the locale, and to the attract loop. It was
none of those.
one `screenshot` call, emulator running: 10.8 s
one `screenshot` call, emulator killed: 0.117 s
92x, measured at a 1-minute load average of 1.80 -- so it is contention
with the emulator through the X server, not background load.
skip_intro.sh takes two grabs per iteration plus a numpy import, giving a
median sampling interval of 41 s in the last run (38/82/41/20/30/30/35/
47/46/45/44/43/21/19). The title lasts "a few seconds" before the attract
loop reclaims it -- wait_title.sh's own header says so. The harness was
sampling slower than the event it was waiting for. That also explains why
runs at 16:43-18:05 the same day succeeded.
Withdrawn as CAUSES, though the observations stand: "the JP run never
reaches the interactive title", "neither locale reaches it without a pad
press", and "the game sat in the attract loop for 604 s". The English
control did control for locale -- it just shared the same defect.
Also recorded: I set kernel_display_gamma_type = 0 for the gamma control
run, which brightens the frame (mid-attract mean 122.8 vs 52.5/82.8 at
type 2) -- and skip_intro classifies movie-vs-static on an ABSOLUTE rmse
threshold, so the gamma change biased the very classifier the run
depended on. Changing a display setting and a capture behaviour in one
run confounds both. Config restored to type 2.
The fix is not applied: make the probe cheap enough to outpace the title
window (small region, no convert round trip, one long-lived process).
Every remaining emulator-side question is waiting on that.
Two iterations framed the JP title capture as possibly locale-specific.
It is not. Same flags, same oracle, English locale: 75 samples over 734 s,
every one glyph = 0. The English boot does not present the interactive
title either. Canary was alive throughout, polling
XamInputGetKeystrokeEx (1801 calls); the frame at 734 s has content but
correlates 0.15 with build 4's render and 0.05 with the main menu -- an
attract-movie frame, not title art.
So neither locale reaches the interactive title in ~12 minutes without a
pad press. What is actually untested is the pad: pad.py and
nav_to_flight.sh exist, and title_states_capture.sh reaches title states
with a flag set this probe did not replicate (--log_ui_draws
--ui_draw_capture_frames, plus xdotool window focus). It is a capture
problem, not a locale one. MISSION.md updated.
Also resolved a caveat I had given the port agent without checking it:
"rotation is decoded but not rendered" does NOT affect their five screens
at rest. Title, main menu, EXTRAS and both splash halves have ZERO
top-level elements with a non-zero rotation. The only rotations on any of
them are the title's two nested ptloop records (r = 30 and -45), and at
rest those sit at x = 1521 and x = -839 -- a 399-wide sprite entirely off
both edges of a 1280 screen. The caveat now applies only to animating the
title build-in, where the sweeps cross the screen rotated.
Two METHOD lines: run the control before theorising about the difference,
and log every sample so a failure is a measurement rather than a silence.
Emulator stopped, lock cleared, locale English.
Last iteration's "never reached the title in 787 s" was a broken tool
reporting on the world. wait_title.sh was still sampling the single pixel
(625,618) that is_title.py had already been written to replace -- its
docstring says why: a 1280x720 coordinate sampled against the 1279x675
game surface, so it always reads the copyright line. The replacement sat
in the same directory. wait_title.sh now delegates to it.
is_title.py passes its own controls before being trusted here: 753
green-glyph pixels on the committed English title capture, 327 on the
main menu, threshold 400.
Re-ran with the working oracle and the profile flag the English captures
use. The game STILL did not present the interactive title -- but that is
now a measurement rather than an artefact: not one frame showed a single
green-(A) glyph pixel, and content correlation against either build-7
render never exceeded 0.22. Canary was alive and polling
XamInputGetKeystrokeEx (601 calls), sitting in the attract movie.
So the open question narrowed again, and is written into MISSION.md:
whether the attract loop returns to the INTERACTIVE title without a pad
press. title_states_capture.sh claims it does on the English boot with no
pad input; if that holds, the difference is the locale.
Nothing decided about the keyframe-time association or the rest() rule.
Emulator stopped, lock cleared, locale restored to English.
Last iteration I wrote into MISSION.md that a Japanese-locale capture is
impossible here, because user_language is DECLARE_int32 at four call
sites with no DEFINE and no entry in xenia-canary.config.toml. That is
true, and it was not the question. The language is PERSISTED:
kernel_state.cc builds XConfig over <storage_root>/xconfig.settings,
SetDefaults() only supplies a value when the file has none, and the file
is writable. Checking where a setting is stored rather than where it is
configured turned "blocked, needs a human decision" into a two-line edit.
Withdrawn from MISSION.md; METHOD and REFUTED lines added.
The field is located from struct landmarks rather than a hard-coded
offset, and the check re-runs on every invocation so it fails loudly if
the layout moves:
music_volume 0.7f at User+449 -> BE float at 2727 -> User base 0x8e6
language at User+44 -> reads 1 (kEnglish) at 0x912
country at User+64 -> reads 103 (US) at 0x926
XLanguage::kJapanese = 2 (xbox.h:307). set_console_language.py wraps it
with a backup and a --restore.
The capture itself is still NOT taken, for a smaller reason than I
claimed. A run with the locale set to Japanese booted fine but never
reached the title in 787 s: wait_title.sh's green-(A) oracle never fired
and burst-sampling found no frame correlating above 0.18 with either
build-7 render -- the run sat in the attract loop. So it needs a longer
or pad-driven run, not a rebuilt emulator. Emulator stopped, lock
cleared, locale restored to English.
Nothing is decided about the keyframe-time association or the rest()
rule; this only changes what standing between us and deciding them.
blit() and fill_quad() both opened with
let sx_pct = if kf.scale_x == 0 { 100 } else { kf.scale_x };
so an element whose pose is collapsed to nothing rendered at FULL SIZE.
I first described this as "a 1-pixel sliver", reading the .max(1) in the
size arithmetic two lines below -- the guard above it meant .max(1) never
saw a zero. Read the whole function, not the lines you went looking for.
Control run before the change: if 0 meant "unset", something on the disc
would use it throughout. Nothing does.
elements with a keyframe group 15 493
at least one keyframe at scale 0 2 166
EVERY keyframe at scale 0 0
zero is a transient the element grows out of 1 762
ptlogo_eff3.t32 runs 0% -> 200%. An "unset" marker no element ever uses
throughout is not a marker.
Both functions now return without drawing when either scale is 0.
Reach: 24 of 24 renders byte-identical across GP_TITLE (all 16 builds),
GP_PAUSE_MENU and GP_OPTIONS -- additive on every screen the port needs.
195 elements have a guessed rest pose at scale 0 and 126 of those have a
non-zero alpha, so the old code painted them; all 126 are in
GP_READY_ROOM.pak, which S1 already declared a no-go. They are
tactical-map sprites at sx=0/sy=100, mid-horizontal-wipe, that the
coercion drew full-width.
So: a real correctness fix with no visible effect on the menu port.
Suite green, 122 passed / 0 failed across 3 suites at commit time.
Follows up the defect found last iteration: rest()'s dwell fallback is
guessing whenever it runs. Two things were open -- how big it is, and
whether "rest = the last keyframe" is the fix. Both are now answered, and
the second is answered no.
plateau_census.py walks the placement regions directly instead of going
through `screen info --geometry`, which decodes every texture and cannot
do a disc-wide pass in reasonable time. Its control reproduces GP_TITLE
build 7's three fallback elements and names ptlogo_eff3.t32 among them
before counting anything.
elements with a keyframe group 15 493
no plateau -> rest pose is guessed 3 807 (24.57 %)
... current rule returns invisible 1 711 (44.9 %)
... current rule returns scale=0 195 ( 5.1 %)
the two candidate rules agree 1 911 (50.2 %)
195 elements get a rest pose with scale 0%, which is not a pose. And
disc-wide the choice of rule is not cosmetic: the candidates agree half
the time.
But the port's exposure is one element. Across main menu, EXTRAS, title
and the developer splash, 14 elements are plateau-less and the two rules
agree on 13. The single disagreement is palogo_anima_eff.t32.
And "last keyframe" loses there, on a control that needed no new capture:
the splash carries three sibling glows with identical structure and
identical times --
palogo_gamearts_eff 15:a=0 30:a=255 45:a=255 -:a=0 plateau -> visible
palogo_seta_eff 15:a=0 30:a=255 45:a=255 -:a=0 plateau -> visible
palogo_anima_eff 15:a=0 30:a=255 45:a=212 -:a=0 no plateau
-- differing in one byte. "Last keyframe" makes anima alone invisible
while its two siblings stay lit. The capture agrees weakly: box-mean
ratios capture/render are gamearts 0.717, seta 0.723, anima 0.772, and a
glow we drew that the game does not would put anima below its siblings,
not above.
So the defect is measured and the fix is still undecided. Nothing in the
decoder changed.
Last iteration I reported a 13.1% render change in GP_TITLE build 7 as
evidence AGAINST the keyframe-time shift, arguing that language twins
should match in brightness. Withdrawn. Localising the diff to an element
shows it is not about the time association at all.
The element is ptlogo_eff3.t32, a transient bloom:
46: (98,42) 100%,100% a=0
61: (108,72) 0%,0% a=0
103: (108,72) 200%,200% a=255 r=80
-: (108,72) 0%,0% a=0 r=150
No two adjacent poses are equal, so there is no plateau, and rest() falls
through to its longest-dwell fallback. The longest gap is 61->103, during
which the sprite grows from nothing to 200% at full alpha and then
collapses. The rule returns whichever end of that movement the indexing
lands on: the invisible frame as decoded, the 200% peak shifted. An
896x389 sprite at 200% is larger than the screen, which accounts for the
entire 13.1% and the entire 4.9-unit luminance gap. I was comparing a
heuristic against itself.
The defect generalises, and structurally. A dwell gap is time spent
interpolating FROM pose k TO pose k+1; neither pose is held during it
unless the two are equal -- which is a plateau, and the plateau path has
already returned by then. So the fallback is guaranteed to be guessing
every time it is reached, under any reading of the times. Every element
with no two adjacent identical poses has a guessed rest pose, in our
renderer and in anything built from it.
Consequences: the case FOR the shift (26x on the hold:fade-out ratio) is
no longer opposed by render evidence -- 10 of 11 builds byte-identical,
the 11th differing only through this heuristic. It is still not adopted,
now because it flips this element to the visibly wrong answer, so the
shift and a decision about plateau-less elements must land together, and
neither half has a capture to verify against.
Default-mode suite green (122 passed, 0 failed across 3 suites so far).
Follow-up on last iteration's unadopted candidate (+36 holds the NEXT
pose's time, not its own). Two new results, pointing opposite ways, and
both are reported.
FOR, and calibration-free: the observed full-alpha hold : fade-out ratio
on palogo_gamearts is 83 : 13 frames = 6.38. The shifted reading predicts
8.00. The current reading predicts 0.25 -- off by 26x. With the glow's 2
units/frame fixed and nothing else free, the current reading says the
logo holds full alpha for 2.0 frames; the capture holds it for 83. This
is no longer the shape argument the candidate rested on.
Also for: rest()'s plain max-dwell fallback picks a=0 -- a transparent
pose, for a publisher logo -- under the current reading, and the visible
a=255 hold under the shift. Only the rest_plateau special case rescues
the render today, and that is the case the port agent reported a bug in.
AGAINST: rendering every build of six UI paks under both readings, 10 of
11 compared are byte-identical and one changes -- GP_TITLE build 7, the
Japanese twin of build 4, by 13.1% of pixels. Build 4, the one verified
against a live capture, is unchanged either way, so the single build the
shift moves is the one with no capture to adjudicate it. The proxy goes
against the shift: language twins are the same artwork, and build 7 reads
70.94 mean luminance as decoded against build 4's 71.41, but 76.32
shifted. Correlation does not separate them (0.6206 vs 0.6201).
These constrain different things -- timing versus pose selection -- and
rest() is a heuristic layered on the times, so moving the times moves its
tie-breaks. Adopting the shift means revisiting that heuristic in the
same change, with no build-7 capture to verify against.
Default UNCHANGED. Experiment reachable via SYLPHEED_KF_TIME_SHIFT=1.
Shifted-mode suite: 122 passed, 0 failed across 3 suites at commit time.
Q1's gate asks whether the ramp is linear. It is, and that result stands:
it rests on the splash's _eff glows, which reproduce exactly. This adds
the part that does not.
The test is a calibration, not a fit. Fix the clock on
palogo_gamearts_eff -- declared 15-unit fade-in 0@15 -> 255@30 against
captured alphas 34,68,102,136,170,204,238, a constant step of 34, giving
t = 2f - 171 -- then check that against the glow's own next landmark: its
declared hold ends t=45, predicted frame 108.0, observed last full-alpha
frame 107. Then apply it to palogo_gamearts in the same bundle and the
same frames, with no free parameter left:
declared a=232 at t=206 -> frame 188.5, observed alpha 255
declared a= 32 at t=210 -> frame 190.5, observed alpha 255
The logo is still at full alpha nine frames after it should read 32; its
fade-out runs ~17 frames late; its declared 80-frame fade-in is never
drawn. Not culling -- the same element is submitted down to a=7 on the
way out. Calibration-free version: the declared fade-out spends 12 of 16
units dropping 23/255 of the alpha, and the capture has no such plateau.
Candidate, offered and NOT adopted: if +36 held the NEXT keyframe's time,
the fade-out shape fits (RMS 4.05 vs 12.13, two elements) and the
decoder's "last block's time is unreadable" special case disappears --
the last block would simply have no successor. Rejected for now because
it explains neither the missing fade-in nor the lateness, and because the
_eff elements cannot discriminate between the readings at all (with four
blocks the shift only relabels the phases). Decoder unchanged.
Also withdrawn, mine, within the iteration: "the _eff glows hold a
constant alpha 33". They ramp 34 -> 255 in steps of 34. I printed the
series minimum and read it as its range, with a "14 distinct colours"
column sitting next to it saying otherwise.
Two corrections and one new confirmation, all from finishing the sweep I
left running last iteration.
Refuted, by my own sweep, within the hour: "rotation appears to live only
in nested .rat leaf records". That held for GP_TITLE, GP_BUNK and
GP_CHALLENGE -- the three archives the sweep had reached when I wrote it
-- and fails on GP_DIALOG and GP_DEBRIEFING_PILOTLOG, which rotate
top-level elements. Scoping the claim made it cheap to withdraw, but the
sentence should have waited for the sweep.
Those top-level cases are the best evidence on the disc, and they show up
in `screen info --geometry` without a hex dump. GP_DIALOG build 0:
pceff03/pceff04 ramp r = 90 -> 30 -> 10 -> 3 -> 0 while alpha ramps
0 -> 255 and they slide into place -- a swing-in settling upright. Build
6: pzeff02 ramps 43 -> 61 -> 75 -> 90 while scaling 112% -> 200% and
fading to 0 -- a spin-out. Rows committed as reference data.
New: the pivot-anchored scale term in blit, kf.x - pivot*(scale-100)/100,
was implemented and reasoned about but never measured, because every
element previously examined sits at 100% scale where the term is exactly
zero. The ptloop pair scale 600% and 800%, where it is worth 450 and 630
px. Formula predicts centre y = 360.0 for both; capture measures 359.1
and 360.0. Top-left anchoring predicts 810 and 990. Horizontally it makes
t-from-position agree with t-from-alpha to 0.33 / 0.65 units against ~8
without it.
That agreement does NOT prove linear interpolation -- both fields were
inverted through the same linear map, so a shared easing curve cancels.
It shows position and alpha ride one shared parameter. Recorded as such.
The disc-wide sweep is still running; the count is incomplete, the
existence is settled.
The rotated quads on the title screen come from the keyframe block after
all. The earlier negative -- "every GP_TITLE build 4 element has all three
angle words at zero" -- read the right bytes over too small a region: it
walked the top-level declaration table, and the rotated elements are the
nested leaf records ptloop01.rat / ptloop02.rat.
Confirmed against the framebuffer rather than against our own renderer.
The two records declare +12 = 30 and -45; the GPU capture submits their
quads at +30.26 and -45.28 degrees -- magnitude and sign, two different
values. Corroborated by shape in GP_BUNK 117ca14f, where +12 ramps
0 -> 360 with position, scale and alpha constant: a spin in place.
Identifying which draw it was needed edge lengths, not bounding boxes:
400x1076 and 400x1444 against pteff03/pteff03a 399x180 at the elements'
two different declared scales, 600% (1080) and 800% (1440). The same test
names three known-positives in the capture (ptlogo1, ptcopyright,
ptbtn00), so it passes its own control.
Keyframe gains rotation_deg plus unknown_4/unknown_8, carried rather than
dropped. NOT rendered -- ui_layout::blit is axis-aligned only, so the
reference renderer and the port will both draw these upright until a
rotating blit exists.
The census tool ships with the trap that broke its first version: nested
RATC blobs are not 4-byte aligned, so an aligned scan found 0/3 of its
own control blocks and missed 16 341 blocks. Disc-wide +12 is non-zero in
14.50 % of 83 862 blocks.
sylpheed-formats tests, SYLPHEED_DISC set: 131 passed, 0 failed across the
6 suites finished at commit time; the run had not yet completed.
iterations of work were built on
I flagged last iteration that "draw 2 is the swoosh" was an inference I
had never checked. Checking it took one line, and it is wrong.
Converting the quads from NDC to screen space: quad A spans y -209 to
925, quad B spans y -292 to 1012. Both cover the full screen height and
run well off it. The swoosh, ptlogo_back2, rests at (71,126) and is a
234-pixel band at y 126 to 360. Draw 2 is not it.
A candidate offered as one: the two ptloop sweeps. pteff03a is 399x180
with element scale (100,800), so 399x1440, and pteff03 carries (100,600)
-- two long thin sprites, two quads, and the drawn vertex alphas 0xC3 and
0xB6 sit inside the 0x80 to 0xff ramp those records declare. Not
confirmed; no texture or position match was made.
If that is right it matters, and I have written it conditionally rather
than acting on it: our renderer parks those sweeps at their final
keyframe, both off screen, and draws nothing, while the game draws them
across the screen. That would mean the capture caught them mid-sweep
inside their t=150 to 600 window, and that the "groups hold" reading
needs re-examining for those two elements.
What this cost is worth stating plainly. Five iterations -- the additive
blend test, the pivot analysis, the vertex-colour capture -- were built
on an identification made by elimination on one screen and never checked
against the draw's own coordinates. The eliminations themselves stand,
because each was measured against the capture rather than against the
identification. The chain that pointed them at ptlogo_back2 did not.
METHOD takes the general form.
title's rotation
Looking for where the rotated quads come from, the obvious candidates
were the three keyframe words at +4, +8 and +12 that ui_layout.rs
documents as zero.
They are not zero. Across 72287 keyframe blocks disc-wide they are
non-zero in 4.81, 4.56 and 15.82 percent of blocks, and read as signed
values clustering on 180, -180, 90, -90, 120 and 22 -- degrees. Three of
them, so plausibly rotation about three axes. I have marked that amber
because it is the shape of the numbers and nothing more; no observed
rotation has been tied to a value. The doc comment is corrected
regardless: "0 on every frame seen" was an artefact of the sample.
And they do not explain the screen I was chasing. Every element of
GP_TITLE build 4 has all three at zero, element by element, while the
game demonstrably submits rotated parallelograms there. So the title's
rotation comes from outside the keyframe data and is still unidentified.
One correction to my own last write-up, flagged rather than left: I
stated that the skewed draw IS the swoosh. It is the only skewed geometry
in the capture and the swoosh is the only diagonal element on the screen,
so the inference is reasonable -- but I never confirmed it by matching
the draw's texture or screen position to that element, and I should have
said so the first time.
blit cannot
Ran the route I corrected last iteration, and it answered the question I
had classified as undecodable.
First the ninth candidate died: every vertex colour in the whole capture
is <alpha>FFFFFF, white RGB with only alpha varying. The game passes no
colour, so that hypothesis is refuted from the running game rather than
from the file.
Then the geometry gave it away. Draw 2 submits TWO PARALLELOGRAMS,
neither axis-aligned -- edges (0.54,-0.56) and (0.44,0.79), roughly 45
and 61 degrees, both extending to y=+/-1.81 in NDC, well off screen.
That is the diagonal Z stroke. And ui_layout::blit walks rows and columns
of an axis-aligned rectangle; it has no rotation at all. So we blit the
sprite upright where the game draws it skewed, which is exactly the
signature I measured: right on average at +1.83, right in position with
the correlation peaking at zero shift, wrong in structure at 0.70.
Nine candidates, and the answer was never going to be a field, because
the difference is not in the sprite -- it is in how the quad is built.
Two things left open and written as such. The decoded keyframe carries
fade, scale, tint, x, y and time and NO rotation, so where the rotation
comes from is not decoded. And the pink-versus-white reading I have been
carrying was a visual comparison of two differently-shaped renderings; it
should be re-checked after geometry rather than treated as a separate
defect.
iteration
Two loose ends, both on my own work.
The disc-gated suite finally completed: all six binaries, 131 passed,
0 failed, one pre-existing ignored. So the rest() fix and the T8adImage
flags field are fully verified rather than verified-on-three-binaries.
The reason earlier attempts produced empty logs was that the tool
timeout's SIGTERM reached the whole process group; setsid nohup survives
it, which is now a METHOD line.
And a correction that matters more. Last iteration I closed the swoosh as
undecodable from the disc and named a per-draw GPU capture as the next
route, "because it reads the actual blend state". It does not. Reading
command_processor.cc, each captured draw records primitive type, index
count, index-buffer address, VS and PS ucode hashes, the pixel shader's
texture bindings, and vertex attribute 0 of binding 0. There is no
RB_BLENDCONTROL dump.
So the route splits, and I have said so rather than leaving the wrong
version standing: the capture can test a per-draw VERTEX COLOUR today
with no code change, which would explain white-versus-pink directly, and
getting the blend mode itself needs a Canary change to dump the blend
registers. Either way it is instrumentation rather than another field.
The general lesson goes in METHOD too: validate a recommendation before
leaving it as advice. A named next step is a claim like any other, and I
made it without checking.
the disc
Testing my own caveat instead of leaving it hanging. I had worried the
plate-free title capture at t=4.0s was too early, since elements carry
keyframes out to t=600 and I had judged "settled" from mean luminance,
which cannot see a thin sprite still moving.
The plate sits at y 550-600 and the swoosh band at y 112-225, disjoint,
so a LATE capture works even with the plate present. The band correlates
0.7342 at t=4.0s and 0.7353 at t=21.5s -- identical to a thousandth over
17.5 seconds, with the band mean flat at 127. So the band is settled by
4 seconds, the capture I handed the port agent is sound, and my caveat is
withdrawn. It also corroborates the earlier finding that groups hold:
nothing crosses that band in 22 seconds.
That was the seventh candidate for the swoosh and the last one that was
not a format field. Pivot is inert at scale 100 and there is no
displacement anyway; fade is white-with-alpha; tint is white; the texture
is blue-leaning; additive via +0x04 bit 0x02 makes everything worse; the
capture is settled. The residual is stable and modest -- band mean +1.83,
edge-correlation 0.70 against 0.92 frame-wide.
So I have classified it rather than generating an eighth candidate:
undecodable from the disc, with the reach written out. And I have named
where a next attempt should start, which is not another field -- a
per-draw GPU capture of the running guest reads the actual blend state,
vertex colours and draw order for that band instead of inferring them
from the file. That is where static RE stops here.
candidate with the check I named
Last iteration I offered the pivot mismatch as a quantified candidate for
the swoosh rendering too thick, and said explicitly that whether it bites
depends on which pivot the compositor uses and that I had not checked. I
checked.
blit takes the drawn size from the TEXTURE and uses the pivot only as the
scale anchor, kf.x minus pivot times scale-minus-100 over 100. At scale
100 that term is zero. And every one of the seven swoosh elements --
ptlogo_back2, its glow and the five eff segments -- reports a single
scale of (100,100) across all its keyframes. So the mismatch, real as it
is in the data, cannot move or resize the swoosh in our render. Refuted,
under my own name.
It is not harmless everywhere, and I have said where it does bite:
ptlogo1 and ptlogo2 run 100, 101, 103, 112, 150 during the build-in, so
there the wrong pivot really does displace them -- during the animation
rather than at rest.
What that leaves is the blend, by elimination rather than by evidence.
Position and size are the texture's own and right, fade is
white-with-alpha, tint is white, the texture is blue-leaning rather than
pink. Seven overlapping sprites at 5 to 23 percent opacity stacked with
plain alpha-over saturate toward opaque white, which is what we draw and
would read as thicker beside the game's thin coloured stroke. Recorded as
the remaining candidate and explicitly not as a diagnosis -- no blend
mode has been identified in the data.
Closing the residual I left explicitly unidentified last iteration, and
correcting my own claim twice over in the process.
The cheap test first: is the oscillation global? No -- the bottom-right
corner is flat at sd 0.003 and uncorrelated with the whole frame. So a
per-tile amplitude map over an 8x6 grid, which localises it hard: sd 7.65
in the band x 318..954, y 560..672 at lag 2.3 s, against 0.06 on the
wordmark. That band is the PRESS (A) BUTTON plate's rest position.
Decoding ptbtn00f.rat, the plate's highlight variant, gives the pulse
itself: alpha 0x00 -> 0x06 -> 0x4a -> 0x50, held, then back down through
0x4a to 0x06 across t=6..105. A glow that fades in and out, which is
exactly what a press-start prompt does.
So "the title screen loops at 2.2 s" was wrong in both halves. The title
ART is near-static apart from the two decoded sweeps at 7.5 s and 9.5 s;
what pulses is the PLATE, and the plate is build 2, not build 4. REFUTED
carries it under my own name.
Still amber, and said so rather than rounded off: the cycle LENGTH is not
readable. The keyframe group's last block has no time -- that slot
belongs to the next group -- so the declared span is at least 105 units,
1.75 s, against a measured 2.3 s. Consistent with a final block
extending the tail. Not confirmed.
Two of my own claims from last iteration die here, and the decode is the
thing that kills them.
Searching for the SHAPE instead of an assumed count found it immediately:
fade words matching 0x??ffffff at alternating strides of 24 and 16, which
is a 40-byte block carrying fade at +0 and tint at +24 -- exactly the
build layout, starting at +0x68. So "a leaf record's keyframes are not in
the build's 40-byte layout" is withdrawn. They are. My scan failed
because I demanded 29 strictly-increasing times, having read the word at
+0x004 as a keyframe count; the records hold THREE.
Decoded, both are horizontal light sweeps. ptloop01 runs pteff03.t32
from x=-639 to x=1521 over t=150..600, which is 450 units or 7.5 s;
ptloop02 runs pteff03a.t32 the other way over 570 units or 9.5 s.
Which refutes the second thing. Last iteration I attributed the measured
2.2 s oscillation to these elements because build 4 declares them --
association, not evidence. A 7.5 s cycle would give about three peaks in
a 22 s capture and I counted eight. So the loops are not what I measured,
the 2.2 s stands as a measurement, and its source is now explicitly
unidentified rather than wrongly assigned.
METHOD gets the general form, because it is what cost the previous
iteration: do not hard-code an expected element count into a structure
scan. Scan for the shape and let the count fall out.
Continuing the one gap from last iteration -- turning the title's
measured 2.2 s loop into a decoded cycle. It did not turn.
The record itself is found and partly confirmed. ptloop01.rat is an
opt-linked leaf RATC at 0xbb5966 in build 4, placing pteff03.t32, with
0x1e -- thirty -- in the high half of the word at +0x004, which is the
~30 keyframes ui-rat-layout.md predicts for a loop record. Its pivot
fields read 200 and 90, matching exactly what screen info prints for
ptloop01, so this is the right blob and the header offsets hold.
The times are not there. A build's keyframes are 40-byte blocks with the
time at +36; scanning every 4-byte-aligned start across a 0xC0 window for
29 strictly increasing values at that stride finds nothing, in either
loop record. So a leaf record's keyframe layout is NOT the build
placement layout, which is now a line in REFUTED because it is the
obvious first assumption and it is wrong.
The 2.2 s stays measured and the port hardcodes it. What the next attempt
inherits is the record's address, a confirmed pivot, and one layout ruled
out -- which is the useful part of a negative.
Two pages of this corpus disagreed. MISSION says the splash "is the RATC
screen, which already renders"; ui-paint-order-key.md says it "cannot be
rendered by screen render at all". Running the tool both ways settles it.
screen list --all shows all 16 GP_TITLE entries instead of 12, and the
four the default listing drops are the splash, each half shipped twice:
entries 10 and 13 are the white SQUARE ENIX publisher logo, entries 11
and 14 the GAME ARTS / SETA / studio anima developer logos. Entry 11's
seven elements are the three logos, their three _eff glows and the
palogo_eff0.prm backdrop -- exactly the composition ui-paint-order-key.md
had measured for the splash without being able to draw it.
So the "cannot be rendered" line is wrong and is corrected in place. What
is true is narrower and worth keeping: the splash is invisible to the
DEFAULT listing because is_build wants a .rat child, so anyone who does
not pass --all concludes it is missing. That goes in METHOD -- a default
filter can hide a whole screen and the corpus will record it as absent.
The payoff is for the port rather than for the RE: the first of the five
screens now has a reference composite, which it did not have. Marked
amber on one point -- there is no framebuffer capture of the splash in
this repo, so the match to the running game is by description against the
milestone-1 notes, not by pixels.
because the test probed the wrong state
Nine stores to the phase field this+132, control-gated on the 136 query
returning at least the 18 known ones. Attributed to the handler each
lives in, they give the phase graph: entry -> 2, then 2 -> 0 (splash),
2 -> 3, 3 -> 4, and 4 -> 2 on event 0.
That last edge is the problem with what I did last iteration. The event-0
block sets TWO fields one instruction apart -- stw r28,136(r30) for state
0 and stw r11,132(r30) with r11 = 2 for phase 2. So B from the menu lands
in PHASE 2, the same phase as the boot title, and the test I ran
believing it probed phase 4 state 0 probed nothing of the kind.
So the refutation is withdrawn. Worse for me and better for the idea: A
working on the B-returned title is exactly what the hypothesis predicts,
since phase 2 is the phase that references BUTTON. The hypothesis is back
to untested, now consistent with two observations rather than one, and
what it still needs is the attract-returned title's phase -- which no
test so far has read.
The measurement from that iteration stands, because it does not depend on
the phase: the B-returned title accepts A, so only the ATTRACT-returned
title is inert. I have said so explicitly in the page rather than letting
the withdrawal take the good half down with it.
METHOD gets the lesson: before testing "state X behaves like this", check
your route actually reaches state X. When a transition writes more than
one field, read the whole block and not just the store you were looking
for.
this corpus thought
Last iteration I proposed that the boot title is phase 2 and the
attract-returned title is phase 4 state 0, and that only phase 2 wires up
A -- a candidate answer to the oldest open question in
canary-scripted-input-traps.md. I marked it untested and named a test
that needed reading this+132 at runtime.
There was a cheaper one. The decoded graph says B from the main menu is
the edge 4 -> 0, so the B-returned title IS phase 4 state 0. If that
state were the inert one, A would not work there.
A works. Control first -- A on the boot title opened the menu -- then A
on the B-returned title opened it too. So phase 4 state 0 handles A and
the hypothesis is refuted, one iteration after I raised it. Recorded in
REFUTED under my own name, and the conditional stated: the edge is
decoded, but which title it produces is my inference.
The test earned something better than it cost, though. That page frames
the problem as "the boot title accepts A, a later title does not", and
that is too broad -- the B-returned title accepts A perfectly well, a
case nobody had tried. Only the ATTRACT-returned title is inert. Whatever
causes it is tied to the attract cycle specifically rather than to the
title having been shown before, which is a much smaller haystack.
Also reproduced in passing: 4 s after B the plate is absent and fades in
after, the same near-miss this corpus has already been caught by once.
confirmed three ways
Chasing where the event code comes from, sub_821C5580 turned out to do
two things worth having.
For Q6 it arms the outer gate: li r11,3 ; stw r11,16(r28) is exactly the
value sub_821C7850 tests before dispatching, which answers the "what does
this+16 == 3 gate on" question I left open last iteration. It then passes
r5 = [r27+4], an event read out of a structure, so the vocabulary is
still not enumerable. Of the four callers of sub_821C7850, three pass
constants -- 0, 0 and 5 -- and this one does not.
The accident is a few instructions earlier: li r5, 1103 into a sound-play
call. 1103 is a BGM cue id, BGM_103. That closes a residual I had written
off as undecodable: which bank the menu plays. The cue table genuinely
cannot say, since its BGM entries are numeric -- but the code can, and it
checks out three independent ways. The census says BGM_103.slb is two
waves of 3876864 and 3930112 bytes; the XMA probe at the main menu saw
two stereo streams of 3876864 and 3930112 bytes. Byte for byte on both.
That also corrects me. bgm-two-stems.md said those observed sizes matched
no bank's declared waves and inferred the decoder gets a window rather
than a whole wave. Wrong on both counts -- I had checked only the BGM_0xx
rows of the census because that was the block on screen. METHOD gets it:
check a measurement against the whole reference set, not the part you
happened to be reading.
Chasing where the event code comes from turned up the level above and
corrected my own framing three sections running.
sub_821C6458 has exactly one caller, sub_821C7850, and that caller is
itself a dispatcher -- on a SECOND, outer state field at this+132, five
phases, gated on this+16 == 3. Phase 4 is the title/menu machine I have
been decoding. Phase 0 is sub_821C5690, which the iterate3E notes had
already fingered as the splash mechanics from a completely different
direction. That agreement is the useful check here.
So GamePart_Title has two nested state fields: a phase at this+132
choosing which sub-machine runs, and the state at this+136 inside phase
4. Everything I decoded over the last three iterations -- ten states,
eighteen edges, the event dispatch -- is phase 4 ONLY. Phases 1, 2 and 3
are untouched. I have called that out in REFUTED against my own earlier
wording rather than quietly restating it.
And the event code is forwarded, not created: all five phase handlers are
called with the same (this, r29, r28), where r28 is sub_821C7850's own
third argument passed through untouched. So the event vocabulary is
defined at least one level further up, and finding what 3, 5, 8, 10 and
25 mean means going up again. Not done.
own corroboration from last iteration
Last iteration I listed a candidate name vocabulary from "upper-case
strings the callers reference", marked it amber because those are not
proven arguments, and named the check: confirm per call site which string
actually lands in r5. Ran the check. It fired.
The first extractor FAILED ITS OWN CONTROL -- a backward scan for
addi r5 recovered 7 of 48 sites and missed both title sites I had already
read by eye, because there the name is built in r27 and reaches r5 via a
later mr. Rewritten as a forward register simulation it reproduces all
three known sites and recovers 46 of 48, giving 28 distinct names.
Two of my own claims die. DIFFICULTY and EXTRA_MENU are NOT arguments at
any site -- they are strings that merely live in the same functions, so
the "independently corroborated by measurement" line I wrote was an
artifact of the loose filter. Only TUTORIAL_MENU survives. And
sub_821CC860 is not a screen factory at all: its real arguments include
BG, BLACK, FADE, FILE, KEY, PAD, SOUND and GAMMA_RGB. It is a generic
name-keyed lookup, mostly config, which the title part happens to call
with its three state names.
The title finding itself is untouched -- sub_821C6458 really does
transition by calling this with TITLE_SCREEN, TITLE_MENU and LOADING.
What is gone is the inference that the 28 names are a screen vocabulary.
Two METHOD lines, both earned here: strings a function references are not
arguments it passes, and a backward scan for a destination register
misses arguments staged through a scratch register.
TITLE_MENU, LOADING
Taking the one residual I had ranked as worth doing: sub_821C6458, which
the previous note left explicitly unread.
It is the title part's screen-state function -- 4460 bytes, has EH,
called from exactly one place, sub_821C7850, which sits beside the
creator the registration site points at. Its string references are
TITLE_SCREEN, TITLE_MENU, LOADING and BASE_INFO. Those are precisely the
states measured off the running game, in the game's own words, and
BASE_INFO is this corpus's own marker for a screen-config lookup rather
than a table read.
That sharpens Q6's negative rather than overturning it. The title part
DOES ask for configuration keyed by TITLE_SCREEN and TITLE_MENU, and
config.ini -- the disc's only config -- has no such sections, only an
empty [SYSTEM] and [LANGUAGE]. So the game asks the question, the shipped
disc does not answer it, and the defaults live in code. The port can
transcribe the state names even though it must author their contents.
Marked amber, not green: this is characterisation from string references
and call counts, not a read of the control flow. Two self-references
inside the function look like jump tables, which is what a state machine
compiles to, but I did not confirm that and nothing here shows which
state leads to which.
One refutation recorded because it looked like a find for several
minutes: the {func, func, ptr} triples at 0x820a3b48 read convincingly as
a state/handler table and are nothing of the kind -- the bytes before
them are the tail of a RegisterToFactory diagnostic string and the data
column is zero-filled descriptors. Static-init records, not dispatch.
extracted", and the tool was in the build all along
Two iterations ago I closed Q8 by declaring the SE audio undecodable:
Static.slb has no wave boundaries, there is no XACT container anywhere,
and I said the index "exists only at runtime" as though that put it out
of reach. The disc half of that stands. The conclusion did not.
This build of Canary carries a cvar called xma_param_probe, added by this
project, whose own comment says it logs each XMA stream's parameters and
head bytes so raw sound.pak entries can be matched to real decode params.
It has been sitting in the startup CONFIG DUMP of every log I have read
this session.
Run with it, driving the main menu: a d-pad move spawns a new mono 48 kHz
stream of 4 packets / 8192 bytes, and B spawns a different one of 2
packets / 4096 bytes. Searching their logged head bytes in Static.slb
finds each at exactly one offset -- 0x1ec0 and 0x0ec0 -- and the two are
contiguous, 0x0ec0 + 4096 = 0x1ec0. So the bank is a packed run of whole
2048-byte packets with no delimiters, which is precisely why the seek
scan found nothing: there is nothing to find. A wave is (offset, packet
count) and nothing else.
That splits Q8's binding cleanly. Event -> WAVE is now measured: the port
can have the audio. Event -> cue NAME is still a name match on the
authors' identifiers, and the page says so.
The same run settled something for Q10 too. Sitting on the main menu, TWO
stereo 48 kHz streams were decoding simultaneously. bgm-two-stems.md said
that observation was what it needed and that this container could not
make it; it can, and a music bank's two waves are now measured as
simultaneous rather than only inferred.
METHOD gets the general lesson, because it cost two iterations: check
what instrumentation the local build already has before declaring a
question blocked on tooling.
closing Q8
Last iteration I named Pj_Silph.xgs as the route to the SE cue index, on
the reasoning that XACT keeps cue->wave there. That route is dead, and
the negative is properly bounded: its 533-byte entry is high entropy with
no XGSF magic, +/-8 KB around it has none either, all 1.08 GB of
sound.pak contains zero XGSF, SDBK and WBND, and the executable has no
XACT or .xgs string anywhere. Control run first -- the same scan finds
RIFF in BGM_001 exactly where the bank structure says it should be, so
the scan works and the magic genuinely is not there.
So the .xgs and .slb extensions came from the authoring tool and survived
into shipped data that is not XACT at all. That goes in METHOD as a
general trap: check for the magic before committing to a parser for the
format an extension advertises. Writing an XACT parser would have been
wasted work.
What that leaves for the SE audio is undecodable-with-reach rather than
pending: Static.slb has no wave boundaries, there is no project file, and
none of the sound table's five records carries an offset. The bank is
certainly the right one -- 8 353 472 bytes over 322 cues is 0.84 to 1.01
seconds each at the bitrates the disc uses elsewhere, exactly the shape
of a bank of short effects. Only the index into it is missing, and it
exists only at runtime.
Q8 closes: vocabulary and bank decoded, the event binding a name match on
the authors' own event names, and the audio explicitly something the port
must source or author rather than extract. With this every MISSION row is
answered.
rather than leaving it amber
Q6's second half asked what the game READS to decide the boot order. The
answer is nothing, and the value here is the reach rather than a find.
Four places checked, the order in none of them: config.ini's [SYSTEM] is
empty and it is the disc's only config; the movie manifest carries the
boot-side assets but no transitions; the requested GamePart id was
already shown to exist only as a stack argument in flight, with no
persistent field and no literal store; and the string GP_ADVERTISE_DEMO
has zero xrefs of any kind, so nothing in the code reads the attract
entry of the id table. A transition is a call with an id argument.
Traced as far as it goes cheaply: the RegisterToFactory<0,
GamePart_Title> string is referenced from exactly one site,
sub_8280E148, which also takes the address of sub_821C7D98 -- where a
factory template puts its creator. Marked amber, because that is position
and convention rather than proof, and sub_821C7D98 has zero .rdata
references, which fits a new+ctor thunk and not a state machine. The
substantial function in that neighbourhood is sub_821C6458 and I did not
read it.
So Q6 closes as answered with the driver classified as code rather than
data, which means the port AUTHORS the sequence -- and that is fine,
because the sequence itself is measured end to end and the handoff now
carries it in one line.
GameParts exist at all
Q6's second half, advanced but not closed, and the negative is the point.
config.ini is the disc's ONLY config file -- one find over the whole
extract -- and its own Shift-JIS header calls it the "Application /
GamePart initial settings table". Its [SYSTEM] section, which that same
comment says holds what the game and every game part share, is EMPTY. So
the boot order is not in disc-side configuration at all, and that search
space is now closed rather than merely unexplored.
What the file DOES carry is the language: XC_LANGUAGE_* -> eng/jpn/deu/
fra/esp/ita, defaulting to eng. That is the mechanism behind the EN/JP
build pairs in GP_TITLE and the <lang>.pak families -- a question the
corpus had described but never traced to its input.
Then the registry. Pulling every RegisterToFactory diagnostic string
binds 24 of the 29 GamePart ids to a C++ class, and five ids have no
registration site: 1, 2, 16, 18, 28. Id 1 is GP_ADVERTISE_DEMO, which
agrees with what I measured two iterations ago -- the attract loop is the
TITLE replaying ADV.wmv, not a transition into an advertise part. Marked
amber, not green: it is an argument from an error message, not from code.
Two things fall out for Q4: ids 3 and 4 are the same class
(GamePart_SaveLoad, one part with two ids), and the menu buttons' ids now
match the executable's own class names rather than a list of table names.
Still a name match, one level closer to the code.
What is still missing is the transitions themselves, and I say so: the
manifest gives the boot-side assets, config.ini the language, the
registry which parts exist. What decides to advance is in
GamePart_Title's code and that dig has not been started.
Two corpus claims died in one boot, and both were costing us.
Q9's last row: a movie IS skippable with a single A. One tap ~45s into
the boot put the title up at ~57s, against a ~193s no-input baseline
measured over three boots, and Canary's own keystroke counter went 3->4
so exactly one press was delivered. The skipped-to title is fully
functional -- it draws the PRESS A plate and a second A opens the main
menu. What actually breaks the boot is hammering: the 88-press run in the
traps doc. The scripts' "tapping breaks the title" comment is too broad
and costs every scripted boot two and a half minutes.
Q4's last row: A on NEW GAME does not hang. It opens DIFFICULTY
(EASY/NORMAL/HARD/BACK, focus on NORMAL), then SELECT DATA, and only then
does the guest throw -- at PC 0x82307128, which is inside sub_823070B0,
the cache-manager STL erase this corpus already documents and which has
nothing to do with the menu path. The screen sat unchanged for 90s
because it was a menu waiting for input from a loop that never pressed
anything. That is now a METHOD line: a screen that never changes is not
necessarily hung, and the fix is to look at it and press something.
Also METHOD: never run ps -ef in this container -- all three long-lived
processes carry the entire loop prompt as argv.
not
Q8. The cheapest thing nobody had tried was to look at the SE cue names,
and they are semantic: SE_UI_CURSOR (2), SE_UI_DECIDE (3), SE_UI_CANSEL
(4), SE_UI_IMPOSI (5, the error buzz), SE_UI_SUB_WIN_OPN/CLS, and
SE_UI_SPLASH_IN/OUT. That is exactly the move/confirm/back/error
vocabulary the question asks for, named by the authors after the EVENT
rather than the sound. 322 SE cues committed as reference data.
They all live in one bank: BANK_SE is a single field reading Static.slb,
and the disc-wide check agrees -- 0 of the 322 has an entry in FILES, the
5135-path list that names every voice, briefing and BGM bank.
I am calling the binding a NAME MATCH, not a measurement, and the page
says so in the same breath as it gives the table. It is a stronger name
match than Q4's GamePart ids -- these are the authors' own event names --
but nobody has watched the game emit cue 2 on a d-pad press, and this
container is muted against a dummy device so there is no audio path to
watch either.
One new negative, with its reach. Static.slb's 8 353 472 readable bytes
contain zero RIFF, zero seek and zero WAVE, scanned over the whole
buffer. The boundary marker that works for all 7 620 other banks is
absent, so the cue is named, the bank is named, and the wave inside it is
not locatable. The UI sound effects cannot be exported yet, and that is a
gap in the assets rather than in the naming. The named next step is
Pj_Silph.xgs -- the XACT project SETTINGS points at, which IS in
sound.pak at TOC 9454, though its 533 bytes carry no XGSF magic and its
region's phase says they are probably the previous bank's tail.
new-game intro has a name
Q9, answered twice over -- once from the disc and once from frames I had
already captured and not used.
The movie manifest's first eight slots are the whole boot-side flow, and
the slot key is the role: LOGO1-4 -> logo1-4.wmv (not on the disc, which
is why the splash is a screen), ADVERTISE_MOVIE -> ADV.wmv, STAFF_ROLL ->
the credits reel, MS00A -> S00A.wmv, MS01A -> S01A.wmv. So the new-game
intro is S00A.wmv, decoded -- which is the half I could not test at
runtime, because A on NEW GAME hangs the emulator.
And there is no separate boot-intro slot: ADV.wmv IS the advertise movie,
the boot just plays it first. Confirmed independently by matching 19
attract frames against five candidates -- 15 hit ADV with a playhead that
advances monotonically at the sampling rate and ends at 137 s, its full
length, with the title back on the next sample. So the attract movie
plays to its end; nothing cuts it short.
That corrects me. Two iterations ago I recorded the attract movie as
"~85 s, so probably not ADV.wmv" -- arithmetic on a start I never
observed, since sampling began 39 s in. REFUTED and METHOD both take it,
along with the matcher's real failure mode: a near-black frame has no
signature, and its runner-up is not evidence.
Skippability I did NOT settle, and the corpus contradicts itself: one
page says A skips a movie every time, while the boot harness deliberately
never taps during one because it breaks the title. Named the one-boot
test rather than picking a side.
question expected
One iteration, as the mission gates it, and no Ready Room work started.
The two criteria it named both pass. It is NOT 3D: the corpus already
holds a capture of the running screen and it is a flat six-item menu in
the same visual language as the title. And unrecoverable paths do not
mean unrenderable screens -- content enumeration finds 60 builds in the
1106 entries.
What kills it is something the question did not ask. GP_READY_ROOM.pak
does not appear to hold the Ready Room's own menu at all. Two signals
agree: the four largest builds composite to briefing panels -- a planet
readout, unit markers, direction arrows -- and, much harder, the 902
bundles declare 317 distinct element names of which NONE matches any of
the six labels on the screen (takeoff, brief, hangar, pilot, option,
mission, ready all return nothing). The pak is the briefing /
tactical-map content behind the BRIEFINGS item. If this is ever picked
up, the first job is finding where the menu lives -- GP_BUNK.pak is the
obvious place -- not naming this pak's keys.
The probe also reaches back into something already delivered: kind ==
0x3002 finds ZERO buttons across all 902 bundles here, while GP_TITLE has
16. 0x3002 is one member of a 0x3000 family with sub-bits, and this pak
uses 0x3000/0x3004/0x300c/0x3008. Nothing in the milestone changes --
every screen in scope is GP_TITLE -- but the handoff now says not to ship
that equality as a general button test.
not something to concatenate
Q10, and it starts by withdrawing the question's own premise. BGM_001 is
not three sub-waves of 10 KB / 4.47 MB / 4.67 MB: the 10 KB is the bank
header. A bank is exactly TWO waves, and across all 32 BGM banks on the
disc the two always have the SAME duration -- equal to 0.01 s over
lengths from 37 s to 277 s. That alone kills intro+loop and kills two
halves of one piece, both of which require unequal lengths.
Four banks appear to break the rule and do not: BGM_106-109 are the known
leading-region straddle, and the giveaway is that the entry named
BGM_107.slb contains BANK id=1108. The seek packet counts pin each join
exactly, so they realign to the same two-equal-waves shape rather than
being dropped as noise. That trap goes in METHOD.
Then the roles, by decoding both waves to PCM. They are
sample-synchronous: transient-envelope correlation searched over +/-5 s
peaks at lag +0.00 s, and both waves stop at the same millisecond,
167.663 s. Two stems of one performance, meant to sound at once. Wave 1
is quieter, has almost no bass and is far more L/R-decorrelated, which
reads as a surround-rear pair or a second intensity layer -- I cannot
separate those two from the file, and say so: ChannelMask is 0x0002 on
both, and this game's channel metadata is already documented as
meaningless.
Two things the port needs that are NOT on the disc, both marked as
authored: the track is not a seamless loop (BGM_001 fades out and is
followed by 6.15 s of silence, no loop-point field found), and nothing
names which bank the menu plays -- all 32 BGM cues are numeric.
its timing is on the disc
Q7. Every title-side screen carries a full-screen black .prm quad that
paints last, and its keyframe group IS the transition: black at T0, clear
by T1, clear until T2, then back to black on exit. Read with the corpus's
start-of-a-ramp rule and Q1's time unit that gives 0.87s for EXTRAS,
0.97s for the main menu, 4.08s for the title -- from the file, not from a
stopwatch.
The disc-wide check is per-pak all-or-nothing rather than the 41% the
headline count suggests, and GP_TITLE's 6 of 12 is the useful row: the
six builds carrying a fade quad are exactly the six SCREENS, and the six
without are exactly the six overlays. GP_DIALOG is 0 of 133. That is
independent corroboration of the overlay finding from two iterations ago.
One piece is NOT on the disc and says so: the fade-OUT length. The fourth
keyframe has no time slot, because a group's last block stops four bytes
short. Measured instead, at 30fps, ~0.4s and the same both directions.
And a warning I earned: the luminance rise after a transition is NOT the
quad's ramp. The incoming screen's own elements animate in after the quad
has cleared -- 1.47s observed against a declared 0.97s. Time the fade
from where the frame is pure black.
Rig: screenshot samples at 0.5 Hz and cannot see a 0.4s fade at all,
which is why an earlier burst called this an instant cut. ffmpeg x11grab
at 30fps instead; both go in METHOD.
method lines withdrawn
Q5, measured off two boots: up/down move one item and WRAP at both ends
on the 5-item main menu and the 3-item EXTRAS alike; left/right do
nothing; B goes up one level and restores focus to the item you came
from (4/4); B on the main menu returns to the title; B on the title does
nothing. The menu opens on TUTORIAL -- the middle item -- 2/2, though a
third recorded run implies NEW GAME, so that one is reproducible rather
than invariant and says so.
Q4 by driving: LOAD GAME opens the save-slot list, TUTORIAL the lesson
list, OPTIONS the settings menu, EXTRAS build 6, MISSION SELECT the stage
list. NEW GAME is not tested -- A on it hangs the emulator and this
iteration needed the session. The GamePart ID behind each is NOT
measured: it is the entry of the decoded id table whose name matches the
screen I saw, and the page says so rather than wearing the badge.
And the withdrawal. Last iteration I wrote that these menus drop d-pad
presses shorter than ~0.3s. They do not. Once wrap is measured, every
press count I had is exactly right -- four presses moved four steps
THROUGH the bottom, which lands one above where a non-wrapping menu
would. I invented hardware flakiness instead of testing the ends of the
list. METHOD keeps the withdrawal rather than deleting the line.
Also: label brightness is not a cursor oracle here -- the background art
outshines the highlight on some rows. menu_focus.py reads the focus ring
instead, 254 vs <82, no tuning.
Q2. The archive is eight screens shipped twice, English and Japanese --
not the "build 4 title, 5 main menu, 6/8/9 submenus" the handoff claimed.
Build 8 is the JAPANESE main menu; 6 and 9 are the EN and JP EXTRAS, and
EXTRAS is the only submenu GP_TITLE holds. The PRESS (A) BUTTON plate is
its own build (2/3), composited over the title art and faded in a beat
later, not a state of build 4.
Confirmed by booting to the main menu and walking it: title, PRESS (A),
main menu and EXTRAS each match their render element for element. Builds
0/1 and 10/11 -- a DELTASABER / SYLPHEED A.I. plate -- were looked for in
the whole boot filmstrip, every title-side screen and the attract loop,
and appear in none of them; the reach of that negative is written down
rather than filled in with a guess.
Two rig traps went into METHOD: the menus drop d-pad presses shorter than
~0.3 s, and a grab 2.5 s after a transition can catch a screen mid-fade
-- which nearly wrote "the returned title has no plate" into the corpus.
The autonomous agent's loop prompt had grown to ~9 000 words of accumulated
findings, refutations and traps. That is a symptom, not a style: it was doing
the job the corpus should do, in the one place that dies with the container.
Three concrete failures followed from it, and each gets a structural fix rather
than a louder instruction.
1. IT REPEATED WORK IT HAD ALREADY DONE.
The "do not revive" list and the method traps existed ONLY in the prompt, so
nothing pointed a fresh iteration at them. Extracted verbatim into two
tracked files:
docs/re/REFUTED.md 105 claims tested and dead, grouped by subject so a
grep for your noun finds the neighbourhood
docs/re/METHOD.md the traps already paid for -- controls, inference,
searching, reading data, runtime
Both are linked from INDEX.md, and the loop prompt now opens by requiring
them to be read. This is the fix for "re-derived something already known":
the knowledge is now where the next iteration looks, not in a context window.
2. IT FORGOT TO ARM THE NEXT WAKEUP.
The prompt tried to solve this by shouting at itself in the first line. The
real fix is to stop asking: `loose` now defaults to a FIXED interval (45m),
so the harness owns the cadence and a forgotten ScheduleWakeup cannot end the
run. SYLPH_LOOP_INTERVAL= (empty) restores self-pacing.
3. IT COULD NOT PUBLISH, SO THE WORK ONLY EXISTED IN THE CONTAINER.
New `push-work`, plus a read-only credentials mount
(SYLPH_GIT_CREDENTIALS, default ~/.sylph-git-credentials). It pushes the
CURRENT branch only, refuses anything that is not auto/*, and never
force-pushes -- so the consolidated line stays a human's decision and a
confused iteration cannot rewrite history. The loop prompt now requires a
push on every iteration that commits, rather than at the end of some longer
arc, which is exactly when a container dies.
The prompt itself drops from ~9 000 words to 85 lines and, more importantly,
stops accumulating: findings go in the corpus, and the prompt points at it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
