"still when I lost a single point and try to serve again every further tap
does hurt me more until I am dead". Reproduced tap-for-tap this time, and the
serveNum fix was a red herring — a real bug, but not this one.
tossBall never stamped ball.lastHit. After any rally the zombie touched last,
the freshly tossed ball still said lastHit 'O' — and the previous pass moved
the server behind the baseline, past the passing-shot line. So the moment the
toss went up, the game saw an opponent ball beyond your baseline: "its
passing shot ends the duel", point to them, blood. Every tap = a toss = an
instant lost point, faster the more he panicked. Two fixes composed into a
death loop; the stale field was harmless while serving happened inside the
court.
One stamp in tossBall — ball.lastHit = sv.side — fixes every caller. The duel
and the gore mode carried the same latent bug for anyone tossing deep behind
the line after a racket-zombie touch.
Verified with the exact field shape: lose a rally the zombie touched last,
then eight badly-timed panic taps — one honestly lost point, zero
toss-bleeds. Before the fix every single toss bled.
41 tests, 15 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"still very weak and slow opponent. it also spawns at places where it is
impossible to react to my serve. so i win like always. in the cases i dont
win there is a bug. when trying to serve after i got hurt i get hurt again
and again until i die".
The death loop was G.serveNum. Tennis's pointTo is the only thing that ever
reset it, and the match bypasses pointTo entirely — so the first fault of the
match armed "second serve" permanently, and every missed serve after it was a
double fault, a point, and blood, repeating to death. Exactly the spiral a
rattled player serves into. matchReady(), the match's own point boundary,
resets it now; a regression pins it.
The receiver starts every point on its mark: matchReady drops it onto the
exact receiver position the tennis branch computes for O, deuce/ad mirrored,
instead of a random spawn x or wherever the last rally left it. An opponent
that cannot reach the serve is not weak, it is absent.
And the base opponent is a player now: matchSpeedBase 1.4 (the duel's 2.8m/s
lumber was tuned for one sudden-death point), matchFlub 0.08 (0.2 gives away
a fifth of all returns). Measured with the same autopilot that previously won
seven straight games: it now loses the match. Possibly too hot — the two
dials are named, Oliver's thumb decides.
40 tests, 15 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"the player is too weak in this mode. also the killing by smashing occurs too
often. i cant really control it. maybe obsolete? also the serving after the
first serve does not happen from behind the groundline".
matchPlayerHp 8 — the match brings its own pool. The gore mode's 4 was an
inheritance, not a decision, and a real game has rallies you deserve to lose
without the run ending.
Regicide is out of the match (kept in the duel, where any-weapon-counts is
the point of sudden death). In a full game every hard serve that happened to
meet the body was a free kill nobody aimed. A match ball never damages the
opponent — it passes through like a soft ball and racketAI plays it — so the
wound stages are exactly the scoreboard. That also made hp-per-opponent
meaningless, so the ladder escalates in racketAI instead: +12% speed per body
felled, whiff chance x0.82 per rung.
And serving resets to behind the baseline with real deuce/ad mirroring:
setupPoint's "serve from wherever you're standing" branch now excludes the
match, which falls through to the tennis positioning. Verified live: second
serve from z 24.3, x mirrored to the ad side, opponent unwounded by anything
but the score.
38 tests, 15 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Triage failed the only test that matters — "i think the normal player does
not grasp this" — and the lesson across four drill designs is that every
invented rule needs teaching, while the rules the scene already implies need
none. So the new mode is the game's own oldest rule: tennis.
MATCH TEST in the back room (#match). One game per opponent, real 15/30/40
scoring. Built on top of the duel machinery: G.match runs a permanent
interlude, which is already what makes serving tennis-like, hides the rack,
quiets the spawn director and wakes racketAI. The one change is in
resolveInterlude — a settled point feeds matchPoint instead of ending
anything. About 90 lines for a whole mode, because the deepest system in the
repo was already built.
The score is the body: a quarter of its pool per point won, floored at 1, so
a love game walks the racket zombie through every wound stage and the game
point fells it. Its points draw your blood — hp--, eaten at zero, matchHeal 2
per game won is how you last. Regicide is inherited: kill the body mid-game
and the game banks by hand.
The timing is drawn, per Oliver's ask: an approach ring around the incoming
ball that kisses a fixed inner ring exactly at the sweet contact distance
(gold inside the band), and an apex band beside the toss spanning the same
0.12m keyPressed scores as perfect, with a tick riding the live ball. Both
cues use the mechanics' own numbers, so the picture cannot lie.
Verified by autopilot: 7 games in 90 simulated seconds, opponents 1..8, pools
4/6/8/10, wounds tracking points exactly.
36 tests, 15 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"the marker flips all the time even during a ball being midair.. i am not
sure that a player would get the mechanic intuitively let alone without the
marker".
The flip had a specific cause and it was a design bug, not jitter. With lanes
held and speeds constant, arrival differences never change while bodies walk
— the queue only reorders on a kill or an arrival. The marker flipped because
it showed only q[0] while the rule accepts any near-tie within the slack:
every kill made it jump between tied bodies, so the marker was stricter than
the rule it was explaining. drillLegal() marks the whole slack set now,
bright head and dim ties, which is stable by construction. Measured: zero
head-changes across 12s of walking with no kills.
Learnability, two answers. The rule gets one modal, once per run, on the
mode's own calm court — CLEARANCE DRILL / DROP WHOEVER WOULD REACH YOU FIRST
/ FAST BEATS CLOSE / STREAKS PAY — because "most urgent first" is not
guessable from a chevron alone. And the markers are training wheels that take
themselves off: hidden at streak >= 8 (the streak is the evidence you can
read the court), back on any fault. The teacher steps in exactly when you
demonstrably need it.
31 tests, 15 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The sequence drill tested nothing — the strip was always visible (no memory),
the batch was slow (no urgency), the aim was easy (no execution). A checklist
with gore, and Oliver said so: "not interesting enough to keep me playing".
Settled in discussion, he picked triage + streak. One rule: always kill the
body that would reach you first. Nothing is called, nothing shown — the court
is the puzzle. arrival = distance / speed down a held lane, and the roster's
speed spread means the answer is not the closest body: a runner at the far
baseline beats a roller at the service line. Court sense — and it trains the
exact target-priority read the gore mode rewards.
Wounds are free: only the kill is judged, so pre-softening a heavy while it
is third in line is legal and eventually necessary. The slack (0.9s) makes
near-ties legal both ways, or the rule is a coin flip. The streak is both
punishment and payout: x1..x5, a misread shows OUT OF TURN and resets it, and
nothing else happens — the queue keeps walking. Caught is the real setback.
Best score banks to localStorage; the world board can carry it if it ships.
drillMark draws a chevron over the queue head for playtest calibration. It
answers the exact question the mode asks, so it must be off in anything
shipped.
Deleted: the alphabet, the batch, the strip, the eaters. Shamblers are
first-class stream members with pts 20; a roller is 100.
30 tests, rewritten around arithmetic riggings (known speeds and distances)
instead of luck; 20 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"one-shot is bad - way too easy - also we're losing all the nice gore
animation sprites" and "I noticed the last kill of a game can carry to the
spawn of a new zombie of the next round".
drillOneShot is off. The second half of that note is the part worth
remembering: a 1 hp body never passes through a wound stage, so every skin in
ZWOUND — arm off, ribs out, still walking — is skipped and the roster's best
art never renders at all. Kept as a switch, never a default.
The carry-over was a real unfairness, not a curiosity. A ball keeps pace
through a kill and travels UP-COURT, which is exactly where the next batch is
born — so the shot that finished your round flew on into bodies that did not
exist when you took it, and faulted you on a round you had not been shown
yet. The shot that ends a round is spent now, and drillCaught does the same
since it also refills the court up-court. Pierce within a batch is untouched:
that is honest occlusion you could have read, this was not.
Verified with real aimed serves through a whole round: round banks, ball goes
inactive, the next batch walks on with 0 faults and a clean sequence, and the
gang carries roster HP again (brute 5, ball boy 3).
33 tests, 15 seeds clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"please make the zombie killable with a single hit for testing" and "remove
the fake zombies for now".
drillOneShot drops every body on a drill court to 1 hp, applied at spawn to
the batch and the eaters alike. Worth knowing what it removes: the roster's
HP is what makes the called ORDER matter — a roller is normally seven serves,
so being told to take it third means starting on it first. At one hit that
planning layer is gone and the sequence is pure recall. Flagged in PROJECT.md
so it gets decided on feel rather than left on by drift.
drillEaterMax 3 -> 0 switches the plain shamblers off. The mechanism is
untouched — spawnEater, the wide entry away from the gang, the never-a-fault
rule — so one number brings them back.
The eater test now asserts whichever contract the knob describes (at 0:
nothing but the puzzle ever appears; above 0: they arrive, capped, always
plain shamblers) instead of being deleted and rotting. And the never-a-fault
rule gets its own test with a hand-placed body, so the rule the eaters depend
on stays covered while they are switched off.
31 tests, 15 seeds clean. Verified live: brute/steward/roller all at 1 hp
where the roster says 5/5/7, zero strays over 40s, one aimed serve puts down
the roller.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The pipeline caught two flaky tests of mine that sixty local runs had not.
Both were bad tests rather than bad code, and both were mine from this
session:
- the eater test counted SURVIVORS, not arrivals. An eater crosses the whole
court in ~24s and despawns, so a snapshot at 40s can legitimately show an
empty court — that read as "the eaters never came". It tracks arrivals over
the window now, and asserts the cap as a running peak.
- the auto-aim test never actually aimed at the body it claimed to: fy was
pinned at -0.95 and fx was a scaled x offset, so the ray leaned up-court and
the called body could genuinely be the best-aligned thing in the cone. That
is the magnet working. It now normalises a vector straight at the wrong
body, picks the one furthest in x from the called one so the two are
separable, and asserts the magnet takes what you pointed at.
Measured 8 failures in 40 runs before, 0 in 60 after.
And the durable fix: Math.random is seeded at the top of test.js. This suite
is a Dockerfile build stage, so a red run blocks the deploy, which makes a
1-in-5 flake worse than no gate — it trains you to re-run it. The game leans
on randomness everywhere, so the stream is made reproducible rather than the
randomness stripped out. SEED=1234 node test.js re-rolls it; 25 seeds pass.
The cost is noted in PROJECT.md: under one seed a guarded test can silently
take its early-out forever, so I checked all four guards still run their
bodies.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver's insight, and it retires a whole subsystem: "we don't need colors at
all! we already have different zombie types.. so the sequence is the zombies
themselves".
The bib and then the ground disc were both inventions bolted on to give a
body an identity it already had. A brute does not look like a runner, the
game is built on reading those silhouettes at speed, and the strip can show
the body's own walk frame instead of a swatch — no legend, no mapping, no new
art ever. It also deletes the accessibility problem outright rather than
deferring it, because a silhouette is not a hue. DRILL_COLS, drawColourRing,
drawBib, PIP_FACES, blip(), drillRingR and drillTones are all gone.
Two rules keep a type unambiguous: the plain shambler is never in the
alphabet (it is what the eaters are, so "not the puzzle" needs no marking),
and one of each type is on court at a time.
The new failure mode this created is instructive. Types move at different
speeds and drill bodies hold their lane, so one in a lane you are not in
walks the whole court and off the back untouched — stranding the sequence
asking for a body that no longer existed, i.e. the un-winnable state the mode
exists to prevent, reintroduced by a change that looked cosmetic. A batch
body crossing your baseline counts as caught now ("GOT PAST YOU"), which is
also just better design, and the speed spread is narrowed to 1.35..0.70 so a
batch stays coherent. Their real difference is HP and that is untouched: a
runner is one shot, a roller is seven.
Verified: 14400 frames of simulation, zero frames where the sequence named a
body that was not on court.
29 tests.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"remove the mob boy - make the normal zombies rarer and not so close to the
color gang".
sendBoy refuses during a drill, guarded at the function rather than at the
call sites — same shape as its existing G.interlude guard, so every present
and future caller inherits it. He is one more body crossing the lanes you are
reading colours down, and unlike everything else out there he is neither the
puzzle nor the threat. Measured: 0 boy-frames in 90 seconds. Gore no longer
gets mopped in this mode; the existing 40/30 caps are what bound it.
The eaters are rarer — every 12s falling to 5s, capped at 3 on court. At the
old rate they were a second crowd rather than an interruption, and a threat
you break off to deal with only works if breaking off is occasional. 8
arrivals in 90s where the old numbers gave ~13.
And they enter wide of the colour gang: best-of-9 lane sampling against the
bibbed bodies, the same trick the batch uses. Narrowest clearance measured
across 8 spawns was 0.91 m. They still converge on you afterwards — that is
what they are for — this only decides where they walk in, so they stop
screening the bodies you are trying to read.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"the colored zombies should also explode like they did before and make the
normal splash sound on destruction.. no bleeps please.. can we make the color
as a color circle the walk on? the sign they're holding is not nice. also
remove the ball slicing in this mode.. this leads to unwanted hits".
The silent deaths were my bug: the hook in killZombie did drillKill(zb) and
then RETURNED, and everything below that return is the death — decal, pile,
36-particle burst, splat, headshot crack. It resolves after the splice and
without returning now. Both halves are load-bearing and each was a bug in
turn: splice first or a freshly drawn combination names the corpse; don't
return, or the body simply vanishes.
The colour is a disc on the court the body walks on, drawn before the sprite.
The bib was a flat quad pinned to a walking sprite — UI stuck onto a
character, covering the art it sat on. On the ground it takes the court's own
perspective, measures 12.4px across at the spawn line against a chest patch's
few, and hides none of the sprite. A hard dark rim plus a bright inner line
is what keeps it from reading as another gore decal.
No slicing in the drill: a cut ball leaves the corpse and strikes whatever it
finds next, which in a mode where the wrong body is the failure is the game
choosing your target for you, badly.
And the mode has no sound of its own now, deliberately. The umpire reading
the list was vetoed, the arpeggio that replaced him was vetoed. blip() is
deleted. A kill bursts and splats like every other kill in this game; the
strip carries the list.
28 tests.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"make the game that it is actually winable meaning that all required colors
are visible on the far side before when a color combination is shown.. for
now no colors that are not in the current combination but occasional zombies
which don't have a color and are converging to the player trying to eat him".
Two rules define the file now. Every colour on court is IN the current
combination — no coloured decoys, so the strip can never name a body that is
not there. And a combination is only shown once its bodies are all on the far
side, which is free because the batch is BORN there: it is a property of the
construction rather than a check that could be wrong.
The pressure moved to the eaters: plain unbibbed shamblers on their own
clock, deliberately not flagged `drill` so they keep the gore mode's homing
and converge on you while the bibbed batch walks its lanes. Killing one is
never a fault; letting one reach you is the ordinary caught. The puzzle plays
fair and the danger arrives separately, which is a better trade than a puzzle
that cheats.
The batch is the round: kill the last of it and the next walks in on the same
frame with its combination already called.
One real bug, caught by the new test: killZombie resolved the drill BEFORE
splicing the body, so a combination drawn on the spot could name the corpse —
a round asking for a colour that is not on court, i.e. exactly the unwinnable
state this pass exists to prevent. Splice first, resolve second.
28 tests.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"what are the dice dots for?" -> "drop them", and "the dodge swipe should not
reset to the middle when firing".
The pips were colourblind redundancy so the mode did not rest on hue alone,
but they were noise to everyone who can separate the colours and they shrank
to nothing on a body at distance — exactly where the channel had to work.
Gone from the bib and the chip. Recorded in PROJECT.md as a known gap rather
than pretended away: if the mode ships, the answer is bibs that differ by
shape at sprite scale, not dots.
The dodge one was never a drill bug. The auto-assist recovered toward HOME.P
including its x, so firing walked you back to the centre line: measured, a
step out to 1.94 m was dragged to 0.77 m over one ball's flight — most of a
dodge, handed back for free, which is the welded-in-place complaint the
sidestep exists to answer arriving slowly instead of all at once. Splatter
recovers in DEPTH ONLY now and holds whatever lane you stepped to. Tennis
keeps the full recovery: recovering to centre is the real sport's footwork
and there is no sidestep there to protect.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The doc edit threw on its second assert last commit, so the code shipped
without its notebook entry. Also records the two ways a drill test lies to
you: one that never shoots gets caught and the sweep looks like a stalled
stream (use eatenR = 0), and zombies[i] is not the same body across frames
because the array shifts at the baseline.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver: "daniel announcing is maximum annoying" / "don't let the rush in
position just let them coming from the back in steady speed.. try to space
them not too close" / "when one combination is done just let the new one in..
in a continuous stream that increases in speed and difficulty" / "the rounds
can stay, they should be continuous".
The round survives as the unit. Every seam around it is deleted: no walk-on
to a mark, no announcement beat, no between-rounds pause, no deal. Bodies
walk in from the far baseline forever and a round is drawn from whoever is on
court — finish one and the next is drawn in the same frame, off the same
rolling population. The bodies you were ignoring a second ago are the ones
being called now.
- Daniel is gone from this mode. Each colour has a pitch; a round announces
itself as a four-note figure in half a second (blip()). drillTones 0 mutes
it. The recordings stay in assets.
- PLACEMENTS is deleted with the marks — nothing rushes anywhere and nothing
stops, so there are no shapes to place.
- A DRILL BODY HOLDS ITS LANE. updateZombies homes bodies onto the player's x
past the net, so they arrived spread and bunched into a knot right in front
of you; spawn spacing cannot survive that. Their lane is the puzzle here.
- One ramp on TIME SURVIVED drives speed, spawn gap and round length. Speed
is stamped per body at spawn, so the ramp reads as newcomers outpacing the
ones already walking, not the court accelerating under you.
- The empty-court watchdog is gone and nothing replaced it: with a stream
there is no state to be stuck in. One line covers every case.
27 tests, rewritten around the stream.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
"the drill mode immediately goes into some horrible audio loop" — on a phone,
never on desktop, and that is the tell. Three of the four defects can only
happen on a phone, and every one of them is a rule voice() had followed for
months that my new function quietly broke.
- A SUSPENDED AudioContext HAS A FROZEN CLOCK. Scheduling four clips at
currentTime +0/+0.9/+1.7/+2.4 while the context is suspended puts all of
them in the past by the time it starts, so they fire on one instant through
a 1.3s reverb tail. Resume first, re-read the clock after the decode.
- A second call layered instead of replacing. voiceSeq SET voiceUntil and
never checked it. It stops the outstanding list now.
- ...and cancelling a list has to give back the time it booked, or the
replacement queues behind a list that will never play. Measured at 6.6s of
silence before the call the player is waiting on.
- The amplifier was in drill.js: the empty-court net redealt on the spot,
making the repair rate the frame rate — and every deal calls the umpire. It
goes through the ordinary between-rounds beat now.
Verified by tagging each decoded buffer with its source file and counting
start() calls: one drill_call, three colours, spaced, zero overlapping voice
lines. The earlier "clean" reading counted fetches, which proves nothing —
VOICE[] caches, so a repeat play never fetches again.
27 tests; the new one empties the court every frame for six seconds and
asserts at most one umpire call per beat.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver played it: "the auto aim just mean no skills involved at all. this
should just go away.. there is no penalty at the moment.. make some
variations of the position on the court and the color".
The first two notes are one finding. With the shot locked to the called body
it was unmissable, so the fault could never fire — there was no penalty
because there was no way to earn one. Both snaps are deleted. What replaced
them is one line: aimTie is 0 during a drill. That is not auto-aim, it is
removing the guess — the ordinary magnet lets a nearer body inside the same
cone steal a shot aimed past it, which is right for a crowd you are mowing
down and exactly wrong where hitting the wrong body IS the failure. An
unaimed tap now lands on grass.
And the variation:
- six colours (purple, white) with the court dealt three to five of them, so
which colours turn up and how many decoys there are are live questions. At
four out of four every round dealt every colour and only the order moved.
- PLACEMENTS: fan, cluster, column, stagger, wing — weighted, mirrored per
round, lanes shuffled against the bib order. column stacks bodies in one
lane, and the ball hits the first thing it touches, so feet matter.
- pips are dice faces now, shared by bib and HUD chip. A row of five dots has
to be counted; a die is recognised.
Measured over 30 deals: all five shapes, counts even, all six colours, and
the first target screened by another body in ~10% of rounds. Verified end to
end in the browser that FAULT and DOUBLE FAULT both fire.
26 tests, including a named regression for the auto-aim staying gone.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The late game is repetitive because the optimal move never changes: kill the
nearest thing, now. This is the first swing at a mode where it does — a
coaching drill where the umpire CALLS a colour sequence aloud while bibbed
bodies take their marks, and you must put them down in that order while they
walk at you. Uncalled colours are decoys you have to ignore. Two faults is a
double fault and the round is theirs.
Back room only (DRILL TEST, or #drill): this is a feel-test for one question —
tense, or just noisy? — so it keeps its own round count and touches neither
tennis scoring nor the run.
- voiceSeq: voice() drops a line that would start while the umpire is still
talking, which is fatal for a list. Decode first, queue at accumulated
offsets, and the anti-overlap rule is satisfied instead of fought.
- Four colours with PIPS as well as hue — red/green alone makes the mode
unplayable for ~8% of men.
- drawBib is a ponytail-marked overlay that knowingly breaks the sprite rule;
it is the thing to delete if the mode survives.
- BOTH target pickers snap (flickTarget and zombieTarget). Snapping only the
slingshot left fifteen tap-serves in a row scoring nothing.
- Contact costs the point, not the run — permadeath stays in the gore mode.
- startDrill disarms the ordinary schedule; updateDrill redeals and logs an
INVARIANT if anything empties its court.
Eight new tests (24 total) pin the ownership contract both ways, the call
beat, the win, the double fault, the un-winnable-round guard and the death.
PROJECT.md carries the four open questions for the playtest.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The gun, beacon, flamer and rocket fell silent the moment they cleared the court:
firing and steering were both gated on phase 'rally', and every beat raised by the
kill that empties the court (a point, a won game, the mop-up) parks the phase in
'over'. A burst is its own control mode for its whole len, like a horde — its own
clock is the only thing that ends it. Not fixed by forcing the phase back to rally:
a parked 'over' beat that never expires is how pendingHorde gets stranded.
The countdown moved off the tile's bottom edge (the rounded frame, and the bottom of
the screen) into the tile: the armed slot's well drains as a gold wash, clipped to
its own rounded path, under the icon and labels.
16 tests.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
THEY KEEP COMING is a 2.4s 'over' beat again (warnHorde in beats.js) instead of a
flash raised on the same frame the wave lands, and the chapter boss walks on behind
an overlay — his name in red over ENTERED THE COURT — held for 3s by the new
self-dismissing modal clock (m.hold) rather than announced on a routine-set card.
startHorde returns a bool: if the court has been taken when the warning expires the
pendingHorde latch is re-armed rather than the set losing the horde it paid for. The
warning ticks on its own clock, not on the 'over' beat, so a beat that moves the
phase underneath it can't strand it (the mgT lock, again).
Three tests, 15 total.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Tapping a tile and landing a thumb-width off it returned -1 — free court — and
the tap became Space, which instantly armed whatever burst weapon was selected.
Reaching for the flamer and missing by 10px fired the gun: 22px tiles, a second
row the thumb hasn't learned, and a selection that moves under you (a won duel
re-equips the tin it tops up). rackHit now reports a dead margin around the
band (-2), and only the TAP dies there, decided at touchend — a drag or hold
starting in the same spot still steers and still serves, so burst steering
parked low keeps working.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
dropStale trusted e.touches — but a finger stolen for a system edge gesture can
stay listed there forever, so the stick survived and the player stayed welded to
the edge (field-proven: escape taps still swished, meaning dropStale ran and
found the id). The zombie is indistinguishable from a parked thumb by data, so
it's told apart by intent: in the absolute-steering modes, a new finger that
drags past the deadzone while the stick has been silent over a second inherits
the stick. Taps still swish and steal nothing.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
"They just overrun you without a chance" — two reasons, and the second one only
existed because of the first.
IT FOLLOWED YOU. The rush tracked your x every frame it was running, so the
sidestep was decoration: it simply came across and took you anyway. The lane is
LOCKED when the rush starts now. You get the whole 2.2 s run-up to leave it, and
what it runs down is that line, not you.
AND IT LANDED ON YOU. A rush that stopped at your feet left the boss standing
there afterwards, and nothing gets you out from under that — dodgeCool is 0.5 s
against a 0.28 s i-frame, so the second contact kills you while the first is
still fading. It stops on a STRIKE LINE 0.4 m short instead, reached on the same
frame the rush timer expires, so it turns and withdraws the instant it arrives.
The version of this that ran the rush THROUGH you and out the far side is why
bossStrike is a short stop and not an overrun: a body past the baseline is
deleted and scores a point, so the boss quietly despawned itself mid-fight and
left the fight running with nothing in it.
Measured, 45 s against the seller, the champion's third phase and the physio:
stand still and the first rush kills you (4.1 s); step out of the lane and you
survive all of it — seven or eight rushes, one sidestep each, the boss on court
the whole time and out at its line between them where you can hit it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
THE LEGS. Measured as lower-third silhouette difference between the two stride
frames: the groundskeeper, whose walk has always been right, is 0.80. The seller
was 0.017 — the same legs twice, which is why he hopped along on one of them.
Both sheets regenerated with the stride as the entire subject (one foot planted
ahead, the other trailing, and they SWAP): seller 0.42, champion 0.67.
Three things the champion's re-rolls proved, all of them the same lesson —
THE REFERENCE IS WHAT IT COPIES. Anchored on his own one-legged frame it drew
that pose again. Given a working walk from another boss alongside him "for the
legs only" it blended the two and returned a slim side-on walker, then a
pot-bellied man in cream. One character in the reference, one character out —
and pick the keeper by EYE, because the metric happily crowned the side view.
THE CHARGE. It rushed, then withdrew a fraction of what it covered, so it ended
the fight parked on top of you: dodge or die every two seconds with no window to
hit it, and nothing but a hitscan laser quick enough to use the window there
was — Oliver, "they come too close, they seem unbeatable". A rush now RETREATS
to its line afterwards, and to keep it from being toothless the other way (from
the band it ran out of clock 8 m short) the rush is TIMED rather than paced: it
covers whatever ground the retreat left in bossChargeT, so every charge arrives
and every charge has the same readable run-up. Measured over 30 s: five rushes,
each one reaching you, 13% of the fight in your face instead of all of it.
And the back room moved off the whole splash onto an invisible box in the TOP-
LEFT CORNER. Five taps in the middle of the screen — which is where your thumb
goes on the way to PLAY MATCH — now do nothing.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The weapon-test cheat sat on the SCOREBOARD, where a single tap is the miss-log
overlay — so the thing you tap to read a log was also the thing that opened a
sandbox mid-run, three taps in, by accident. That tap does one job again.
The cheats now live behind five taps on the splash logo, where nothing marks
them and nothing in play can reach them:
WEAPON TEST the old sandbox. It stocked only the BARREL weapons, on a screen
called the weapon test — it stocks every weapon in the game now,
eight of the nine slots.
BOSS TEST a portrait of every chapter boss, built by asking bossArt() what
each chapter gets, so the list cannot drift from what a run
actually spawns. Pick one and you land in a real fight at that
chapter's game number with the full arsenal at ×99.
Neither is a special mode: both ride the transition out with laserScene's trick
(force the turn, then step the ten-second cinematic out in one frame) and hand
you the ordinary update loop, ordinary scoring, ordinary death. A test that
plays by different rules tells you nothing about the game.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The roster ran out at chapter 6 and then repeated itself; game 70 got the
groundskeeper again, bigger. bossArt() now walks nine designs, one per chapter,
and hands out THE CHAMPION on every tenth — the floor(ch/10) term takes his
chapters back out of the count so chapter 11 opens the second lap on the
groundskeeper rather than skipping him.
70 THE STRINGER — leather apron, bandoliers of gut string, racket frames on
his back. He takes your lane and puts two dead-straight ones on it.
80 THE TOURNAMENT DIRECTOR — blazer, lanyard, radio. `adds` is new: he calls
stewards onto the court while you fight him, which is the one thing a boss
fight could not do before, since the stage is cleared for it.
90 THE STRAWBERRY SELLER — tray of cream, apron soaked red, ladle. A tank
that sprays four at once and then comes for you.
100 THE CHAMPION — torn whites, a laurel of strapping tape, the trophy still
in one fist. `rally`: anything under splatter quality comes straight back
at your head, so he is the one body you have to out-HIT rather than
out-shoot. `phases` lays three rows over his own as his health falls — he
holds the baseline and rallies, then runs the line drilling volleys, then
drops to a charge — and each turn is an event: cry, shake, the stand up.
Thirty-six frames through assets/_regen.py, the mascot's three-batch recipe
(2 stride, 3 wounds, 4 throws). The trap that cost the re-rolls: a throw batch
references its own wound frames, so a wounds batch that fails leaves the throw
sheet drawn against the SEED — two of the four came back half groundskeeper.
Wounds first, verify, then throws.
And the physio, who Oliver reported "does basically nothing but move towards
the net": he kept 4.2 of every 6.3 m he charged, so the fight was always over
before he arrived. `withdraw` is his own number now (0.2) and he charges half
again as fast — measured, he is on top of you in 11 s instead of never — and he
drills three balls a cycle on the way in.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The right edge is where iOS keeps Control Center, and it is exactly where the
horde's absolute finger-steering parks your thumb. When the system takes that
gesture it delivers neither touchend nor touchcancel: `stick` survives forever,
so every later touchstart falls through to the extra-finger branch (swish, and
only swish) while touchmove ignores the new finger's id. steerPX stays frozen at
0.97 and the player is pinned to the screen edge, melee firing, for the rest of
the run. Any new touch event proves the old finger is gone if it is not in
e.touches — drop the stick then. Verified against the bug: pinned at x=6.11
before, steers to -5.20 after, and a genuine second finger still only swishes.
And wCap 9 -> 99, which is no cap in practice: game 70 wants a bigger stack, not
a rationed one. The barrel weapons move from "paid out only once all four tins
are full" (a branch an uncapped rack would never reach again) to rungs of the
same ladder, so a late duel still hands out a rocket instead of a fifth bomb.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver, watching a sliced ball circle a shambler it had only wounded: "an
orbiting tennis ball midair, nobody knows what is going on." He is right,
and not only about that shot. Every ball in this game flies a parabola you
can read, so the one that thinks reads as broken even on the shots where it
works. The hook was the wrong primitive, not a hook with two bugs in it.
So there is no hook. At the instant of a kill the ball is struck again —
ONE velocity out of solveShot, the same solver the serve uses, aimed at the
next body's chest — and from the next frame it is an ordinary ball under
ordinary gravity. It cannot orbit, cannot hover and cannot outlive its shot,
because after the strike nothing about the slice is still running.
That also deletes the wound bug rather than fixing it. Six sites spend the
ball (lastHit = 'X') and not one of them cleared the hook, so a ball that
merely WOUNDED its target was left steering at 30% pace toward a point that
body had already walked off — the loop he saw. With no persistent state
there is nothing to clear: a ball spent on a body it failed to kill just
drops, like every other spent ball.
The hop draws its own band, white, so a chain leaves the same chalk streak
every other shot leaves. sliceTurn and sliceHookT are gone with the steering
that needed them; sliceArc stays, no longer as a turn-radius limit but to
keep a carom from doubling back down-court, which reads as a stray
deflection rather than as a cut.
Net 24 lines lighter than the thing it replaces.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A slice hooking onto a body two metres to the side flew a closed loop around
it forever: 285 frames of the same circuit, never arriving, never falling.
A constant turn rate has a constant turn radius (v/w = 2 m), and anything
inside that circle cannot be reached however long the ball steers — while
the hook pins vy at 0, so nothing else ends it either. A crowd is nothing
but close bodies, which made the one place this weapon should shine the one
place it hung.
The rate is no longer a constant. An arc leaving along the current heading
and passing through a body at distance d, bearing a, has radius d/(2 sin a),
so the ball needs 2 v sin(a) / d — solved every frame, because the body is
walking while the ball travels. sliceTurn stays as the FLOOR, which is what
it was always really for: how wide the curl looks when there is room.
Turning faster than needed is safe, the aim being re-solved each frame.
Two things the maths cannot cover, so they are ruled out instead: a body
behind the ball, which no tangent arc reaches at any rate (sliceArc), and a
target that walks out of reach mid-hook (sliceHookT, a hard 0.7 s licence to
steer). An endless orbit is now impossible by construction rather than by
the geometry happening to work out.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Slice is not a tin and not a rack slot. It is what a perfect serve DOES —
the toss ritual's 45 ms window, or a pull that stops inside perfectBand.
Every other weapon is handed to you; this one is a technique, and it hands
the plain ball a real route to the DOUBLE/TRIPLE KILL ladder that until now
only a bomb or a lucky line-up could reach. Banking your specials stops
meaning playing worse in the meantime.
Only a kill hooks, so a wound ends the chain and no brute, steward or boss
ever extends one. It is a crowd skill shot, which is the half of the arsenal
the tins do not cover.
The pull had to learn a third state to say this. Quality SATURATES at
pullFull, so an overpull and a perfect pull drew an identical band: the game
was asking you to hit a window it never showed you. The band now goes chalk
white inside it and the mark on the target grows a second ring — the fill
says where the ball lands, the ring says it will not stop there.
The chain path is deliberately NOT previewed. It is a scallop laid over the
exact patch of court you need to read, it solves the shot before you take
it, and the bodies walk — any chain drawn at pull time is a lie by the time
the ball arrives, which is the one thing the band exists not to be.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The band work from f352612 rides along; everything else is this round's list.
- proj returns a NEGATIVE scale behind the camera, so four ellipse call sites
could be handed negative radii — ctx.ellipse throws on those, and the throw
killed the rest of the frame (no HUD, no rack, no band). All four now skip
points behind the camera.
- a burst tin is no longer a commitment: tapping another slot aborts it, spends
it and hands back the serve loop. Lives in selectWeapon, so the rack tap and
the 1-0 keys both get it. A horde still locks the rack — that is the one place
your finger is already spoken for. Spending splices the rack, so the tap
resolves to a weapon id and re-finds it rather than trusting the index.
- beacon 1.6 -> 2.4 rounds/s: a lob cannon you could not leave AND could not
feel working was two problems.
- FIRE. Flamer throws fat embers instead of a one-pixel thread; burning bodies
get nine tongues over a heat bloom plus smoke; the fire ball gets a halo and a
real tail; an ash death gets twice the motes at twice the life over a ground
bloom and a column. The flamer had no sfx case at all and was using the gun's
dry tick — it has its own roar now.
- chapter bosses carry an HP bar over their head. The no-HP-bar rule stands for
everything else and the comment says why a boss is the exception.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Every transient message — milestone card, RESUMING clock, one-line flash — now
centres its ink on CARD_MID and only one is ever up. Three shared pieces do all
of it: inkSpan measures, plate() sizes the fill to what it holds, slideDraw
swipes it in from the left and off to the right with a four-ghost smear sampled
back along its own easing curve.
- weapon card is icon + name + count, one centred row, no prose
- plain cards centre as a block instead of hanging off the band's top edge
- the clock left the net for the band; it takes over when a card or flash clears
- band moved up 20px into the flash's old row, which it can now that they never
draw together
- hold times +0.8s: the swipe eats ~0.55s that the old numbers didn't budget for
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
- card: padding 20, smaller icon (46x44 box), band up to 88
- RESUMING IN: Bangers over the net tape, no panel — the countdown
belongs to the court the pause is holding
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- opaque canvas: the compositor stops alpha-blending 2.5Mpx at 60Hz
- settled modals and the death summary idle at 20-24 fps; play untouched
- TUNE.lowPower caps RES at 3 (-44% pixels/frame), flippable live
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- icon LEFT of the text: every card is 62px at a fixed top, fully over
the crowd — the stacked layout hung its lower third over live court,
and the art draws bigger this way, not smaller
- nothing draws or ticks under a modal: a card raised in the same beat
used to burn out dimmed beneath it — it holds and plays after
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- praise was firing at RELEASE, half a second before impact (measured:
32-frame flight, word on frame 0) — it now arms on the ball and pops
on the first body it meets, or its first bounce
- DOUBLE KILL / MASSACRE / RETURNED rise off the thing that earned them,
as world-anchored pop texts — the announcement band is for milestones
- the band got rows: a card can no longer print over the flash line
- GAME N flash deleted (the scoreboard bar already says it), CHANGEOVER
folded into the countdown panel
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- x_hand.png generated to Oliver's reference pose (index up, curled
knuckles, thumb) — drawGhostHand draws the art, vector stays as fallback
- fingertip anchor measured off the PNG, not guessed
- RESUMING panel top no longer floats above the glyphs
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- one definite band for cards, flash and the countdown — never on the court
- the card heading down a size; RESUMING IN finally on a readable panel
- the gesture-teaching strings are gone, the ghost hand is the teacher
- ghost hand v2: bigger, bigger type, and knuckles that make it a hand
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- opening CPU beatable in a game or two (softer cap, more whiffs, rubber band)
- modals are for calm courts only; live-court news is a card now
- the transition howto is gone: a ghost hand teaches the pull, the toss,
the apex and the swing, each on the exact struggle it answers
- a shot barrel keeps your weapon and glows its slot instead
- switching weapons clicks, the live tile is unmistakable
- barrels come faster the deeper the run goes
- the umpire triple-tap cheat is deleted
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
LATER gains two items. Item 5: how a viewer actually reaches the game — bio
links cost two taps and convert badly, Shorts descriptions and pinned comments
click, and most people search later instead, so the handle, an on-screen name
and a resolving domain matter more than any link. splattertennis.com is
registered.
Item 6: a bio-link tap opens TikTok's or Instagram's webview, not Safari, so
that embedded browser is the first-run environment for most players the
campaign delivers and nothing has ever been tested there — autoplay audio and
canvas perf at RES 4 being the two that would actually break the hook.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The PERFECT spark ring is out — the hitstop, the click and the praise word
already carry the moment. The net-cord wobble drops to ~1 px. And the
floodlight stutter is gone entirely: on the iPhone a random dark dip over
the stands reads as the crowd flickering black — a rendering bug, not
ambience.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Kills stutter the world a touch more per streak; a massacre borrows the
boss's slow-mo; a boss death gets the transition's zoom punch. PERFECT
contact flashes a spark ring. The sidestep and every real bounce kick up
court-tinted dust. The net cord ripples where the ball clips it. The
landing ring contracts on time-to-touchdown. Hot rally balls drag three
fading ghosts. And every 8-15 seconds the floodlights stutter.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
EATEN_GRAB (write-only marker), ball.firstLand (written five places, read
nowhere), zombies' unused VOLLEY import, three scoreboard palette fields
from the pre-PNG board, the x_tin.png nothing references, and a dozen
export keywords on functions no module ever imported.
Kept on purpose: o_run3/o_run4.png — the SPR list excludes them so both
tours run on two frames (the women's set has no 3/4 twins), per the comment
at the preload list.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- The opaque menu no longer has the whole court rendered behind it at full
resolution — minutes of battery burn nobody could see.
- Stands, ads, rot tint, court, lines and the run-off are one cached blit,
rebuilt only when the palette actually moves; the clap ripple stays live.
- RES derives from devicePixelRatio (capped at 4), so a 2x phone stops
painting double the pixels it can show.
- One shared noise buffer for every synthesized sound — applause alone was
allocating ~200 fresh buffers per cheer.
- The touch-steer path stops re-measuring a fixed canvas rect per move event.
Skipped as not-quick: the z-sort closure churn and swap-remove splices —
real but small, and each is a behavior-preserving refactor for another day.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- The duel misread a spent ('X') ball: bounce out = instant win, two bounces
= a loss plus a surge, for a shot that should just be re-served.
- Duel serves routed through the temp weapon riding slot 0 — a laser fired
instead of the ball. Tennis rules now mean plain balls, temp or no temp.
- A tap during the death animation tossed a ball that hung frozen midair.
- The EPIC boss card was clobbered by the SET card when game 30 was both.
- The telegraph glow divided by the wrong wind: negative alpha, no fade-in.
- The crowd clapped forever over the death screen.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A first-seen tin announced itself twice after a won duel: the modal that
teaches its rules, and then the delivery card right behind it, same title
same icon. weaponModal now reports whether it spoke, and the card (and the
barrel's flash) only speak when it didn't.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
assets/_sheet_*, assets/_pose_ref_* and the generator scripts are what the
frames were CUT OUT OF, not what the game loads — 34 of the 45 MB in
assets/, in the image and publicly served, fetched by nobody. A
.dockerignore on the underscore prefix takes assets/ to 16 MB, with
nothing to maintain per sheet: the SPR list in render.js has no
underscore-prefixed name in it.
Also keeps .git (83 MB) out of the build context.
Not verified against a real build — no docker daemon on this machine.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The cold-start complaints — late fonts, zombies missing from the stats
screen — were one header. server.py sent no-cache, must-revalidate for
everything, so a fresh launch revalidated ~230 sprites and three fonts
before the game had a full set of frames. Assets under /assets now cache
for a year when the URL carries ?v= (changing the file changes the URL,
which is what SPRV is for) and a day when it does not; the fonts get a
?v= too. HTML and JS stay no-cache, so a deploy is still visible.
Two client-side halves of the same report: a dropped sprite request used
to be permanent for the session — the sheet stayed undefined and the
zombie drew as some other design — so each one gets one retry. And the
splash now holds its buttons back behind a loading bar until the last
sheet has landed and both faces are ready, instead of letting you open
the stats screen while half of it is still in flight.
The bar is driven by setInterval, not rAF: rAF does not fire in a hidden
tab, and an app that loads while backgrounded would come back to a frozen
bar and a menu with no buttons on it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The hang: I could not reproduce it or pin it on one latch, so this is two
things. A root-cause fix — iOS does not guarantee a touchend when you switch
apps mid-drag, and the held G.aim it leaves behind blocks the uncaught-toss
retoss forever — and a watchdog for whatever else is out there. The watchdog
does not time how long you have been blocked, because a horde blocks you for
fifteen honest seconds; it watches whether the clock that should release you is
moving, and logs the frozen state before letting go.
The duel: past about game 20 all four tins sit at wCap and a won duel handed
over literally nothing. It falls through to the barrel weapons now.
The prize also stops borrowing the praise channel — arriving in the same slot
as PERFECT made it read as another compliment. It gets a card with its own icon.
The turn into zombie mode was unwatchable because the zoom started ON it, and
the curve render applies is zero at t=0, so the camera only arrived half a
second after he had changed. It peaks on the turn now, and the swap lands on a
hitstop, a flash and a burst so it reads as an event rather than a glitch.
Laser down from 0.55 to 0.22 — zap() sums four voices, so its vol was never one
voice's level.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Nothing secret is in it — the .env never enters the image and the client is
public JS regardless — but handing out server.py and the asset-pipeline
scripts to anyone who guesses a filename publishes the exact validation rules
for free.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Ranked on KILLS IN A SINGLE RUN. Games survived is the same information at a
tenth the resolution — most players land between 5 and 20, so that board is a
wall of ties broken by who got there first. Games ride along as the tiebreak
and as a column.
The image is python:3.12-alpine running server.py now, not nginx. The board
needs one process that can write, and everything the nginx conf did here — the
no-cache headers and the manifest MIME type — is a few lines of stdlib. /data
must be a persistent volume or every deploy resets the ladder to empty.
A name belongs to whoever took it first: the client mints a random id once and
the row remembers it, so a stranger cannot overwrite your score. The id never
leaves the server. That is the whole security model and it is enough.
The prompt to sign the board appears on exactly one occasion — a run that beat
your own best kill count, on the tap that leaves the death screen.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The transition turns your opponent into a zombie with variant 'c', and the
roster table had no row for it — so the stats screen fell back to the raw key
and the shambler sprite, and the list appeared to hold two shamblers. He is
THE OPPONENT now, and he is not the racket zombie, who keeps coming back.
Type across the menus was sized on a desktop and read tiny on a real phone:
tabs, opponent names, counts and labels all go up a step, and the roster
sprites go from 34x46 to 48x62.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The translucent panel fill was invisible on the list screens, which sit on
flat ink, and wrong on the splash: the court's white lines ran straight
through STATS and RANKINGS and the buttons stopped reading as buttons.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The court art had three faults: a hard-edged pale panel sitting on the near
half, mown stripes that stopped and restarted where the court began, and the
blood pool stranded behind the baseline where a portrait phone cropped it off.
Redrawn with the stripes running unbroken the full depth and the stain out on
the court where it can be seen.
The picture also read as zoomed in, which was not the art but the crop: at 4:6
`cover` was throwing away a third of the width on a phone. The sky is extended
to 1:2 instead, so the whole arena survives the crop.
Menu palette rotated off olive towards a cool desaturated blue-green with much
quieter borders, and the boxed BEST RUN / WORLD RANK bar is gone from the
splash — it sat on the best part of the art to say what the two buttons under
it already said.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The menus were placeholders next to the art. Three screens now: a title over
an empty, dimmed, blood-stained court; a MATCH SETUP screen where each court
wears meters read straight off its own bounce numbers; and BODY COUNT, which
needed the game to start measuring anything at all.
stats.js keeps one capped array of per-run records — all time, last match and
seven days are three filters over it rather than three counters that can drift
apart. Kills file under the body AND the weapon: killZombie only ever knew an
index, so every damage loop stamps the cause on its way past.
The tutorial is gone. Marvin explained what the in-game modals already teach,
and his stage is where the stats live now.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver kept a webhook per branch rather than the single unfiltered one, so the
note now says that. The part worth keeping is why a filtered hook's Test
Delivery always looks dead.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Coolify's manual Gitea endpoint picks the resource by repo AND branch, so the
single existing webhook already serves staging and a branch filter would only
break it. Noted alongside why Test Delivery can never confirm this.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A subdomain means a separate origin, so a staging run can no longer overwrite
the localStorage the testteam is playing on. That makes the /staging alias
pointless, so the Dockerfile goes back to exactly what is deployed today.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The image now answers at /staging/ as well as /, via an alias to the same
root — that way the staging deploy works whether or not Coolify strips the
path prefix, and the staging branch needs no diff of its own to merge.
The manifest MIME moved from an exact-match location into the stock types
map: an exact match would have missed the /staging/ copy, and a regex
location would have out-ranked the prefix and resolved against root.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Four things Oliver caught on the phone.
The screen title is the ANNOUNCEMENT face now. A menu heading is a thing that
shouts, the same register the canvas gives SET and CHAPTER — it was wearing the
HUD font, which is the voice for scores and key hints. Bangers carries its own
weight, so it takes size and air instead of Silkscreen's letter-spacing, plus
the padding under it he asked for.
Her toss and smash were shrinking the whole woman. The slice anchors frame 0 to
224 px, and frame 0 was the toss — whose arm is straight up, so the BODY got
squeezed to make room for it. This is the trap already written down for the
raised racket; it just moved to a different sheet. Re-sliced the retained sheet
at the men's toss height (250) instead, so toss/smash/pull come out 250/293/191
against his 250/320/213 and her body matches her own idle again. No re-roll.
The pre-serve ball did not follow the MIRROR. drawChar flips the whole sprite
when a server faces left, so dodging during the bounce swapped his hands over
and left the ball hanging out by the racket — one `* sv.facing` on the offset.
It was wrong in the men's world the whole time, it just took a second body
standing next to it before anyone saw it.
And she reaches further out and higher than he does, so BOUNCE_IDLE's pair is
overridable per tour now (0.46/0.78 against his 0.36/0.68). Measured off the
drawn hand by eye, which is the only way those two numbers have ever been set.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The court screen picks WHO as well as WHERE now: two tiles above the court
rows, men's or women's singles, remembered like the court is. No new stage —
three rows and a pair of buttons already fill a phone, so the tiles go side by
side and the tap count from splash to play does not change.
Everything downstream is one line: drawChar already built every frame name off
a single prefix, so SK() appends '_f' when the women's tour is picked and falls
straight through to the men's frame for anything without a twin. The horde, the
officials and the ball boys are shared — they are staff, not the draw — but the
turned CPU is not: a male corpse opposite a female player reads as a bug.
Her frames came back as a different woman in every batch — blonde in one,
auburn in the next, greyscale line art in a third — because each was anchored on
the MAN whose pose it was. "newart" repaints the character and "state" is welded
to the corpse palette, so neither could say "the same living woman, a new pose".
That mode now exists, and every batch anchors on her own settled frame.
The death is drawn to frame 3 and recoloured after it. The generator will not
draw a dead woman in any wording worth using; frames 4 and 5 survive that
because by then she is flat and then a heap of cloth, so they are the men's
drawing with the blue shorts painted into her white (see assets/_die_f.py). The
collapse still falls 224/159/117/103/43 straight through the handoff.
Her first bounce frames reached to the shoulder, which is a catching pose, not
patting a ball — redrawn low rather than bending BOUNCE_IDLE to fit a wrong
drawing, so both tours share one apex and the ball meets her palm.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Two separate faults, both of which read as "the duel gave me nothing".
The first is mine, from making weaponModal first-sight-only. G.flash ticked
ABOVE the `if (G.modal) return` in update(), so a flash raised in the same beat
as a popup spent its whole 1.1 s counting down invisibly behind the dim. That was
harmless while every tin opened its own modal — but once the modal went quiet the
label flash WAS the announcement, and the trainer's WOUNDS TREATED popup ate it
every single time you won a duel while wounded, which late in a run is every
time. Measured: flash null after 1.5 s of reading, nothing left on dismiss. A
modal freezes the world and a transient HUD message is part of the world, so the
tick moved below the return. It now holds at 1.1 for as long as you read, and
starts counting when you tap.
The second is older and Oliver felt it directly. deliverTin CYCLED the ladder —
TINS[G.deliveries++ % 4] — handing you whatever rung came next even when that
type was already at wCap, where grantWeapon's Math.min clamp makes the delivery a
silent no-op. Bomb and fire fill in three duels each, triple in one, so past
about the thirteenth duel three rungs of four are full and most late duels paid
out in literally nothing. It walks to the first rung with ROOM now. Simulated
twenty wins with nothing ever spent: every one delivers through #16, where before
#10, #13, #14 and #15 were already empty handovers. Past that it says RACK FULL,
which is a real state and not a bug, so it says so plainly instead of miming a
delivery — no boy, no umpire line.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The laser was playing boom() — the BOMB — so the showiest weapon in the game
sounded like the least showy one. zap() is its own: a highpassed crack, two
detuned saws diving through a resonant lowpass, weight underneath so it lands on
a phone speaker, and a triangle that rings after the plane has gone. The dive is
the whole reading; a rising sweep is a charge-up and this thing never charges.
And the NORMAL hit — ball meeting a body without killing it — was borrowing
pock(), the racket thock. That is a bright, dry, stationary-band click: exactly
right for gut on felt and exactly wrong for felt on a torso, which is why hitting
a zombie sounded like hitting a wall. thud() falls instead of sitting still,
which is where the wetness comes from, with one quiet bright tick just behind the
slap: something gave. The machine keeps its pock — it is furniture and it rings.
Both thud() and splat() now take a size, zombieH/1.8, so 1 is a shambler, 1.5 a
brute and 2.04 a chapter-4 boss. Everything pitches down and lengthens with it.
Measured through a patched AudioParam rather than assumed: the body thump under a
hit runs 150 Hz on a shambler, 100 on a brute, 59 on the boss.
splat() got the same treatment while I was in there — a resonant band collapsing
down the spectrum instead of a plain lowpass, plus a second brighter burst a beat
later. One clean hit is a thud; a thud followed by spatter is a body.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
endBurst looked its tin up with [0, 1, 2, 3, 4].find(...) — correct for exactly
as long as the rack held five slots. Growing RACK_MAX to 9 meant a burst weapon
picked up into slot 5..9 could never be found, so it was never spent: you fired
it and the count stayed where it was, forever. Oliver hit it with the beacon, the
rocket and the flamer, which together with the gun are precisely the four
kind:'burst' weapons — endBurst is the only spend path that searched for its slot
instead of being handed one.
The serve weapons were never affected, and that is the tell: bomb, triple, fire
and the laser all spend through spendSlot(G.weapon), the live index. Checked the
laser end to end at slot 8 rather than reasoning about it — 6, 5, 4, 3, 2, 1,
gone.
Fixed at the invariant rather than at the call site. The rack has always had a
comment saying everything reads through the slot accessors so nothing else needs
to know how it is stored; slotOf(id) is now the fourth one, and it is the single
place that knows the range. Verified all four burst weapons arm from a real tap
in slots 5-8 and leave the rack when the burst expires.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The band made PERFECT earnable, which was the fix — but you learn a band, and
then the HUD is talking on every other serve again whatever the word says. So
the praise goes behind a shared cooldown (TUNE.praiseEvery, 14s), the same model
the umpire's asides have used all along for exactly this problem. Eight
band-perfect serves back to back now yield one word. The bright click in the pull
is the reliable feedback for a good draw; the word was only ever garnish, and
garnish on every plate is just food.
And the free-floating slot drops 24 -> 18. It shares that slot with CLEAR THE
COURT, DOUBLE KILL, RETURNED and GAME N — half of which are a footnote to
something you already watched happen. Sharing the loud FACE with the milestone
cards is right; sharing their SIZE made every footnote read as a milestone.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
PERFECT fired on almost every serve, and it was not a threshold that had drifted
— it was the wrong test entirely. pullFrac CLAMPS at pullFull, so pullQuality is
exactly 1.0 for any pull of 130 px or more, and 130 px is an ordinary comfortable
thumb drag. "quality > 0.999" therefore meant "you pulled far enough", which is
not an achievement and is why it read as noise.
So hitServe takes `perfect` as its own argument now. The toss ritual keeps
earning it the honest way, on the 45 ms perfectMs window. The slingshot earns it
by STOPPING in a band at the top of the draw (pullFull..pullFull*perfectBand,
130-166 px) — a target you can miss, rather than a ceiling you cannot. The pull
already clicks when it crosses the splatter threshold; it clicks brighter when it
enters the band and goes silent when you pull past, so the window is learnable by
feel instead of being invisible skill.
And the word rotates: PERFECT, TEXTBOOK, IMMACULATE, OH LOVELY, UNPLAYABLE, SHOT
OF THE DAY. Cycled rather than random — the same word twice running is what makes
one read as a stuck HUD element rather than as praise. The umpire has still never
acknowledged the massacre, so a clean shot gets the line it would get at Wimbledon.
Also: the card band grew 72 -> 80 px, because the second sub line ended 4 px off
the bottom edge and looked like it had fallen out. And the quiet face floats as
much as the loud one does, so it gets the same 1 px shadow — the control hints,
the gesture-demo labels, the card subs, the modal body and prompt, the death
summary lines. The scoreboard and the rack keep theirs off: both already sit on
dark panels, and a shadow under 6 px type on a flat fill only smears it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
He died at every chapter boss because he could not move, and there were two
independent reasons for it. The movement block is 'rally'-only — a settled feel
decision the six boss AI styles were written around — and a boss is fought from
'pretoss'. On top of that canAimServe() is true through a boss fight, so his one
finger was already spent on the slingshot: there was no gesture left over that
could have meant "move".
So it is a SIDESTEP, not a walk. One committed step on a cooldown, carrying
invulnerability that beats both damage channels — the chip damage in hurtPlayer
and the instant-death contact test. The rule about walking stands untouched.
The gesture is picked by WHICH WAY the drag leaves the deadzone, decided once and
never revisited for that touch: down-ish is the pull the game already teaches,
sideways is the step. No timing window, no invisible mode, so "one gesture means
one thing" still holds. The finger-steer modes are untouched — stick.aimable is
false in a horde, so the drag is the absolute steer it always was. The apron demo
now alternates PULL DOWN and SWIPE TO DODGE, and Marvin has a page for it.
The rest of the playtest list:
- Bangers is bundled beside Silkscreen, and the register IS the category. The
quiet face keeps the HUD, the rack, the control hints and RESUMING IN; the loud
one takes SET, CHAPTER, EPIC!, PERFECT and every free-floating flash. Floating
text gets one shared y and a drop shadow, so it survives crowd, clay and blood.
- Modals queue instead of being dropped. That was why the trainer and the weapon
had to be crammed into one card after a won duel; the heal gets its own popup
with the x_cross tin over it — a sprite loaded since that merge and drawn by
nothing.
- weaponModal's first-sight gate moved inside the function. The barrels gated
themselves and the tin ladder did not, so four weapons re-explained themselves
every three games while a barrel weapon explained itself once.
- Ten rack slots. Past five it grows a second row upward rather than shrinking
ten tiles under the thumb target; slots 0-4 stay welded to the bottom edge.
- The boss win holds for its own 3.6s and says EPIC!, with no GAME n NEXT line —
the bar under the scoreboard already prints that permanently. That bar went
from 6px to 7px, which is the smallest type in the game becoming words.
- The turn punches the camera in on him and the ground shakes harder at each beat.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver's tuned numbers become the defaults: beamY 0.95, beamLen 23, beamW0 0.08,
beamW1 3.1, beamOriginLift 0.44, beamBorder 0.18.
Three knobs earned their keep on the way there.
beamLen is the DRAWN length now, not only the kill range. The drawn length was
solved every frame so the far end landed on the middle of the ad strip, which
made this a control that visibly did nothing. One meaning beats one clever
behaviour -- and it pairs with beamY, since a lower plane meets the ads sooner.
beamOriginLift replaces beamOriginY: the apex is a lift ABOVE the plane rather
than an absolute height, so lowering beamY brings the near end down with it
instead of leaving it hanging at chest height with the beam sloping away.
beamBorder exposes the edge width, which was hardcoded.
And the playground is back, deliberately NOT paused. The first version froze the
world so the picture would hold still, and then nothing redrew, so every value
changed did nothing -- a still frame you cannot change is worth nothing. The
bodies are held still individually and spawning is off instead.
Also: a boolean living in TUNE has to be tested with >= 0.5. The panel steps
every key by 0.02, so one nudge set beamBand to 0.02 -- still truthy -- and the
renderer switched variants under him mid-tune.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
TUNE.beamBand flips between the two looks live from the T panel. 1, the new
default, draws the laser the way the BALL's aim trajectory is drawn: a path down
the lane whose width comes from each point's own projection scale, so it narrows
into the distance like everything else on the court, with a shadow on the lawn to
pin it there. 0 is the solid widening cone.
The cone was a true projection, which is why it was not obviously wrong to me --
but it GROWS in world width, so on screen it opens out while the court recedes,
and the eye reads that as a shape stuck to the screen rather than a plane lying
in the world.
The kill follows whichever look is on, in both fireBeam and drawBeamCuts.
Otherwise the comparison would be a lie: a narrow band that deletes a cone-wide
crowd tells you nothing about how the narrow band plays. Measured down one lane,
the cone takes 9 of 9 and the band 5 of 9.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The cone is fully opaque now: solid colour, deep blue at the edges and near-white
down the core, dimming as it runs away. Depth is carried by BRIGHTNESS instead of
by alpha, so it still reads as receding without anything showing through.
That settles an argument the translucent version could not win. Its opacity and
how well you could tell above-the-plane from below-the-plane were the same dial
pulling opposite ways: readable enough to show the split and it veiled the whole
cone, subtle enough to look right and there was nothing to see. Solid, a body
above the plane is drawn on top of it and a body below is hidden behind it. The
trade is that you cannot see the crowd through it, which is the thing to judge on
the phone.
The `alpha` parameter went with it rather than being left in the signature lying
about what it does.
laserScene is gone. T.pause(true) stays -- it freezes the world but keeps drawing,
which is worth having whenever a still is what you need to look at.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
T.laserScene() (or #laser on the URL) drops into splatter with six bodies of
different types at known depths, the laser armed, a pull held, and the world
paused.
T.pause(true) freezes the WORLD but keeps drawing, which is the whole point: a
value changed in devtools shows on the next frame while the scene stands still.
The old recipe halted rAF, which freezes the picture too and is no use for
judging how something looks.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Lower alpha through the whole fill (0.34 -> 0.19 at the strings, falling with
depth) and the core is a LIT BLUE rather than near-white, which is what was
washing it out. The stroked edges come down and turn blue with it.
Worth recording, because it will come up again: the cone's opacity and how well
you can tell above-the-plane from below-the-plane are the same dial. The plane is
flat -- a body above it is drawn clean on top, a body below is seen through it --
so the distinction just IS the cone's alpha. Fainter cone and sharper split pull
against each other; the fainter cone wins and the cut line carries the signal.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The blinking was drawZombie validating its frame count against the LAST frame
only: a set with (say) z_h2 missing but z_h4 present kept N = 4, the walk cycle
stepped onto a name that had never loaded, and drawStatic silently drew nothing.
The body vanished for that frame of the cycle and came back on the next one,
forever. One cached 404 -- this project's oldest trap -- is enough to cause it.
It holds frame one now: a body that stops animating is a bug you can see, one
that flickers in and out reads as the game breaking. SPRV bumped to flush
whatever his phone has cached.
And no ctx.filter anywhere. The cut line was briefly the sprite redrawn clipped
to a band and flattened by a canvas filter, which put it exactly on the
silhouette -- and on the iPhone the line did not draw at all. Unsupported or
leaking past its restore, a rendering feature that fails silently on the target
device is not worth a pixel of polish. It is a plain bar again, brighter, and it
crosses the odd gap between the legs, which is the trade.
The cone gets the reference's look: a horizontal gradient across the width with a
near-white core down the centre line falling to blue at the edges, plus the depth
falloff along its length, plus stroked outer edges. Canvas has no 2D gradient, so
it is filled in eight slices each carrying its own across-gradient -- enough that
the seams vanish, which is what the old nested wedges got wrong.
Test mode: barrels gone (the rack already holds everything they would give you)
and the crowd raised to 16 bodies at a spawn every 0.9s, so it does not run dry
between kills.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The beam is a real plane now, holding TUNE.beamY (1.15 m) all the way out instead
of sloping to the ground. That one change is what makes the rest possible: sloping,
it lay flat on the grass across the whole far half, so there was nothing to cut
exactly where the crowd stands.
One flat translucent triangle, apex at the player's CHEST, opening out down the
court with the alpha falling away. It was four nested wedges composited 'lighter'
and that layering WAS the banding -- additive light blowing out to white where the
layers overlapped. Flat alpha also means a body under the plane is tinted rather
than brightened, which is what lets the cut read at all.
Bodies are sliced by it: the part above the plane is redrawn clipped so it sits on
top, the part below stays tinted underneath, and it happens while AIMING as well as
on the shot. Measured, the plane takes a shambler at 64% of its height, a brute at
43% and the roller man at 36% -- bigger bodies visibly take it lower.
The cut line follows the SPRITE rather than a rectangle. A fillRect spanned the
body's bounding width, so it crossed the gaps between the legs and under a held-out
arm and read as a bar lying behind him; the line is the sprite itself, redrawn
clipped to a thin band and flattened to one colour by ctx.filter. It thins with
distance on s squared, because this camera's scale only spans about 1.5x across the
court and a linear falloff was invisible.
Aiming a filter at a specific colour turned out to be a measurement, not a guess:
brightness(0) invert(1) goes to WHITE, and sepia adds almost no chroma to white, so
the first chain rendered a grey. Leaving a mid grey for sepia to tint lands on
(32,154,255) against the laser icon's (29,153,232). The probe fills a rect and reads
the pixel back.
The pass is bounded to the eight nearest bodies above 14 px, because it draws each
one twice more through a filter and runs every frame you are aiming. 0.27 ms/frame
with 20 bodies on desktop, against a 16.7 ms budget.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The mg racket is gone, not moved. It was a procedural racket drawn over a sprite
that already holds one: straight up-court it projected onto the player's back as
a floating oval, and offset to the side it just stuck out.
The laser's far edge was pinned to the middle of the ad strip two commits ago and
nothing appeared to change, twice. The geometry was always right -- each wedge
layer faded to alpha 0 along its length, so the beam went invisible before it
reached its own far corners and the VISIBLE end sat well short of the pinned one.
The far stop is lit now, so the two are the same line. Verified by recomputing
the projection: far corners at y 113.8, the ads' middle, against 110.2 unpinned.
It also falls off with distance now, brightest at the strings and thinning down
the court -- but never to zero, which is the trap above.
Fixes a third instance of the empties-under-the-iterator bug: a rocket blast can
bank the last kill of a game, and the duel or chapter boss that follows clears
pellets[] mid-loop. shots, pellets, extras and the body loop all carry the guard.
Also records that two consecutive render() calls do not produce the same frame --
render mutates camera smoothing as a side effect -- so frame-diffing to find where
something was drawn measures the camera drift instead. It claimed the beam reached
16 px above where its own geometry could go.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A burnt body does not burst. Fire kills by cooking, so there is nothing left to
spray: a body that dies alight -- flamethrower or flame ball -- leaves dark grey
ash instead of blood. Same decal, mound and particles, recoloured and smaller,
with a puff of smoke and a dull thud instead of the wet splat and the headshot
crack. The flame BALL has to mark itself, because that kill lands before the body
ever catches.
The flamethrower is a lighter, not a gun: no impact damage at all, it pierces,
and it only IGNITES -- the burn does the killing, so a lit crowd walks a few
steps and goes down together. 26 m of reach against a 22.8 m court, so it
crosses to the far baseline. The guard matters: hp -= 0 fell straight through to
killZombie, which made the cone delete the crowd on contact.
Every burst weapon had no progress bar and the cause was the spend. Arming spent
the tin at once, so the rack entry vanished -- and the bar draws on the slot
holding G.gunId, which no longer existed. The tin is spent when the burst ENDS
now, so icon, count and bar are all there while it runs, and every path that
ends a burst routes through endBurst or it would hand out free ammo.
Auto-fire movement again: the previous fix read moving as mx||mz||assist, and
finger-steering zeroes mx/mz, so sliding left and right still counted as standing
still. And the oval on his back was the mg racket -- its head projected almost
exactly onto the sprite; offset by facing it reads as held out beside him.
The laser's far edge is pinned in SCREEN space to the middle of the ad strip
rather than running out to beamLen on the ground, which landed on their top line.
beamLen goes back to being the kill range instead of a number tuned to one camera.
Beacon thumps, rocket swooshes.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The auto-fire jog was one guard. gameplay skipped the pose update in mgSteer mode,
so the pose stuck on 'idle' -- and render covered for it by faking a run cycle off
G.idleT, which animated whether he was moving or not. That fake is what read as
"this hectic run animation... odd all the time". The pose follows MOVEMENT in every
mode now and the fake is deleted, so auto-fire, the horde and the melee racket all
use the ordinary run animation and standing still is standing still. The comment
that justified the fake -- two rackets when idle -- was solving a smaller problem
than it caused; at sprite scale the overlap does not read.
The laser, second pass, all of it "it must read as light he fired":
- it draws AFTER the net and every body. It was drawn before the far-side pass,
so the net and the crowd painted over it, which is exactly why it looked like
it was lying on the floor and sitting under the net.
- beamLen 26 -> 34: at 26 it stopped just past the far baseline and read as cut
off in mid-court. It carries into the courtside ads now.
- beamW0 0.35 -> 0.06, a POINT at the strings. At 0.35 it left the racket as a
0.7 m bar hanging in mid-air instead of something emitted from it.
- it DELETES ordinary bodies, hp irrelevant. A plane of light that left a roller
man standing made it a worse ball, not a weapon. A boss is the exception and
takes dmgBeam 8 -- severe, survivable.
- four nested wedges composited 'lighter' (violet bloom, blue, cyan, white core),
each fading along its length. One quad with a gradient was a coloured shape.
Rocket: speed 17 -> 38, and a real jet plume laid tight ON the flight line rather
than scattered sparks, which looked like a firework.
Beacon: a finer one-pixel thread and a much brighter ball, so the BALL is the
bright thing and the trail only says where it has been.
And the bug in the middle of that: firePellet stamps a whitelist of fields onto
the round, and jet/trailR went onto the WEAPONS rows without being added to the
list -- so the rocket flew with no plume and nothing errored. Any new pellet field
has to be added in both places.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A test mode for trying the new content without playing twenty games to reach it.
Three taps on the scoreboard toggles it, three more turns it off, and a single
tap is still the miss-log overlay. Same shape as the umpire triple-tap cheat it
sits next to.
It OVERRIDES the difficulty director rather than editing it, so the wave is the
same whether you opened it in game 1 or game 40 and the run keeps running
underneath: 8 bodies, a fixed 1.15 m/s shamble, a spawn every 1.6 s, and
pickVariant pretending to be at game 20 so every body type can turn up. All four
barrel weapons sit on the rack and are topped back up each frame, because
committing a burst spends the whole tin and a sandbox that runs dry is not one.
A barrel rolls out every 14 s so the pickup itself is testable too.
The third tap must NOT toggle logShow: taps one and two already toggled it twice
and it is back where it started, so toggling again left the miss-log overlay
switched on behind you every single time.
Also records the trap that cost a debugging round getting here: index.html loads
main.js?v=N, and fetching 'main.js' does not bust that URL -- so main.js stayed
stale through a reload while every other module was fresh, and the tap handler
appeared to do nothing.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Oliver's playtest of the barrel weapons. Four notes, all four real defects.
"The bigger guys were way too easy to kill at the beginning." A bug, and an old
one. The ball-vs-body chain handled boss, g/m, s, f and a/b, and everything NOT
named in it fell through to "splat-quality kills outright" -- so the STEWARD has
died to one serve since the day he shipped with stewardHp 5 and a comment calling
it a brute's worth of hp, and the three new bodies inherited that the moment they
existed. Pellets and melee always read hp; only the ball skipped it. One generic
hpMax > 1 branch now, and measured after: shambler 2, steward 5, brute 5, ball
boy 3, bouncer 5, roller man 7 -- each exactly its hp.
"I have seen all of them by round 10." The new three move 3/8/15 -> 5/12/20. The
original four keep their gates: that curve was already tuned through game 12.
"The laser fired only when I tapped, that was confusing." The whole tap-per-round
kind is gone. A barrel weapon is now either the normal ball gesture made
spectacular or something that fires itself -- nothing waits for a tap.
- laser is a serve, and the RAY is the weapon: no ball anywhere in it. The
preview IS the wedge, it leaves the RACKET rather than lying on the grass, the
release erases the plane along its own path behind a bright front, and the
edges are hairlines instead of the heaviest lines on the court.
- beacon and rocket are LOBBED bursts -- `arc` turns a flat pellet into a thrown
one, solved onto the nearest body ahead, and landing is itself a hit so a
rocket that misses still goes off where it lands. The beacon changes colour
every ball and drags that colour behind it.
- flamer got "way more pronounced": 30 rounds/s, a ~10 m cone, embers.
"I got the laser almost every time." A flat 1-in-4 over four items does not feel
uniform. BARREL_WEAPONS carries weight and a `from` gate now, never repeats twice
running, and the beacon is what you meet first. Measured: laser 16-28%, rocket
unlocked at game 12 so it is genuinely a late find.
And the barrel keeps its pink halo -- that was always procedural, painted by
drawBarrel, and never went near the chroma key. Three rings and a pool on the
lawn now, because one flat disc read as a smudge.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The rack had five hardcoded slots where the index WAS the weapon, spelled out in
five places: SLOTS, slotStock, WDESC, drawRack's three parallel arrays,
deliverTin's w+1, and bare wsel === 1|2|3 / G.weapon === 4 tests across five
files. config.WEAPONS is the one table now, G.rack holds up to four {id, n} in
the order you GATHERED them, and slot 0 stays BALL. Everything reads through
slotId/slotStock/slotEntry, so nothing else knows how the rack is stored.
A fifth weapon is always allowed, per Oliver: it rides slot 0 as G.temp until
spent, then slot 0 goes back to the infinite ball. A barrel arms on the spot and
is spent quickly, so refusing it would mean a barrel you shot did nothing. And a
spent weapon LEAVES its slot -- with gathered slots, an exhausted weapon holding
one is a quarter of the rack you can never pick up into. (The old rack kept its
empties dimmed, which was right when a slot meant "BOMB goes here".)
Four new weapons, and they are all the ONE existing pellet loop. firePellet(w)
stamps the WEAPONS row onto the round -- r, dmg, hp, pierce, boom, ignite, col,
speed, life -- so the beacon cannon, flamethrower, laser and rocket needed no new
array and no new update; the break after the first body was the only thing making
a pellet single-target. All four go through the same isBoss branch, which is why
none can silently no-op on a boss the way three of five once did. Measured:
gun/beacon/laser/rocket reach past 16 m, the flamer is a ~6 m cone that ignites.
G.mgT stays the mode timer for all of them, so the eighteen !(G.mgT > 0) guards
across five files did not have to change. kind:'burst' auto-fires at its own rate
for len seconds; kind:'shot' holds the same braced stance and fires one per tap.
The barrels reuse the ball machine's shape: a rolling, shootable non-body. All
three of the registrations that fail silently are done -- blocksClear (loot must
not gate the hand-clear), the contact guard (furniture cannot eat you), and
hitR/GORE_MUL/zombieH. It scores nothing.
The pink had to become procedural. The chroma key deletes every pixel where red
and blue both beat green by 40, so a magenta drum and a magenta laser pistol both
came back as hollow outlines -- 19% and 15% solid against 50-56% for the icons in
other colours. The drum is grey steel with hazard chevrons and every bit of its
pink is the halo drawBarrel paints; the laser went electric blue. Both are solid
now (92% and 58%), and the rule is written down in HANDOFF.
tinTriples 3 -> 10 as asked, and the four dead TUNE ammo constants are gone --
nothing read them once the table landed, which is a panel value you can tune all
day for no effect.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A chapter boss had no AI at all: it matched no branch in the variant chain and
fell through to the crowd's aimless wobble, so "the boss AI owns its x" was a
comment describing code that did not exist. Movement is now data on the BOSS_ART
row -- pace, shadow, surge, charge, bounce -- and bossAI is the one place it
happens. Every behaviour field is optional, so a row carrying only art behaves
exactly as a boss did before.
Three earlier drafts of those styles failed the same way: written as "walk toward
the player's x", which in splatter means arrive in four seconds and stand still,
because the player is rooted in pretoss for nearly all of a boss fight. Every
style now moves under its own power and only biases toward him. Measured over a
simulated minute each: 50-136 m travelled, none static.
Three new designs -- THE HEAD COACH (a 5-shot drill 0.14s apart), THE PHYSIO
(charges, and withdraws less than it charges, so it reaches you at 38s and that
is the fight's clock), THE SPONSOR MASCOT (gap 0, so the whole wall of five
leaves in ONE frame and is dodged rather than parried). Harder to kill: 16 -> 20
hp, 9 -> 10 per chapter, plus per-design multipliers.
Three new bodies: the ball boy (3 hp, 2 skins, from game 3), the bouncer (6/3,
game 8) and the roller man (9/4, game 15), each with a 4-frame walk and a drawn
two-frame lurch for every wound stage.
Three wound rules became one. The brute had a hardcoded stage array, everything
else a single skin behind 'abd'.includes(...), and the boss a fraction of hpMax.
The boss's rule reproduces the other two exactly -- verified: the brute still
draws z_gw/gw2/gw3/gw4 at hp 4/3/2/1 -- and keeps working when hp is not
one-per-skin, which all three new bodies need. So a multi-hit body is now a row
in config.ZWOUND and no code. bleed() reads the same call, which also gave the
steward the trickle his wound skin had always implied.
Fixes a real crash on the way past: killPlayer empties shots[] while the shots
loop is mid-iteration, so every remaining index read undefined and the frame
threw. The mascot's five-at-once volley makes it easy to hit; the shipped 3-shot
volley could always have hit it.
_regen.py gains two style blocks and a check tier. One constant used to say
"keep every colour exactly as the reference" -- right for a remaster, and it
fights every word of the subject otherwise. newart is for a new character, state
for the same character wounded or mid-throw; using newart for the wound sheets
licensed a repaint and got one, five came back with living tan skin. And the
detail gate ratchets against the file it replaces, so check: any exists for
re-rolling already-fine art to fix something the gate cannot see, such as colour.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The switch and the losing branch are gone — a fired band is a short streak
behind the ball, full stop. TUNE.shotTail, WAKE and the tailMode branching all
removed; the aiming band is untouched as ever.
THE IDLE DEMO NO LONGER POINTS UP. It grew chevrons up-court as the pull grew,
because the shot going the opposite way to the thumb was meant to be the
lesson — but an up arrow on screen is an instruction to swipe up, and swiping
up is the one gesture this game does not have. The only direction drawn now is
the one you actually make: the dotted track, the thumb sliding down it, and a
down arrowhead at the end, clear of where the thumb lands at full draw.
And it says so in words. The mime alone was too abstract to read as "drag your
thumb down", so PULL DOWN sits under the arrow in the pixel font.
Which makes the text slot below it redundant while the demo runs, so it shuts
up rather than printing the same instruction twice 40 px apart. The wording
everywhere else moves BACK -> DOWN too: "back" is the ambiguous word that could
be read as up-court, which is exactly the confusion here.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
shotTail 0 -> 1. The panel is keyboard-only ('T'), so on a phone the default
IS the setting — noted next to the knob so the next person doesn't assume it
can be flipped in play.
Verified on the default build: the drawn slice runs from z 21.4 back at the
racket side to z 18.84, which is the ball to the centimetre — entirely behind
it, ending on it. Nothing ahead. 600 frames clean.
Keeping the losing branch until a real playtest confirms the pick.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The laser look came from carrying the AIMING picture unchanged into the flight.
While you pull, a bright saturated line spanning you to the target is exactly
right — it is the information you are deciding on. Once fired it is spent, and
a projectile drawing its own future in a continuous saturated line from itself
to what it is about to hit is the visual definition of a targeting beam. No
real ball shows where it is going.
So a fired band now desaturates toward dust, collapses, and shrinks to a short
streak. TUNE.shotTail picks which side of the ball that streak sits on:
0 a WAKE ahead of it, dissolving — your erase idea, beam cut down to a wisp
1 a TAIL behind it, nothing ahead at all — conventional projectile language
The shadow follows the same slice, because "nothing ahead" has to include the
ground or it still gives the future away. The target mark fades with it: a mark
still burning on the shambler is the other half of the laser.
Both leave the aiming band untouched.
Built the collapse squared first and it was wrong — both variants were gone by
a third of the flight, which is not "less like a laser", it is no streak at all,
and the strobing at release is what this trail exists to prevent. Linear now.
The SHORT SLICE is what kills the beam; the fade only has to make sure nothing
is still burning when the ball lands.
Flip it live on the phone: T panel, shotTail, type 0 or 1.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
PACE 1.3x -> 1.2x on all four numbers: 19.2 / 14.4 rally, 22.8 / 26.4 serve.
The strength-to-velocity relationship you suspected was already there is: a
splatter serve runs 26 m/s at the minimum pull, 36 at the kill threshold and
49.2 at a full draw. The draw very nearly DOUBLES the pace (1.89x), because
serveQualitySpeed carries more of the pair than serveSpeed does. Nothing to
change unless you want it steeper, and that is the one knob.
THE RACKET-ALWAYS-UP LOOK WAS A LATCH, not the art. hitServe sets pose 'serve'
with poseT 0; that only expires after POSE_TIME (0.38 s); and p_pull, the
wind-up frame, is only ever selected from pose 'idle'. So firing faster than
0.38 s reset the timer before it could expire, he never returned to idle, the
pull frame was never drawn once, and he stayed welded into p_smash — arm
extended overhead with balls appearing out of the strings. Exactly the late
game, because that is where the fast successions are.
serveToss now clears a stale 'serve' pose: a new wind-up ends the last shot's
follow-through. At ~4.6 shots/sec the cycle is now pull x5 frames, smash x8,
pull x5 — a real swing. Slow fire still plays the whole ritual: bounce, pull,
smash, idle, bounce.
Caveat: past roughly 8 shots/sec each pose gets 1-2 frames and will read as a
flicker rather than a swing. There is no fix for that short of drawn
follow-through frames; the poses are stills.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
baseShotSpeed 16 -> 20.8, qualityShotSpeed 12 -> 15.6, serveSpeed 19 -> 24.7,
serveQualitySpeed 22 -> 28.6. All exactly 1.3x, so nothing about the relative
feel moves: rally and serve, tennis and splatter, his ball and the CPU's, which
share baseShotSpeed.
Swept the full-pull serve down the court: 53.3 m/s against the old 41 at every
range, flight times ~23% shorter (30% more pace is 23% less time in the air).
One exception, and it is geometry rather than tuning: a shambler standing right
on the tape is loft-limited. That shot has to come almost straight down, so the
net-clearance loop trades the extra pace straight back and it ends up only ~14%
quicker. Nothing to fix unless it reads as sluggish next to everything else, in
which case serveSpin is the knob — a harder dip buys pace over the tape.
Tennis serves 8/8 into the box at the new pace; 900 frames clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
THE BALL WAS SOLVED TO THE GROUND. solveShot's terminal height was a hardcoded
0.08 — ball-on-the-grass — so a splatter serve came down past the shambler and
registered at the ankles however the mark was drawn. It now takes an arrival
height, and a lane shot aims at the body: zombieH x Z_CHEST, scaled off the
same height table render draws with, so a brute is struck on the brute rather
than at the height a shambler's chest happens to be. Measured at the frame the
wound registers: shambler 1.11 m of 1.8, brute 1.64 m of 2.7, and an empty lane
still lands on the grass. The aim mark reads the same `ty`, so mark, arc end
and impact are now one point instead of three guesses.
AND A BUG FOUND ON THE WAY. The net-clearing loop ran for shots that never
reach the net. A shambler on your own side sits short of the tape, so `tN` is
still positive — the net is ahead in the direction of travel, the ball simply
lands first — the height test there is far below the tape, and the loop spent
all twelve cuts. Every same-side shot left the racket at 28% of nominal pace as
a slow loop. That is exactly the shot that needs to be fast: those are the ones
about to eat you. Guarded on the shot actually crossing the net, in serveArc
and in doHit's copy. Flight time to a body at z=16 drops 0.71 s -> 0.20 s.
THIS ONE IS A FEEL CHANGE YOU DID NOT ASK FOR — near shots are now fast and
flat where they used to loop. Say if you want the lob back.
Tennis and the duel still serve into the box, far-side shots still clear the
tape, 900 frames of splatter render clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
wuwiandClaude Opus 5