Every figure in CLEAN GROUND's step-5 material now resolves against
tools/step5-split.mjs, which enumerates rather than stores: the countdown rows, the
two-case table and its blend, the Act Four reasoning, the counterfactual pair and
the POW×5 targets.
No prose changed -- stripping every HTML comment from the result diffs
byte-identical against HEAD, which is the check worth having when editing another
session's document.
Proved live: raising Okonkwo's POW to 13 in a worktree makes Braithwaite the lowest
in the room and eleven of the thirty-two markers go red at once, each naming the
path and saying nothing can be re-recorded. The three figures that went bad in this
document were all in prose rather than tables, which is why the prose is marked and
not just the tables.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The work landed in a5c4378 under the subject "R-294", which was already
taken by the peer session's derived-source resolver. R-295 was taken too,
so renumbering to 295 as I first proposed would have collided with
something already in the log. Read the log before writing rather than
accepting the correction: it runs to R-295, both of those entries are
theirs, and mine is R-296.
The commit subject stays wrong; rewriting a pushed subject to fix a number
is worth less than the entry pointing at it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Matching the step-5 prose against the derived source before placing markers left
five figures unmatched: 21.9, 74.3 and 3.8, twice each. They are the thing against
Ashcroft's undiminished 85 -- the contest that never happens, which Act Four prints
on purpose to price what THE OFFER sold.
Derived now as hadHeRefused, so the counterfactual is held to the rule like
everything else and carries a name that cannot be mistaken for an event. Quoting
those three as odds a GM meets is what went wrong in two sentences and a table.
Markers not placed: CLEAN_GROUND.md is c0's and they are in it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
c0 argued a baseline for the step-5 split would be a cache of the rule and a guard
over it would mostly assert that arithmetic has not changed. Right objection,
wrong conclusion: do not store it. ARTIFACTS now takes a { derive } entry as well
as a file path -- enumerated on this build, nothing stored, and no --update able
to silence a real disagreement between the document and the game.
tools/step5-split.mjs enumerates all 10,000 pairs through opposedContestFor with
its targets derived: 55 and 60 are the lowest POWx5 in the six and in the cut, 42
is applyDifficulty(85, "difficult"). Two different rules produce those three
numbers -- the accepted row is a named exception, not the lowest of anything -- and
a test fails if anyone unifies them. A tie in "the lowest POW in the room" is
fatal rather than silently resolved; it fired for real in testing.
Proved four ways, including raising Okonkwo's POW to 13: the lowest moves to
Braithwaite, refused.held goes 48.1 to 52.4, and the citation that was correct a
moment earlier fails. The figure follows the rule.
Landed unused -- the markers are c0's to place in their own file.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Counting the readers that had broken on their own format turned up one nothing had
caught. docs/scenarios holds scenarios, eight playtest records and two art prompt
sheets, and every guard treated all three as scenarios -- harmless for citations
and skill spellings, false for reachability. Six quotations across passes 4, 6 and
7 were checked as live rolls, so a record of a session already played could fail
the build over a skill nobody can reach. Planting Science (Physics) in pass 4 fails
before the split and passes after; the same skill in CLEAN_GROUND still fails.
Classified by the document's own H1, not its filename, because tools/scenario-*
naming is what swept a tools file into this corpus in R-268. An unclassified
document is fatal: an allowlist that silently drops what it does not recognise
would take a new scenario out of reachability checking on the day it was written.
89 rolls across 18 files becomes 83 across 8.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
c0 asked whether tagsIn should return indices so beatLikeIn could take spans from
it. No -- that widens this file's signature to serve another, and R-291 already
fails when the two disagree. But detection is not prevention, and there is a third
thing to share.
The two patterns differ in their bodies for good reasons: tagsIn captures the whole
tag for skillsIn to split, outcome-coverage stops at the em-dash because the
outcome follows. What was copied into both files, and what drifted, is the opening
-- the literal \[CUS:, widened by R-290 here and left behind there. TAG_OPEN is now
exported as a string with tagRe(body) composing a fresh matcher onto it; fresh
because a shared /g regex carries lastIndex between callers.
22 tags before and after, c0's body composed onto TAG_OPEN gives the same 22 its
own pattern does, 19 guards green.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
c0 asked whether R-290's widening of tagsIn reaches past outcome-coverage's loose
counter, which would zero its unparsed count for the wrong reason and retire a
canary. It does not -- 22 beats from each reader on CLEAN_GROUND.md, nothing that
tagsIn accepts invisible to the other guard -- but "does not today" decays quietly,
so it is now a test importing both real functions rather than a copy of either
pattern. Mutation-checked: widening tagsIn to accept [CUS Spot] turns it red.
Measuring it found something the question did not ask about, recorded in the log:
for [cus: ...], [CUS : ...] and [ CUS: ...] the two guards disagree -- tagsIn reads
the tag, beatLikeIn calls it unparseable -- because beatsIn's strict half kept the
case-sensitive literal. The build stops, which is the safe direction, but names the
wrong thing. outcome-coverage.mjs is c0's file and the call is theirs.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Swept every reader that parses a human-written marker. Two more had R-289's
fail-open.
tagsIn matched a literal [CUS:, so [cus: Spot], [Cus: Spot], [CUS : Spot] and
[ CUS: Spot] all read as nothing -- and it is the corpus reader behind
check-scenarios' skill validation and check-rollable's reachability, so a beat
spelled any of those four ways was checked by neither while both printed OK. The
corpus contains no such tag today; the 88-to-89 roll count during this work was c0
writing v0.18, verified against HEAD's reader on the same tree.
The POWER: marker had it with nothing covering it. bestiary.mjs and check-powers'
own scan both used the literal, so a creature added with "Power:" generates no
entry line and is never reported unclassified -- check-powers passes green on it,
measured on a clone with its dodge skills stripped so the dodger-count assertion
could not fire instead. Reader now takes POWER\s*: and stays uppercase, because
power: occurs in ordinary prose; an asymmetric counter names the rest, and was
measured against content.mjs first (41 strict, zero loose outside them).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Both earlier cast defects were shown by editing CLEAN_GROUND.md in the shared tree,
which is how this session came within a git checkout of c0's uncommitted work and
is also the weaker test. Seven cases now live in check-behaviour as strings.
Writing them found a live one: "<!-- cast : ... -->", one space before the colon,
matched nothing -- not an empty declaration R-288 would refuse but no declaration
at all, so check-rollable widened to the whole duty roster and printed its usual OK
line. check-cited's three spellings again. The reader now takes cast\s*:.
castLikeIn adds the asymmetric half: anything comment-shaped containing cast\w* the
reader did not consume is named by check-rollable, so a spelling nobody anticipated
fails the build instead of silently declaring nobody. Looser than the reader on
purpose -- a false alarm costs a reword, the opposite error costs a guard that
checks the wrong six people and says OK.
Tests then mutation-checked for being load-bearing, each mutation asserted to have
applied after a first pass where three silently did not.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
c0's point on R-286: the reader is unambiguous now, but the roster fallback that
made the failure look like success is still reachable. Narrowly closed -- sixteen
of the seventeen scenarios declare no cast and are rightly checked against the
whole roster, so only a marker that is PRESENT and names nobody is refused, with
its line named. Replacing the declaration with <!-- cast: --> passes before this
change and fails after it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
indexOf returns -1 for a name that is gone and slice(0, -1) is everything but one
character, so a scope anchored on a moved name does not shrink or fail -- it
becomes the file. Renaming the end anchor took one test's body from 3,184
characters to 30,825 with the assertion still passing. The other site windowed
burstAttack at 4,000 characters over a function that runs 4,305.
bodyOf asserts the anchor and ends at the next top-level function. A missing anchor
now says which anchor and what would have happened, where the old code reported
"burstAttack still calls rollWeaponDamage" -- a claim about a call when the truth
was a claim about a name.
Also records the rest of the sweep, including the guards deliberately left alone.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-284 scoped the factored-rating rule to the creature's entry and left the
defence-stacking rules matching anywhere on the page, so the exemption sentence
and the "Across N creatures that spend defences" count were accepted wherever they
happened to sit. Moving the exemption line out of its section into the redcap's
statblock -- the page silent exactly where a GM reads the stripping advice --
passes at R-284 and fails here; confirmed by running HEAD's copy against the same
tree.
The owning slice is not always the creature's entry. A factored rating belongs to
the creature; the ladder exemption is an answer to the paragraph it sits in and is
generated into "## Shooting at something that moves", so scoping that rule to
"### Redcap" would have failed a correct page. sliceOf now takes a heading at any
level and each rule names the slice that owns its claim.
A renamed section fails by name rather than scoping to nothing, which is the shape
of every guard that passes because it found nothing to check.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-283 left the intact-sentence-wrong-number case falling through to omission, so
the guard said BESTIARY "does not state" a rating the page was stating. reads()
now has a shape tier between strict and loose: the strict pattern with its value
slot loosened, reporting which of the two numbers moved.
Scoped the per-creature rules while adding it. They read the whole page, and each
is the only rule of its kind today, so a page-wide match found the right line by
luck; a second attackFactor creature would have let the courier's rule match that
creature's sentence and report the courier correct. They now read the creature's
own "### Name" entry -- proved by deleting the courier's line and planting an
identical one under the redcap: still omission, where before it would have passed.
Five discriminations plus the decoy, proved in a worktree with the message read.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-282 matched the page with String.includes, so rewording the exemption sentence
reported "BESTIARY never says it is off the stripping ladder" -- sending a
maintainer after a sentence that is sitting right there, and never naming the
real problem, which is a pattern that has silently stopped reading.
Each textual rule now reads twice. Strict is the sentence as it stands and is
tighter than before (the bold and the full stop, not the bare clause a substring
accepted); loose is the same claim in any wording. Strict passes, loose-only is
reported as a reword with the line quoted and the page presumed right, neither is
the omission.
The loose anchor was wrong on its first pass in the way that matters: "a sentence
with 40% and 80%" also matched the courier's own statblock line, so deleting the
sentence reported a reword and quoted the statblock back. It now excludes that
generated marker, which makes it a test of the claim and not of the digits, and
degrades to omission rather than to a false reword.
Four discriminations proved in a worktree with the message read in each.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
check-bestiary proves the page matches its generator, and the generator had never
heard of powers.mjs -- so the redcap sat in "the ones that take the most
stripping" under eighteen green guards while its own entry said it never spends a
defence. Two files agreeing with each other while both disagree with the engine
is a quorum, not a check.
check-powers now asserts per effect kind what the page must say: defenceStacking
requires the creature off the stripping list, named as exempt, and the "across N
creatures that spend defences" count reconciled against powers.mjs; attackFactor
requires the rating the simulator actually uses printed as a number, which the
courier's entry now carries.
The clause that matters is the failure on an unknown effect kind -- a wired effect
with no DOCUMENT_RULE fails the build, so the next one cannot arrive without
somebody deciding what the document owes it. Without that this would guard the
mistake already made and nothing else.
Proved three ways in a worktree, exit codes read directly.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-280 left the claim satisfiable by a gain of 0.2. The artifact now records how
many measurable packs clear their own noise -- reliable: {aboveNoise: 36, of: 38}
-- and the check refuses if that share falls. It may rise freely.
A ratchet rather than a threshold: any threshold here would be a number I chose,
and choosing one just under the current value is what produced MEASURABLE =
[15,85]. A share rather than a count, so widening admission cannot pay it off.
Proved three ways in worktrees: making focus fire actively bad fires the drift
check first, which is correct; making it unreliable and re-recording fires the
older claim at 37 of 38; and claiming a better past, 38 of 38, is refused by the
ratchet itself. The ratchet bites exactly where the old claim does not -- between
"still helps everywhere" and "helps as reliably as it did", which is where a slow
degradation lives.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
MEASURABLE = [15,85] proxied for "can this fight move at all". The direct form,
in the units the guard already uses: the spread rate must stand clear of both
ends by more than its own noise. Deliberately blind to the gain -- admitting the
sizes where focus fire clears its noise would make the guard's claim true by
construction. Size is still picked on nearest-an-even-fight.
31 measurable became 38. the_arrears returns at 6.3, and switchboard arrives at
8.1 -- the second largest gain in the artifact, thrown away for being one point
past a round number. Four of the eight carry effects larger than most rows the
band already admitted.
Two of them do not clear their own noise: the_choir has 0.7 points of headroom
and used 0.2, the_stanchion has six and used 0.2. Above-noise falls 31/31 to
36/38 and the bestiary prints 36. That is two measurements reporting no
detectable effect, which the band suppressed by refusing to take them.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
It claimed focus fire is worth 6.3 points against two of them, 14.7% to 21%.
Measured independently at 4000 runs x 3 seeds: 14.2% to 20.4%, gain 6.2, which is
2.4x its own noise and 44% of the base. The claim was true and reproduces.
What failed is a threshold. MEASURABLE is [15,85] and the scan's estimate of a
boundary value moved 14.7 to 14.4. And the creature is a step -- 85.7% at one,
14.2% at two, 0.4% at three -- so no pack size gives an even fight and the band's
endpoints fall in the gap. The band records nothing about a creature whose
defining property is having no middle.
Not moving the band to 14: fitting a threshold to the datum it excludes is how a
guard stops being a test. But the band is a proxy for "can this fight move", and
the direct test -- does the gain clear its own noise -- is already in the
artifact and answers yes. Replacing the proxy is a decision about all 47, not a
fix, and not mine to take unasked.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Asked to raise the scan until the redcap's pack size stopped flipping. Measured
the threshold -- unstable at 3000 and 4000, stable across twenty seeds at 6000 --
raised it, and the re-record took fifteen seconds, which was impossible.
winRate takes three parameters and pickSize passed SCAN_RUNS as a fourth.
JavaScript discards it, so every scan has always run at RUNS and SCAN_RUNS has
never been read by anything. The fix I was asked to make was inert in the same
way as the thing it was fixing.
winRate takes runs now. The redcap is still n=3 with gain 6.3, arrived at stably
rather than luckily; the_arrears drops out of the measurable band at an honest
scan, 31 packs to 30; the_committee moves 2 to 6 and stays pinned. Claim check
still passes, bestiary regenerated.
The only signal was a number being too small. A fifteen-second re-record is good
news, and good news is what nobody investigates.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
check-focus picks the pack size nearest an even fight. For the redcap that is
n=3 at 28.0% against n=2's 73.4% -- correct, and also where focus fire is worth
most. But the margin is 1.4 points and the scan is 400 runs: run across eight
seeds it picks 3 seven times and 2 once, and the recorded gain would move 6.3 to
4.9 with it.
R-276's explanation was wrong. Its table was measured against the CLEAN GROUND
cut, which it never named. Against the frozen party a lone redcap is worth 0.4
rather than 6.1, because that party wins 98.9% and nothing shows against a
ceiling. The power is worth most where the fight is in doubt -- 3.8 at n=2, 3.1
at n=3, nothing at either end. Not outnumbered. Undecided.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Asked to re-record the lethality baseline against a lone redcap, I read the tool
and found it already does: measure(party, [spec]), with a comment saying "Solo,
because a creature is the unit under test". My claim that it fights packs came
from memory of check-focus, whose redcap is n: 3.
The lethality figure was also not hiding the power. Wipe rate moved 0.7% to 0.9%
because one redcap cannot wipe four agents whatever it ignores -- that column is
at its floor. Agents down moved 0.55 to 0.74 of 4, a 35% relative increase, which
is the column I did not look at.
No baseline re-recorded: it is already solo and was re-recorded in R-275.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Played four agents against one redcap. Two dodges in one round, both at 75 --
DEFENCE_STEP is -30, so the second would have been 45 and the roll of 70 would
have failed. The wiring fires.
Measured with and without, 2000 x 3 seeds: the power costs the party 6.1 points
against a lone redcap and 0.1 against three of them. It is a rule about being
outnumbered -- a lone defender spends four defences a round, a pack spends one
each -- which is why check-lethality moved only 0.7% to 0.9%. The baseline fights
redcaps in a pack, the configuration where the power is worth nothing, so that
figure is the floor rather than the effect.
Not presented as evidence: at seed 3 the party wins with the power on and is wiped
with it off, which is stream divergence rather than direction.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
41 statblocks carry a POWER in their tactics and simulate.mjs read none of them,
so a redcap that ignores the cumulative defence penalty has been measured as a
creature that tires -- in check-lethality, in check-focus, and in every figure
published about it. The defect was not that the powers were unimplemented, it was
that nothing said they were not.
powers.mjs classifies all 41: 2 wired, 14 notSimulable with a stated reason, 25
not fight rules. check-powers refuses an unclassified POWER and refuses a
notSimulable without a reason -- and it does not test that the harness imports a
power, it fights the creature with and without and requires the two to disagree.
Moved: the courier 9.4% to 1.0% wiped (it attacks at half while carrying), the
redcap 0.7% to 0.9% (small, because these fights rarely spend a second defence).
ARGENT AND GULES was wired and then un-wired: it tripled the supporter's wipe rate
to 75.2% because the harness has no ground and applied the borough-ground condition
unconditionally. Same reason THE PULL is not wired. I had wired one and refused the
other on identical facts.
Lethality and focus re-recorded, bestiary regenerated.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Played the hollow man against the cut. THE FILE IS YOURS NOW never fires --
simulate.mjs reads statblocks and weapons and has no concept of tactics, so
R-273 closed a gap in rules.mjs and left the same gap one layer out.
Resolved by hand it behaves: every roll pair enumerated for both beats. The
tie-break carries it -- at POWx5 100 against 60 the aggressor still only takes
them 48% of the time, because equal bands go to whoever is being acted upon.
Act Four prices THE OFFER: accepting it moves Ashcroft from the safest person in
the room to the least safe, 21.9% to 48.7%, past Braithwaite's 37.6%. And that
once-only row no-ops 14.5% of the time against him, which Act Three can absorb
and a permanent countdown beat may not.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The content has called for 'opposed POWx5' since before the rule existed -- the
hollow man's tactics, CLEAN GROUND twice, slice-text three times -- and rules.mjs
defined none. A scenario carried a local ruling that said in its own text it was a
ruling and not a rule.
Generalises that ruling rather than inventing another: both sides roll, the better
band wins, only the ladder the game already has. Ties go to whoever is being acted
upon, which is what defenceOutcomeFor has always said; the scenario's 'favour the
agent' gave the same answer only because no agent ever initiates one. Neither side
succeeding leaves the contest unsettled rather than won, which the two beats need
in opposite directions.
Two exports at the scenario session's request: opposedOutcomeFor compares graded
levels and carries both, so a caller can price a fumbled attempt without this file
deciding what a fumble costs; opposedContestFor runs it from ratings and rolls with
per-side difficulty, so 'resists at Difficult' does not put applyDifficulty back
into a document. Spot-checked against the real Act Four beat: 55 against 85 at
Difficult, which is 42.
Page 1 states it by asking it -- the tie-break and the margin are computed from the
rule at build time, so the book cannot drift from the engine.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Three things Tim asked for together: measure the long end of a fight, make the
swing check bite on stale prose, and record in rules.mjs the coincidence that
hid an invented rule from two readers.
GUARD 15 — check-fight-tail, on tools/fight-tail.mjs.
Fourteen guards measured this scenario's combat and none could see a long
fight, because every one of them averages. The cap is the measurement here, so
the tool passes its own (400, not runFight's default 40) and --update refuses
to record anything reaching it. Verified by lowering it back to 40: the cut
loses 20 fights to the ceiling, the six-a-side 117, 107 of those ending
neither won nor wiped, and the recorder stops.
Two claims, both read from a fresh measurement rather than the baseline, so
--update cannot silence them. "The long end is twice the median" was rejected
as a claim because it is true of both encounters and so distinguishes nothing.
Instead: a fight past 15 rounds wipes the party materially more often than a
short one, and the SIX-A-SIDE fight is the longer one (median 14 vs 10.7) --
R-270 showing up as duration, since a disabled fighter keeps fighting 30
points down. The cut is shorter because it is decisive, not safer.
GUARD 16 — check-cited, on tools/check-cited.mjs.
check-firstblood and check-attackers catch the game changing; neither reads
the document. Re-record after a re-cast and the artifact updates, the guard
goes green, and the prose keeps printing the old number under a citation
saying where the new one lives. So citations are now machine-readable --
**40**<!-- cite: first-blood cut.swing --> -- and resolved on every build. 25
of them. It failed three times on its first runs, all real: a config keyed
"column" that the prose called "line", two figures rounded 32.3 -> 32, and a
vacuous pass on zero citations, now fatal in its own right.
It also refuses citation of unstable fields. fight-tail.longest may not reach
prose: same party, same seeds, same runs, and renaming a config moved it 71 ->
90 rounds, because seedFor derives the stream from the id. Across seven
labels -- median spread 0, p95 1, p99 3, longest 21. A sample maximum reads
like a bound and is a property of the label. EXPOSURE states p99 instead.
Same discipline on the deadlier ratio: 2.42 with seed spread 1.1, so the
document gives its direction and declines to quote its size.
RULES.MJS — one comment, no rule change.
Over resolveLocationHit: its two thresholds are unrelated and usually agree.
disabled is a fraction of the pool per location; majorWound is ceil(hp/2) and
feeds only dyingLimitFor; neither removes anyone from a fight, which is
conditionFor at 2 hit points or a destroyed head. At 10 hp, leg/abdomen/chest
capacity is 5 and majorWoundFor is 5, and those locations take 12 of 20 melee
results and 15 of 20 ranged -- so two readers reconstructed a rule that does
not exist, checked it against the log, and were confirmed by it. The note
says to test an arm, the only place the difference shows.
Guards verified to bite, not assumed: drift, re-cast, censoring, the longest
refusal, the rounding catch and the vacuous-pass catch were each forced and
each failed the build with the right guidance, then restored.
npm run check: 16 guards, exit 0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Fourteen guards all average, so none can see how long a fight might run. 6000
fights per configuration: the cut runs a median of 11 and a 95th of 22, the
six-a-side line a median of 14 and a 95th of 32, and the longest seen are 71 and
79. The line is the LONGER fight, because more bodies means more of them fighting
on at reduced skill rather than dropping. Length predicts death: 61.6% of cut
fights past fifteen rounds are wipes against 42.0% of shorter ones.
maxRounds = 40 is the default every guard runs at and a fight reaching it is
scored as neither wipe nor win -- censoring 0.2% of cut fights and 1.87% of the
line's. Measured before proposing anything: uncensored, the swing moves 39.8 to
39.9 and 14.8 to 15.1, against a recorded noise of 2.1. The cap stays, documented
rather than corrected, because the cure is four re-recorded baselines.
Also R-270 addendum: the hit table is weighted toward the locations where the two
thresholds coincide -- 12 of 20 melee results, 15 of 20 ranged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-266 said a fighter goes down at half maximum hit points and called it one rule
in both directions. conditionFor puts someone down at hp <= 2; majorWoundFor is
ceil(hp/2) and feeds only how long the dying last; a location is disabled by
resolveLocationHit at its own locationMaxHp capacity.
The reason it survived reading: for a 10-point neighbour majorWoundFor is 5 and a
leg, abdomen or chest holds exactly 5, and for a 12-point agent both are 6. On the
locations that get hit most the invented rule returns the real one's answer, and
the narration prints MAJOR WOUND and disabled on the same line. Arms and heads are
where they part, and I had not looked at an arm.
Found by the scenario session going to rules.mjs to verify a different correction
of mine and reading the next function along. Both errors made the fight look
easier than it is.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Seed 16 is the missing corner -- party takes first blood in round 1 and is wiped
anyway -- and it runs 24 rounds. Its last five are Neil dropped by a critical
through armour, then Dominic alone at 1%, failing three times, then dying. That
is what three effective attackers costs at a table when a fight goes long.
Corrects a mechanism I had written twice: a disabling hit does not remove an
attacker, it charges 30 points off physical or manipulation per locationEffectsFor.
Bhattacharya keeps swinging at 10 for four rounds. A slope, not a cliff.
Neither guard fired and neither was wrong: the fight sits in the 25.5% the swing
does not cover, and no guard measures the tail because all of them average.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
CLEAN GROUND prints that the four-player cut is three effective attackers and
that Ashcroft is not one. Nothing checked it, and a GM reads it aloud to decide
who a real player spends four hours being.
effective-attackers.mjs measures per attack, not per fight: counted per fight
Braithwaite leads on disables, but only because his armour buys him a third more
swings -- per attack he is the weakest of the three. Both rates are recorded and
only the per-attack one is reasoned from. Asserts exact drift, then the sentence:
three clear 5% of attacks disabling, one does not, and that one is Ashcroft.
Re-recording does not silence the claim check; verified in a worktree.
declared-cast.mjs holds the cast marker reading both scenario guards need, rather
than a copy in each. It was briefly named scenario-cast.mjs, which check-scenarios
sweeps into the scenario corpus -- its own example marker was read as a real cast.
update-readme dropped any guard that exited non-zero, so check-rollable vanished
from the README and the count word fell to thirteen while fourteen guards ran. A
missing line is now fatal.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
I told the scenario session that this guard turns a silent re-cast into a build
failure, and it accepted the coupling on that basis. The cast was hardcoded, so
a re-cast would have left it measuring the old six and reporting success.
Reads the <!-- cast: --> marker check-rollable established, drops the two the
scaling note drops by name rather than by position, and refuses to run if the
marker is gone. Verified by re-casting CLEAN GROUND in a throwaway worktree and
watching the guard name the change.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Another session verified R-266's measurement, agreed the frame was better than
its own, and declined to publish it because the figure had no guarded lineage a
doubter could re-derive. It was right, and the fix costs 0.7s of build time --
which is the number I should have checked before calling it a reading tool and
not a guard.
first-blood.mjs gains --update/--check and is now both the reader and the
measurement of record; check-firstblood.mjs is a thin wrapper over it, the same
shape as check-bestiary. Compares the recorded figures exactly, then asserts only
what a page would claim: first blood lands within 5 points of even in the cut,
and its swing exceeds the six-a-side line's by more than both noises.
Recording it moved the swing from the scratch run's 43 to 40 against a
seed-to-seed spread of 2.1 -- high by more than its own noise, which is the
argument in miniature. update-readme's COUNT_WORD could not reach thirteen.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Reading both ends of the four-player cut — seed 5's sweep and seed 2's wipe —
says the fight is not decided by round five but by whoever lands the first
disabling hit, on average in round two. Both sides disable on one good blow, so
each one thins the return fire and makes the next likelier; nothing pulls a
fight back toward the middle.
tools/first-blood.mjs measures it: 49.7/50.3 on who strikes first, and 23.1% vs
66.5% wipes on either side of that. Six against six is the control at 8.6% vs
22.4%. A reading tool, not a guard — it records nothing, and no figure from it
goes on a generated page.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The round-five race line read "zero in 2000 fights", which invites the reading
that no fight in the run was a wipe. It is zero out of the 84 that reach round
five with two of the three down — a small bucket, and the sentence should say so
before the figure goes to print in a scenario.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
tools/playthrough.mjs exists to show a reader why a measured number is what it is. I cited
it to another session — read seed 2, see the wipe the 45% row is made of — and the
reproduction failed in front of them. They reported different outcomes at both seeds and
guessed the cause correctly from outside: the two tools were not drawing from the same
stream.
They were not. simulate.mjs used a private mulberry32 makeRng; playthrough.mjs had its own
LCG written to look like it. Both deterministic, both reproducible alone, and "seed 2"
named a different fight in each — which breaks the only thing the tool is for. Its own
comment claimed a seed here names the same fight there. check-focus carried a third copy of
that LCG, so the two guards described the same game with different dice.
makeRng is exported and both files use it. A playthrough seed is now exactly the first
fight of simulate.mjs --seed <n>: --runs 1 --seed 2 and the playthrough give 9 rounds, 4 of
4 down, 1 dead, both.
The other half was my citation rather than the code: the command I sent omitted --mode, so
it plays both targeting arms and prints two fights. They read the last line, I quoted the
first. The summary line now names the arm.
check-focus re-recorded under the shared stream; figures move a point or two. What it buys
is that the bimodality analysis now reproduces the published means exactly — 2.31, 3.03,
0.40, 0.42 against the four rows CLEAN GROUND publishes. Under the old LCG it agreed to
within a decimal, which looked like corroboration and was two experiments landing near each
other.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Reading a narrated supporter fight showed it being hit on armR and head. It has wings; it
has never had arms. Three defects, all in the pipeline every published number comes from.
1. It located hits by species, not body plan — simulate.mjs read defender.species where
the game writes spec.bodyPlan ?? spec.species ?? "baseline" into speciesProfile
(build-packs.mjs:294). The barghest, kelpie and church grim were fought on two legs
with arms, and the supporter with no wings, so the fight its tactics call the one the
agents can win could not occur in a measured fight.
2. Armour was one scalar for the whole creature, and the harness's locations had no
armour field. Bare wings, the grounded halving and the vital exemption — R-254 and
R-255 — were invisible to every number. Worn armour was summed the same way, which put
a stab vest on a cleaner's head.
3. Nothing was ever grounded: a wing could be ruined and the creature kept flying.
Fixed in the game's order — locate, then apply what that location carries, every term
imported from rules.mjs. A ruined wing calls groundedPlanFor and the wounds carry across
by severity through remapLocationDamage, the function _preUpdate uses.
A bug of mine no guard would have caught: remapLocationDamage returns { damage, moved,
rescaled } and my first draft passed the whole object where a damage map was expected, so
every wound a creature carried was forgiven the moment it came down. check-lethality would
have passed it — fewer wounds means a longer fight, which reads as a number moving, and
this commit moves numbers. Found by probing a landing by hand.
20 of 47 creatures moved, 18 deadlier and 2 less. The supporter goes 69.3% -> 25.1% wiped,
3.38 -> 2.16 down, second deadliest to fourth: it was being measured as a 30-hit-point
creature in uniform armour 9 that could not be grounded. The small rises elsewhere are the
party's armour no longer covering locations it never protected.
check-focus then caught the page overclaiming, which is what it is for: focus fire still
helps in all 30 but only 28 clear their own noise where 31 of 31 did. The guard was
asserting more than the page needs — the bestiary prints that count from the artifact and
cannot overstate it — so it now checks only what the page asserts outright, and the page
rewrote itself to "in 28 of them".
Both baselines re-recorded. Minor version, not a patch: the published numbers changed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Every guard here reports a number per creature, and a number cannot say why a fight went
the way it did. That gap is what produced R-258 through R-260: each written from a side
harness built to watch a fight, each modelling something slightly different from the game,
two of the three wrong in ways no guard could catch because no guard was involved.
runFight takes an optional `say` sink. It reports what the simulator already decided — the
attack roll and its target, a defence and the penalty it was made at, the landing level
after a dodge downgrades it, damage against armour before and after armourAgainst, the
location, major wounds, disablement, death. It never touches the generator, so a narrated
fight and a silent one are the same fight; check-lethality and check-focus both still
match their baselines exactly with the hook in place.
tools/playthrough.mjs (npm run play) is its consumer, in the same commit deliberately: a
hook with no reader is the exact defect this project keeps finding in its own rules, and
adding one to the measurement pipeline with only a scratchpad file calling it would have
been committing the thing I have spent the session removing. Pack size comes from
focus-baseline.json so the fight you read is the fight check-focus measures; creatures
recorded as pinned are played solo.
Three redcaps, seed 20260913, same seed both ways. Spread fire: wiped in 10 rounds, 3
dead, and all three redcaps still standing — 39 hit points spread three ways so that none
of it finished anything. Focus fire: same opening, diverging at one target choice in round
1, party wins in 20 with two up. The +13.2 points check-focus records, seen once instead
of averaged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Two sessions numbered an entry against the same committed log, which ended at
R-260, and both picked R-261. Mine was written first but was uncommitted, so the
other session could not have seen it and had every reason to think the number was
free. When it staged docs/REVIEW_LOG.md my entry was sitting in the file, so it
went out inside 1d915c5 under that commit's message, and HEAD reached origin with
two R-261 headings in it.
Mine renumbers to R-262 and moves below theirs so the log stays monotonic. Theirs
keeps R-261 because theirs is the one that was committed; this diff moves and
renumbers nothing but my own prose, and their entry appears in it only as context.
The hazard is now a matter of record in both directions: their note at the foot of
R-261 caught check-rollable arriving from another session mid-build and handled it
without clobbering anything, and this is the same collision on a file whose
convention is to append to the end. A shared working tree makes the end of
REVIEW_LOG.md the likeliest place for two sessions to meet, and an uncommitted
entry there has no claim on its own number.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-258 put a tactic on a GM-facing page from an unguarded harness, R-259 found it worth
nothing, R-260 found the replacement right for the wrong reason. Three corrections, each
landing as prose with nothing checking it — the arrangement that let ddc4f99 describe a
game nobody was playing.
tools/focus-baseline.json records, per creature, the pack size and the spread-fire and
focus-fire win rates. check-focus re-measures all of it on every build and compares
exactly, with the party, seeds, run count and seeding scheme recorded alongside so numbers
taken under different conditions are refused rather than compared. The bestiary READS the
artifact instead of restating it, and check-bestiary refuses a page that has fallen behind
it. Page, guard and simulator cannot disagree.
The pack size is recorded rather than re-chosen: a fight at 0% or 100% cannot show an
effect, and a guard that picked again each run would let a changed creature move quietly
to a different question and pass. 31 of 47 creatures land in the measurable band; the
other 16 are recorded as pinned, with the rate that pinned them.
It guards the claim as well as the numbers. The page says focus fire helps in every fight
in doubt; check-focus fails if any row's gain reaches zero or stops clearing its own
noise. That failure means rewrite the page, not re-record the baseline.
The run count was chosen by evidence. 1000 x 3 seeds costs 8.5s and takes the suite from
2.5s to 13.7s. I tried 500 to halve it and the claim-check failed — at 500 runs one row
no longer clears its noise, so "without exception" is not supported by that much
sampling. Recording twice at 1000 gives byte-identical files.
Negative-tested four ways, all firing: a creature quietly made nimbler (redcap dodge
75 -> 85, spread 40.6% -> 27.6%), a baseline under different seeds, a hand-edited page,
and the claim failing at 500 runs.
Guard eleven (check-rollable) arrived from another session mid-build; this is twelve.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Measured in simulate.mjs. For each creature, the pack size whose spread-fire win rate
lands nearest 50% — a fight at 0% or 100% cannot show an effect, which is why R-259's
first sample was three-quarters useless — then focus versus spread, 2000 fights across
three seeds.
31 of 47 creatures had a pack size that could show anything. Focus fire helped 31 of 31,
by more than that row's own noise 31 of 31 times, mean +12.8 points. Best is 6x The
margin, 40.4% -> 61.1%. The other 16 are unmeasurable rather than unaffected: eleven are
trivial six at a time, five are hopeless in pairs.
R-259's decomposition was confounded, and I wrote it while correcting a confound. It
credited the defence ladder with 4-5 points by comparing dodge <=30% against dodge >=55%
— but the low-dodge samples were PAIRS and the high-dodge ones TRIOS. Held at a fixed
pack size, dodge is worth 1-2 points. Pack size is the driver: 8.0 points for a pair,
14-15 for three or more, then flat. What focus fire buys is that a dead creature stops
attacking and a half-dead one does not attack any less.
Two failures, one cause: comparing groups that differ in more than the thing being
measured. The control is cheap and I did not run it until the third time.
The bestiary advised concentrating fire and gave the ladder as the reason — right advice,
wrong reason. It now gives the reason that survives measurement.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-258 put a tactic in the bestiary on the strength of a table-side harness — cheapest
weapon first, best weapon last, 10.7 points of win rate across 24 orderings. Measured in
simulate.mjs, the pipeline every published number comes from, it is worth +0.27 points,
with per-seed differences of -1.5 to +1.8 against 2.6 points of noise on a single arm.
Zero.
The toy lied for two reasons, both mine. Its party had one uniquely valuable weapon (an
anti-materiel rifle the borough does not own, so unhalved, worth double anything else) so
there was something to sequence around; the frozen party is 4.09, 3.98, 1.43, 1.43 —
two near-identical pairs. And the toy fixed the order every round, while the game
re-rolls Reaction every round and has no rule for holding an action, which I checked
before measuring. The order is not a decision the rules offer, and I had written table
advice for it.
The page is corrected: the ladder stays, because it is derived and true, and it now says
you cannot choose who goes first.
What is worth doing, measured the same way: committing a round's attacks to one target,
in fights that can actually move (a fight at 0.2% or 100% cannot show an effect and three
of my first four samples were pinned there). Against dodge <= 30% focus fire gains 7.6
points, which is just killing attackers sooner; against dodge >= 55% it gains 12.2, so
about 4-5 points is the defence ladder and the rest is arithmetic unrelated to dodging.
Separating them needed contested fights at both ends of the dodge range — without that
control I would have credited all 13.8 points against three redcaps to the ladder, which
is R-255's mistake again.
runFight gained two hooks, off by default and unused by the baseline: partySequence
rearranges the party within the slots initiative already gave them, leaving enemies where
they fell so the experiment isolates agent order from who acts before the creature; and
partyTargets "focus" concentrates fire. Both roll the same dice, and check-lethality
confirms all 47 creatures still fight exactly as recorded.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
DEFENCE_STEP is -30, so a creature's Dodge is not a percentage it has for a round — it is
one it has once or twice. The median dodge in the book is 45% and goes 45 -> 15 -> 1
inside a single round. No player-facing document said so, which hid a real decision:
what order the party shoots in.
The bestiary now has a section on it. Only a landing hit spends a defence, so a miss
strips nothing; each landing hit makes the next attack of that round easier; therefore
the cheapest reliable weapon fires first and the one you most need to land fires last —
against the supporter, the weapon the borough does not own.
Everything it states is generated from DEFENCE_STEP, resolveBands and the statblocks: the
step, the ladder, the median, the spread (13 of 47 creatures stripped by one landing hit,
25 by two, 9 by three) and the list of the hardest to strip. Negative-tested by setting
the step to -15 and -20; the page rewrites itself both times. No win rates are published:
the ordering measurements come from a table-side harness that ignores wound penalties,
bleeding and dying, so the page states the mechanism and quotes no percentages.
My first draft used the dodgiest creature in the book as the worked example — 80%, three
landing hits to strip — directly beneath a claim that the first attack of the round eats
the dodge. True of a modest dodge, false of a good one, and both were on the same page.
tools/bestiary.mjs --check has existed since the document did, with the comment "(for the
guards)", and no guard ever ran it: a check written and wired to nothing, the same shape
as every rule found computed and never read. That was tolerable while the page restated
statblocks and is not now that it derives from rules — change DEFENCE_STEP and the
document lies to a GM. check-bestiary.mjs is the tenth guard, negative-tested both ways
it can fail: a rule change that staler the page, and a hand-edit of the page.
update-readme flagged it as "run by the build, not advertised" before I listed it, and
the README's count word went from nine to ten by itself.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Left open by R-256. Two rules decided how a hit meets armour — armourAgainst (a critical
ignores it, a special halves it, rounded down) and damageAfterArmour (what is left never
goes below zero). Both lived in ringbrp.mjs, which is the game. Neither lived in
rules.mjs, which is the authority, so the guard that forbids redefining a rule had
nothing to forbid. tools/simulate.mjs applied armour from its own inline copies of both,
as it had since it was written.
They agreed, which is the whole point. Nothing published was wrong and no guard could
have said they were two things rather than one. But every number in BESTIARY.md, every
row of the lethality baseline and every measurement quoted from R-251 onward comes out of
that harness: change how a special hit meets armour and the game changes, the numbers
describing the game do not, and all nine guards still pass. Same shape as ddc4f99, with
no tolerance to blame and no reason it would ever have surfaced.
Both rules now live in rules.mjs. ringbrp.mjs imports and re-exports them because they
are public API at game.ringbrp. The simulator imports them. Verified live in the world
that game.ringbrp.armourAgainst and the rules.mjs export are the same function object,
not two that agree.
Proof it changed nothing: check-lethality replays 47 creatures over 2000 fights each and
every one fights exactly as recorded — no tolerance, no drift — across roughly four
million resolved attacks, all of which now go through the moved rule.
check-rules fails any file outside rules.mjs that halves armour or subtracts it inline,
naming file and line. Negative-tested by restoring the harness's original three lines
verbatim (both patterns fire) and by redefining armourAgainst in ringbrp.mjs, which the
name-based guard now catches because the rule finally lives somewhere it belongs to.
Arriving in rules.mjs also tripped the spot-check requirement immediately: eight new
spot-checks, and what a special does to hide 9 is now written down once.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Verifying R-255 on a live sheet found the rule working everywhere it was built and
contradicted in the first place a GM looks. The header's ARMOUR field is the hide the
creature HAS; on a grounded beast the hit-location table under it read 4 and the header
still read 9. Nothing computes off that field, so nothing in the code was wrong. It was
wrong at the table.
The field stays editable and stays 9 — that is the creature's hide and what a GM types
into. Under it, when system.grounded is set, the sheet now says what is in play:
DOWN: 4 · CHEST 9, derived from exposedHideFor rather than from a second opinion, with
the reason on the tooltip. Verified live: no badge in the air, and grounded the badge and
the location table agree at 4 with the forequarters at 9.
check-rules forbids redefining a rule by NAME, which misses how this goes wrong in
practice — nobody redefines exposedHideFor, they write Math.floor(hide / 2) in the sheet
because it is cheaper than an import. I wrote that version first and check-rules passed
it. It now fails any file outside rules.mjs that halves a hide inline, naming file and
line; check files are exempt because stating the expected value independently is what a
test is for.
That guard also turned up something I have not fixed, recorded in R-256: armourAgainst
lives in ringbrp.mjs rather than rules.mjs, and tools/simulate.mjs has its own inline copy
of it. They agree today so no published number is wrong, but every figure in the bestiary
and the lethality baseline comes out of that simulator.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-254 shipped the belly rule and called +6.4 points "modest on purpose". That read the
wrong statistic. +6.4 measured whether grounding helps; the question was whether the
borough still matters, and the gap between fighting the supporter on borough ground and
off it had fallen from 60.6 points to 44.9. Half damage from every departmental weapon is
meant to be the fight's central problem, and I had quietly taken sixteen points off it
while reporting a small number about something else. Same error as R-251.
Three changes that only work together:
- hide 7 -> 9, restoring the weight on the axis that should hurt
- grounded still halves it, 9 -> 4, which an ordinary round beats
- exposedHideFor now takes the location's kind and leaves `vital` unhalved: a beast on
its belly is not presenting its chest to the floor
Measured: 28.4% on borough ground flying, 68.4% grounded, 84.6% off it — a 56.2 point gap
against 44.9 before. Predicted 28.4 / 84.3 / 55.9 before building.
The composition of the three hide functions lives in prepareDerivedData where no test
could reach it, and check-anatomy's guard 6 is deliberately one-sided so it would not
have caught the vital losing its exemption. check-behaviour now composes the same
expression from the real supporter spec and the real tables. Negative-tested four ways:
removing the exemption, un-baring the wings, removing the halving, and inflating the hide
past what a round can beat each fail it with the right message.
Cost: the supporter goes from fifth to second deadliest creature in the book (2.84 -> 3.38
down, 46.8% -> 69.3% wiped). check-lethality refused the build until that row was
deliberately re-recorded.
BESTIARY.md printed the new hide and said nothing about it halving, because the body-plan
blurb predates the rule. The README said "eight guards" in three sentences above a
generated block listing nine. Both are the signature defect in prose: a number written
down that nothing reads. Both now generate.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-95 made losing a wing change the animal. R-96 took the hide off the
wing so losing one was possible. Separately each was right and each was
measured. Together they gave the party a soft target and deleted it the
moment the party hit it:
winged 6/20 ranged faces land on armour 0 (a bare wing)
quadruped 0/20
A service round on borough ground averages 3.75 after halving — 3.75
through a bare wing, 0 through hide 7 — so grounding the creature took
the party's chance of winning from 44.9% to 1.4%, while its own tactics
told the GM that grounding it was the fight they could actually win.
Found by measuring something else. Every aimed-shot configuration made
the party worse, and the most aggressive was the worst: aiming at a wing
at -20% grounds it in 99% of fights and HALVES the win rate to 10.1%.
Buying the objective reliably was the fastest way to lose, which only
makes sense if the objective is a trap. So no aimed-shot rule; the
problem was never the lack of one.
It also retires the "+31 points" I reported for grounding at 1.7.3. That
was win-rate-if-grounded against win-rate-if-not — selection bias, since
the parties that grounded it were the ones shooting well. Forcing the
grounding gives the causal value and it was -43. A correlation of +31
and a causation of -43 out of the same mechanic.
The fix: a heraldic beast is armoured the way it is drawn, across the
back and the flanks. On its belly half that hide is no longer in the
way, so exposedHideFor halves natural armour while system.grounded is
set. Hide 7 becomes 3.
flying grounded worth
on borough ground 45.3% 51.7% +6.4
off borough ground 90.6% 97.9% +7.3
Modest deliberately. Grounding should help, not decide — the borough is
still the real answer. Losing its damage modifier was worth +26.5 and
would have made the wing the whole fight.
check-anatomy gains a sixth property: a flyer must not be harder to hurt
on the ground than in the air. The first version probed at a single
4-point round and declared the fix broken, because 4 sits inside the one
band where a bare wing beats a halved hide — below 5 damage the creature
really is better off grounded, above it much worse. Integrating across
1-12 gives 2.83 flying against 3.75 grounded and reads it correctly.
Negative-tested against 1.7.8's behaviour, which it names.
The lethality baseline is unchanged: its simulation never grounds
anything, so system.grounded never comes up there.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
R-252 established that a recorded figure carried several points of slack
at 200 runs and that only more runs narrows it. This is that.
Measured by taking every creature under four different base seeds and
looking at how far the four answers sat apart, which is the noise a GM
reading the bestiary is unknowingly trusting:
mean spread worst creature
200 runs 1.28 points 11.5 points
2000 runs 0.36 points 2.7 points
The worst case is what mattered. One creature's published wipe rate
could be eleven points from the same creature rolled with different
dice, on a figure printed in docs/BESTIARY.md for somebody to plan an
evening around. Under three now.
It corrected 19 of 47 published figures, mean 0.35 points: the arrears
6% -> 8.9%, the supporter 44% -> 46.8%, the nuckelavee 13.5% -> 16%.
Each was a sampling artefact printed as a property.
The cost is the part worth recording, because it is why this was never
done. I guessed "maybe a minute of build time" when I suggested it,
which was wrong by a factor of thirty and would have been a fair reason
to say no:
check-lethality alone 0.29s -> 1.88s
whole nine-guard suite 2.55s
BESTIARY.md regenerates from the baseline, so its run count follows
automatically; its header now also says the seeds are derived per
creature, since "at seed 11" stopped being the whole truth in R-252.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Each creature now derives its own seed from its key, FNV-1a mixed into
the base seed, instead of every one of them being measured against seed
11.
This is the second reading of the request and the one I had not tested.
What I checked before was position independence — whether a creature's
numbers depend on its neighbours — and that was already true and still
is: inserting a creature ahead of the barghest changes 0 of 47 existing
rows under the new scheme, same as the old. What was NOT true is that
each creature had its own dice. All forty-seven faced the same two
hundred sequences.
Measured, because the argument for doing it is better than the result:
shared 11 per-creature
mean wipe rate across bestiary 5.67% 5.67%
mean agents down 0.538 0.522
rows changed -- 34 of 47
Seed 11 was not biasing the book. There was no systematic luck to
remove, and the aggregate is unmoved to two decimal places. What the
change buys is decorrelation: the error in each row no longer comes from
the same draw as every other row.
The useful number fell out of the comparison rather than the change.
Individual creatures moved up to four points of wipe rate purely from
being handed different dice — the courier 3.5% -> 7.5%, the long walker
66.5% -> 62.5%, quarantine unit 84.5% -> 88.5%. That is the sampling
noise inside any single recorded figure at 200 runs, and it means these
numbers are an exact regression fingerprint and a loose description of a
creature at the same time. Only more runs narrows the second; more seeds
does not. R-252 says so on the page.
seedMode is recorded alongside the numbers and checked, because changing
how a seed is derived moves every row without changing SEED itself. A
baseline from the old scheme is now refused rather than compared against
this one silently and wrongly — verified by running the new code against
the old file before re-recording.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The baseline needed no re-recording — e5dc9b5 already captured these
numbers, it matches HEAD's code exactly, and it differs from 4b71859's
in precisely the 26 entries ddc4f99 moved. What was missing was any
record of WHAT that change did, since it was folded silently into a
commit about wings.
Isolated by replaying 322389b and ddc4f99 against the same party, seed
and counts, so the figures are ddc4f99's own and not contaminated by the
four commits after it:
creatures whose numbers moved 26 of 46
deadlier 3
less deadly 8
unchanged wipe rate, moved elsewhere 15
mean change in wipe rate -0.17 points
Routing every blow through a hit location REDISTRIBUTED lethality
without raising it. Creatures with one big attack got slightly less
dangerous, because a blow that used to come off a single pool now has to
beat the armour covering wherever it landed; creatures that attack often
got slightly more dangerous, because they now accumulate damage in a
limb and disable it. Across the bestiary those cancel to within a fifth
of a point. Largest moves: the courier 6% -> 3.5%, the arrears 7.5% ->
9.5%, the stanchion 26% -> 28%, the long walker 68% -> 66.5%.
The log entry also records the two corrections I owe this thread, both
from explaining before measuring: the seeding was always per creature,
and the claim in check-lethality's header that I called untrue was true.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Ten pages a player reads before session one, in their own compendium at OBSERVER
ownership: nine of setting, one cheat sheet.
The draft called the organisation "the Bureau"; everything else in the game calls it
the department or the agency, so that is what it is called. Thin areas filled rather
than padded: the seven character principles as a list, the real Coherence band table
instead of a paragraph describing one, and "what continuity is not" pulled out of the
middle of a page because it is the most important sentence for a new player.
The cheat sheet's numbers are computed from rules.mjs and lang/en.json at build time.
A hand-typed one would have been the sixth time this project shipped a number that
disagreed with the code, and the one the players were holding.
R-250: the design tokens were never on :root. Journals rendered as default parchment,
and four things were wrong in sequence — v14's hook is renderJournalEntrySheet; its
`element` is the node, so `[0]` returns the first CHILD and put the class on a header
button; there is no .journal-page-content in v14; and then the real one: with the
selector matching, font-size, line-height and padding all applied while every colour
did not, because the palette is declared on .ringbrp-sheet/.ringbrp-chat/.ringbrp-window
and never on :root. A journal inherited no tokens, so every var() silently became
nothing. A rule that HALF applies is not a specificity problem — I spent three attempts
treating it as one, including a version bump on a cache theory that was wrong.
Branded: primer, rulebook, starter GM text and handouts. Scoped by flag, so other
modules' journals are untouched. Version 0.9.0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
