Ruling — Le Havre

You asked

When does the game end?

Answered counted as answered

Asked with: Le Havrebg-35677-en-1e

The game ends immediately after the final actions [1]. The sequence leading to the end of the game is as follows: 1. The round stage ends when the last Round card is resolved [1]. 2. The final stage begins [1]. 3. Each player takes exactly one more turn to carry out a final action (a main action) [1]. 4. The starting player begins their turn, followed by each other player taking a turn [1].

[1] RULEBOOK Le Havre — Rulebook · p.9 · F I N A L S TAGE · F I N A L A C T I O N · GAME END AND WINNER

🔧 debug — where this ruling came from

outcome: ANSWERED — Answered
corpus profile: live
answerer pin (deploy-frozen, attribution only): gemma4:26b think:false 5571076f3d70
timings: search 264 ms · ask 10240 ms · total 10514 ms
single-flight (PR-5): queued 0 ms · generation 10514 ms
retrieval refs (raw): [{"page": 9, "score": 0.04918, "snippet": "The round stage ends when the last Round card is resolved. The fi nal stage begins. Each player has exactly one more turn and carries out a fi nal action a main action (see page 5) . Supply and Buying actions may not be taken, and Interest is no longer paid. Players may still repay loans and sell bu", "headingPath": "F I N A L S TAGE · F I N A L A C T I O N · GAME END AND WINNER", "sharpsignalDocId": 233}, {"page": 1, "score": 0.046642, "snippet": "Le Havre is a French city, containing France's second largest harbor (after Marseilles). The city is notable not only for its size but also for its unusual name. The Dutch word 'Havre', meaning 'Harbor', was adopted into French in the 12 th century, but these days it is considered archaic and 'le po", "headingPath": "ABOUT THE GAME · C O M P O N E N T S", "sharpsignalDocId": 233}, {"page": 10, "score": 0.046432, "snippet": "No Token Limit The supply of game tokens (goods and Francs) is unlimited. If the supply of one type of good runs out, there are 5x multipliers on the back of the 2 food production tokens. If a token is placed on one of these, it counts as 5 tokens. Building Proposals Emptied If any of the three Buil", "headingPath": "RULE CLARIFICATIONS · LOAN CARDS · A player receives 4 Francs for each loan taken.", "sharpsignalDocId": 233}, {"page": 4, "score": 0.045956, "snippet": "During each player's fi rst action, new goods arrive in the harbor and are placed on the corresponding offer spaces. The player taking the turn places their ship marker on the next free Supply tile (in the direction of the arrow). In the fi rst round, the Supply tiles are face down until they are oc", "headingPath": "S UPPLY A C T I O N", "sharpsignalDocId": 233}, {"page": 11, "score": 0.044337, "snippet": "Who Takes the First Move in a Round? Whenever a Round card is turned to the ship side, the number at the right of the new Round card shows which player will take the fi rst move in the new round: '1' is the starting player, '2' is the next player, etc. This ensures that players remember to turn over", "headingPath": "TIPS F O R SMOOTHER GAME PLAY", "sharpsignalDocId": 233}, {"page": 4, "score": 0.032258, "snippet": "During the round stage , players play a fi xed number of rounds (depending on the number of players). A round consists of 7 individual player turns and, following those, the resolution of the applicable Round card. In the fi nal stage of the game, each player carries out one fi nal action. Afterward", "headingPath": "PLAYING T H E G A M E · ROUND STAGE · A PLAYER'S T URN", "sharpsignalDocId": 233}, {"page": 12, "score": 0.031746, "snippet": "· We suggest that players rotate their Building cards away from themselves, so that other players can read them more easily. Players need not wait for the previous players to announce that they have fi nished their turn before taking their own turn. Rarely, players might choose to sell or buy a buil", "headingPath": "TIPS F O R SMOOTHER GAME PLAY · Rotate Buildings · · After the Main Action, It Is the Next Player's Turn · · Round Number · · Final Actions", "sharpsignalDocId": 233}, {"page": 15, "score": 0.029851, "snippet": "A player may only sell buildings and ships outside of their turn if they cannot meet a forced payment (like interest or food). The player may sell as many buildings and ships as is necessary to meet that payment. On their own turn, a player may sell as many buildings and ships as they like.", "headingPath": "S E L L I N G B UI L D I N G S A N D S HI P S", "sharpsignalDocId": 233}]
citations (raw): [{"page": 9, "quote": null, "headingPath": "F I N A L S TAGE · F I N A L A C T I O N · GAME END AND WINNER", "sharpsignalDocId": 233}]
boost provenance — errata + base rulebook (E2/E3): no registered errata card matched this query; base rulebook doc 233 boosted; rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base rulebook doc 233 boosted
answered-before recall: this ruling was answered, so past rulings were not offered
similarity threshold in force: 0.55 (rulesage.answered-before.similarity-threshold; the exact-question tier is not gated by it)
debug row retained until: Nov 6, 2026 17:27

retrieval — what the search arm returned

score = the fused retrieval score (reciprocal-rank fusion over the search arms) — NOT a cosine similarity. This ruling ran the THREE-ARM fuse (full-text + vector + the doc-priority boost arm, which carried the base rulebook and/or the errata corrections this query matched), whose arithmetic ceiling is 0.04918 — all three arms ranking the same chunk #1 (ceiling = arms/(k+1) at k=60). The measured healthy band 0.016–0.033 was measured on TWO-arm asks and does not bound this one: its high end is the two-arm ceiling, and a third arm has room above it. A chunk only one arm ranked highly lands near 0.016 (that one arm's #1 vote), so 0.03 here is an excellent hit, not a 3% match.

per-arm ranks (the rank each search arm gave a chunk): NOT DISCLOSED by the gateway. Fusion happens gateway-side and only the FUSED score crosses the wire, so rulesage cannot show them without a gateway change. The rank column below is the FUSED rank (position in the returned list), which is real.

token count per chunk: NOT DISCLOSED — chunk token counts are not reported on the ask path. The character counts below are of the bounded snippet/quote rulesage actually received — a real measurement of what we hold, never an estimate of the chunk.

fused rank	document	heading	page	fused RRF score	snippet chars
#1	RULEBOOK Le Havre — Rulebook (corpus doc 233)	F I N A L S TAGE · F I N A L A C T I O N · GAME END AND WINNER	p.9	0.04918 (at the 3-arm ceiling (0.04918) — every arm ranked it #1)	300
#2	RULEBOOK Le Havre — Rulebook (corpus doc 233)	ABOUT THE GAME · C O M P O N E N T S	p.1	0.046642 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#3	RULEBOOK Le Havre — Rulebook (corpus doc 233)	RULE CLARIFICATIONS · LOAN CARDS · A player receives 4 Francs for each loan taken.	p.10	0.046432 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#4	RULEBOOK Le Havre — Rulebook (corpus doc 233)	S UPPLY A C T I O N	p.4	0.045956 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#5	RULEBOOK Le Havre — Rulebook (corpus doc 233)	TIPS F O R SMOOTHER GAME PLAY	p.11	0.044337 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#6	RULEBOOK Le Havre — Rulebook (corpus doc 233)	PLAYING T H E G A M E · ROUND STAGE · A PLAYER'S T URN	p.4	0.032258 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#7	RULEBOOK Le Havre — Rulebook (corpus doc 233)	TIPS F O R SMOOTHER GAME PLAY · Rotate Buildings · · After the Main Action, It Is the Next Player's Turn · · Round Number · · Final Actions	p.12	0.031746 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	300
#8	RULEBOOK Le Havre — Rulebook (corpus doc 233)	S E L L I N G B UI L D I N G S A N D S HI P S	p.15	0.029851 (healthy, below this ruling's 3-arm ceiling 0.04918 (the measured band 0.016–0.033 was taken on two-arm asks and does not bound a 3-arm one))	291

chosen chunks — what the answer actually cited

#	document	heading	page	quote chars
[1]	RULEBOOK Le Havre — Rulebook (corpus doc 233)	F I N A L S TAGE · F I N A L A C T I O N · GAME END AND WINNER	p.9	no quote on the wire

generation — timing and the path taken

timing split: retrieval 264 ms · first delta 9748 ms · total 10514 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
full breakdown: queued 0 ms · retrieval 264 ms · ask 10240 ms · gateway work 10514 ms · total 10514 ms
tokens in (the prompt the answerer read): 3182
tokens out (what the answerer generated): 97
tok/sec: 138.4 tok/sec — 97 output tokens in 0.701 s of generation
live seat (read from the running child): answering — gemma4:26b think:false · num_ctx 32768 (live-verified 32s ago)
stream path: streamed — the gateway pushed at least one accumulated draft
watchdog: first-token watchdog: 30s per arming window (rulesage.table.first-token-watchdog-seconds), re-armed once by the passages frame — so the worst case before a reader falls back to polling is 60s. Whether it actually fired is a BROWSER-side decision that never reaches the server, so it is not recorded here. The stream path above is the server's half of that story: whether a delta was emitted at all.
num_ctx · think posture: both are the CHILD's settings, not rulesage's, and no ask reply carries them — an answer never states which context window or think posture produced it. The live-seat line above reads them from the running child instead, where the gateway can answer for itself; it describes the child NOW, not this ask.

disposition

outcome: ANSWERED — Answered
abstain rationale (n/a — this ruling answered): answered — no abstain rationale applies

setup-shape — why the checklist band did or didn't render

decision: NOT_SETUP_SHAPED — no setup-shape pattern matched this question

errata registry — what the gate above had to match against

this game registers NO errata cards (edition 159), so "no registered errata card matched" above could never have matched one

Errata scope: the gate above matched this ruling's own edition only. When an ask also reads an expansion, that expansion's own registered errata cards stay outside the match — a stated phase-1 limitation, not a failed lookup.