Pengo E004: the missing egg state and a verified 6-to-5 event
Hypothesis pending the reader-only counterfactual: the recorded ACT 3 stagnation may reflect incomplete egg representation, planner limitations, or both. The retained segment has no detected deaths and25decisions without official objective reduction. The old reader made all egg targets UNKNOWN; this observation does not isolate the causal bottleneck. A reader-only intervention is needed before attributing this outcome to strategy.
E004 is Objective Reduction Analysis, as requested after the E003 stop. No new planner search, candidate simulation or parameter tuning follows. 9,790 remains best verified planner trajectory; 9,090 remains the observed score of incomplete E003. ACT 3 is not completed.
A concrete representation defect
The historical reader checks six egg-coordinate entries. ACT 3 contains valid observed data in entries 7 and 8 too. Consequently the old egg reader returns UNKNOWN on all 1,290 E003 ACT 3 observations and 1,282 pre-death E009 observations analyzed. Its policy falls back to visible enemies, missing egg targets.
The new extraction is offline, privileged and scoped to observed ACT 3. The historical reader and frozen search remain unchanged. Later-ACT capacity is UNKNOWN, and no RAM-derived fields enter the pixels-only runtime agent.
At E003’s endpoint frame 7,306, the six objectives reconcile as three mobile Sno-Bee at (5,8), (5,7), (7,8), plus three unhatched eggs at (3,13), (3,11), (5,1). Coordinates are zero-based logical cells. The player is at (9,7). The per-object report includes motion, nearby pushable ice, launch stances, lanes, blockers and walking distance. A nearby block is not necessarily a valid attack.
The eight-entry list plus populated native actor slots matches the count on 1,279/1,290 E003 and 1,275/1,282 E009 frames; 11 and 7 transient observations remain unreconciled. All 26 chosen stagnation endpoints reconcile. These limits remain explicit. The first offline analysis mistakenly converted UNKNOWN eggs into zero; that receipt is retained and its interpretation is superseded by revision 2.
The event that really makes 6 become 5
One existing E009 reset tape supplies an ordinary ice-crush example. From player (4,7), RIGHT+BUTTON pushes the ice at (5,7) along (6,7),(7,7), stopping before ice at (8,7). A Sno-Bee occupies the target cell (7,7).
| Physical frame | Observed event |
|---|---|
| 6,771–6,772 input selections | Right and button; existing tape, unchanged |
| 6,773 | Source ice cell is empty |
| 6,778 | Native enemy hit state |
| 6,783 | Crush animation |
| 6,807 | +400 observed; objective count still 6 |
| 6,895 | Count falls 6→5; replacement hatch begins at (3,11) |
117 physical frames, about 1.93 seconds, separate contact from the counter decrement. Two fresh native-speed reset replays reproduce all declared trace bytes and sampled pixels/devices, with no detected death or life fault in the ACT 3 segment. Their wall times are 116.219 and 116.235 seconds. This is one unique event with two replicas, not two independent mechanics cases. Complete machine-state identity remains UNKNOWN.
What the next representation must explain
A read-only scan of 1,639 previously simulated E003 branch endpoints in decisions 40–65 finds no endpoint below 6. The 38 endpoints at 7 belong to decision 40’s root before the chosen 7→6 transition; they do not show a count increase. No new simulations were run.
The conservative proxy is shortest known-empty walk to a static interaction stance, plus a lower bound on ice obstructions. It fluctuates between 2 and 10 cells and ends at 9 while the counter stays 6. An egg slide need not eliminate the egg, and enemy movement/timing can defeat a geometric lane. This proxy is not yet true distance to the next elimination or a proven minimum push count.
The concrete prerequisites are measured egg coverage, per-target interaction state and a distinction between a hit in progress and its delayed count confirmation. Fixing the egg list alone has not been tested as a planner improvement. No new planner policy is installed or launched here. Diamond mechanics and values remain separate and unverified; E005 scoring and the numeric 304,000 target remain gated. Original N1 preregistration still has priority; N1b is not adopted.
Evidence
- Full E004 report and six-objective table
- Frozen diagnostic protocol
- Revision 2 objective census, geometry and proxy data
- Two fixed native-speed replay receipts
- Event inputs, lane and causal timeline
- Read-only scan of existing branches
- E003 scientific interpretation
- Updated score-target provenance
- Video publication receipt
- Project Pengo and E003 incomplete outcome
Local MAME 0.289, pengoja, 3 lives, Medium, Upright, bonus 30,000, rack test Off, DSW1 176, DSW2 204 on doppia-3090. Historical Pleiads and frozen Pengo sources/results remain preserved. Only selected editorial evidence, derived observations and hashes are published; ROMs, raw RAM/state dumps, extracted templates and secrets remain excluded.
Progress diagnostics6October2026
E005 is now Objective Reduction Planner, targeting a healthy ACT3 prefix6 to0 without optimizing score or time. The earlier unexecuted scoring assignment is superseded and preserved as history.9,790 remains best verified planner trajectory,9,090 the incomplete E003 observation; no new full-game record or completed ACT3 is claimed.
The first E015 native latency batch shows97–121physical frames across14 complete attributable events after strict deduplication.117 is not constant. A separately frozen32-recipe timing extension is in progress; the20-event gate is not yet passed in the published first batch. E016 matched egg controls distinguish hatching, which preserves objectives, from two direct ACT3 removals with immediate official decrements. Later death during one neutral wait remains visible. E014 reproduces a skipped cell target; E013 retains the earlier failed routes.
No scoring search or N1b is launched. Egg display/score timing, residual cell control and event acknowledgement remain separate validation work. Diamonds are prerequisites for future scoring, not a presumed dependency of this non-scoring E005.
Review correction6October2026
The causal interpretation above is a hypothesis. The next controlled trial is E020, E003 with only the egg reader changed, preserving the original controller, heuristic, seed and budgets. The redesigned E005 is deferred. No new scoring or autonomous-agent result is claimed. Future stagnation is NUL for its registered configuration; infrastructure failures are separately incomplete. Frozen historical results retain their original labels.
Y1–Y7 response and exact mismatch explanation. The11E003 and7E009 original arithmetic discrepancies are explained by egg transport and one retirement/hatch boundary. This compares RAM-derived counts with a counter in the same RAM; it is not independent pixel-reader accuracy. Per-act initial counts measured so far are6,6,8 forACT1,2,3; ACT4+ remains UNKNOWN despite the finite Rack diagnostic. The new reader preserves two additional E009 header-update UNKNOWNs.
Current9,790 editorial v2 already applies W1–W6; the originals remain unchanged and superseded.9,390last alive /9,790tape end, death7299, ACT3incomplete.