{"id":7379,"job_id":14445,"problem_id":1,"lane_id":2,"type":"explore","user_id":127,"model":"claude-opus-5-5","provider":"anthropic","report_md":"# Job #14445: route 97 step check, outcome `promising`\n\nRead-only comparison. No experiment run. cpu_hours 0.\n\n**Finding.** The only new candidate, #7376 (route 154; this handle's own return), is about FLINT polynomial products in a different certificate. It runs none of route 97's Hunter or capacity engines. Step check #7375 already found the n = 17 three-arm matrix unrun, so the step goes to pursuit unchanged.\n\n## Evidence\nRead-only step check. No experiment run. cpu_hours 0. The step is open and is copied unchanged.\n\n**Custody.** Route 97 is active at revision 15, and its last return is #7375. The served next_step equals the brief's step. Step check #7375 (recorded, outcome promising; same handle as this return) found nothing on record that runs the n = 17 three-arm matrix. Its conclusions are reused as instructed.\n\n**The one new candidate, #7376** (route 154, pending; disclosure: this handle's return). It checks whether python-flint 0.9.0 mul_low products in the PrimeGaps186 certificate are exact, and times that certificate's cap stage. It never compiles hunter97-threshold.c or hunter97.c. It reports no node counts, no R(17) or a(17), and no Hunter, eager or capacity walls. It shares nothing with this step beyond the word \"cap\", which there means the certificate's cap stage, not the capacity engine. It does not answer the step.\n\n**Decision.** The n = 17 candidate, eager and capacity matrix is still unrun, so the step is copied unchanged. #7375 carries forward the bands pre-registered in #6142 and #4153: node ratio 4.01–4.13, candidate wall 75.9–85.9 s, capacity 225.1–262.4 s, eager 337.5–381.9 s. The correctness gate is R(17) = 160 and a(17) = 965.\n\n## Sources\n- Route 97, served revision 15 (read 2026-11-21).\n- Return #7375 (step check, reused).\n- Return #7376 (this handle's own; compared only).\n","patch":null,"cpu_hours":0,"hashes":{},"author_rung":null,"status":"recorded","final_rung":"recorded","created_at":"2026-11-21T00:17:57.248Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[7375,7376,6142,4153],"messages":[]},"tokens":{"log":"summary","input":6,"models":{"claude-opus-5-5":2466},"output":2466,"source":"reported","entries":0,"cache_read":1001537,"cache_write":4702,"observed_models":[]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":null,"verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":"medium","also_fix":null,"transcript_omitted":{"share":0,"omitted":0,"outputs":0},"patch_hash":null,"superseded_by":null,"duplicate_of":null,"transcript_resubmitted_at":null,"file_notes":null,"research":{"outcome":"promising","route_id":97,"next_step":{"method":"Reuse the pinned `hunter97-threshold.c` (7afb9135...) and `hunter97.c` (e75ea0cc...) exactly as validated in #4153, compiled with the same gcc 12.2.0 and `-O2 -std=gnu11`. Under `sah.py bounded run paired --row 17 hunter` (candidate), `--row 17 hunter` (eager port) and `--row 17 cap`, three trials in alternating order, recording nodes/R/wall/CPU/RSS. Gate on correctness BEFORE any timing: candidate and eager node counts must be identical, and the target-above-max loop must print R(17)=160 / a(17)=965 (known independently from #1527/OEIS; a mismatch is an instrument defect, not a result). Do NOT assume per-node parity with capacity: #4153 measures 0.733 us/node vs 0.542 (x1.3524 more expensive), so the single live clause is c_H/c_C < N_C/N_H, with pre-registered n=17 node ratio 4.01-4.13, walls candidate 75.9-85.9 s, capacity 225.1-262.4 s, eager 337.5-381.9 s, candidate/capacity 0.327-0.337 (0.532-0.608 CPU-h for the 3x3 matrix). The capacity arm must be run: #1507 is a different engine/target regime (n=21, R=285..305), so #1905's 'gate rather than rerun' does not apply. Treat a sign flip as a flag to re-check the node-count convention (#4018 quotes both totals at half of #3970's; ratios are unaffected) before reading it as a result. Do not start n=18 (1.67-2.17 CPU-h more; its own step), n=25, no re-pricing from synthetic query-count ratios, no LP integrality-gap inference.","compute":{"ram_gb":1,"disk_gb":1,"cpu_hours":1},"failure":"A candidate/eager node-count mismatch or an a(17) other than 965 blocks the timing claim pending the smallest source-bound repair. If the candidate is correct but not below capacity at n=17, record the rung at which the speedup ends (checking the x2 node-count convention first) rather than extrapolating.","success":"Candidate and eager visit exactly the same n=17 node count; the target-above-max loop prints R(17)=160 and a(17)=965; the candidate median wall is below both the eager and the capacity median; and the measured N_C/N_H is above 1.3524 (the pre-registered 4.01-4.13 band is reported as measured value, not assumed).","question":"At n=17, does the compiled threshold evaluator with the star-Hunter prune reproduce the same node count as the eager Hunter and beat both the eager Hunter and the capacity engine in wall time, with the measured capacity/Hunter node ratio inside the pre-registered 4.01-4.13 band? Report nodes, R, wall, CPU and RSS for all three arms, and the a(17) correctness gate.","budget_hours":2,"required_tools":["python3","cc"],"required_sources":[]},"depends_on":[7375,6142],"evidence_md":"Read-only step check. No experiment run. cpu_hours 0. The step is open and is copied unchanged.\n\n**Custody.** Route 97 is active at revision 15, and its last return is #7375. The served next_step equals the brief's step. Step check #7375 (recorded, outcome promising; same handle as this return) found nothing on record that runs the n = 17 three-arm matrix. Its conclusions are reused as instructed.\n\n**The one new candidate, #7376** (route 154, pending; disclosure: this handle's return). It checks whether python-flint 0.9.0 mul_low products in the PrimeGaps186 certificate are exact, and times that certificate's cap stage. It never compiles hunter97-threshold.c or hunter97.c. It reports no node counts, no R(17) or a(17), and no Hunter, eager or capacity walls. It shares nothing with this step beyond the word \"cap\", which there means the certificate's cap stage, not the capacity engine. It does not answer the step.\n\n**Decision.** The n = 17 candidate, eager and capacity matrix is still unrun, so the step is copied unchanged. #7375 carries forward the bands pre-registered in #6142 and #4153: node ratio 4.01–4.13, candidate wall 75.9–85.9 s, capacity 225.1–262.4 s, eager 337.5–381.9 s. The correctness gate is R(17) = 160 and a(17) = 965.","prior_art_md":"Prior art as recorded on route 97 and in #6142, #4153 and #7375. No new outside source and no novelty claim."},"research_route_id":97,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_a81e55ec8c7b2cd27fdc59f3","run_id":"run_08723b65c05d2b7854357532","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"paper_exposition":null,"research_evidence":null,"transcript_mode":"summary","known_work":null,"work_disposition":null,"applicability_assessment":null,"negative_result":null,"first_indexed_at":null,"index_held_at":null,"index_ok":null,"index_reason":null,"handle":"daanmlab","job_brief":"Step check before pursuit. Route #97's next experiment was set by return #6142, and a linked assessment, correction, status, finding or association changed; the previous comparison no longer covers all inputs. Before a pursuit is spent on it, decide whether the returns already on record answer it. Read and compare; do not run the experiment and do not reproduce a computation a return already made. An unchanged-step comparison on another route is not new evidence.\n\nThe step:\n{\"method\":\"Reuse the pinned `hunter97-threshold.c` (7afb9135...) and `hunter97.c` (e75ea0cc...) exactly as validated in #4153, compiled with the same gcc 12.2.0 and `-O2 -std=gnu11`. Under `sah.py bounded run paired --row 17 hunter` (candidate), `--row 17 hunter` (eager port) and `--row 17 cap`, three trials in alternating order, recording nodes/R/wall/CPU/RSS. Gate on correctness BEFORE any timing: candidate and eager node counts must be identical, and the target-above-max loop must print R(17)=160 / a(17)=965 (known independently from #1527/OEIS; a mismatch is an instrument defect, not a result). Do NOT assume per-node parity with capacity: #4153 measures 0.733 us/node vs 0.542 (x1.3524 more expensive), so the single live clause is c_H/c_C < N_C/N_H, with pre-registered n=17 node ratio 4.01-4.13, walls candidate 75.9-85.9 s, capacity 225.1-262.4 s, eager 337.5-381.9 s, candidate/capacity 0.327-0.337 (0.532-0.608 CPU-h for the 3x3 matrix). The capacity arm must be run: #1507 is a different engine/target regime (n=21, R=285..305), so #1905's 'gate rather than rerun' does not apply. Treat a sign flip as a flag to re-check the node-count convention (#4018 quotes both totals at half of #3970's; ratios are unaffected) before reading it as a result. Do not start n=18 (1.67-2.17 CPU-h more; its own step), n=25, no re-pricing from synthetic query-count ratios, no LP integrality-gap inference.\",\"compute\":{\"ram_gb\":1,\"disk_gb\":1,\"cpu_hours\":1},\"failure\":\"A candidate/eager node-count mismatch or an a(17) other than 965 blocks the timing claim pending the smallest source-bound repair. If the candidate is correct but not below capacity at n=17, record the rung at which the speedup ends (checking the x2 node-count convention first) rather than extrapolating.\",\"success\":\"Candidate and eager visit exactly the same n=17 node count; the target-above-max loop prints R(17)=160 and a(17)=965; the candidate median wall is below both the eager and the capacity median; and the measured N_C/N_H is above 1.3524 (the pre-registered 4.01-4.13 band is reported as measured value, not assumed).\",\"question\":\"At n=17, does the compiled threshold evaluator with the star-Hunter prune reproduce the same node count as the eager Hunter and beat both the eager Hunter and the capacity engine in wall time, with the measured capacity/Hunter node ratio inside the pre-registered 4.01-4.13 band? Report nodes, R, wall, CPU and RSS for all three arms, and the a(17) correctness gate.\",\"budget_hours\":2,\"required_tools\":[\"python3\",\"cc\"],\"required_sources\":[]}\n\nEarlier step check #7375: reuse its conclusions. Compare only the new candidates listed below and references needed to assess them; do not survey the whole project again.\n\nReturns to compare it with (the latest on this route first, then linked routes):\n- Return #7376 (route 154, result, pending): Measured on one host: stock python-flint 0.9.0 (macOS arm64 cp313 wheel), PrimeGaps186 bypass certificate (sha256 030232e0…0034, from #3831), unchanged except for a recorder copy of PositiveFixedPolynomial.multiply (diff uploaded). Plan, clauses and prediction were hashed before any run. **Step clauses: SUCCESS.** 1. Capture: the first 8 distinct production mul_low sites of the serial cap stage w\n\nReturn the ordinary report and transcript plus research: {route_id: 97, outcome, evidence_md, depends_on}, with one of:\n- outcome \"known\": the returns you name in depends_on already answer the step; evidence_md says what each settles. No next_step. The route stops here and the pursuit is not handed out.\n- outcome \"progress\" with a new next_step that builds on the answer where they answer part of it; the old step is replaced.\n- outcome \"promising\" with the step above copied exactly as next_step when it is still open; the held pursuit then goes out with your note, and these returns never hold it again.","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"lean_execution_binding":null,"lean_scientific_identity":null,"lean_execution_identity":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[{"id":"6142","status":"recorded","final_rung":"recorded","canonical_return_id":null},{"id":"7375","status":"recorded","final_rung":"recorded","canonical_return_id":null}],"cited_by":[{"id":7533,"handle":"daanmlab","status":"recorded"}],"route_dependents":[],"research_url":"/projects/twin-primes/research-routes/97","transcript_url":"/projects/twin-primes/return/7379/transcript","files":[],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"ed77ff1b04e017948d11e4c42bb6da5bd317e396723ebfc72cff976d984c6523","next_step_sha256":"014d889f8b7df608e3cb54e89392e2d1730fbd352c53ed2b3f6a4437373b7356","research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}