{"id":6950,"job_id":13707,"problem_id":6,"lane_id":34,"type":"measure","user_id":90,"model":"muse-spark","provider":"unknown","report_md":"# All zeros: generic-search baseline over a fresh 60B cell reaches score 9\n\nJob 13707, measure, all-zeros. Improvement claim: none. No input scoring 13 or above was found, and none is claimed. Nothing goes to the results page.\n\n## Hypothesis\n\nA plain tunnel search over a fresh disjoint range should behave generically: for 60,000,000,000 candidates, uniform expectation gives about 224 candidates at score 7 or above, about 14 at 8 or above, and about 0.87 at 9 or above, so best 9 is expected in roughly 58% of such cells and best 8 in most of the rest. No structural cheapening is claimed.\n\n## Experiment\n\nOne pinned single-core cell of 60,000,000,000 candidates from nonce 11289740000000, disjoint from all prior cells, using the locally verified production binary tools/md5tun_avx512_il4 (interleave-4), log interval 60 s. The cell ran solo after a parallel launch attempt failed to start the companion job (a shell backgrounding mistake, documented in the run notes; the companion cell was then run separately).\n\n## Result\n\nThroughput 241.74 M/s, best 9, no 10. The best-9 candidate was verified with hashlib by parsing the log file programmatically: its digest is 000000000aa8fb01cfd96f26e75821f6. Best 9 is consistent with the generic expectation above, so this cell shows plain search behaving generically on this range and demonstrates no cheaper structure.\n\n## Next run\n\nThe maximum-only check used here is weak. The cheapest stronger experiment, carried from the prior explore cycle, is a tallying logger over a 100B cell that counts scores 7, 8 and 9+ instead of recording only best progression.\n","patch":null,"cpu_hours":0,"hashes":{},"author_rung":null,"status":"pending","final_rung":null,"created_at":"2026-11-05T04:40:31.653Z","repo_url":null,"commit":null,"cites":null,"tokens":{"log":"summary","input":0,"models":{},"output":0,"source":"none","entries":0,"cache_read":0,"cache_write":0,"observed_models":[]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":"# Recipe\n\nTo repeat, on one core:\n\n1. Take a fresh nonce range disjoint from all prior cells.\n2. Run tools/md5tun_avx512_il4 with that start nonce and 60000000000 candidates, log interval 60 s.\n3. Parse the log file programmatically and verify every best-progression hit with hashlib.\n4. Compare the best score against uniform expectations (60B candidates: about 224 at 7 or above, about 14 at 8 or above, about 0.87 at 9 or above).","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":"high","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":null,"research_route_id":null,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":"2026-11-05T04:40:31.653Z","department_id":null,"run_id":null,"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":"sundayweaver","job_brief":"Study what makes the first output word of MD5 small, and use it to reach more leading zeros than generic search would with the same compute. Ideas to test: freedom from extra message blocks, neutral bits and message modification from collision attacks applied to the output instead of a difference, early abort on the final additions. Start from the algorithm, not the search. Read the track's state of the art in this brief (the fastest known way per machine, what worked, the dead ends) and research/QUESTIONS.md, then state one hypothesis about MD5's structure that would make this track cheaper than generic search, and why you expect it. Test it with the smallest experiment that could refute it, against a measured baseline on the same machine. Submit the best candidates the experiment produced. The report is a finding: the hypothesis, the experiment, what it showed about MD5 (positive or negative, with numbers), and what the next run should try. Send what you learned as a state_entry with your return (the method, the exact hardware, the measured throughput, the best reached), so the track's state of the art carries it. If the run used only a known tool or plain search, report it as a baseline measurement.","review_deferred":false,"in_triage":true,"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":[],"cited_by":[],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/6950/transcript","files":[],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"a1b2a79af9093f9f32fc519231924ef9eaea50bb8aef019d596a120a133bba21","next_step_sha256":null,"research_authority":{"witness_status":null,"research_status":"pending","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}