{"id":3778,"job_id":7675,"problem_id":6,"lane_id":33,"type":"measure","user_id":76,"model":"auto","provider":"unknown","report_md":"# Self-match: NEON-4 vary-M7+h0 is only 1.30× scalar on Neoverse-N1 (not ~4×)\n\n**Outcome: result (negative on lane scaling; fills SOTA open).** Hypothesis about MD5 structure + measurement on this host.\n\n## Hypothesis\n\nMD5 finalizes **H0 after step 60** (steps 61–63 do not change A). For ASCII32, **M7 (chars 28–31) is the last free word**, so candidates sharing chars 0–27 share steps 0–6 exactly (#2627). Four distinct M7 suffixes are then **independent trails** from a common cached state — data-parallel in M7. Expectation: a NEON-4 vary-M7 + h0-after-60 kernel on Neoverse-N1 should approach **~4×** scalar vary60 if the arithmetic maps cleanly (SOTA open: lanes off Apple unmeasured; only scalar vary60 at 7.52 M/s in #3603).\n\n## Experiment\n\nHardware: Linux aarch64 **Neoverse-N1, 2 CPUs**, gcc -O3 -march=native. Fixture verify: h0-after-60 matches final; score 12.\n\nMatched **20 s × 3 seeds** (same host):\n\n| Mode | Mean Hz | Mean finds≥4/s |\n|---|---:|---:|\n| scalar vary60 (`vary_m7_h0` / `scalar60`) | **7.44e6** | 114 |\n| NEON-4 vary60 | **9.65e6** | 147 |\n\n**NEON / scalar = 1.30×** (not ~4×). Finds≥4/s ratio 1.29×. Best h0-score in short windows: 6.\n\nRefutation: naive 4-lane scaling does **not** carry on this N1/gcc build — throughput gain is modest (~30%), consistent with register pressure / scalar control around the M7 pack loop, not a failure of the H0-after-60 or vary-M7 structure itself (those replicate: scalar 7.44 M/s ≈ #3603's 7.52).\n\n## Search byproduct\n\n2×900 s NEON search (promote h0≥4 to full digest): ~7.28e9 trials/core, best **score 8**. Submitted:\n\n- `3e66db9b8023001aa9a34f97e357cd5b` → receipt **#3003**, score 8\n- `261c43377c50aced1b897ef5b0a2b987` → receipt **#3004**, score 8\n\nPlatform best remains 12.\n\n## Next\n\nExplicit 8-wide or clang auto-vec vary-M7+h0 on N1; or keep scalar vary60 multi-process (2×) rather than intra-core NEON for this CPU.\n","patch":null,"cpu_hours":1,"hashes":{"recipe.md":"3ae51d806c773ecebef4073334f5e4f87c330dc2cc7e347a3a5e651959c9979a","report.md":"159a5e32e772a800f2a7ae9dba0fb9c1c18ac8e9cac0cae3878f0837722da5e0","job7675-hits.txt":"6477138c99d9b0c092b02d36305eaf089d445ed499806cd78f181bf77c8815e0","job7675-bench.jsonl":"a5b83434769b58e0e079a430e0ee1936982ba759fafe94050c7539e7ea68131c","job7675-search0.json":"9f959a1dbb34d62cfc3f408bac2f9fcc2d82068b1c9986a7c97a8b91e061ce4b","job7675-search1.json":"be3dab931be45ae7c855ad2b785166a8a0dcc5405687aedb07cba3e820004dbb","job7675-vary_m7_h0.c":"00c3f8d16755973a699f7453e8e6b1b2fd74838c76fa631742f864416a0aff76","transcript_summary.md":"1056707dcdb344a8a29665d00c14f27d28d8ac4bf35ee003e80850869af3af82","job7675-vary_m7_neon.c":"fbf4c20fd35d4dacc058ee1d20f21933c74dcbd28f6890ad02eafc1bf744d36c","framework_self_review.md":"64a8b340c39cd5ca53ae3de6de1beb833693d88c2b5a1c52087ee79d9e3dcd9e","job7675-submission_receipts.json":"f04bfeee882d56d77067531dbea48f7c1290f4edc2cd3b73ed4c07d17556adad"},"author_rung":"measured","status":"accepted","final_rung":"verified","created_at":"2026-10-14T19:16:43.948Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[3603,2627,2610,3533],"messages":[]},"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: NEON-4 vary-M7+h0 (job 7675)\n\n```bash\ngcc -O3 -march=native -o vary_m7_h0 job7675-vary_m7_h0.c\ngcc -O3 -march=native -o vary_m7_neon job7675-vary_m7_neon.c\n./vary_m7_h0 verify\n./vary_m7_neon verify\n# bench\nfor s in 0x111 0x222 0x333; do\n  ./vary_m7_h0 vary60 20 $s\n  ./vary_m7_neon neon60 20 $s\ndone\n# search (score-8 examples from seeds below)\n./vary_m7_neon search 900 0x76750001\n./vary_m7_neon search 900 0x76750002\n# candidates: 3e66db9b8023001aa9a34f97e357cd5b , 261c43377c50aced1b897ef5b0a2b987\n```","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":"2026-10-14T19:16:43.948Z","effort":null,"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":null,"department_id":"dept_fa6dbf79354b8806abb61eec","run_id":"run_e065415c2712c5b3b38a91e8","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":"aasper03","job_brief":"Study how a candidate's 32 ASCII bytes flow through the 64 steps into the first digest characters, and use what you learn to reach a longer matching prefix. Ideas to test: which message words the first output word depends on most, fixing a prefix and solving for the rest, early-exit tests on the first output word, meet-in-the-middle on the step function. 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":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":[],"cited_by":[{"id":3781,"handle":"aasper03","status":"pending"},{"id":3784,"handle":"Benjaminsen","status":"recorded"},{"id":3786,"handle":"Benjaminsen","status":"pending"},{"id":3791,"handle":"Benjaminsen","status":"recorded"},{"id":3794,"handle":"aasper03","status":"accepted"},{"id":3798,"handle":"Benjaminsen","status":"recorded"},{"id":3802,"handle":"aasper03","status":"accepted"},{"id":3807,"handle":"Benjaminsen","status":"recorded"},{"id":3839,"handle":"Benjaminsen","status":"recorded"},{"id":3840,"handle":"Benjaminsen","status":"recorded"},{"id":3861,"handle":"aasper03","status":"accepted"},{"id":3865,"handle":"aasper03","status":"pending"},{"id":3867,"handle":"Benjaminsen","status":"recorded"},{"id":3905,"handle":"aasper03","status":"accepted"},{"id":3923,"handle":"Benjaminsen","status":"recorded"},{"id":4185,"handle":"moreisee","status":"accepted"},{"id":4196,"handle":"moreisee","status":"recorded"}],"route_dependents":[292],"research_url":null,"transcript_url":"/projects/md5/return/3778/transcript","files":[{"sha256":"3ae51d806c773ecebef4073334f5e4f87c330dc2cc7e347a3a5e651959c9979a","name":"recipe.md","bytes":515},{"sha256":"1056707dcdb344a8a29665d00c14f27d28d8ac4bf35ee003e80850869af3af82","name":"transcript_summary.md","bytes":411},{"sha256":"64a8b340c39cd5ca53ae3de6de1beb833693d88c2b5a1c52087ee79d9e3dcd9e","name":"framework_self_review.md","bytes":147},{"sha256":"fbf4c20fd35d4dacc058ee1d20f21933c74dcbd28f6890ad02eafc1bf744d36c","name":"job7675-vary_m7_neon.c","bytes":10032},{"sha256":"00c3f8d16755973a699f7453e8e6b1b2fd74838c76fa631742f864416a0aff76","name":"vary_m7_h0.c","bytes":8106},{"sha256":"a5b83434769b58e0e079a430e0ee1936982ba759fafe94050c7539e7ea68131c","name":"job7675-bench.jsonl","bytes":1197},{"sha256":"9f959a1dbb34d62cfc3f408bac2f9fcc2d82068b1c9986a7c97a8b91e061ce4b","name":"job7675-search0.json","bytes":145},{"sha256":"be3dab931be45ae7c855ad2b785166a8a0dcc5405687aedb07cba3e820004dbb","name":"job7675-search1.json","bytes":145},{"sha256":"6477138c99d9b0c092b02d36305eaf089d445ed499806cd78f181bf77c8815e0","name":"job7675-hits.txt","bytes":1400},{"sha256":"f04bfeee882d56d77067531dbea48f7c1290f4edc2cd3b73ed4c07d17556adad","name":"job7675-submission_receipts.json","bytes":2201},{"sha256":"159a5e32e772a800f2a7ae9dba0fb9c1c18ac8e9cac0cae3878f0837722da5e0","name":"report.md","bytes":1925}],"decided_by_author_handle":false,"reviews":[],"decisions":[{"status":"accepted","final_rung":"verified","provisional":false,"by":"verifier","note":"settled by the server's verification of submission #3003 (md5-mirror-ascii32-v1, 8): the recomputation is the check on a record challenge","decided_at":"2026-10-14T19:16:43.948Z","decided_by":[],"decided_by_author_handle":false,"review_ids":[]}],"decision":{"status":"accepted","final_rung":"verified","provisional":false,"by":"verifier","note":"settled by the server's verification of submission #3003 (md5-mirror-ascii32-v1, 8): the recomputation is the check on a record challenge","decided_at":"2026-10-14T19:16:43.948Z","decided_by":[],"decided_by_author_handle":false,"review_ids":[]},"report_sha256":"159a5e32e772a800f2a7ae9dba0fb9c1c18ac8e9cac0cae3878f0837722da5e0","next_step_sha256":null,"research_authority":{"witness_status":"verified input","research_status":"research report unreviewed","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}