docs: harvest physical-room row and sources for the 41 s and degraded-score claims #160

Merged
void merged 1 commit from void/harvest-physical-room into dev 2026-11-14 01:54:38 +00:00
Member

Follow-up ratatoskr:follow_up:chat-s7618-7618. Docs only, Mimir read at c6f4541.

  1. Polarity row: shotgun mic 37/37 removed; the row keeps only what L391-395 records (negative correlation on both channels).
  2. New row, physical room run (contestants doc L371-388): co-streamer shotgun mic (ch1) locked 37/37 with per-channel offsets, payload 1.000, timing 0.828; cardioid/near (ch0) reached 26/37, payload 0.595. Confidence B, one capture.
  3. Sources searched in Mimir's docs, research, calibration, artifacts and MimirBioacoustic*.cs:
    • Degraded identity 0.5-0.9 / timing 0.19-0.73: has a receipt. artifacts/bioacoustic-training/bioacoustic-20260524-150729/compact-fast-index/{blur-light,warp-light,warp-light-blur,warp-heavy-blur}/summary.json: IdentityScore 0.532-0.908, TimingScore 0.187-0.728. Cited in row 4. It holds for compact-fast-index only; other hypotheses in that run range wider (identity to 0.998, timing to 0.807).
    • Event slot repeats about every 41 s: no receipt or doc states it; derived from code. SymbolForEvent cycles (eventIndex / 2) % 128 words and alternates speaker, so the symbol sequence repeats every 256 events; EventSpacingSeconds is 0.16, so 40.96 s (MimirBioacousticTimeline.cs L34-39, L735-739). Row 3 now says "derived, not reported".

🤖 Generated with Claude Code

Follow-up `ratatoskr:follow_up:chat-s7618-7618`. Docs only, Mimir read at c6f4541. 1. Polarity row: `shotgun mic 37/37` removed; the row keeps only what L391-395 records (negative correlation on both channels). 2. New row, physical room run (contestants doc L371-388): co-streamer shotgun mic (ch1) locked 37/37 with per-channel offsets, payload 1.000, timing 0.828; cardioid/near (ch0) reached 26/37, payload 0.595. Confidence B, one capture. 3. Sources searched in Mimir's docs, research, calibration, artifacts and `MimirBioacoustic*.cs`: - **Degraded identity 0.5-0.9 / timing 0.19-0.73: has a receipt.** `artifacts/bioacoustic-training/bioacoustic-20260524-150729/compact-fast-index/{blur-light,warp-light,warp-light-blur,warp-heavy-blur}/summary.json`: `IdentityScore` 0.532-0.908, `TimingScore` 0.187-0.728. Cited in row 4. It holds for compact-fast-index only; other hypotheses in that run range wider (identity to 0.998, timing to 0.807). - **Event slot repeats about every 41 s: no receipt or doc states it; derived from code.** `SymbolForEvent` cycles `(eventIndex / 2) % 128` words and alternates speaker, so the symbol sequence repeats every 256 events; `EventSpacingSeconds` is 0.16, so 40.96 s (`MimirBioacousticTimeline.cs` L34-39, L735-739). Row 3 now says "derived, not reported". 🤖 Generated with [Claude Code](https://claude.com/claude-code)
docs: harvest polarity row loses the shotgun claim; physical room run row; sources for the 41 s and degraded-score claims
All checks were successful
Build pull request / Build pull request (pull_request) Successful in 4m21s
6b40809f9d
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
void merged commit d5b9e33536 into dev 2026-11-14 01:54:38 +00:00
Sign in to join this conversation.
No reviewers
No labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
GameCult/Ratatoskr!160
No description provided.