Fault Line: bass and guitar phrasing VII

David liked the Style VI bass in isolation but found it disconnected from the rhythm guitar in the mix. This pass changes only the bass performance. Every patch, instrument level, drum part, guitar part and lead bend stays as in VI. The moving entrance and finite 49.524-second layout remain intact.

The previous bass often attacked during guitar releases and missed the open chord accents. Its repeating two-bar shape also ignored the guitar's different fourth- and eighth-bar replies. Simplifying a line alone did not make it support the arrangement. This revision follows the actual guitar phrase variations.

• The call shares attacks at rows 0, 10, 16 and 28, with the same gate lengths. The guitar alone plays its quick row-6 strike and third/fourth decorations.
• The reply shares rows 0, 12, 18, 26 and 30. Its third follows the guitar's third instead of landing a root or fifth against that note.
• The more spacious fourth-bar reply shares rows 0, 10, 18 and 28. A quiet fifth at row 24 answers inside the rest, then resolves with the guitar at row 28.
• The eighth-bar turnaround follows the guitar's D pickups. The moving entrance shares its A, C, E and D accents without adding an independent running line.

The bass has 100 attacks rather than 90. Of those, 97 share guitar onsets, compared with 47 of the previous 90. The three independent notes are deliberately quiet fifth responses in longer gaps. Every shared attack agrees with the guitar pitch one octave below; gates in the revised sections end together. This is a rhythmic foundation with selective responses, rather than a duplicate of every guitar strike. These checks establish the intended relationship, not whether the musical result works for the listener.

Listening and reproduction

The full demo is 49.524 seconds. The comparison plays VI first and VII from 0:19.048. Each half covers four pocket bars, the two-bar moving entrance, two solo bars and two release bars. Both halves use the same delivery gain.

Run from the album worktree:

sigil docs/music/tools/fault-line-bass-lock.sgl \
  --motif "$PWD/build/dev/bin/motif" --output /tmp/fault-line-vii-replay
sigil docs/music/tools/fault-line-bass-lock-audit.sgl \
  --motif "$PWD/build/dev/bin/motif" --output /absolute/external/artifacts
sigil docs/music/tools/expression-render.sgl --stage native \
  --input "$PWD/docs/music/alternates/fault-line-bass-lock-vii" \
  --motif "$PWD/build/dev/bin/motif" --output /absolute/fresh/artifacts
sigil docs/music/tools/expression-render.sgl --stage delivery \
  --input "$PWD/docs/music/alternates/fault-line-bass-lock-vii" \
  --gain-db -0.9 --output /absolute/fresh/artifacts

The audit compares all seven non-bass compiled tick streams, the complete instrument bank and the bus. It checks the increased proportion of shared attacks, their octave relationship and matching note-offs in the changed sections. The finite-score audit checks the playback clock and released gates.

Renders, source snapshots, command arrays, renderer identity and measurements: ~/artifacts/crash-the-stack-album/fault-line-bass-lock-vii-delivery/. This is still an audition; maintained game/album arrangements and public files are unchanged.

Validation completed: all seven generated source/map/manifest files reproduce exactly; all fifteen maintained songs regenerate. The finite audit reports 719 attacks, zero sample-clock error and all final gates released.

For native render headroom, BOTH references use compressor makeup 0 dB instead of VI's +1 dB. This is common output attenuation; compressor threshold, ratio, timing, patches and instrument gains are unchanged. The final chord reached full scale in the first native preflight. An interim bass velocity reduction did not fix it and was reverted. Final bass velocities are those of the intended VII arrangement. The two rejected preflights remain in the external fault-line-bass-lock-vii/ and fault-line-bass-lock-vii-final/ folders; only fault-line-bass-lock-vii-delivery/ is delivered.

Final delivery

Source commit: bc75f26. Native full-after peak: -0.9 dBTP. The common export gain was capped to -1.1 dB from the requested -0.9, in addition to the common 1 dB native attenuation. Decoded OGG: A/B -19.9 LUFS-I / -3.7 dBTP; before full -20.0 / -2.2; revised full -19.9 / -2.0. Duration, decoded peak and final release checks pass. No additional limiter or per-version normalization.

Telegram confirmed the comparison as message 1190 and the complete revised demo as message 1191. Encoded hashes are in encoded.sha256. The previous full reference remains available locally. These are listening copies, not masters.