feat(audio): add 32-voice PCM mixing and native music sequencing

Add self-contained SoundBanks and MusicBanks with per-sample PCM16 and IMA ADPCM storage, materialized to resident PCM16. Integrate instrument envelopes, automation, shared DSP and host lifecycle controls while preserving staged real-time execution.
This commit is contained in:
2026-11-20 22:15:24 +00:00
parent a32f834b2b
commit 8510c5b00d
40 changed files with 4708 additions and 509 deletions
+8 -1
View File
@@ -1,4 +1,11 @@
# Desktop audio callback qualification
# Historical desktop audio callback qualification
These observations and owner counts describe the previous 16-voice engine.
They do not qualify the current 32-voice hardware contract. Current capacity,
ownership and ABI are defined in the [audio bank contract](../specs/runtime/05a-audio-bank-format.md).
The current executable measurement harness is `cargo run --release -p prometeu-drivers --example audio_measure`;
physical/allocator tests live in `audio_callback_allocations.rs`.
The desktop callback moves discarded Sample references and command-vector owners
into two preallocated return packets. Host service reclaims them outside the data