Block Gain
Fully SIMD-optimized gain plugin: block-rate processing plus a vectorized envelope precompute for the slow-path smoothing case.
Source: examples/truce-example-block-gain/.
#What it demonstrates
- Fast path on converged smoothers (
is_smoothing() == false) vs. slow path during active smoothing. truce_simd::math::db_to_linear_blockfor the slow-path envelope precompute. With many smoothed gain knobs in flight, this closes the gap between block-rate and the pre-vectorization scalar baseline.- Trade-off: one extra
[f32; MAX_BLOCK]stack scratch and one extra pass over the envelope. Net win measurable when N smoothers ≥ ~4.
#Parameters
| Name | Range | Unit |
|---|---|---|
| Gain | -60 to +6 | dB |
| Pan | -1 to +1 | pan |
#Build and test
cargo build -p truce-example-block-gain
cargo test -p truce-example-block-gain
cargo truce install -p truce-example-block-gain