testing-dsp-analysis

name	testing-dsp-analysis
description	Deep DSP-verification techniques — FFT-based aliasing measurement, THD/SNR and harmonic analysis, click/LPC/spectral-anomaly artifact detection, frequency estimation, and deterministic test-signal generation. Use ONLY when verifying the numerical/spectral correctness of a DSP algorithm (waveshaper aliasing, oscillator anti-aliasing, saturation THD, click/discontinuity hunting, golden spectral references). For general test writing, Catch2 patterns, build-before-test, integration, and VST3 validation, use the broader testing-guide skill instead.
allowed-tools	Read, Grep, Glob, Bash

DSP Analysis Testing

The narrow, deep companion to testing-guide. testing-guide fires broadly for any test work; this skill fires only when you are measuring DSP correctness in the frequency/statistical domain — the work that needs an FFT, a THD figure, or an artifact detector rather than a REQUIRE.

Reach for this when the question is one of:

• Does this waveshaper/oscillator alias? → FFT the output, measure energy above Nyquist-image bands.
• How much harmonic distortion does this saturator add? → THD / SNR measurement.
• Did this change introduce clicks or discontinuities? → click/LPC/spectral-anomaly detectors.
• Is the spectral output still golden? → reference spectra + A/B comparison.
• What's the true fundamental of this output? → frequency estimation.

Contents

• DSP-TESTING.md — test signals, THD measurement, frequency estimation, deterministic RNG
• SPECTRAL-ANALYSIS.md — FFT-based aliasing measurement for waveshaper and ADAA tests
• ARTIFACT-DETECTION.md — click detection, LPC analysis, spectral flatness, signal-quality metrics, parameter-sweep testing

Helpers these techniques use

In tests/test_helpers/:

• spectral_analysis.h — FFT-based aliasing measurement
• signal_metrics.h — SNR, THD, crest factor, kurtosis, ZCR
• artifact_detection.h — ClickDetector, LPCDetector, SpectralAnomalyDetector
• statistical_utils.h — mean, stddev, median, MAD
• golden_reference.h — reference comparison, A/B testing
• parameter_sweep.h — automated parameter-range testing

Non-negotiable

• Assert the timbre/spectrum you expect, not merely that output changed. A "the signal is different now" check passes even when the change is a regression. Name the band, the centroid, the THD threshold. (See testing-guide/ANTI-PATTERNS.md.)
• Never REQUIRE/INFO inside a sample loop — collect the metric across the buffer, assert once.
• Cross-platform tolerances: MSVC/Clang diverge at the 7th–8th decimal — use Approx().margin(), never exact equality, on spectral magnitudes.