PASS Envelope defaults are 0.5 ms / 5 ms
PASS All band faders at MUTE produce exact silence
PASS Band fader applies the requested -6 dB gain
PASS Silent input produces exact silence (no carrier leakage)
  pink octave-energy difference=0.132666 dB
PASS Pink carrier has approximately equal energy per octave
PASS Noise input also excites new noise rather than passing through
  bands=1 RMS=0.108937
PASS All requested band counts produce finite, useful output
  bands=2 RMS=0.104412
PASS All requested band counts produce finite, useful output
  bands=8 RMS=0.0828757
PASS All requested band counts produce finite, useful output
  bands=16 RMS=0.116675
PASS All requested band counts produce finite, useful output
  bands=32 RMS=0.111297
PASS All requested band counts produce finite, useful output
  bands=64 RMS=0.110065
PASS All requested band counts produce finite, useful output
PASS Noise width 0% gives sample-identical mono
PASS Mono input can excite independent stereo noise
PASS Input width 0% uses true M/S mono summing
PASS Input stereo side information drives the analyser
PASS Frequency range rejects far out-of-range input
  coherent tone power fraction=5.56747e-05
PASS A pure tone becomes noise rather than a retained oscillator
PASS Bypass restores the original input exactly
PASS Sample-rate limits and crossed endpoints remain stable
PASS Rapid band/range/width automation and invalid samples stay finite
PASS DSP processing and bank changes allocate no heap memory
64 bands: 589.673 ms for 5333 ms of stereo audio (11.0564% of one realtime core)
PASS 64-band processing is faster than realtime on this test machine
Failures: 0
