... and that they have equal width on the linear scale
... also simplify the logic involving zero padding factor
This is more preparation for constant-q.
I confirmed that the currently used real FFT code in RealFFTf.cpp is faster
than the old one with a quick benchmark that calls PowerSpectrum() on 4-minute
audio file, with different sizes of computation windows:
Window_size: 256 method: new FFT time_s: 0.393
Window_size: 256 method: old FFT time_s: 1.065
Window_size: 1024 method: new FFT time_s: 0.38
Window_size: 1024 method: old FFT time_s: 0.958
Window_size: 4096 method: new FFT time_s: 0.413
Window_size: 4096 method: old FFT time_s: 1.084
Window_size: 16384 method: new FFT time_s: 0.518
Window_size: 16384 method: old FFT time_s: 1.338
Window_size: 65536 method: new FFT time_s: 0.655
Window_size: 65536 method: old FFT time_s: 1.524
Window_size: 262144 method: new FFT time_s: 0.735
Window_size: 262144 method: old FFT time_s: 1.873
... also make WaveTrack responsible for storing and validating the bounds
... also let the bounds vary per-track even though other settings are default
... also change some code names to mention "period" not "undertone"
... not per track,
and the preferences or View Settings page has a separate static box for
global settings as opposed to track settings. This is the only global setting
for now.
... SpectrogramSettings does that instead, and Preferences or View Settings
are the user interface for changing it.
Handle invalidation of spectrogram pixel cache for scale type changes,
just as for other changes of settings. No more
TrackArtist::InvalidateSpectrumCache().
View type of track now switches to Spectrum when applying or OKing the
View Settings... dialog and the Spectrogram page is open (and for now
it is still the only page)
... writes global preferences explicitly, and only when it is the default
settings object.
Also impose validation of settings when constructing from preferences.
