FIR high-pass digital filtering
Ilustrates WaveWarp's FIR filtering (in this case, a high-pass design). The filter is designed "on-the-fly" by a
variety of methods or can be read in from ascii files (e.g. generated by MATLAB).
The direct implementation in this example is fine for filter lengths below 100 or so. However, for higher filter lengths, the
direct implementation is computationally inefficient. To appreciate this, try reducing the stop gain to -200 dB and making
the filter transitions very narrow. The filter design will work but the audio output will be "jumpy" because the CPU will be
exhausted. For long filters, the fast convolution (based on the Fast Fourier Transform) must be used. WaveWarp has a
complete set of FIR filters implemented both directly and by Fast Convolution. [See the "FIRfilteringFastExample2.dwb"
DrawingBoard for an example of a high order high-pass filter implemented via the FFT.]
