[sdiy] Polybleps and arbitrary waveforms

[Richie Burnett]

Yes, provided that you are happy with the waveform being played back in a stepped zero-order-hold fashion.  So there will be considerable imaging distortion for low pitches, due to the steps.

The bleps band limit the energy from the steps to the nyquist frequency at the system's output sample rate, not the playback rate of the wavetable. They don't interpolate the wavetable waveform in the way that we would normally want to produce a single band-limited instance of the waveform's spectrum when played back. So you get multiple images of the wavetable waveform's spectrum repeating over between DC and Fs/2 for low pitched notes.

If you want gritty low-Fi PPG type wavetable sounds with lots of imaging distortion but without any aliasing distortion this is how to do it.

In practice though it becomes very computationally intensive for high pitches where you will have steps due to many wavetable pointer increments between two adjacent output samples, and therefore many overlapping step corrections to compute.

It's likely more computationally efficient to use another band limiting method for notes that are pitched up at playback, and your method for waveforms played back at lower pitches where you want that authentic sounding gritty imaging distortion. 

Hope this helps, 

-Richie,
Sent from my Xperia SP on O2

---- Gordonjcp wrote ----

>I was thinking about polybleps and it occurred to me - if you could compute the polyblep fast enough, could you do that for every sample read from a wavetable and thus antialias any wave you play back?
>
>-- 
>Gordonjcp MM0YEQ
>
>_______________________________________________
>Synth-diy mailing list
>Synth-diy at dropmix.xs4all.nl
>http://dropmix.xs4all.nl/mailman/listinfo/synth-diy