Following on Hordijk’s fluctuation modulation, here’s a demo that I saw a while ago that’s been popping into my head, on how to program the lead sound for Vangelis’ Blade Runner theme. There’s a section on vibrato that starts around 4′23″.

I also took another look at the effects of uni and bipolar waveforms on linear FM in Robert Syrett’s Audulus Bahama oscillators – unipolar FM alters the perceived pitch (which makes sense for audible frequencies ¹ since the modulation is only in an upward direction) while bipolar waveforms keep it centered. I noticed that the Moog Model 15 App has AC and DC frequency modulation inputs as well. In that case it’s simply a question of how strong the (exponential) effect is.

I was also curious to take a look at how the Model 15 App handles sync. Hard sync is as one would expect, but the weak (soft) sync is quite sublte and beautiful in the way that it gradually matches up the phase of the waveform while attempting to preserve its shape.

And finally a look at the Make Noise DPO which very much takes both linear and exponential FM into timbral texture territory. Robert Syrett has made an Audulus version of that as well.

Phase Art: The picture above is of the Wikipedia 3:2 phase modulation example ¹ and corresponding comparison with frequency modulation as displayed by an oscilloscope in X-Y mode.² This is a mode typically used for testing phase relationships between signals and is well known as the method for displaying Lissajous curves – a sine wave on the X axis and a cosine wave on the Y creates a perfect circle, for example.

What I hadn’t realized is that a phase modulated signal against the same (unmodulated) reference frequency could create the kinds of shapes above. (The frequency modulated signal against the same unmodulated reference frequency produces a rotating oval.³) In this case the phase and (linear) frequency modulated signals are presented simultaneously, resulting in the bands of lines that look something like a staff in music notation.

Further down the phase/frequency modulation rabbit hole: I’ve played around a little more with the PM vs FM patch (uploaded to the Audulus forum) and got the levels set to recreate the Wikipedia phase modulation gif. The Audulus LFO waveforms are unipolar and I experimented at little, comparing the results with using a bipolar modulating waveform on the linear¹ and exponential FM inputs.

I’ve also been looking at how the shapes change depending on whether the modulating frequency is higher or lower than the one it is modulating. The characteristic phase shapes (with amplitude sub-peaks within them) appear when the modulating wave is higher than the carrier, having a much stronger effect in this area than linear or exponential frequency modulation.

It’s interesting how one can keep circling around a topic, slowly covering all it’s aspects and letting them sink in.

I’ve been taking a look at frequency modulation again – specifically the difference between phase and frequency modulation that I touched on in a previous post. Phase modulation (as used in the Yamaha DX synthesizers) has the same (linear) effect as linear frequency modulation, but while mathematically equivalent, the two sound very different. I came across a thread in the Audulus forum that I’d somehow overlooked before, in which Robert Syrett provides an example patch that helps clarify the differences.

Here’s my own slightly altered version of his patch that I’ve been playing around with. I haven’t quite managed to match the nice phase gif from the Wikipedia phase modulation article, but that can be explored further on another day.