Harmonic Generating Oscillator
  • That's pretty cool! Thought I'd do a bit of creative phase shifting. I built a little phase-adjustable Phasor because I got tired of working out how to shift phase in Phasor based oscillators. So I'm phase-shifting a phasor to phase-shift the phasor based harmonic oscillator. I guess it's just a phase I'm going through :)
    harmonic scanning phase shifted.audulus
    Screen Shot 2017-11-12 at 11.48.29 AM.png
    2690 x 1466 - 619K
  • @stschoen Set that phasor on stun! That sounds a LOT like a verbos harmonic oscillator. I think you are onto something. Now to get spectral tilt up and running
  • @RobertSyrett Beautiful!

    Not exactly the same thing but gets me thinking of the sweeps at the end of this 1976 demonstration of the EMS Synthi 100: https://twitter.com/BBCArchive/status/927883455714746369
  • @stschoen This is wonderful! @Rudiger I remember reading that there are only 6-7 EMS Synthi 100's left in existence. How fun would that patch matrices be to play with?!
  • P.S. Here's some trivia (no using Google to cheat): Who came up with the theme music for the original Dr. Who? Hint: There's a link to them on this page. lol
  • @Zenji I thought it was Delia Derbyshire, is it really Roger Limb? Well I love all those old BBC Radiophonic Workshop :)

    @Rudiger TY! I loved that video, thanks for the link.
  • @Zjenji I generally like the patch cable approach, but watching the demonstration I could suddenly see the beauty of that matrix :-)

    @RobertSyrett You’re welcome. Interesting how some of the waveforms seem vertically asymmetrical. There was also an interesting link re. the Synthi’s dual output envelopes in the comments on that tweet: https://twitter.com/suitandtieguy/status/928233580706217984

    The dual envelopes are also described in this Synthi 100 manual: http://www.thesynthi.de/data/Synthi_100.pdf
  • @rudiger Let's start en EMS modeling thread
  • @RobertSyrett Yes! I can see you’ve done that already :-)
  • I like your implementation of the smooth mux. I wish I'd known about it when building my xmux for the wavetable nodes!
  • @jjthrash, well that would have been hard because I made that after looking at your xmux and taking inspiration :)
  • @RobertSyrett D'oh! Sorry.. it's been so long, I've just taken to assuming most things have been done. :) Anyway, I like the width factor.. that's a nice touch.
  • Just thought I would share a patch I am working on that started off with the harmonic generator. It combines the tiltatron with a smoothstep demux made from cascaded panning nodes. There is a limit on the shape of the tiltatron so it is more just a simple seesaw and the demux is only ever sending a signal to 2 input tops. So there is a contrast of one harmonic against the coherent sweep of the rest. Would love feedback on how best to turn this into a single module.
    harmonic 16 sketch 2.audulus
    Screen Shot 2017-11-19 at 1.56.52 PM.png
    841 x 1000 - 240K
  • The tiltatron currently folds the waveform when it exceeds 1. I added the mux after the fold but before the even/odd inverter.
    Tiltatron M4.audulus
  • Here's something I've been wanting to do for a while – another little modification to the harmonic oscillator, this time taking it in a different direction so that the frequency of each harmonic can be individually tuned instead of adjusting the phase. The perfect tuning is very beautiful but sometimes slight deviations can add a little movement and character to the sound – also visually – see Jerobeam's video above in this thread from around 5m30.

    Each harmonic can now be adjusted up or down to the frequency of the next harmonic.

    It could also be interesting to add some kind of spread function with a curve that adjusts all harmonics from a single control. The individual knobs could also be given a curve so that the frequencies closer to the pure harmonics are given a little more space, but I thought I'd stick to a straightforward start for now.
    STS-RM harmonic osc model 3 V1.1t.audulus
    2224 x 1668 - 2M
  • Neat idea! I'm surprised that adding the additional phasers doesn't increase the CPU load more. You can certainly get some interesting sounds by mis-tuning the harmonics
  • Ahhh, all the knows are -1 to 1! Seriously though, harmonic spread is an interesting idea and one that is found on the BIA.
  • I've been experimenting with ways in which to slightly spread the pure ratios of the harmonic frequencies, and listening to the timbral changes that result. Fortunately the Tiltatron has provided a good starting point!

    By reducing the range of some of the Tiltatron controls, fixing others, and adding a further offset control I manged to set up a controller that could give me the spread shapes I was after with just two parameters – the ‘shape’ control adjusts the intensity and direction of the curves, and the 'pivot point' makes it possible to select a fixed harmonic around which the shape can fold. For example, a fixed root with gradualy increasing deviations as the harmonics increase, or vice versa. Or a central harmonic as a fixed point with deviations (up or down) towards the higher and lower frequencies.

    Rather than automating the spread and pivot I’ve left them standing for manual interaction in the tryout patch. The option to invert the even/odd harmonics can lead to some nice bell like tones, and the ‘scan’ row of controls on the Tiltatron provides a nice way to highlight specific harmonics.

    A next step could be to strip it all down and put together a compact module with only four controls. A shape and pivot for adjusting the frequencies of the harmonics, and a shape and pivot for adjusting their intensity. The cosine output is nice for oscilloscope music but perhaps it could also be fine with a single sine out.
    STS-RM uHarmonium M3 V1.1t - Double Spread Tryout.audulus
    2224 x 1668 - 1M
  • Finally got a chance to sit down and play around with your harmonic spread. You can get some pretty cool variations in tone as you sweep the pivot point and spread. It's particularly interesting that the perceived pitch sometimes changes even when the fundamental tone is invariant. I definitely think it would be worthwhile to condense this into a single module. As I had mentioned, it's already pretty reasonable as far as CPU usage, By removing the meters, knobs etc I'm sure you can make it even more efficient. It's always a kick to see others building on your work. As they say, "two heads are better than one". Did you ever see the saw variation I built of the harmonic oscillator? I'd love to see how it would behave. I attached the patch in case you're tempted.
    Tiltatron V3 demo.audulus
  • OK here’s a first version of the Stretch Oscillator.

    I’ll try and get round to a micro version soon, although I must say that I enjoy having the meters.

    @stschoen thanks for the saw version, I haven’t had a chance to look at it yet but will soon.
    STS-RM Stretch Oscillator V1.0.audulus
    2224 x 1668 - 1M
  • Nice work! I like the meters too, they provide an interesting display (particularly if you are modulating the controls). After thinking about it, it would probably be easier to modify your existing design for saw waves rather than adding in all those phasors again.
  • I haven't read this whole thread, but I downloaded the latest version to make today's video with it and I love this thing! It's crazy deep and once I get more time, I'll dive into it more :)

    Here's the video and patch:

    Blang Blorp Bleep Blop.audulus
  • This is a perfect example of the great things that can happen with an open environment like Audulus. It started with a pretty simple concept. Build a waveform from a set of harmonics. Pretty basic audio principles. Robert suggested building a "tilt" module to adjust the harmonic values which led to the Tiltatron and later added a sweep function to the module. Rudiger added a cosine out which made for some cool o-scope graphics as well as providing a quadrature output, and most recently modified the design to allow the frequencies of the harmonics to be shifted relative to the fundamental pitch. I don't think any one of us would have come up with all these ideas on our own. It's not designing the modules that I find difficult, it's coming up with an original idea worth exploring. Almost everything I have done in Audulus has been the result of suggestions by others or inspired by other's designs. I love to see other Audulus users taking my work and using it to build something new I never would have considered. In that light I'm posting the latest revision of the core harmonic generator module for your building pleasure. I recently optimized it somewhat to reduce the CPU load. There are two variants, The first takes an external phasor input (w) and a relative phase (0-1) and the frequency (Hz). The frequency input is used to provide the anti-aliasing and can be derived from a zero-crossing node if desired. The second has an internal phasor and takes an octave signal, phase and sync. For each harmonic you have a level (l) and relative phase (p). There are outputs for each harmonic as well as a mixed output. Let me know if there are any questions.
    harmonic generator V2.audulus
    Screen Shot 2018-03-08 at 5.41.54 PM.png
    1280 x 1310 - 266K
  • Ok, here’s a version with a saw output instead of the sine. With a just a little bit of detuning curve it starts to approach a quite nice ‘analogue’ sound.
    STS-RM Stretch Oscillator V1.0s.audulus
  • @Rudiger, you can get some really nice low frequency sounds with the pivot toward the bottom and the harmonics spreading upward, particularly with just a bit of LFO modulation. Of course I always liked saws. Thanks for putting this together.
  • Here's a mico version of the Stretch Oscillator. I find the visual representation of the curves quite useful, especially when getting to know the module, so I've added them inside the micro module for reference.

    STS-RS-RM uStretch Osc. V1.0s Drone.audulus
    2224 x 1668 - 813K
  • Looks good! I kinda like the meters as well, but screen real estate is valuable, especially on the iPad.My iMac is a 27" model so size is not quite so important, but I really like the more compact approach of the uModules. (more FLPSI) (That's flashing lights per square inch)