Waveforms in motion

This is about the idea / observation / hypothesis that theremin waveforms from responsive heterodyning instruments, are actually modified in real-time by the thereminists actions:

I have moved this from the theremini thread rather than continuing a hijack:

The most interesting thing to come out of the original ( theremini) thread for me, from an educational perspective, was the "operator oscillation" angle which Kevin brought up and Dewster named / identified. The whole latency issue suddenly took on an even higher priority for me with the realization that, unlike a portamentoed keyboard where you know the note will get to the intended key eventually, with a theremin this delay actually disrupts the essential 'control loop' formed between the player and the instrument.

But on a more esoteric level, it got me thinking about more 'subliminal' effects - and one of these is with regard to the real-time bending of a waveforms shape that a player can exert on a fast / heterodyning instrument.

Ok, simulation of anything truly realistic is difficult, as is capturing meaningful samples without having a great thereminist living next door (or probably more to the point, even if I had one next door, I would probably have by now become someone who caused them to be "out" whenever they saw me coming ;-) -

I decided to do the crudest simulation - two HF square waves into an XOR to produce a triangle wave, and subject the variable oscillator to some rapid pitch changes, and look at the waveforms. The following picture shows waveforms which are almost perfect triangles when the pitch is constant, but the kind of dynamic waveshape modulation possible with a players hand:

The lower the frequency, the more distortion is possible - the example above I think would be the most extreme 'modification' available from a really rapid movement or vibrato, and I dont verify the "reality" of the above.. (due to the length of time it takes for the simulation to run, I picked higher difference frequencies than are most effective - But down at 100Hz (10ms) the kind of distortion seen above is, I believe, quite possible.)

I do think it clearly shows that some degree of actual harmonic variation as a result of playing style DOES happen - Something that wont happen on a high latency instrument or single cycle updating wavetable playback instrument.

Fred.

It also strikes me that more rapid pitch change in the bass end is possible if the far field is not too compressed - so, in fact, it could be that linearity actually has an effect on the tone and not just the playability.

Could it be? ... The "classic" theremins possibly (?)  had more oscillator coupling than modern instruments - this in itself gave higher harmonics in the bass difference frequencies- add to this a more compressed bass field and the increased player control of harmonics... ?

So I continued looking for real-life examples -

The sound from Clara Rockmore's theremin played by its present owner does nothing for / to me - but there are quality samples available from these performances, and performances without any other instruments (solo theremin) so I used Dalit Warshaw playing Rachmaninoff's "Vocalise"

Not much in the bass registers, and finding a section where amplitude was hopefully constant (amplitude modulation will also affect waveform of bass notes) the best I could find was a note at about 400Hz which had changing harmonics that I suspect must be the result of  player induced frequency modulation:

One striking thing to me about the waveforms from the Claramin (as all Lev theremins) is the huge variation in waveforms one sees over a few octaves.

I think (hope) I am slowly working through the samples I can analyze and getting to identify which of these are due to the spectrum of the instrument(s) (frequency response, formants / resonances etc), which are source waveform related, and which of these are player instigated and not repeated on a slow sweep.

There are massive "distortions" to the waveforms as the pitch is changed rapidly through any mechanism (vibrato etc) and these only occur while pitch motion is happening (or volume motion, but thats perhaps a different story - or also part of the "wave crafting" mechanism available to thereminists.

Spectrum analysis shows these distortions clearly, and is actually a better way to visualize whats going on - but its also more difficult to present in a few pictures and more subtle (at least when using the Aidition spectral display)

I really would love some owner(s) of classic instruments to put a set of samples together - slow sweeps, fast sweeps, sustained pitches, volume sweeps, vibrato, tremolo etc ... Digging this stuff out of performances is really unreliable and difficult!

Fred.

Fred maybe we can finally agree on something, change the word "pitch" in this phrase to "tone."

"Beyond 4 inches (10cm) the belief hand capacitance alone controls
 the RF heterodyne theremin pitch is a  factoid."

These are the words from a hobbyist 10 years ago. . .

It seems to me a mostly digital theremin will lose this effect.

Christopher

Christopher,

If anyone said ""Beyond 4 inches (10cm) the belief hand capacitance alone controls the RF heterodyne theremin tone is a  factoid" they would, IMO, be talking nonsense anyway.

Sorry, I would love to find some common ground - some rational basis upon which a foundation of common understanding and language could be established..

But the fact (regardless whether people believe it or otherwise, and regardless of whether there are contradictory factoids or not) that have been established through experimentation and verified (as far as one can within the scientific paradigm that everything is "open" if other evidence or explanations appear) is:

1.) Pitch of a theremin is determined by capacitance alone - other factors may have extremely minor effect, but such effects are so marginal as to be irrelevant / lost below the "noise" threshold.. Really, any such effect if it exists, is so utterly irrelevant that it is meaningless unless one was trying to build an extremely long distance (100 meters say) sensor.

Then, with regard to theremin tone - the idea that the thereminist actually controls the tone (as in, waveform / harmonic content) to any extent has not been (to my knowledge anyway) widely, if at all, postulated - "obvious" factors such as vibrato depth and speed and control of amplitude and timing, and the importance of the thereminist input have been repeated many times - but I have not seen mention of the implication of these actions on the WAVEFORM mentioned by anyone else - And my suggestion that the waveform was actually altered by the thereminist has been scoffed at and derided by you in past posts.

2.) As capacitance ALONE determines the theremins pitch, and as it is variation / modulation of this capacitance which changes the pitch and thereby acts to change the tone (waveform / harmonic content), I am sorry - but your statement is still completely wrong.

And there would be no problem with your statement being wrong IF you had presented it as a hypothesis, IF you didn't state it as a fact, and if you didn't call sound scientific theory a "Factoid".

I could be wrong - there could be a completely different mechanism in operation for everything - there could be a single electron in a quantum 'multiplexed' super-string 'state' - there may only be one electron in the entire universe, and everything we believe could be a complete set of models that have no bearing on "reality"... But these models are all we have to work with, and if these models predict behavior accurately, then they are useful - Any model which fails this test is of no use.

And anyone who argues against a model which works (as the theremin capacitance model DOES) needs to show a flaw in the model, and anyone who presents an alternative model needs to do this as a hypothesis, with the explicit purpose being to deconstruct that hypothesis or have others find flaws in it - ONLY when flaws are not found can the hypothesis perhaps move to the status of a theory..

But to present a flawed, half baked hypothesis as "truth" and declare established theory "factoid" without presenting ANY fault in said theory... These are the actions of someone who is deluded.

I do not blame you for falling into this trap though - You are not alone, it is an easy trap to fall into.. We are probably the most complex thing in the known universe, and the factors determining our personal realities are vast.. Every day new data appears that reveals even more complexity, and the more that's revealed, the more astounded I become that we are able to function as "well" as we do, and how little of what we do or thinkor say or how we behave is really fully under our control.... We even have low-life gut bacterium which can, unknown to us, influence our behavior - they can make us do things that THEY want us to do, without us even knowing that they exist!

It is my opinion that the scientific method is the only defense we have against delusional ideas - It is a structure which gives us the ability to test ideas against each other, and whilst it cannot reveal "truth", it can, at least, reveal untruth.. IF used correctly in an unbiased, honest way.. Like all great ideas though, it is subject to corruption when abused.

Fred.

I think that another area that bears some research and/or experimentation is the power supply -- particularly if it is operating at the edge of its capacity.

In the case of the volume circuit, a current-starved (tube) circuit will produce variable distortion according to the gain (which is varied by the volume section of the instrument).  Likewise, a somewhat current-starved power supply will negatively impact the isolation of the oscillators.

I have heard reports that when old (weak) power supplies are replaced with modern, regulated supplies that this can effect the tone quality (in amplifiers, etc) and this may have to do with the way the circuitry distorts outside of its normal operating range.  (The reason guitarists prefer vacuum tube amplifiers over solid-state).

Also, given that a theremin's oscillators are in the RF range, the parts placement would influence how they couple, too.

My general impression is that the interaction of the power supply and the parts placement are complex and subtle, are likely difficult to model, and the result is achieved through experimentation.

Fred I think you and I interpret words (across the pond thing) “differently” like Voice, Pitch, Sound, Frequency, Tone etc. In my quote, not being a musician, I am referring to the unique character of the theremin voice that has “multiple” sources of influence.

I call this fluid motion of sound which sets it apart from a keyboard. I thought that is what you were seeking input about. When I mention Pitch it is to avoid saying Voice which needs human interaction to develop. Frequency of the audio wave can be measured but Pitch is a bit more complex? I completely agree that hand-capacitance controls the frequency of the RF Oscillator. Pitch like Sound is a human experience. imho

Christopher

Edit: I could never find the authentic sound with solid-state but with the vacuum tube it is magic. For the guitarist and myself I think the secret is in overdrive which sounds more musical, the tube is very forgiving.

Hi Kevin,

The whole issue of interactions, particularly on something like a theremin where separate circuits like oscillators can (will) interact through any path and directly affect the audio, is extremely complex - as you say, impossible to model accurately or at all -

And here I think lies some of the ambiguity, and for me often doubt, about hobbyist circuits or circuits developed by "incompetents" .. Because it is possible that a badly designed circuit can accidentally give desirable results.

And this is a gulf between professional designers and "hobbyists" - Things like power supplies - I want mine to be as stable as possible under worst and best case situations, its one of the things that as an analogue designer is way up there, possibly at the top of the list of things I must define and get right before I do anything else.. I feel really uncomfortable with the kind of via-power-rail HF coupling I see on some classic theremin designs - Yes, I can see their validity and economy, but I can see other ways to achieve the same result and I pick these.

Also, simulation and modeling now play a big part in the design process - I was an analogue designer before these were available, and used some of the first simulation packages when they became available - In truth, because I am bad at maths, simulation became an essential tool- And simulators are pretty damn hopeless if one depends on obscure coupling routes like power supplies!

So, for me, its about identifying the mechanisms of "known" influences on the sound, and about discovering any influences, even if the cause is unknown or indeterminate, and devising some probably completely different means of reproducing this effect if it is desired.

Yeah, its probably boring ;-) .. But I like each 'block' to do its thing, all interactions to be deliberate and quantifiable, power supplies to be clean and stable, audio coming out to be audio devoid of HF remnants - when one has that, then one can add tubes and audio transformers and oxygen free gold plated wire and snake-oil saturated transistors if one wants ;-)

The distortion as a result of analogue pitch change is something I have been aware of for many years with reference to analogue synthesisers - I am starting to think it has even higher significance for theremins - but its a hypothesis which wont be liked by many digital instrument developers, and it is only a hypothesis - I know the distortion occurs, and I believe I can hear it, but I have no proof that I or others can - And I have no objective measure of its significance.

Fred.

ps - I agree the significance of the PSU is worthy of more research - Alas, there isnt a theremin research fund I know off.. Cannot even get a set of samples from the most important remaining original instruments, let alone get to them with probes and audio analyzers to measure anything!

Hello Christopher,

I agree about over-driven tubes sounding sweeter than over-driven BJTs, ( and over-driven digital is an offense!)

But the mechanisms for this are mostly agreed on and understood, and only apply when over-driving! If its the 'over driven tube sound' you are after, build a BJT theremin and drive it hard into a tube amplifier!

;-)

(or make a JFET amplifier or use a H11F1 to get similar clipping)

Also, tubes are microphonic - I can see that a tube oscillator could give FM when exposed to loud audio background - My oscillators have a voltage control input to allow feeding audio back to the RF stage, as this really does give increased musical harmonics IMO.

IMO, we know enough about the mechanisms now that for something like a theremin, we dont need tubes in the front end - If we need them (and I really dont believe that we do) then the output stage should be enough... Its an icing at best - If the player cant play, then all the icing in the world wont make the cake taste any better!

Fred.

Fred,

I've noticed this when running an oscilloscope plugin on my Etherwave. Here I set a stand near the theremin to have it play a fixed tone without my hand wavering. Then I set my DAW to monitor and ran an oscilloscope plug-in in looped mode. The theremin waveform is always in constant motion. Pitch and amplitude should be pretty much fixed here as I was 10 feet from the theremin.

Maybe this is meaningful. Maybe not.  

Etherwave Waveform on Fixed Tone

Rich

Hello Rich,

Ok - You got me! ;-)

pitch is constant - but waveform is certainly changing..

Ok, that didnt do a lot ;-)

(just captured the audio from the video so I could listen to it and view/analyze  it with Audition, and got most puzzled - until I plugged the cans in and listened to you tube .. I keep my big PC with all my tools and good soundcard off-line, and only transfer stuff to it that I know is safe)

I haven't seen anything like that ever on EW boards I have played with, and haven't seen such a waveform either - Doesn't look like any EW I have seen before! - I imagine it has a really rough tone, phasing even harmonics in periodically, but being predominantly odd harmonics.

Thierry - Help!  ;-) ... Because I dont recognize this as an EW! - So either my EW's are odd, or this one is!

Fred.