Synthesis : subtractive, additive, ring modulation, fm... ++?

Hey there…
I’ve been watching some videos from Kaitlyn Aurelia Smith lately… And somehow I’ve getting slowly hooked to the idea of maybe get a few module to experiment other ways to make sounds. But in the same time I know I have way too many instruments already, and am convinced I’m too fragile to get exposed to modules…
So the idea would be using my current gear to experiment and discover different types of synthesis.
I have read Synth Secrets, but I should do it again for sure : now that I have been playing with synths for 2 years, things should get clearer… Still but less a noob :tongue:

In the meantime, I’d like you guys to throw some ideas on the way to achieve different types of synthesis with our gear…


Subtractive synthesis is quite evident, every Elektron machine offers at least one filter per voice.
But from the beginning of this year I’ve been feeling that the good old fight between analog and digital is a mountain hiding a whole lot of other way to make sounds…

Plus there is this idea that by filtering you remove sounds, like cutting the branches of a tree instead of growing new ones…


FM synthesis : the addition of note-following FM-range LFOs on A4 and @Ess’s impressive demonstration and @cuckoo playing in NAMM with the Volca FM got me interested in this type of synthesis.
Before that, I had always been considering FM as cheesy, and always hated the synth sounds from the 80’s…
The following video was the clearest lesson on what FM was and could achieve, and now that I spent a few months experimenting with this I understand how rich this kind of synthesis is.

  • A4 can be used to get interesting FM tones, and while writing I figure that you may chain the tracks to achieve quite complex routing of operators, I still have to experiment on this matter.
    Oh, and I’ve just discovered that LFO1 was following oscillator 1 while LFO2 was following OSC2… I now understand why I had results not as clean as what I would have expected. I also recommend to recalibrate the A4 (after 2 hours of warming).
  • I have nice sounds on the Monomachine but I don’t quite understand what I’m doing here, does anyone have pointers on a clear explanation on what’s going on with MM 3 different FM synthesis types ?
  • On MD, the FM-based sounds are impressive, but I haven’t messed that much with LFOs to add FM character to the sound. Any ideas/pointers on this matter are very welcome.
  • So far I only used quick fading out FM-range LFOs on AR to add dirt to transient…
    Maybe I should consider getting back to experimenting in this area…

Got me a Volca FM and PreenFM2, I know that I have still a lot to learn about FM synthesis but it’s getting clearer and I love it.


About Ring modulation : I know A4 can do AM, although I don’t get it really.
I mean, I know how it works theoretically (I might have to read again Reid’s lessons on this one btw). Thing is I don’t really see how to

  • use AM properly enough in A4. I’d like advices to understand this part more precisely.
  • achieve Ring Modulation such as what Mutable Rings can achieve (very, VERY impressive resonating sounds). I really wonder if one can reach such sounds with any of the Elektron machines… Can you ?

Additive synthesis seems to me like getting parallel oscillators that would be related with each other in some way.
Most obvious being fondamental and harmonics.
Now we can think of having slight variations of the harmonics to get weirder timbres… Modulating the relationship of fondamental and harmonics over time… Maybe even get some weird relationships instead of constant ratios between the oscillators.
I wonder how to achieve this in A4 or MM… Any ideas ?

FACT video on Kaitlyn Aurelia Smith was displaying some very interesting sounds… Some I have no idea how to synthesize…
These Buchla systems are way to expansive for me, there is no way I can get the hands on one.
Modules are a hole I don’t want to get into before knowing more on what I would like to achieve with them.
Well, I now I could use a Braids module to expand A4 possibilities but getting a case for this seems to me like an obvious step forward temptation…
And as I say, I have already so many machines and I first need to understand and experiment more with what I have.


There are other kinds of synthesis I am not very aware of, and I’m slowly diving into the previous threads about synthesis, discovering things I had overlooked…

Anyway, any idea / lead for experimenting new kind of synthesis on whatever machine would be very welcome :slight_smile:

4 Likes

Physical modeling.

I think the posterboy is Karplus Strong - which, if I remember right, is just a very short delay line with near infinity feedback, and the delay time controls the pitch. Then you just throw noise bursts at it and you have a pretty stringy sounding instrument.

I remember Allerian doing some Karplus Strong on a Machinedrum in YouTube, if I’m not mistaken. Or was it a Monomachine?

The G2 Nord can do all the FM stuff very straightforwardly (pricey s/h), but if you’re a G1 user like myself, all is not lost … the creative possibilities in that range are mighty - especially if you want to explore a variety of synthesis techniques from 1st principles

This site is a great resource for an academic breakdown of synth techniques adapted/applied on the original G1 range - the micro modular is surely the single best bang for buck device still buyable … and so small !!

Advanced Programming Techniques for Modular Synthesizers

There are probably countless ways to do budget exploration on a computer

FWIW - The Nord G2 editor also permits using the patching program to generate a single voice, even if you do not own the hardware

2 Likes

Found @futureimage’s take on KPS synthesis and @darenager’s one :slight_smile:

As I understood A4 is linear FM, but what is exactly the difference with FM ?

Any help with how AM and FM works in A4 would be appreciated:)

Oooh, Daren Ager it was. One of the OG’s, nonetheless :slight_smile:

If I understood correctly, different kinds of FM where used.
First there is a difference between Phase modulation (DX7, Volca FM) and real Frequency modulation (PreenFM2).

You can also directly use an oscillator to modulate another oscillator (which I assume, but not sure at all, is what is referred to as linear FM) or you can apply a mathematical operator between the modulation operator and the carrier (which would be, if I’m not talking nonsense, nonlinear FM).

You can also apply feedback for an operator to modulate itself (each algorithm of the Volca FM does this).
The thing is feedback or intermediate transforms are mostly interesting when you have only sine oscillators available for your operators, and you want to add harmonics and get close to e.g. saw oscillators.
When you have any kind of oscillators to start with, feedback or mathematical transforms may reveal to be uselessly difficult to implement for real FM, which is why they are not available on PreenFM2.

Please people correct me if I’m wrong :slight_smile:

1 Like

Here my few cents … I hope to contribute something useful … :wink:

FM synthesis:
If you have got the PreenFM2 already, you just have one of the best little boxes to make full blown digital FM-synthesis. Six operators, all required modulation capabilities and some more compared to the DX7 … with this little box the entire world of loved, weird, and special FM sounds is open to you.

There are synths, which provide FM between two oscillators too, but it makes a difference whether a synth is FM dedicated like the PreenFM2 or not. My experience is that the sound is different.

Once you go modular and look to west coast style synths, you will find specialized “complex oscillator” modules with two FM dedicated VCOs. The analog FM synthesis is mathematically equivalent to the digital FM synthesis, but from my experience both have quite different sound character. I use them for different purposes.

Using a general synth with a keytracking LFO in audio frequency range for FM is FM … absolutely, but I use this more for FX, or for very special noises or sounds. IMO the typical mallet like FM sounds or the FM basses are best generated with a FM-synthesizer.

Okay … much of the said is a matter of personal taste :wink:

Ring Modulation:
Well it’s not a kind of synthesis I would say … it’s rather a “multiplier” module. It was one of the old school tricks to get disharmonic sounds from input of two harmonic oscillators, which should have a harsh, weird, or metallic sound color.

Additive Synthesis:
The real fun starts, if you have 512 to 1024 oscillators at hand. With such a synth you can realise everything soundwise. You may already guess that this is more in the domain of digital-software synths rather than an analog instrument.

The A4 has 4x2 oscillators, if we neglect the sub-oscillators, which makes 8 in total. Technically this would be additive synthesis, if you tune all the oscillators harmonically … but it would more sound like an organ or a supersaw depending on the waveshape choosen.

If you are interested in additive synthesis, then you should consider synths, which give you power over the entire frequency spectrum, or so called sonograms or spectrograms. Here are some recommendations to check out:

  • Image Line Harmor
  • Image Line Morphine
  • iZotope Iris2
  • Native Instruments Razor
  • Rob Papen Blade

All those synths provide methods … often differently … to shape the spectrum or the sonogram (that’s a series of frequency spectrums, changing over time). Iris and Harmor allow to drag and drop a user sound file in the synth and use the “spectral-tools” to make the weirdest things. Only creativity is the limit … :wink:

3 Likes

Thx dear for the input…
Amazing how synthesis is vast and deep.

I’m really into the sounds you can get from additive synthesis. I’m also super into using hardware over vsts and I couldn’t really come up with much for additive hardware synths aside from some older Kawai synths and going modular.

Could be wrong, too. The concept is still pretty foreign to me. Waveshaping and multipliers and such.

If additive synthesis is supposed to generate realy comlex sound structures, many oscillators are required. Just an example. A synth like the Harmor from Image line operates up to 500 oscillators per voice. With such many oscillators the engine can re-synthesize almost every sound we can imagine. If an audio file is dropped in the synth and the “sample” is played, it’s like playback of a recording … BUT it’s synthesis. Tools allow the change and modulation of the sound spectrum. We can steal the melody of a singer and now he is talking, or underlay a talking with a melody and it becomes singing. We can define the “surgical filter” just by drawing it as we want it to be and so much more …

Now imagine this as a hardware synth … 500 groups of knobs for all oscillators, or a menus to dive in to get everything done? …This just wouldn’ work. The additive synthesis is IMO best implemented with computer based engines with a graphical interface. Check out Iris, where we “paint” our sound. It’s very creative and very different workflow, but a child can learn it in minutes … despite the complex mathematics of Fourier Analysis/Synthesis in the background … :wink:

1 Like

You should also check out the Iris (if you haven’t already). It’s pure fun to “paint” sounds on the screen … :slight_smile:

The logic of the “Harmonizer” and “Prism” of Harmor is something that took me some time to get my head around. But what I love is that we can “import” audio and waveshapes and that we can “draw” our own filters.

If I could just get a wave folder or multiplier, I’d be happy. I guess what I mean is west coast style synthesis. Not much for hardware in that arena.

If you don’t shy away from digital … :wink:

I use the Harvestman Piston Honda for waveshaping. What is great about it is that you can combine wavetable morphing with waveshaping. That is awesome :sunglasses:

Alternatively there would be the Shapeshifter of Intellijel.

That’s what I was saying earlier, pretty much the only way to get those additive elements is going modular, which I don’t think I’m prepared to do.

Agreed if you think about hardware … but there are many plug-ins out there … Trash 2 from iZotope would be one of those.

Have you considered to get one of those Doepfer A-100MC mini cases (including power supply) and start with just one “waveshaping/wavefolding” module?

I have a mother 32 and have thought about getting one of their cases and a dual stand. That would probably work out alright. Pretty sure I’d end swapping modules like a madman though.

This is true only if you work with sine, the lowest unity in sound. But there are other ways to add harmonics : you can transform the sine e.g. with pulse modulation or distortion… Adding simple sine is not the only way to go towards additive synthesis, and the proof is that the modulars don’t have 500 knobs :smiley:

:smile: of cause … you have a point here … :smile: … it’s “additive” … absolutely.

IMO the term “additive synthesis” is more often used in the context of methods based on Fourier-Synthesis rather then “mixing” different waveshapes ,or add harmonics with waveshapers, or special VCOs like the Harmonic Oscillator from Verbose (which is a great sounding module by the way).

I just had a too narrow view of “adding harmonics” :wink: