Very weird sound from Cybertech (1993)

How to make that sound...
Post Reply New Topic
RELATED
PRODUCTS

Post

This sound has intrigued me for years.
https://www.mediafire.com/file/6h8k8b1u ... h.mp3/file
It's especially the "dynamism" of it - the way it changes - that I like, rather than the texture of the sound itself. Does anyone have tips for how to this sort of thing?

Post

It's amplitude modulation. Oscillator A (modulator) modulates the amplitude of oscillator B (carrier) at audiorate. Mathematically signal A x signal B.

AM results in sum and difference frequencies of all partials in both signals (carrier and modulator). That can soon lead to many frequencies, approaching noise. That's why mostly relative "poor" spectra are used. Like sine and triangle, less often squares and saw or noise. In this case it's oddly rich, probably something like a square modulating a triangle or vice versa. But I have not tried.

So something will happen as soon as you detune the oscillators. Usually only the modulators pitch is modulated. If you do a pitch sweep, it wil result in frequencies sweeping downward (the differenence) and a frequencies sweeping upward (the sum) at the same rate.

Anyway, you can hear the effect on any synth that can do AM. Just modulate the frequency of the modulator. And keep a bit of the carrier (the not pitch modulated "note") audible as in the example you gave.

Post

Just tried Ichiro Toda - Synth1 (free)
https://www.kvraudio.com/product/synth1-by-ichiro-toda

It's a pretty basic, but versatile and rightfully highly praised synth. And it does have basic AM options. It also has a true pulse/square waveshape (as in min/max only, with straight lines and 90 degree angles only). Which is what you need here (I guess). You can also adjust the pulse width (P/W, where max = pure 50% pulsewidth square) of Osc1 (if Square is selected). And use Mix to set the relative Oscillator/Signal levels.

It sort of comes close...to the sweep...sound wise...using the above recipe.

Sadly it has just an Attack and Decay envelope for Pitch Modulation. And only for Modulator Osc2. Enough to have a experimental sweep on the pitch of modulator/osc2. But not enough to get the delayed sweep of the example. It also lacks the option to delay the start of an LFO as a workaround. But to overcome that, the Pitch of Osc2 could be automated.

Post

Thank you for these helpful reples, Kwurqx. I have to say I didn't realise there would be so much to this effect! I play around with Synth1 and see how close I can get to the original sound.

Post

Since you are exploring Amplitude Modulation (AM) or as it is often called, Ring Modulation (RM).

Generally both oscillators frequencies are at some musical (semitones) interval, both at frequencies based on the frequencies of the notes. There's also AM using non-musical intervals (in between semitones...) which gives non-harmonics content (e.g. for bells).

Then there's a different aproach where the modulator has it's own separate note-independent frequency. Fixed or pitch modulated. Usual for external gear like guitar pedals. Unusual in synths.

A FREE synth that offers this somewhat unusual option is

Krakli (Ian Webster) - Arminator 2
http://www.krakliplugins.com/

It's a brilliant and FREE emulation of (or at least insired by) the iconic Yamaha CS80 (think Vangelis). Named after sound designer Armin Kujashi as a tribute. Check out the presets.

Off course AM/RM is also available in (external) effects.
https://www.kvraudio.com/plugins/free/r ... tor/newest

Post

It could also be ring modulation, or also just playing so high sounds that they start to alias a lot at the converters or other pieces of the equipment chain. You might want to try sample rate reduction to get the same kind of effect with just high sine sweeps.

Post

Taika-Kim wrote: Tue Apr 09, 2019 6:49 am It could also be ring modulation, or also just playing so high sounds that they start to alias a lot at the converters or other pieces of the equipment chain. You might want to try sample rate reduction to get the same kind of effect with just high sine sweeps.
Digital aliasing.

This occurs when you try to push frequencies above the Nyquest frequency (half the samplerate). To obtain a sample of a periodic signal, you need at least 2 samples. 1 above zero/offset and 1 below. If the samplerate (frequency) is lower/slower then the actual frequencies you're trying to sample, you will miss your samplemoments (because you can't sample fast/often enough). The result is that these frequencies get divided by some number and end up back into the audible spectrum (below Nyquist) at the wrong (lower and possibly inharmonic) frequency.

This is what "anti-aliasing" tries to prevent. E.g. by using a steep lowpass filter to eliminate frequencies above Nyquist. Or by using FFT, eliminating all partials above Nyquest (e.g. amplitude to zero) and iFFT back.

Many older digital synths "suffer" from this phenomenom (especially at higher notes). Though sometimes it actually adds to the character. Like on the iconic Supersaw machine: Roland JP-8000. Or the Yamaha DX or SY synths. FM (or PM...) is notorious for being able to quickly generate many very high partials. And without the (lowpass) filter of the subtractive synths, they tend to moreoften reach the DAC.

Post Reply

Return to “Sound Design”