Possible to get rid of "phasey" osc-beating in soft-synths?

[Dillinger]

By hardware you mean analog...correct?
It’s the nonlinearity, the difference you hear.

Edit:removed quote

[mdstudio]

turn off free running osc, and any mods/drift options

I think this is the correct answer. In Serum, for example, the Rand knob changes this behavior.

[BONES]

The entire purpose of detuning the oscillators is to get that interference pattern going. If that's not what you want, don't detune your oscillators. Simple.

[cthonophonic]

One thing to consider is that asymmetric waveforms can produce less beating. A pulse wave with 20% width will sound less phase-y than a square wave does when detuned against itself, for instance. This is because the amplitude of the beating pattern depends on the shape of the waveforms: you get a single strong regular beat with detuned square waves, but a couple of smaller beats with narrower pulse widths. Instead of WaaoooaaWaaoooaa, you get waowaooowaowaoo, if that makes any sense. Maybe there's some sort of quantitative measurement like an autocorrelation function that one could use to predict the beating patterns of different waveforms, but the best thing to do is probably just a side-by-side listening test.

Though it's possible that the differences you're hearing have something to do with the ways that analog waveforms depart from a mathematical ideal, I would assume any VA model of an oscillator that's worth a damn (or hell, even a single-cycle sample of an analog oscillator) would capture those too.

[GRUMP]

My Focus are Pads, I know the mentioned Problem quiet well.

For my Ears DUNE 3 delivers a good Solution for it. You can adjust the Volume of the detunded OSC Voices. Lower Volume = less "Beating".

If you stack seperate OSC in an other Synth you´ll have this Option, too. The Mix should not be higher than really necessary. Mastering Tools esp. a Limiter/Maximizer will maybe help you asses the Leveling, too (my Experience).

Concerning this Problem FM is advantageous by the Way.

[Gamma-UT]

For my Ears DUNE 3 delivers a good Solution for it. You can adjust the Volume of the detunded OSC Voices. Lower Volume = less "Beating".

"Doctor, it hurts when I do this."

"Don't do it then."

[GRUMP]

For my Ears DUNE 3 delivers a good Solution for it. You can adjust the Volume of the detunded OSC Voices. Lower Volume = less "Beating".

"Doctor, it hurts when I do this."

"Don't do it then."

Or get clear about your Scope and try to find a different Solution!°

[DJ Warmonger]

Slightly modulate the pitch of one osc over time, so beating becomes irregular.

[JJ_Jettflow]

Just use one oscillator.

[imrae]

Just use one oscillator.

Juno 6 wins the polysynth video for me!

[JerGoertz]

One thing to consider is that asymmetric waveforms can produce less beating. A pulse wave with 20% width will sound less phase-y than a square wave does when detuned against itself, for instance. This is because the amplitude of the beating pattern depends on the shape of the waveforms: you get a single strong regular beat with detuned square waves, but a couple of smaller beats with narrower pulse widths. Instead of WaaoooaaWaaoooaa, you get waowaooowaowaoo, if that makes any sense. Maybe there's some sort of quantitative measurement like an autocorrelation function that one could use to predict the beating patterns of different waveforms, but the best thing to do is probably just a side-by-side listening test.

I get what you're saying. I think PWM would probably also help in this regard. I've been playing around with that, and it *seems* to be helping, although it's not always the exact result I want.

My Focus are Pads, I know the mentioned Problem quiet well.

For my Ears DUNE 3 delivers a good Solution for it. You can adjust the Volume of the detunded OSC Voices. Lower Volume = less "Beating".

If you stack seperate OSC in an other Synth you´ll have this Option, too. The Mix should not be higher than really necessary. Mastering Tools esp. a Limiter/Maximizer will maybe help you asses the Leveling, too (my Experience).

Concerning this Problem FM is advantageous by the Way.

Interesting... I use Poly-Ana, and I can get it to sound really, really close (and maybe in fact good enough); it also allows you to adjust the relative volumes of the detuned osc's, so I'll give that a shot. (I do have Dune 3 as well, but it's too digital-sounding for the kind of analog strings I'm trying to get.)

Slightly modulate the pitch of one osc over time, so beating becomes irregular.

Sure, I think this has merit. I suspect that analog synths have some of this naturally with oscillator drift, which probably contributes to the difference I'm hearing.

[JO512]

If you don't like the beating, one thing to avoid is saturation or distortion, as this will accentuate it. I love it myself! I like to play dyads and add lots of saturation just to sharpen that wonderful beating that emerges in the interference between the notes! I often just use pure sine tones, interfere them, and then add lots of saturation. Great stuff! Or put a reverb or delay before the distortion/saturation. Same effect. Or use long release times in the amp envelope so the tail of one note interferes with the next.

If the distortion is per-voice, as you have the option of doing in Falcon, it doesn't have the same effect on chords, as you then aren't sharpening or clipping the interference pattern itself. Both ways are useful for different ends.

[cron]

Another idea to minimise beating might be to detune saw against square. Only half of the harmonics present will beat against each other, but you've still got a 'fused' timbre with a detuned character.

[JerGoertz]

If you don't like the beating, one thing to avoid is saturation or distortion, as this will accentuate it. I love it myself! I like to play dyads and add lots of saturation just to sharpen that wonderful beating that emerges in the interference between the notes! I often just use pure sine tones, interfere them, and then add lots of saturation. Great stuff! Or put a reverb or delay before the distortion/saturation. Same effect. Or use long release times in the amp envelope so the tail of one note interferes with the next.

If the distortion is per-voice, as you have the option of doing in Falcon, it doesn't have the same effect on chords, as you then aren't sharpening or clipping the interference pattern itself. Both ways are useful for different ends.

Heh, okay, I might give that a try sometime.

Another idea to minimise beating might be to detune saw against square. Only half of the harmonics present will beat against each other, but you've still got a 'fused' timbre with a detuned character.

Thanks, this sounds worth trying as well.

---

After a bunch of trial and error, I've pretty much got my sound the way I want. I think anything more would be far past the point of diminishing returns.

I appreciate everyone's helpful input.

[TIMT]

i asked something similar in The Legend thread a year or two back and i never got a decent answer either.i know what you mean though.i also don't get why on VSTI's when you use a lot of voices they just seem to blur into themselves with no real definition between them.almost like you used an FFT smear process over it when you use unison patches or patches that use multiple oscillators(hyperbole,but its in the ballpark of that sound)whereas when i use my analog poly even with extreme amounts of detune,the beating never gets washed out and phasey and the oscillators do not smear into one another to the same degree of VSTI's,which makes it sound more dynamic within a mix compared to an equivalent patch on say DIVA,Legend or insert any VSTI here.the one place i actually notice this happens less is on modular VSTIs like VCV and Reaktor,so if i absolutely have to use a VSTI(travelling and what not),i tend to gravitate to those.