You're not even reading are you. Try to make a vocoder out of RMSC.

That means that RM sidechain ducking technique turns a waveform into an envelope with sample precision.

No, it has attack and release as you yourself said.

 

Last edited: Aug 21, 2025

Usually you have a bus with your kick and snare, limited to 0 dBFS, and a bus with literally everything else, also limited to 0 dBFS. You set up the routing to sidechain your kick and snare to everything else. The result then won't exceed 0. It's not used for EDM, it's used for techier genres. Two famous producers that use it are Current Value (frequently) and Virtual Riot (all the time).

 

Ok thanks for your explanations, actually Virtual Riot has video about this technique that I'm watching right now. He does mention that he also uses regular sidechaining and this ring mod technique in addition. He added sidechain pump after the Ringmod, so there's still traditional pumps there

 

Last edited: Aug 21, 2025

And attack and release on a compressor doesn't work with a sample precision right? How does it work then? Does it skip a few samples? Or there is a variant degree of precision, one sample is precise, next one is 1/3 of precision, then it jumps over and then goes back is that how it works?

 

You're being obtuse on purpose aren't you? The attack/release etc. controls set the compressor response time which is not 1:1 sample precise in relation to the input. There are compressors that respond to signals on sample by sample basis, it's just that we call them saturators and clippers.

 

It would be nice, in general; if people used the term "sidechain" with more precision in sentence construction. It should be a noun, a verb, and then a plugin.

 

Wata??? In digital domain it is "1:1 sample precise in relation to the input" how else would it work then??? The attack and release determine the volume ducking response, attack - it's how fast it will reach the volume target and release - how soon it will fade back. Or are we talking about two completely different things here?

Who are those "we"? What are you talking about? They all "respond to signals on sample by sample basis" every compressor, saturator, clipper or any other plugin for that matter. The plugin type and its controls determine how and what exactly going to happen to the sound while it "respond to signals on sample by sample basis". Ring Modulator for example respond to frequency and volume of those frequencies and that what RM SC does too. Saturation introduces extra harmonics not because it compresses a signal by responding on sample by sample basis but by applying a non-linear transfer function to the signal. Cmon mate..

Again, nope it doesn't, it responds to frequency, that what Ring Modulation is
https://en.wikipedia.org/wiki/Ring_modulation
Although it's volume dependent, both RM and RM SC. That's why it introduces distortion which would be absent otherwise if it just was an "envelope with sample precision".

 

In my imagination the one (sidechain compressor) does gain reduction in accordance with a predetermined set of variables (attack time, release time, ratio etcetera) once the external sidechain signal hits a certain treshold.

The rm-sidechain signal actual subtracts the exact incoming signal from the channel it's applied to. So its the exact waveform of the external sidechain that gets taken out of signal (hence the term "sample-precise", since it subtracts the sidechain signal on sample accurately).

So in a sidechain-compressor the type of gain reduction is determined by the compressors parameters (how quick, how long, how strong) and the gain-reduction curve is not exactly an inversion of the ingoing external signal but a dynamic reaction to it, while in rm-sidechain the external signal is literally being mathematically subtracted from the target signal it's applied to.

Please anyone correct me if I got this wrong.

 

Sidechain became a colloquialism for ducking when everybody figured out the Eric Prydz pumpy sound and it never went away. a sidechain is not really a process in itself as you said and more of a component of usually dynamics processors that doesn't strictly have to do with ducking. plus " externally keyed sidechain ducking with a compressor"is a bit of a mouthful, and most people don't generally use compressors for ducking these days anyways (it never made much sense for plugins anyways as in a lot of them there is too much ramping going on in the gain cell/envelope follower, so it's never following the trigger sample to sample)

 

No, you are correct. The RM SC is bit more complicated than just an extraction of a waveform (because it's frequency dependent it introduces unwanted sidebands, thus distortion) but as a summary of the effect that's exactly what's going on.

 

I was shooting for a little funnier I guess, so it would not be taken so precisely. I actually still like to use a compressor sometimes, and my favorite to do it with is either the Logic stock compressor or Fabfilter Pro-C2. Especially Pro-C2 where there is an internal sidechain which has a filter so we can control what information in the incoming audio we want to trigger the compressor, or switch it to the external sidechain source. If I am starting a project that is dance music where I know I will sidechain stuff, I just make a midi track full of quarter notes and the channel's audio output so it can be muted. Another trick I like is if I want to add more bass to my kick channel by using the Logic test oscillator set to sine wave and a noise gate to create a tuned sub to mix back with the real kick. I still have to try this ringmod trick, but it sounds pretty nice. Even though Logic's plugin delay compensation is pretty good, using a longer fx chain on that channel strip can get dicey; so I will usually end up screwing around with negative delay on that track. It can screw with the lookahead also, so I usually end up setting the compression ratio to infinity in Pro-C2 trying to get it to clamp down fast enough. I think this is similar to what Ableton users complain about with its PDC also.

It can be a lot easier to use Kickstart or Duck, but it's nice to have all these other options.

 

You should really read this link yourself. Ring modulation is a multiplication of two or more functions, which is a wideband operation. It doesn't "respond to frequencies". It doesn't respond to anything strictly speaking, it just takes a few signals and outputs a multiple of them.

 

Seriously it's like saying that a saturator responds to frequencies because it produces harmonics. Either shallow semantics or a grave misunderstanding of underlying maths.

 

I think you 2 are probably fundamentally agreeing but just using different words to describe the same thing

 

https://www.desmos.com/calculator/jfrvbjbbuj
Here's the full math behind ringmod sidechain ducking, illustrated on a simple interactive graph with the simplest functions. You can add harmonics to them if you wish, it won't change a thing.
You can see that it's completely specter-agnostic. There's no parts of it that respond differently to different frequencies, and there sure as shit no FFT. It only takes immediate sample values (i.e. volume) into account, and it only modifies immediate sample values (i.e. volume).

A pure ring modulation would be even simpler - as it's just a multiplication. It's in there.

 

No, the dude just read that ring modulation outputs per-frequency sum and difference of two signals and stalled there.
Meanwhile this is just how sample-to-sample multiplication of two functions works because math is weird and full of patterns like this.

 

Are you fng serious? It's right in-front of your eyes
t1 - drums frequency;
t2 - bass frequency;
What does it says? Oh no frequency, how can that be???

 

Ok I've done bit of dive into this technique and I kind of fail to see the appeal for my kind of music at least. I think the point is actually the distortion and flutter it creates, it can work for some type of music but not so much about the "ducking" itself actually.There's also lot of misconceptions around, one guy wanted to get rid of this "flutter" or distortion that remains with this technique, was kind of sharing a 'secret', used some elaborate chain but then basically it was just then the same as you would just mute the audio from the bass at that point when the kick hits!!! No point at all, there was nothing there anymore same as just muting, nothing you couldn't have achieved with traditional sidechaining (that would've been probably even better) or just muting the bass sound for the duration of kick. And people were applauding him for this in comments (granted some agreed that it was kind of weird what he was doing). But to each to it's own, I think I can create nice energy for some type of music with the flutter / distortion but it is still often used in combination with traditional ducking

 

Last edited: Aug 22, 2025

It's two sinusoides. I made the graph specifically so that you can move sliders around and see how ducking behaves in dynamic. The math is all down there under the sliders and you can just read it all and see that it contains no frequency-specific lines, if the very fact that you can move sliders however you want and it just works doesn't tell you that.

 

Last edited by a moderator: Aug 22, 2025

You watched Zef Parisoto's vids, right? No, flutter is not the point. It's a side effect that some people love and some people don't. The point is to limit drums to 0 dBFS, and limit everything else to 0dBFS, and make these two massive chunks of volume not interfere with each other.