As a Renoise and FL Studio guy, I am jealous of Bitwig’s interface. When I tried the trial, I was able to learn it and flow through and write an entire track in a few hours without any real learning curve. It’s that good.
Just make music and have fun. Don't worry about your skill level and understanding everything. Find some presets that you like, save them and categorize them so you can quickly put frequently used FX chains and sounds on your tracks. Many of my popular presets are from Polarity and helps me not have to worry about sound design and understanding what each knob does. Also, F1 is your best friend. 😉
exactly where thousands and thousands of producer are stuck. exactly where plugin and hardware companies get you. people would (1) learn more if they didn't go to youtube for every problem they run into and (2) produce waaaay more music. fingers crossed you hit that u-turn at some point.
How are you going to learn if you don't make music? You have it back-to-front! Make whatever music you can make every day and you'll learn as you go. There's no point trying to learn music production abstracted away from producing music - it'll never happen. Music production is music practice!
I prefer to use a replacer on kicks and snares on mute and use that to create a midi note that will trigger a mseg on the volume of a tool. It’s precise and basically becomes lfo tool
Appreciate the shoutout it’s always cool to see variations and different takes on this method. I’d like to point out that the RMSC VST is not by me but another developer. I don’t use mringmodulator for this anymore like in the original video but my own M4L device, and sometimes the RMSC vst. But I’m working on a plugdata version now. Keep in mind that using the mid signal with Bitwig’s audio rate modulator can still peak over 0db when summed together if the target has side information; ideally rmsc:ing would be 100% stereo unlinked so both L and R get independent ducking, that’s if you’re looking for absolute headroom advantage. For the sound effect itself it doesn’t really matter.
nice tutorial. I'm an old time poducer. Never cared about such details in the past but it's nice to see the sound quality "granularity" that we can obtain these days.
So this new way of sidechaining introduce distortion and to solve the problem we have to reinvent standard sidechain processing. Well….. looks like those audio engineers from the 50s were not complete idiots after all.
the ducking is what i have been doing, and mainly the reason i got bitwig, i use it also on EQ to dig specific frequencies instaed of ducking the entire frequency range of a sound.
First, all things Polarity are worth a watch!! Always!! RMSC techniques are interesting and informative but ultimately, I personally have never found RMSC useful. When I solo the ducked instrument noise pokes through, because you're ducking specific parts of the spectrum you are left with a waveform that has nothing to do with the original sound. I concluded that for hard duck it's tool with midi note sidechain from a "SC" midi track where I can lengthen or shorten notes and use timeshift to avoid attack 'clicks'. Alternatively, if you're just dying to keep some part of a sound I use Robert NIH linear phase crossover. Then I know exactly what it will sound like during a duck.
Sometimes the distortion is desirable and precisely why someone would use ringmod sidechain. It could add the sensation of intermodulation distortion without chainging the original sounds.
The disconnect, I think, is that a lot of people end up getting steered toward RMSC when what they really want is something a bit different. They don't want the peaks and valleys, but rather a contour that goes from one peak to the next... ideally with a best fitted curve. Follower / audio sidechain isn't quite that either. One way to think of the goal would be that it is the inverse of the MSEG that created the kick. It's also worth noting that different people mean different things by "clean". The distortion seen between the peaks is one thing, but being completely out of the way on the first sample is another issue entirely. If any method fails to achieve that, there is the potential for a burst of up to 200% of full scale in those first few samples which can be far from subtle depending on what follows.
What you're describing is the "convex hull" of the waveform (or more generally an "alpha shape" to allow some concavity). Imagine wrapping a rubber band around it. I'm not sure that it can be properly computed without infinite lookahead (since the next peak can come at any time in the future, or never), but there are approximations. Envelope followers are one of them. ;) In the modular world, you can just modulate both with the same envelope signal. You can probably do something similar in Bitwig somehow.
In theory, the perfect distortion-free sidechaining you describe could be achieved with offline processing of the complete audio files, where you literally do have infinite lookahead. I suspect that offline processing is going to become a next big thing in audio because it turns out infinite lookahead is pretty useful: it makes hard things easy and impossible things possible.
@@reinh It's infinite only in theory, though. In practice, I don't see why it would need to be more than the next inverse peak at the lowest frequency you'd want to capture... or 25ms for half of a 20hz wave. That's for linear. A better fitted curve would probably require double that. I'm not saying it's easy... just that it may be a clearer description of the goal since there's no benefit to the wobbles between peaks, and followers introduce other issues.
@@bongospank8321 Yeah, that's what I meant by "approximations". You can scan your lookahead window for the next maximum peak and lerp to it. There are probably edge cases that make this more difficult than it sounds. Getting the windowing right might be hard, or highly program dependent. But it's a good idea.
damn bro if I get even more precision my tracks are going to be made out of a kick track and a bass track sidechained just to show the absolute perfection
Bitwig's Audio Sidechain has a 1 ms rise minimum. This results in audible clicks before the signal is then pushed down after that 1 ms. I will never understand what possessed the devs to cripple the essence of this tool in this way.
I think the main benefit is at the transient. If both channels peak at 0db, do these methods result in the sum not going over 0db? Last time I looked at this there was a small sample delay and it went over 0db.
What I'm never going to understand is why the people who want that clean sound just don't compose it. It's a synthesizer, you can dial in a precise envelope. Good video btw!
Fascinating flavor of sidechain, thanks for the lesson! I'm not an experienced mixer, but what's wrong with a basic multiband sidechain? Is there some sort of overall volume advantage so that we can make "cigar" mixes?
I was always a bit sceptical about this RM sidechain, feels more like flexing your knowledge than actually doing something that makes your music sound clean. I don't even think it's that good for the sake of making mixes louder, is that crunchy result worth a hustle?
Nice twist to do sidechain ducking. Never thought about this. Thanks Polarity for the constant inspiration!! I will give it a try, but will stick with the CLip to Zero (CTZ) approach which I discovered during Covid on Baphometrix´s channel. Change my way of mixing and mastering 100%. And Baphy´s tutorials where the reason why I said goodby to Ableton at that time....And without THAT I would not have discovered Your deep Bitwig insights, Polarity! 😀
Curious to know what you think. I use a 2 band linear split(sub frequencies + everything else) and use ringmod sidechain on the sub band but I also use regular envelope sidechain at the end for each band. Do you think this is fine or is it overkill?
@PolarityMusic I probably should have mentioned that I do this for dubstep(where we typically don't like overlap)😅 but yeah for other genres that would sound very unnatural. Unless you think there's an issue even in the case of heavy electronic bass genres 😄
@@PolarityMusic Sure I did, but that still doesn't explain why a more complicated approach to side chaining would be needed. I understood when ducking plugins became popular, because sidechain compression is not really ideal and not that precise, but I don't understand what you cannot do with normal ducking to justify all of this hassle.
I really do not understand why people would want to use the ringmod sidechain technique because it is utterly useless and sounds like crap. It seems like rectifying the audio rate signal is cleaner
