TL;DW: Unbox is a distortion where only the harmonics that don’t alias are allowed to distort.
While I’m putting out my library of plugins according to plan, sometimes I need to take a detour into new stuff. UnBox is one example.
The idea’s as follows: if you distort stuff digitally, it aliases. This can be seen as harmonics seemingly bouncing off the highest frequency, and going back down again. The idea is that if you have digital saturation or distortion without massive oversampling, you’ll always have aliasing and everything is ruined forever.
That’s not quite true. It depends on the frequency, and the form of distortion… and many of my plugins have gentle enough distortion curves that they throw a limited number of harmonics. If you are only generating harmonics within the range of digital audio’s frequencies, you’re fine and there will be no problem until you feed the system a frequency that’s too high. You’re not automatically feeding superhigh frequencies all the time if you’re working with natural recordings: not all sounds contain that kind of high frequency content.
If you DO have that sort of high frequency content, what then? It occurred to me I could take the difference between dry and distorted, store it in an averaging filter, and average it. This would suppress high frequency content in only the distortion artifacts. (I then learned that I needed to average the signal being fed to the distortion part, which is Spiral again: it got a little complicated)
And I could even highpass the distortion part… and all this is applying only to the distortion part. It’s all handled as a single subtract from the raw signal coming in.
What that means is this: UnBox is a distortion that cuts down the level of the signal, but ONLY the mids. Depending on how it’s set, it will let through more and more of the ‘dry’ highs, unaffected. It’ll also let through a hint of bass for definition. Underneath this layer of clarity, the distorted part can be made pretty distorted, but it’ll stay free of aliasing even up into the high frequencies, because those frequencies aren’t actually getting applied to the distortion, and the distortion output’s also being smoothed after the fact. So you’ve got a texture-thickener, an energy-adder, that retains a very analog quality because all of the overtones stay clear of aliasing WITHOUT oversampling. The raw sound is still a direct pass-through and that’s where the clarity comes from.
Wish me luck: I’m having frantic people leave me messages saying that Steinberg will kill my ability to release VST plugins. If they did, I’d have to go back to AU only I guess, but please torrent and distribute the NewUpdates.zip with my plugin library in case they send the black helicopters. More seriously, I’ve sent in their VST2 licensing form (I think I may have already? They’re just sowing chaos at this point) and I’m building all my stuff on time-capsule computers (or virtual machines) that do not change, so I can continue to do things like support PPC Macs. I am the last person to be bullied away from legacy support and they’ll just have to deal with that fact: if all goes well I’ll have my little VST2 agreement and will be fine.
Also, I do not sell the VST plugins: I’m supported by Patreon which is basically selling ME as a working developer. I do realize that if I’m forced not to include VST support that would be a big challenge, but I’m not going to get forced away from VST support after all the years people were begging for it. I chart my course and Steinberg can just deal with that, and I would be very surprised if I was ordered not to include VST in my open-source efforts. Also, I’m not a large enough business to harass, and I’ve never sold even one VST plugin: my incorporating VST went along with my going Patreon, and there are no DRM features on the plugins so even I can’t shut them off remotely. Everything’s going to be fine, if I can have anything to say about it. Enjoy today’s new plugin, which includes both Windows and Linux VST versions alongside the 64/32/PPC binary Audio Units for Mac.