It's really nice, minimalistic! Thumbs up. I can imagine someone to rap to this.

...and name them properly otherwise you end up like me, having bunch of custom presets with no f***ing way to navigate yourself to the right one when needed.

Someone once said "music theory is descriptive, not prescriptive". It doesn't tell you what to do, it just describes what's already there, and gives you the tools to do more with it. So if someone tells you it "limits their creativity", tell them they're full of it.

This.

Music is essentially brain having a good time with math around the frequencies it hears. What this "math" is and how it works, that's what you learn in music theory. But there is a point where you should stop. Once you feel you "have to do this or that becouse you were tought this way", you should stop learning. ...to preserve some creative freedom.

...also it's not "required" to know theory. Freaking Hans Zimmer once told to deadmau5 he has hard times reading notes. But it's really useful as you get to know what's actually going on in your track.

On a side note, I always wondered this: There's always this condition in remix competitions to use at least one stem. So ... if you sample a kick drum, or crash cymbal, you can enter with original song as long as this little sample is in?

I'm probably not someone who should give advice on this topic, lol, but I think better you mix the track, less it need when mastering. Owsinski's books are great for that. Esentially you wanna look for balance, stereo width/placement and dynamics. In our generes, we kinda kill dynamics by limiting, though, so the real job is to get it spectrum-wise and width-wise on the similar level as records you wanna compete with.

I don't like all that "woodoo magic" of analog models, tape saturators and stuff. It messes up with frequency spectrum, adds wow and flutter and subtle saturations. Don't get me wrong, those are good things, but I think it's better to apply them while mixing or even better, while sounddesign stage.

So my chain: EQ, Width Expander if needed, Limiter. All in O-Zone, quickly out. (But again, I'm probably not the best example.)

...to watch a lot of Seamless R.

No really. It got it's roots at Reese bass (do I wrote it correctly?), which was originaly set of two really detuned oscilators phase-canceling each other. Widely used by good old dnb scene. By filtering, distortion, and use of some crazy wavetable oscilators (Massive used to be / still is popular for this.) you can achieve a lof of those crazy neuro sounds. Today's synths got few more tricks up it's sleeves though. Serum has really cool wavetable options including awesome editor. Harmor has amazing resynthesis and per-harmonic unisono detune. ...and so on.

There's really no one way to do it. Experimentation is the key. Watch Seamless R and you'll hear what I mean by this.

Drums are hard. Especially now when there is A LOT of cool complex sampled stuff out there. Each drum has it's own way to be made.

Kicks have big fundamental sinewave with pitch doen envelopes and bunch of layers to fatten up the attacks. Snares has that fundamental too, but higher a bit and lot of longer layers with distinct set of resonators and kinda random noise (which is in the real world made by set of springs underneath a real snare.) Hihats have those resonators too, but way up there in high frequencies. It's really dificult and complex so you can spend days, weeks, months by unsuccesfull trying. (Been there.)

Also, a lot of those sounds that sound synthetic have it's roots to sampled originals. Like on 909. Roland tried to synthesize hihats on 808, but it didn't sound good at all, so they recorded real hi-hat for 909. (Which got sampled and used as a base for a lot of today's samplepack's hihats.)

Also talking about pads, when not working with resampled ones, freeze them. Seems like nonsense, but it's vital. That way when you press play, the tail of "previously unplayed" notes is already there. Makes quick descisions way easier.