Adding more instruments/elements...?

[Chew_Bear]

I usually start with drums and percussion first. That way I can establish a rhythm, groove and beat that I think best fits what I am trying to go after or emulate. After I establish the drums and percussion. I will usually add some FX type stuff (one shots, reverses etc). But it seems like I always get stuck here though.

Which element or instrument seems most logical to add next after you have established drums, percussion and some FX? Bass? Pad (ambience)? Lead?

I know most people will work according to frequency. That is, start adding instruments/elements from lowest frequency to highest frequency or vice versa. But I am confused as to which approach would be better/best?

I have been experimenting over the last few days and it seems that I like to go straight to working on a lead instrument with a lead melody, right after I finish the drums and percussion. Is this a good workflow? Should I start with a Pad or Bass instead since they occupy a much wider and lower frequency spectrum? Should I leave the lead instrument and melody last in my workflow, since it is the most important (i.e. save the best for last)?

In general...I am just curious as to what others work flow is like when trying to add new instruments/elements and which instruments you like to start adding after a certain point (e.g. drums/percussion).

 

[krushing]

Frankly I don't think "most people will work according to frequency", or at least I doubt people consciously think they're going to add a certain range of frequencies, even if obviously some things will end up that way (as in, "now I'll do a bassline"). While some people always go through a certain order, I at least try to break things up every now and then and instead of, say, starting with the drums, I'll play a guitar line or do a chord progression with a pad sound or an arpeggiated sequence, or take a random sound out of the library and process it into some kind of rhythmic or textural thing to begin with – in other words, starting with something that sounds inspiring and then build on that instead of going through a series of pre-defined steps.

So in order to answer the "should I" part, I think you should start with whatever you find cool. It's easy to end up doing the same thing over and over again because you think you "have to" start with drums, for example.

 

[localspace]

Yup there ain't never no set way to do nothing

For me personally it does not work at all to have this additive approach of 'first I make this, then add this and that'.
That's a fine way to work yourself in a corner where you might have these amazing drums, pads, chords or whatever... and struggle to find anything to match that awesomeness. Or you'll have really fat drums that sound amazing, but take up so much room you can't get your bass to work with them, without processing the drums again to where whatever made them awesome before gets diminished.
I guess a lot of people do work that way, considering how many tracks are completely structured around a single element... which is boring. That's also how you get stuck in the 8 bar loop thing, like: I have this groove, I have this variation to the groove.. now what? Let's call it the static approach.

I structure my tunes around transitions, moments of change. So while I often start with drums, that's simply because in my music drums are by far the most important element.
I kinda sketch in a few basic drum lines which could be a few pops and clicks at this point, or some 707 samples... nothing too big, or interesting. Definitely not interesting.
Next I try to find something that really sets the atmopshere up, a deep pad, something that does something interesting going into a big reverb. This usually sets the tone for the rest of the tune.
The next step is finding another sound, any sound.. that creates an interesting interaction or contrast with the existing ones. Like an extra drum or bassline that kicks things off when you add it in.

When I have that, I know this track is going somewhere and it's just a matter of taking it where it takes me. Everything up to this point is very simple... they're thin sketches which I draw out (or over lol)

In general, but it's key to my process and I don't see a lot of people do this... when I look for sounds and create patches.. I'm not looking for 'the perfect sound'.. anything but. If it's perfect already, anything you do to
it is only gonna make it less perfect. I'm actually looking for imperfections and weirdness I can amplify and create those interesting transitions with.. or a cool way to just destroy it if all else fails. If you start of with a really shitty or basic sound... say a little unfiltered nasty square wave, and do things to it to make it sound nicer you can use those things to create progressions and tension/release in your tracks. This is why (while I don't have anything against them technically, or ethically) sample packs are utterly useless to me. They're static sounds... they're too nice, the artistry is already built-in and my role gets reduced that of a very slow DJ, lol. My best sounds often come from just init-ing a particular synth on top of my track.. copying a few MIDI notes (from the bassline usually) and then shape the sound along with the track.

But at the same time I'm more than happy to slap on a preset or pre-made sample if it gets me what I want. Like on my track 'Pool', the chords come from a patch that loads up if you start up Reaktor's Akkord synth. It doesn't get more cliche, overused and easy.. lol. But hey, it worked bloody perfectly right away and it's a beautiful kind of sound that I had been hearing on great records for years, had gotten close to myself... and finally found the secret sauce to. I now figure all those producers simply used that exact synth for it, because it's been around for ages now. That as an example that, yeah workflow is hugely important.. but only to you and only so far as it gets you the results you want. No-one else is gonna give shit, they just want nice music and don't care how it's made.