Dubstepforum.com

Title. How do you take your ideas and evolve them into full songs? This is one thing i seriously fail at. Any advice would be great.

I usually battle other pokemon.

Take your first theme/idea/riff/whatever and make variations of it and just keep messing around and making whatever you think sounds cool that is a derivative of your first idea. Just keep trying new variations on the same thing until its different enough to stand on its own or you have enough of what you need to make your original part "evolve". If you get enough things, you can pick and choose the best parts. Kind of like movie editors. They shoot all these scenes, but some of them they decide to scrap for the sake of the whole movie.

My songs usually start out as me experimenting with an idea, it could be anything. My most recent is me discovering music theory *derp* lol. So now that I have this thing from this idea, if I like it I decide to expand upon it and try to implement even more ideas into it.

ChadDub wrote:My songs usually start out as me experimenting with an idea, it could be anything.

This.

I usually get a sub down first and then start building up layers.

Make bassline... this dictates the entire vibe the tune will have, nasty, deep, dark, bright, techy, minimal, dancefloor, headphone...

Vary shit... oh its getting boring? 4x4 time with the drums, change the bassline, switch the hats... constantly have something changing, or rather, create the effect that it is... If you make four 16 bar hihat patterns, And two 8 bar kick snare patterns, and layer them up, its like have 64 unique patterns if your sequencing right... spend some time thinking about this shit... Constantly vary your bassline too, even if its something subtle like automating the resonance a bit here n there and all that, just switch some shit up so its not the same patch constantly, it has some life to it... same with leads and stuff...

Loads of white noise sweeps. have a tiny sweep every 2 bars, a little one every 4, an average one every 8, a big-ish one every 16, huge one on the 32... and just let em layer up with each other... keeps listeners on the edge of their seats.

Make the bulk of the song, then sweat the intro. Chances are, noone will ever hear the intro, this is DJ music, but everytihng that comes after the in, has to be amazing.

Few random, somewhat cliche tips.

Slam my face on the midi keyboard while recording, take all those notes I hit in the piano roll and pick out which ones sound good together, lay them out horizontally, putt at piano roll to a massive or a fm8, pick my bass from my own library or make a new one, bust out some drums on my pads, for the buildup slOwly introduce different elements , dothe same for the second part of the song, then the outro opposite of the intro.

Brostepe track is finished, true story.

Artie Fufkin wrote:Take your first theme/idea/riff/whatever and make variations of it and just keep messing around and making whatever you think sounds cool.

I was about to scratch my head and say good question but this sounds about right, I kinda just GO, and a few hours later I have a lot more there to work with, I usually don't have to really think about it, I just push buttons and bam. Sometimes it helps to get some new sounds in the song if I'm feeling stuck. It is also all about taking a initial sound having it progress melodically or just in how it sounds over a stretch of time, subtle progression can do a LOT in keeping a tune moving. Sometime vocals can totally write the song, or keep a song going (as long as their tasteful, but I guess that's with everything in song writing, it's all about having good taste.)

@bevoh fucking ingenious solution

BevOh wrote:I usually battle other pokemon.

However it starts, it usually pauses with a drum loop, bassline, and a melody or atmosphere. At this point I copy out the loop into the length of a full track, and then start to edit to make it change as it goes. If I don't copy out to a full track size and instead try and make it "grow" to the end, I usually get stuck and don't finish it.