Turning an idea into a Song, how do YOU do it?

[Dabrenn]

I very often times can create a short 4 to 8 bar loop that I really like, but I always have problems extending a loop and creating a polished song out of it, in fact, even after about ~7 (really more like 4) months of sparatically producing (I've been really busy for ahwile) I wouldn't say I've ever made a song I would consider totally "finished".

Maybe I'm approaching music creation wrongly, but how do you guys go from idea/short loop to interesting completed songs? I'm very curious because I don't think what I am doing is totally working, even though I'm getting better slowly but steadily.

Do you start with a loop? do you compose completely linearly? Feel free to share any of your own thoughts or add questions to the mix, I'd like for this to be more about discussion rather than answering a question of mine.

[Motivfs]

Like I've been told (but don't listen to for some reason), that when you make a loop, you build on it, and you, integrate it as the middle of your song (the main section) Then you build from the middle, atleast that's what I've been told.

I have the same problem, I build half a song, but then become lazy/unproductive or just don't get any more open-minded and build on it, just copy/paste which is a problem I've started to see reoccur. I want to get more diverse with my music as well, so I'll be keeping an eye on this thread.

Sometimes I find adding random melodies in works quite well, what I usually do is hum a melody while the lead is playing, and then throw it in there.

[Dabrenn]

just copy/paste which is a problem

. This is my problem. I recently did an experiment where I had one loop and literally copy and pasted to make an entire song in an extremely short period of time, just to see what would happen.

This was the result.
http://soundcloud.com/dabrenn/house-in-1-5-hours


Unfortunately, this is very indicative of the rest of my creations, even though it was an intentional extreme.

[Motivfs]

just copy/paste which is a problem

. This is my problem. I recently did an experiment where I had one loop and literally copy and pasted to make an entire song in an extremely short period of time, just to see what would happen.

This was the result.
http://soundcloud.com/dabrenn/house-in-1-5-hours


Unfortunately, this is very indicative of the rest of my creations, even though it was an intentional extreme.

What I figure the hardest part is, is to break from the melody/sound that you've created, it sits as more of a problem, then enjoyment.

I mean you love it so much, yet you are scared to build around it, ruin it, that's what the problem really is. I still find myself having trouble to break out from it, hopefully we can both in due time do that.

As for your song, indeed it sounds similar to what I do.

[Dabrenn]

I mean you love it so much, yet you are scared to build around it, ruin it, that's what the problem really is.

That's actually very insightful. This is an excerpt from my current project and when you put it that way, that's exactly what I'm doing, I'm scared to ruin it.

http://soundcloud.com/dabrenn/vocals-experiment


For some reason, that little loop meshes with me, but I have absolutely no idea where to go with it.

[Motivfs]

I mean you love it so much, yet you are scared to build around it, ruin it, that's what the problem really is.

That's actually very insightful. This is an excerpt from my current project and when you put it that way, that's exactly what I'm doing, I'm scared to ruin it.

http://soundcloud.com/dabrenn/vocals-experiment


For some reason, that little loop meshes with me, but I have absolutely no idea where to go with it.

So this is just me, and note, I might have it in the wrong key considering I never played Piano or Guitar my whole life, but I think it's in A# Major? Anywho, you'll get the idea regardless, this is how I would've went with it. Oddly enough, it's my kind of style, a very chill beat, so I can work with that.

http://soundcloud.com/motivfsmusic/mlr-dabrenn-loop-example

Arpeggios are always a good way to start, but just don't overdo it like I do, can ruin the song.

[Kopachris]

1. Write random stuff
2. Edit it until it sounds good
3. ???
4. Profit!

Seriously, though, it seems like that's how it's always worked. Even the great master composers of old such as Beethoven, after they came up with an idea, had to work for ages to make a whole piece out of it. Beethoven sketched numerous different versions of various passages for his 5th symphony, sometimes up to 20 completely different versions of the same passage. Even in his "final" handwritten score, there were tons of parts crossed out and rewritten--the ending of the first movement was completely rewritten twice in the final score, and probably more times before he started working on the final score.

[Captain Ironhelm]

Everyone has their own unique way of creating a song; there's no one right way. Here's one method:

Create your melody.
Build the most complex (climax) part of the track first with all your elements.
Mix the elements of your short section of music until it is cleaner than a bar of soap.

This makes it so you can just drop in new sections and rearrange and expand the track to however long your imagination can take it because all of your elements are already mixed perfectly.
(Variation!! Change a minimal element every 4 bars, and something major every 8 bars, for instance. Change up the synths, the drumloop, add automation/effects, add sweeps and other effects that tie sections together, make a calmer section, add syncopation, key change, just use your imagination!)

If you start in the beginning with the buildup and intro, and continue to produce more complex parts, you will find yourself frustrated with trying to cram new elements into the mix. Great way to stress yourself out!

[Motivfs]

Everyone has their own unique way of creating a song; there's no one right way. Here's one method:

Create your melody.
Build the most complex (climax) part of the track first with all your elements.
Mix the elements of your short section of music until it is cleaner than a bar of soap.

This makes it so you can just drop in new sections and rearrange and expand the track to however long your imagination can take it because all of your elements are already mixed perfectly.
(Variation!! Change a minimal element every 4 bars, and something major every 8 bars, for instance. Change up the synths, the drumloop, add automation/effects, add sweeps and other effects that tie sections together, make a calmer section, add syncopation, key change, just use your imagination!)

If you start in the beginning with the buildup and intro, and continue to produce more complex parts, you will find yourself frustrated with trying to cram new elements into the mix. Great way to stress yourself out!

Exactly what I do... Start at the beginning, which is very painful. After this last song I'm finishing I think I'll finally stick to doing the Climax/middle section first, then work from there. Alot easier.

[Dabrenn]

So this is just me, and note, I might have it in the wrong key considering I never played Piano or Guitar my whole life, but I think it's in A# Major? Anywho, you'll get the idea regardless, this is how I would've went with it. Oddly enough, it's my kind of style, a very chill beat, so I can work with that.

http://soundcloud.com/motivfsmusic/mlr-dabrenn-loop-example

Arpeggios are always a good way to start, but just don't overdo it like I do, can ruin the song.

Wow, That's really kind of you to take the time to do that! thank you! Those faster Arpeggio's definitely sounded enticing. I think I tend to not create climaxes with enough elements to stretch into full songs since I tend to value a lot of simplicity, though there are exceptions. I will definitely be incorporating what you suggested, thanks again!

I was actually working on creating two separate loops with a similar sound to turn into one song for this particular project, which is something I have never done before. I think it may work well since I will have more elements to tug from.

I'm still working on this one loop, I don't like it as much.

http://soundcloud.com/dabrenn/vocals-expirement-loop-2

[Motivfs]

Wow, That's really kind of you to take the time to do that! thank you! Those faster Arpeggio's definitely sounded enticing. I think I tend to not create climaxes with enough elements to stretch into full songs since I tend to value a lot of simplicity, though there are exceptions. I will definitely be incorporating what you suggested, thanks again!

I was actually working on creating two separate loops with a similar sound to turn into one song for this particular project, which is something I have never done before. I think it may work well since I will have more elements to tug from.

I'm still working on this one loop, I don't like it as much.

http://soundcloud.com/dabrenn/vocals-expirement-loop-2

Take your time with it, just keep experimenting, it took me 6 hours recently to get a Drum & Bass liquid loop going, and actually get the middle section prepped good enough to work from it. It sometimes just happens magically, and sometimes just doesn't happen from what I've been through.

Like I said, one thing I do that helps alot is hum a backing/support melody while the lead plays, helps so much, gives me an idea of what works and what doesn't.

And with a chill/relaxing beat like that, it's good to think with simplicity, like Ironhelm said, just change up melodies/drum patterns every now and then, it's about keeping the listener entertained.

[Lavender_Harmony]

With hours and hours of stressful, tedious proceedures. I'll be grey by 30 at this rate.

But seriously, my ideas are varied. Sometimes they flow out like liquid out of a cup, and sometimes they're, complex and hard to achieve, almost maddening at times, and often I'll abandon an idea for a few weeks/months and come back to it later.

I never start at a specific point, and I never finish at one either, I work differently for every single project, I have no workflow, I have no order I do things. It's as random as can be.

[Rough Lungs]

What I do is , build up a drum rack get the drums going to a decent nodding head beat . Add a bass in the back ground then make the synth in according to the bars . But a good beat and bass can often be just enough

[itroitnyah]

I don't have any idea what a song is going to sound like before I start it. I have a slight idea of the STYLE of music I want to make (dirty, hard, bass-y, etc.) before I start, but I just open up piano roll and for rhythm, I put down chords until it sounds good, and for melody I place down notes, almost at random, until it sounds good.

[Freewave]

To me the idea and the concept of the song IS the starting point. How is the song going to relate or connect to the show or to the aspects of the show. So THAT starting point to me is the most critical and that's not something that's generally musical in nature. I don't generally start with melodies first although i can see that some people make music that way. The genre style is rather important tho in deciding what direction i want to go. I also don't make music the same way every time as each song has its own cycle of creation and some are a breeze and others are kinda painful.

[Nine Volt]

I start by thinking of an idea for a song, and what general genre I want it to be. I start with the drums almost always. Then I start to add in different synths I know i want to use. Then I make chords, then a melody. I don't always end up staying with my original idea though.

An example would be my WIP song, Blorpx. It was originally intended to be a darkstep/neurofunk track. I had a heavy, dark reese and a cool funky organ sound (similar to one of the synths in Noisia's Split the Atom). But then I made some cool chords with that synth and I kinda let those overtake the dark reese, which started not fitting, so I removed it and replaced it with an almost guitar like reese thing.

I build the track up linearly, and I try to have at least one noticeable change every 8 bars.

[Sonarch]

http://soundcloud.com/dabrenn/house-in-1-5-hours

I haven't listened to the full thing yet, and I am much better at this kind of thing in theory than in practice, but the bit that comes in around the 16 second mark, you could always make into a bit longer of a looped progression by having the original bit, and after that have it start on the same first note, go to a higher note than the first one, and come down to a different note. Still following the pattern of the first part, but varying the notes a little.

As far as my own method, I don't really have one yet. I've got tons of ideas, and i'm working on arranging some brony songs for a classical guitar, but as for electronic stuff, i'm really just learning about how to do it from tutorials and reading advice threads on MLR. I get tons of inspiration when i'm unable to work on music, like when i'm at school or in the car, but I think there are several ways you can go about making a song. If you have an idea or picture in you head about the theme or how you want the song to sound, start experimenting with it until those thoughts are adequately represented by your tune, or your chords, or your bassline, whatever, and build everything around it. if you're making a song for the sake of making a song, be it for one of these community albums, or to get experience, like me, then it could be good to simply lay down a drumbeat of some kind, jam/play around until you get a nice bass or melody bit, and expand from that. But I may be horribly wrong with all this, since i've never completed a song yet, so I don't know how applicable my theoretical ideas are.

Wow that turned out to be longer than expected. Hope I helped someone with this mess of a post. I should be doing homework right now ugh.

[Navron]

I start with the setting. Am I creating an eerie post-apocalyptic scene, an upbeat club, or a serene, peaceful landscape?

From the setting you establish the mood. Is it mysterious, dark, happy, floaty, atmospheric, etc? That's what you use to build your melody, and from there, the harmony and counter-melodies.

Think of it like directing a movie. You can't just throw a sci-fi character into a WWII background. Likewise, you aren't going to have a super happy melody with a very minor and dissonant pad and the distant sound of suffering voices in your song.

You have:
Background (Where are the characters at?)
Middleground (What's in the background that interacts with your characters?)
Foreground (Who are the characters themselves?)

You're telling a story. Even some of the most generic club songs have some kind of story being told through the music. So first think about your story, then think about turning it into sound.

[vladnuke]

I just go crazy with it, man. I mean, not always, but sometimes I just turn on the computer one day, realize I don't have anything to do today, and I'm up on it. I start at the beginning and end at the end. As soon as I realize I'm painting myself into a corner, I take a nosedive to end.

Not so much with remixes, though. They take a bit more to prepare, and give me some time to think.
I can just find the establishing shot, work on that first, cut back to the intro, work on that, then chuck all that out, put it back in piece by piece, and fix it up.

But most of the time, when I'm away from my compy or have too much to do, I just sketch out the idea on piece of paper. A lot of them don't work out, but some of them are just wonderful. "No sunshine allowed" was that one.

[Djeinus]

Well, here's how I do it.
1. Turn on Sibelius.
2. Pick a preset or pick random instruments
3. Pick a key and time signature.
4. Write some notes down.
5. Do they sound good? No=repeat step 4. Yes=Continue.
6. Go crazy, add some shit, change up the melody, add ninjas, go insane with the drums, don't give a damn just do.
7. End, discard or keep for later use.

[XXDarkShadow79XX]

I seriously just walk around and combine horrid noises in my head. (Though usually my ideas come from when I'm bored in school) Then I just come up with the rest when I sit down at my computer. Or sometimes I just screw around with massive until I make a cool patch to base my track around.