How did you learn to vary compositions, and not be repetitive?

[Radiowar]

Experiment.

All I have to add to that is also to be creative.

hmm, a bit controverisal but i like it

[timaeus222]

Alright.

Use a different time signatures.

Use a different tempo speed.

Experiment with different synths.

Experiment with the effects you have.

Experiment with samples you don't usually use.

These are the helpful phrases I found.

It's also a good idea to, like Moseph said, break down a melody or the source tune in many different ways throughout the song.

is an awesome example. Just study how many variations zircon put into Devil's Lab. And in 8 hours, no less.

There's also no inherent wrong to changing genres in the middle of a song, like Kristina said. It helps to change the drum rhythm to adapt to a different "genre". For example, DnB into halftime, then into a breakdown, then downtempo, then build back up to DnB.

[Ryan Jobson]

After 3 years I still don't think I've really figured this out...

[Chimpazilla]

For example, DnB into halftime, then into a breakdown, then downtempo, then build back up to DnB.

Hhhhhmmm what could he be working on here???

[AngelCityOutlaw]

For example, DnB into halftime, then into a breakdown, then downtempo, then build back up to DnB.

Thing is, those examples are all electronic music.

That's the equivalent of saying "This part of my track was death metal, this part is power metal and this part is folk metal." They all have so much in common stylistically that it's almost like you didn't change genres at all.

[dannthr]

Variety is like spice. If all you ever cook is one type of cuisine, then it will always feel repetitive. If all you ever eat is one type of cuisine over and over again, all you'll know is the same thing over and over again, and your cooking will become formulaic and rote. This is something to avoid if you're trying to create something fresh or interesting.

It's not about trying tricks, that's more formula, it's not about eating one other type of cuisine, that's just another formula.

Let me put it more concisely: Your personal listening experience is what informs your spectrum of ideas. Your personal listening experience is narrow and limited if you're only chasing after the same shit over and over again--whether it's just copying and pasting the same drum pattern over and over again, or if you're using the same key over and over, or if you're using the same instrumentation over and over--it doesn't matter what the problem is, the problem is you, your mind, and most importantly, your engagement in variety and commitment to improving yourself.

You need experience and that means studying shit you never studied before.

The most valuable studying you can do is listening and analysis.

You have spent so much time building the scaffolding of a comfortable little box. This box is a safety, it's what you know, it's your comfort zone, and it provides you a foundation for how you approach music. Who knows how you came to design this box, you did it yourself, maybe you just built it over time and realized "hey, this works, let's keep doing this."

It doesn't matter how you came to build your box, you're stuck in it.

You need to expand your experience and what you think is possible, appropriate, or even what works. I will let you in on a secret: there is no end to this experience, it is as vast and as expansive as human knowledge (as limited and growing also) and you will never in your lifetime experience the entire breadth of what music has to offer, your life is too short.

But music has a RICH and DEEP recorded history and empowered by the Internet, you can explore this diamond mine of possibility. You just have to be willing to step outside of your box.

At first, it will be scary, you haven't been outside of your box since you built the one you're in now--so think back to what that was like, before you knew even as much as you do now. That's what it will be like. It will require a person with an open mind, a willingness to learn, and a spirit of adventure. Including roughing it through stuff you maybe didn't like before--well, here's the real trick to learning music: Everything can be liked, you just have to study it to find out what the good parts are.

After you've gathered enough "good parts" you can return to your box and build additions or even build new boxes.

For me, about 10 years ago, I left my little electronica hut behind and started building a modern orchestra house. Now I look back on that little electronica hut, and the shit I thought was impressive back then, what took me years then to develop and build, now I can build blind-folded in a day.

For me, I am always looking to the new horizon--my education NEVER ends and I hope it never does. I'll be learning as long as I live because to me that's what living is.

This is not a problem that will be fixed by tricks or method, this is a problem that will be fixed by a change of attitude and mental state.

Listen to everything.

[Frederic Petitpas]

Take a walk in the woods, find a Y, go right, write a tune at the end of the day.

The next day, just go left...

[Rozovian]

I suggest you approach it from the opposite end of the problem. Some people have the opposite problem - their tracks end up being an eclectic mess of melodies and ideas that don't fit well together. You can solve that problem by throwing stuff out and only keeping the stuff that works well together, whether it's part of the writing, the instrumentation, or something else. You end up with a varied track that has plenty of variety.

In your case, consider taking a break from making cohesive, sensible stuff and make a track that's just any crazy idea you have, thrown together. It'll get your creative juices flowing, and you should end up with a gallery of ideas that you can piece together into a more cohesive track once you get rid of the random junk (save it somewhere tho, it might be useful for another track).

Basically, go nuts with random ideas and make something that doesn't have to be the least bit cohesive. Then see what parts might actually make a cohesive whole in a different context. Use those.

[Palpable]

I find a section of a song I love, play it over and over, analyzing it and figuring out what I fundamentally love about it. Then I shrug my shoulders and steal it outright for my song. The trick is knowing what to steal. (Only half kidding.)

[timaeus222]

Thing is, those examples are all electronic music.

That's the equivalent of saying "This part of my track was death metal, this part is power metal and this part is folk metal." They all have so much in common stylistically that it's almost like you didn't change genres at all.

The difference between what I said and what you said are the extremes. Yes, it's electronic music, but those genres aren't the same genres at all. They all have different drum rhythms. One of the characteristic parts of any genre is the drum rhythm, a lot of the time. They only work together if there are proper transitions.

I will say that Death Metal, Power Metal, and Folk Metal are similar enough, but your earlier example of Metal, then Dubstep, then Country was too extreme to be practical. You can go Metal and then Dubstep+Metal fusion, but we both know shifting to Country really doesn't make any sense.

In your case, consider taking a break from making cohesive, sensible stuff and make a track that's just any crazy idea you have, thrown together. It'll get your creative juices flowing, and you should end up with a gallery of ideas that you can piece together into a more cohesive track once you get rid of the random junk (save it somewhere tho, it might be useful for another track).

That actually worked for me on the few times that I tried that. Gunstar Heroes, for example.

Edited August 6, 2013 by timaeus222

[Infenro]

Thanks for all the replies guys! This was all incredibly helpful!

I'm sorry I would've responded sooner but I've been sick the last 2 days, but all this definitely helps!