next »

[Daid]

Let me start background first:
-I'm a programmer
-With many other skills
-I'm actually a bit competent with photoshop/inkscape/3D studio max
-I've played an instrument (keyboard, not a toy one, a Yamaha 740 PSR to be exact)

Now I want to see if I can expand my skills with some limited music making skills. Now, I've played music, always from sheets, so I can read/understand sheet music. I also have a bit of music theory in the pocket. And I found http://www.musictheory.net to learn the missing bits. (I'm having troubles remembering all the names for everything, but I can look those up)

Now, I've also identified the software I like using. Which is Musagi. And I understand how it works.

But now I'm missing 1 bit. And it must be obvious somewhere, but I guess I miss the "natural music sense gene". Because I cannot seem to get from "music theory" -> "music that doesn't sound like crap". Any tips for me? I understand music takes time to learn, and I won't produce Cave Story  like music quickly. But I want to try to create something that people won't turn off when it's playing behind a small game.

Yes, I know there are many people who will create music for me. But I want to try myself.
Oh, and I can seem to create a half decent drum line with musagi. So I do have a sense of rhythm I guess. But the melody is where it's failing.

[bauer]

Hehe, if you know how to play sheet music on a keyboard, you should be screaming out crazy melodies in no time!

If it's just about "coming up with melodies" then it could be an idea to steal a chord progression and rhythm (simple drum pattern) from some song you like, and then jam on top of it to try to come up with a melody. This is of course easiest if you have a keyboard with midi support, but also doable by playing on your computer keyboard.

That's usually how I do it when I'm out of musical ideas.. setup a simple rhythm and chord sequence and then pick a scale and jam along, increasing the "crazyness" over time (i.e notes out of the scale, switching scales etc). Usually something sticks and I'm able to expand from a small idea that I can use to build up the rest.

[pgil]

First of all, it's hard to judge where you need help without hearing your music.  I'd suggest posting a couple of tunes.  Even if you don't think they're very good, it's nice to let others see where your strengths and weaknesses are.

I don't really know how one learns a sense of melody. I can't read sheet music, and have always struggled to play instruments, but I picked up composition pretty naturally. I often get ideas while messing around with a guitar or keyboard. I'll hear a couple of bars that I like repeating, and so I try to develop that into a song.  I never learned any formal music theory, but I'm constantly thinking about music.

Don't use Cave Story as the standard for good music. Not that it's bad or anything, but there's so much more you could learn from. There's thousands of cultures, and thousands of years of history that you can experience through music. Listen to as much music as you can, even music you don't like.

Edit:  so, yeah, what bauer said.

[s0]

Yeah, I jam on my guitar all the time, and when I play something I think could be used for a song I'll transcribe it into my sequencer software.

But usually, I don't necessarily start writing a song from a melody or rhythm or w/e upwards. I start with a vague idea of what I want it to sound like and work from there.

[MoritzPGKatz]

Hello there,

Best advice I can think of: Listen to a lot of music you like!

The rest should come naturally. Music theory is there to help you organize the ideas your mind comes up with, not to create. (with some exceptions, e.g. abstract music, Neue Musik, some minimalism, 12-tone...)

There are some creative exercises you can do though where theory helps: as bauer suggested, take some changes, could be from a pop song or jazz tune or your favorite cave story track. Now pick out the notes you want to play: e.g. just play the fifth of each chord.

Or do single note lines (a popular jazz exercise and improvisation technique) - start with one note and keep that note played as long as it sounds good to you. As soon as it sounds dissonant, take the note one step up or one step down. Repeat ad nauseam.
Next step would be to do arpeggios.

You're bound to come up with some great melodic ideas that way. Have fun experimenting!

Cheers,
Moritz

[Daid]

The rest should come naturally.

Sadly, it doesn't seem to come for me. I've played keyboard for a very long time, and I can hear when it sounds wrong, but not why it sounds wrong.

I'll keep trying, maybe I just need more practice.

[MoritzPGKatz]

I might be going out on a limb here, but you're probably just doing too much.

Show us some of your stuff!

[Daid]

I might be going out on a limb here, but you're probably just doing too much.

In some way I was. I found out a few things I was trying sounded a lot better with short pauses in them.

Show us some of your stuff!

"Show" is kinda a misnomer with music right? I might show something when I know what I want to make. I've just been trying a few things until now. Not sure what I wanted to make. But now I have an idea for a tune so it might resolve into something.

[MoritzPGKatz]

I might be going out on a limb here, but you're probably just doing too much.

In some way I was. I found out a few things I was trying sounded a lot better with short pauses in them.

As my guitar prof used to say: "Rests are music too."

Listen to John Cage's 4'33" and you'll see what I mean.

[1982]

The rest should come naturally.

Sadly, it doesn't seem to come for me. I've played keyboard for a very long time, and I can hear when it sounds wrong, but not why it sounds wrong.

I believe that is doing it by "naturally". All my personal music composing is based on that. I hear when it is wrong, but I might not know or even bother to analyze why. Usually it is about some notes clashing harmonically in unintended way. Sometimes it is different instruments clashing. If you have the skill to hear the music, without really knowing the theory, you can do pretty much anything. Then there are those people who can play and compose through their knowledge of music theory (and playing skills), but I wonder that it alone might be a bit boring...

[MoritzPGKatz]

Knowing music theory will help you to do the following better:

- Working faster, playing what you have in mind on your instrument / recording that
- Communicating with other musicians
- Copying arrangement styles
- Transposing (and no, I don't mean just putting the notes on the MIDI roll a few steps higher or lower)
- Improvisation beyond the scope of pentatonics
- Bragging on the internets, ticking off fellow musicians

Theory won't help you enjoy the music emotionally though. And it might sound cheesy, but that's the very essence of it.

1. Discover new music, be open-minded,
2. Reflect on the music you heard. Why did you like it/didn't like it?
3. Play with other people,
4. Record yourself a lot.
5. Compose. Produce. Get Feedback.

Repeat these steps in an infinite loop.

You'll know when you're creating something good: it's a special feeling, and people will hear that, no matter what genre.

[1982]

- Working faster, playing what you have in mind on your instrument / recording that

-I don't have anything in my mind, it just comes when I start touching keyboard.

- Communicating with other musicians

-Those are douches.

- Copying arrangement styles

-Copying is boring, but taking inspiration doesn't require music theory.

- Transposing (and no, I don't mean just putting the notes on the MIDI roll a few steps higher or lower)

-If you don't mean that, then I don't know what and I don't care.

- Improvisation beyond the scope of pentatonics

-Improvisation withing the limits of madness.

- Bragging on the internets, ticking off fellow musicians

-Just present bad ass music and screw them.

[s0]

i don't think it's necessary to learn music theory to make like chiptunes or anything else related to (popular) electronic music. that stuff is pretty much music for non-musicians and has a long and artistically prosperous history of being made by people who can't even play an instrument.

i studied musicology for a while but i try to not to think of theory when i make music (and esp not when i play guitar) and just follow my intuition.  

[medieval]

The trick when you don't know what to do is just to do something. make a song out of 2 notes. Make a loop of 1 second. You might not like the sound of it, but.. baby steps. Don't think too much about how music is supposed to sound, instead think about what sound you want. Whatever you feel like. Let it come naturally

[Chromeleon]

I'll keep trying, maybe I just need more practice.

This is absolutely the single most important thing you can possibly do.

That

that's been floating around recently nails it; "It's only by actually going through a volume of work that you actually are going to catch up and close that gap, and the work you're making will be as good as your ambitions."

You mention having many other skills;  think about how long it took you to get to decent with any of those.  Your time spent practicing performance can help inform certain choices, but ultimately it does not carry over as practice composing.

Keep making new pieces.  Keep looking for ways to improve older ones.  Keep working at it.

[krasse]

Although the name of the book "The Complete Idiot's guide to Music Theory" is repelling, I can strongly recommend it.
It is actually very good even after studying some hardcore voice leading and harmony. The next book in the series "... guide to Music Composition" is also very good.

I can also recommend "Music theory for computer musicians" and "Composition for computer musicians" (Hewitt) which is more directed towards the piano-roll style and modern composition tools.

Good luck!