Jump to content

Recommended Posts

I've been really struggling writing melodies recently. I used to write all of my melodies on my guitar, no problem, they would write themselves. Now I'm trying to depart from that and write on the keyboard or write by arranging samples, and it's really tough. Anyone got a nifty trick they use to get kick-started on a good melody? For example, would you start with a simple bassline and then start building on that? Or a simple keyboard lick first? It's weird how shitty my keyboard melodies sound, whereas on the guitar they always sounded pretty good to me. Maybe it has something to do with the layout.

Link to comment
Share on other sites

Make a chord progression first, have your melody start on the root note, make sure it doesn't clash with any of the notes in the chord progression along the way?

 

I dunno I'm shit at music in general.

Link to comment
Share on other sites

Probably not exactly what you're after but change ringing might be of interest. Pick some pitches and just go through the permutations.

 

...or yeah, just start with chords, most melodies are only as good as their implied harmony.

Link to comment
Share on other sites

ok I see. I guess I have a vague understanding of chords when I play the guitar whereas the chords I gravitate towards on the keyboard aren't the same and sound overly cheese to me. perhaps I shall look for a chart of piano chord progressions.... hmmm not a bad idea at all. thanks.

Link to comment
Share on other sites

oh and this is probably common knowledge but i seem to hear a lot of less-than-interesting melodies from people who have never realized that it's ok to play the same note more than once in a row. if you started on guitar you already know that, i'm sure, but man there are a lot of electronic musicians that just frantically put notes all over their piano rolls and end up with these disorienting and noodly melodies that can't seem to actually finish saying anything before hey look at that cat

Link to comment
Share on other sites

oh and this is probably common knowledge but i seem to hear a lot of less-than-interesting melodies from people who have never realized that it's ok to play the same note more than once in a row. if you started on guitar you already know that, i'm sure, but man there are a lot of electronic musicians that just frantically put notes all over their piano rolls and end up with these disorienting and noodly melodies that can't seem to actually finish saying anything before hey look at that cat

 

+1

Link to comment
Share on other sites

forget chords. turn on record, and just improvise melodies. do that for 10 mins.

then listen to what you played. every time you hear something you like, put a marker in the sequence.

go back, and cut out all the melodies you liked.

start stringing them together in different combinations.

 

repeat, with your newly constructed melodies looping in the background. write counter melodies to your established ones, in the same way.

 

repeat until you feel like you have enough.

 

i personally find writing with chords to be confining. I used to do it all of the time, but it is something I actively tried to avoid. if something is odd later, I can analyze my lines, and see what harmonic motion I'm implying, and adjust things from there. just use your ear... you'll get more aesthetically pleasing results, imo.

Link to comment
Share on other sites

forget chords. turn on record, and just improvise melodies. do that for 10 mins.

then listen to what you played. every time you hear something you like, put a marker in the sequence.

go back, and cut out all the melodies you liked.

start stringing them together in different combinations.

 

repeat, with your newly constructed melodies looping in the background. write counter melodies to your established ones, in the same way.

 

repeat until you feel like you have enough.

 

i personally find writing with chords to be confining. I used to do it all of the time, but it is something I actively tried to avoid. if something is odd later, I can analyze my lines, and see what harmonic motion I'm implying, and adjust things from there. just use your ear... you'll get more aesthetically pleasing results, imo.

 

ooh going to try this right now

 

I usually just go da-da-da-dada-dada-etc in my head & figure out what those notes were.

 

I can only do this after I get started with something, like writing the harmonic parts

Link to comment
Share on other sites

forget chords. turn on record, and just improvise melodies. do that for 10 mins.

then listen to what you played. every time you hear something you like, put a marker in the sequence.

go back, and cut out all the melodies you liked.

start stringing them together in different combinations.

 

repeat, with your newly constructed melodies looping in the background. write counter melodies to your established ones, in the same way.

 

repeat until you feel like you have enough.

 

i personally find writing with chords to be confining. I used to do it all of the time, but it is something I actively tried to avoid. if something is odd later, I can analyze my lines, and see what harmonic motion I'm implying, and adjust things from there. just use your ear... you'll get more aesthetically pleasing results, imo.

This is quite good advice.

 

I wish I could give some advice, especially since in my opinion my melodies are pretty AWESOME (hey, I'm not gonna lie, I write for my own satisfaction). But unfortunately I've become so accustomed to writing melodies the 'backwards way' in a tracker that I'm actually better at doing it that way than improvising on guitar or keyboard..

Link to comment
Share on other sites

I used to write all of my music exclusively via drawing in the piano roll. I've never been much of a performer... but I started realizing the stiffness that resulted from that approach. forgetting about being a "performer" I would just try to noodle things on the keyboard. the result, was much more musical phrases, that were more nuanced and varied than anything I could do simply thinking about chord sequences and placement on a time-line. like I said, I'm not much of a performer, so this process does rely heavily on going back through the mess of an improv that you just played, and picking out the gems. inspiration can be short and fleeting... thank god we have MIDI record.

 

I also get inspiration from semi-random MIDI stuff... I'll build a little MIDI system on my nord, or I'll put arpeggiators on a track in live, and mix and match MIDI effects, and then record those MIDI events to another track. let it run for 10 mins... there are all sorts of interesting melodies that come out of randomness. the key is to find the gems, and that is the artistic part... much like finding an awesome sample/breakbeat... the art is hearing a moment, and re-contextualizing it. it may feel like cheating... but its not. this is what composers did before computers... they would find a million ways to mutate an idea... inversions, retrogrades, canons, etc. it's all just techniques for finding new inspiration... none of these composers sat down and in one take composed these dense compositions without thinking about how to create new variations and combinations. we now have computers that make that task much quicker... it just requires a musical mind to do it in a way that "works"

 

dont forget... art is telling the truth with lies.

Link to comment
Share on other sites

also, regarding chords...

 

ive found many times in my noodling, that I'll do something that sounds really awesome... and then I'll try to analyze it with music theory, and have been lost to what exactly I did, or why it works. I'm sure with enough effort and time, I could find some theoretical justification for what I did, but honestly, who cares? It sounds good, and I never would have come to that conclusion had I focused on harmonic progressions as a starting point. analyzing your music should be a last resort, when you are beyond stuck.

 

imo,

Link to comment
Share on other sites

also, regarding chords...

 

ive found many times in my noodling, that I'll do something that sounds really awesome... and then I'll try to analyze it with music theory, and have been lost to what exactly I did, or why it works. I'm sure with enough effort and time, I could find some theoretical justification for what I did, but honestly, who cares? It sounds good, and I never would have come to that conclusion had I focused on harmonic progressions as a starting point. analyzing your music should be a last resort, when you are beyond stuck.

 

imo,

 

Depends how interesting you want your music to be... most theorists make rubbish composers because they lack intuition and a musical ear, but the best composers were able to expand and develop their musical language through a practice of self-reflective analysis. It's what makes Bach better than Vivaldi, Stravinsky better than Shostakovich etc.

 

Noodling is fine, I do it a lot myself, but there's something wasteful about it and I certainly wouldn't recommend it as the best way to create interesting music, especially if you're in any way proficient as an instrumentalist. It can be difficult to stop your hands from going naturally to the positions they know from previous experience and it becomes as much a process of muscular memory as genuine musical creativity.

Link to comment
Share on other sites

for sure... however those techniques are for composers, who have studied theory deeply, and can use it as a tool. most people here don't have that level of musical education, so I'm not going to recommend that they use those tools.

 

saying noodling is wasteful is a bit close minded. id say writing out every inversion and retrograde is wasteful too, if you don't use all of them, no?

 

actually, anything you don't use in your final piece is a waste, by that logic, right? often times you have to create the waste to get the goods.

 

but really, this is an endless topic, and what it comes down to, is really just trying to offer different ways for people to come up with new ideas, using the skill set they currently have. im not going to tell anyone to go read a theory book. i will point out a few simple ways to approach their current methods though, to get results they might not have before.

Link to comment
Share on other sites

I usually just go da-da-da-dada-dada-etc in my head & figure out what those notes were.

 

yeah, i try to do this but often end up (within the constraints of the piano roll and the effect that has on the reatime conversion of mind to computer) just doing something else that is often driven by the piano roll (and the "expected" intervals that come from an understanding of western scales and the visualisation of that within the piano roll structure).... i often - as has been suggested before here - try and start with chords first - letting melodies evolve out of them - but then the chords i write in a computer are (often) lacking (or "cheesy") when compared to say constructing chords on a guitar (as Fred pointed out)... guitar chords involve upto 6 notes - perhaps 2 incidents of octaves plus 2 intervals but then maybe incorporating 4 or 5 different notes..... when i compose chords in a piano roll i rarely write chords with more than 3 individual notes so in doing that am limiting the depth of the chord to half of what it could potentially be when played on a six string guitar.... also the dynamics of chords played in realtime (whether it be guitar or synth) are lost when (re)constructed in a piano roll (unless great attention is played to the velocity and/or envelope of each note within it).

 

somebody ear;ier mentioned that repeating notes in a melody is not a crime (i believe mentioning guitar playing as reinforcement of this statement) - which i absolutey agree with....i too often hear melodies that seem too eager to jump from note to note without forming any kind of emotional consistency (does this make sense?) - often a simple three or four note pattern can have far more emotional resonance than a skittering all-over-the-place melody (as once in the piano roll it is all too tempting to jump from A to B (or A to G (to use musical terms)) willy nilly. At the bottom line I think the main barrier in terms of constructing melody in the virtual format is the distance between what the mind can "hear" and the time it takes to translate that into note data in a piano roll.... ideally, i suppose, the way forward is to directly play those meodies via a synth into a daw / recorder (with further tweaking if needed)....

 

the bottom line though, i think, is that any melody is only as good as it's relationship with the chords / bassline (and rhythm) that underpin it... even the simplest of melodies can sound amazing when layered over the rest of the constituent parts of a song.

 

]not sure if any of this makes sense....

 

[/drunk]

Link to comment
Share on other sites

Kcinsu has a lot of good recommendations, some of which have inspired me to go straight to the keyboard first when trying to write a track.

 

It kind of reminds me of when you told me about that Sibelius composition that Phillip Glass perhaps lifted a phrase from and turned into a really catchy repeating melody of his own. Not that randomly improvising on a keyboard is the same as copying another artists music necessarily, but to me it's using a similar experimental approach to melody composition not grounded in theory

Edited by Awepittance
Link to comment
Share on other sites

that made me want to see if I was right, as I never really confirmed that... and I was!

 

As early as the 1970s, Sibelius could be seen -- if anyone cared to look -- as one of the unsung prophets of Minimalism. The repetitive string figurations and pedal-points that undergird many of his orchestral works clearly point the way to the early music of John Adams (who indeed listened hard to Sibelius when he was young and, according to a recent interview, is now revisiting the music of his old hero). The quietly bustling perpetual-motion strings in the first work that brought Adams into the spotlight, "Shaker Loops" definitely spring from similar passages in Sibelius' Fourth and Fifth symphonies. Sibelius also infiltrated Philip Glass; in "Floe" (from the "Glassworks" album), Glass explicitly paraphrases the big swaying tune for the horns in the third movement of the Fifth Symphony as the basis for the piece's climax.
Edited by Kcinsu
Link to comment
Share on other sites

yeah, i try to do this but often end up (within the constraints of the piano roll and the effect that has on the reatime conversion of mind to computer) just doing something else that is often driven by the piano roll (and the "expected" intervals that come from an understanding of western scales and the visualisation of that within the piano roll structure).... i often - as has been suggested before here - try and start with chords first - letting melodies evolve out of them - but then the chords i write in a computer are (often) lacking (or "cheesy") when compared to say constructing chords on a guitar (as Fred pointed out)... guitar chords involve upto 6 notes - perhaps 2 incidents of octaves plus 2 intervals but then maybe incorporating 4 or 5 different notes..... when i compose chords in a piano roll i rarely write chords with more than 3 individual notes so in doing that am limiting the depth of the chord to half of what it could potentially be when played on a six string guitar.... also the dynamics of chords played in realtime (whether it be guitar or synth) are lost when (re)constructed in a piano roll (unless great attention is played to the velocity and/or envelope of each note within it).

 

somebody ear;ier mentioned that repeating notes in a melody is not a crime (i believe mentioning guitar playing as reinforcement of this statement) - which i absolutey agree with....i too often hear melodies that seem too eager to jump from note to note without forming any kind of emotional consistency (does this make sense?) - often a simple three or four note pattern can have far more emotional resonance than a skittering all-over-the-place melody (as once in the piano roll it is all too tempting to jump from A to B (or A to G (to use musical terms)) willy nilly. At the bottom line I think the main barrier in terms of constructing melody in the virtual format is the distance between what the mind can "hear" and the time it takes to translate that into note data in a piano roll.... ideally, i suppose, the way forward is to directly play those meodies via a synth into a daw / recorder (with further tweaking if needed)....

 

the bottom line though, i think, is that any melody is only as good as it's relationship with the chords / bassline (and rhythm) that underpin it... even the simplest of melodies can sound amazing when layered over the rest of the constituent parts of a song.

 

]not sure if any of this makes sense....

 

[/drunk]

very useful post. thanks ludd.

Link to comment
Share on other sites

I agree with Kcinsu.

 

Also: one of my favorite ways to invent melodies is by singing all the time (as much as you can get away with it). I guess this is the same as playing any instrument you're comfortable with, but when singing it seems especially easy to channel new ideas, they just slide through you.

Link to comment
Share on other sites

Lots of good ideas in here...

 

This isn't really directed at anyone in specific, and it could probably go without being posted at all, but the blue Post button is so very close already...

 

 

 

while I'm not sure the argument against learning music theory was ever mentioned explicitly in this thread, as a preventive measure against seeing it crop up in its full form, I'd like to just say this:

 

 

In order to effectively break the rules, it helps to know what the rules are.

 

that idea goes way beyond the realm of music theory, and while I don't think anyone needs to know too much theory in order to make interesting music, it certainly doesn't hurt to look into new "theoretical ideas" if/when you find yourself in a creative rut. think of it in a broader context. you wish to do something, and do it well. which seems like the wiser choice? A) experiment in the dark and entirely ignore the body of established relationships, knowledge, tips and tricks that others before you have set out, or B) start off experimenting, find somewhere you'd like to start, and then learn about the ways that others have worked out similar solutions to whatever challenge you've presented yourself with.

 

If we were talking about cooking, surely B would be the more effective place to start. If we were talking about science, B would be the correct route. If we were talking about bank-robbing, again, B would be the wise choice.

 

For some reason, a good number of [electronic] musicians seem to think that learning theory will somehow stifle their creativity. This is false. There is a fine balance to be achieved, for sure - it would be unfortunate to learn so much theory that you somehow did stifle all your creativity, but this is not the only way to go about learning something new. You can know the established rules, and find out how to break them in creative new ways. "Educated but not academic," I guess.

 

 

 

 

Anyway, enough spoiled theory shenanigans - Fred, you kinda started saying this in your first post, but one of the easiest ways to find new creative angles in melodies is to build them up over different backing parts of a song... or to build melodies one on top of another (using different voices and dropping out old parts + adding new ones as the song moves along is fun), as has already been mentioned. For me, starting off with a melody is usually a dead-end...

 

Really sparse melodies, or "suggested" notes I guess, are also a really useful and fun trick to pull off. One of my favorite songs from last year was Daniel Savio's tune Pulguita -

 

http://shaarptooooth.com/07%20Pulguita.mp3

 

the lead voices throughout the song seem to be speaking to and agreeing with each other - I'm probably just high - but there really isn't all that much going on in terms of actual melody and the song still manages to have a very peculiar moody atmosphere and grooviness... there's a bass synth that really only hits its lowest note once in the entire song (~3:36) and it is fucking glorious, as if it had been struggling to say that one thing the entire time and didn't even need to say it again once it was done. :cool: I guess my point is, say less to say more. I'm gonna take my own advice and stop typing

Link to comment
Share on other sites

For the record, I have a degree in music composition. I'm not going to tell anyone NOT to study it, but practically, advice on how to change up your work flow, is going to be more helpful to more people, than saying, go learn theory. it takes a lot of time and dedication to get much out of music theory, and realistically not many people are going to go out and study it because I advised them to.

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.