#1
I've spent many years studying guitar and teaching myself as much as I felt I needed to know about music theory, and I'm finally to a point where I feel I've learned enough to comfortably write music.

I'm very into progressive rock and metal. I take influences from Rush, Pink Floyd, Dream Theater, Tool, A Perfect Circle, and a few others. I wish it was a simple as writing along to a 4/4 beat, but that wouldn't be much fun now would it? :P

I'm at a crossroads now. I've spent some time writing the last few months, and I've only come up with a few song ideas. After taking the time to record parts separately, and placing them together and rearranging as I felt it was needed, I'm not happy with what I'm hearing. It doesn't sound like something I would proudly play. It doesn't sound like something I would actually like as a listener. It's close, but it's still not there.

I think my biggest struggle is deciding what aspect of a song to start designing first: the drum beat/time signature, or the guitar riffs/chorus/etc. Being as my music will be progressive, the time signature will change several times in most songs, but there will be a "main" time signature that will carry the majority of the song. So far when I've been playing around with arpeggiated chords and what not, I've been writing a lot of parts that match a 3/4 time signature, however, I formed an entire song around 3/4 and just didn't like the way it felt. I'm being very nit-picky about this, but I want it to be right.

I've tried looking up interviews with a lot of my favorite guitarists to see if they ever gave out some information about their songwriting process, but most of them don't, understandably. But that doesn't help out a new guy like myself. I would like to know what has worked for some people, so I can try their method myself and see if it helps.

So, when you guys sit down and start writing your compositions out, which do you do first? Do you decide what the time signature is going to be, as well as the tempo, ahead of time? Or do you get right into playing around on the guitar, recording/tabbing some bits, and finding out later what time signature/tempo to sync with it?
#2
It sounds like you have decided how the time signature should be structured before you have any idea of how the piece of music will sound. As though you're approaching it with a paint by numbers approach. It might work though I'm not arguing against that method - whatever works.

The point I was getting at is that since you seem to have decided how you want the time signature to go before you have a solid idea on melody or harmony it would make sense for you to start there. Put together a basic rhythmic structure first including all your time changes and then jam along till you find some guitar parts that work.

Work those into something cohesive and if you find you need to change the rhythm or time to fit an idea, there's nothing stopping you.
Si
#3
I appreciate your advice.

I might have let myself talk way too much when I made the original post. To put it simply, I want to know what people find works better for them: Coming up with the beat first(drum track, time signature, tempo, etc) and then writing the guitar parts to match, or picking up their guitar and playing until they find something they like and forming a song based off that.
#4
well most my recordings sucks due to lack of pucussions. What I heard in my head
and what I actaully use as a drum track. Whether time signature changes or
not. I simply hear different drums beat or want to use different beats.
I heard breaks, verious fills...ect
I dont have a decent drum machine or play drums that well.
I actaully get more fustrated when I go into recording.
I have to lay down the drum tracks first....You know how it goes, different
kits of drums will have different tones. I get picky like that or I'll notice
the difference in drum tones. It dosnt sound right to me when that happens.

At the sametime being familar drum machines. I can spend weeks just deciding
what type of drum set to be used.
I'll spend just as much time or more on drum tracks than I do on guitar parts.
Punching in drum sequnce isnt my thing.

It just depends. Sometimes music just comes to me...I'll hear riffs, melody
or whatever. I'll write around that.

Other times. I'll just go through a chord pregression chart. Pick a progression.
Stum the guitar, pick it or whatever the heck I do. Try different ideas.

Other times, I'll just leave my recorder on while Im just jamming to some drums.
I'll go back and listen to it. Then just pick whatever I think sounds cool.

As far as playing with drums...Like Steve Via say. Just let it sink in.
Feel the groove of the drum.
Last edited by smc818 at Jan 14, 2014,
#5
I usually write the guitar parts first, a whole, fully fleshed out song of just guitar, then add everything else. I think it helps that I play drums too, because I usually have an idea of what I want going on behind the guitar as I write. That said one of my favorite recent songs was just a drum track that I programmed, and then put guitars to later. So basically just do whatever seems natural.
#6
I either have a certain riff idea in my head before I pick up the guitar or I noodle around and play something waiting for my fingers to do something awesome. I'll usually take the riff I like and start writing based around that.
I don't think I've ever sat down and thought "I'm going to write a song based in 5/6 timing at 145BPM", doing that doesn't feel natural to me. It feels like I would be trying too hard to write something in that way and it would be too restricting rather than writing what comes naturally to me.
When I was eleven I broke the patio window and my mother sued me... She's always been a very aggressive litigator.
#7
From someone who is also really into progressive stuff:

Sorry, but I think you're doing it COMPLETELY wrong.

You just have to write songs, what you're doing now is writing ideas. It starts with an idea but you have to make a full song out of it. And first of all, if you can't write a decent song in 4/4, you're using time signatures to hide that. There's a pretty big chance you'll notice that the time signature is 'chosen' instead of felt. But you don't see Pink Floyd arguing about Comfortably Numb being 4/4 and just i - VII - VI - iv do you? Only use the time signature you're feeling.

Start with writing a riff or chord progression you like, or with a rhythm that's inside your head. Then work around with that.

I usually work around some theoretical thingy like how I can make iii(DM) - i sounds as smooth as possible just as a starting point, but in the end it's what you like to hear and feel. Even (man, maybe especially) with prog.
#8
don't let your pride get in the way of you writing a good song.
Quote by archerygenious
Jesus Christ since when is the Pit a ****ing courtroom...

Like melodic, black, death, symphonic, and/or avant-garde metal? Want to collaborate? Message me!
#9
Starting song structure with time signature changes is extremely odd, it's like buying a guitar pedal before even deciding what instrument you want to play, you have no idea what you're actually going to do.

You need a proper starting point, whether it be a drum groove that you like, or a good rhythm guitar part, or a good lead riff to build around. Then think of good transitions that work with the main idea that you have. If you're so intent on having time signature changes then you have to make sure that the time signature changes are logical, and not just "hur dur im prog i change time"

One skill that you should work on is visualizing (hearing, since well you can't really see music) backing parts to whatever main idea you are working with. For example if I make a fast guitar riff, in my head I imagine a fast drum beat to complement it, thus making my ideas feel complete even though I am only one person that can only play one instrument at a time. Once you are good with building up full ideas from smaller starting ideas you will be making your own songs with minimal effort.
Peavey 6505+ Head (Left)
Line 6 Spider IV HD150 Head (Right)
Marshall 1960A 4x12
ESP DV-8
Ibanez RG T42DX
Agile Hornet Pro 727
#10
You can start anywhere as long as you end up with a finished song.

I generally start with a jam. It could be chords or a riff on the guitar, and then I'll build it from there. I might do a melody next. I might do drums next. I might do keys next. I'm very passive with my songwriting to begin with (as in I don't structure anything until I have a concept or concepts that I like), so this works best for me.

Of course at some point you have to begin to plan out specific structures, but I find that allowing music to write itself in the beginning brings out the 'you' in the music. If you like progressive stuff and that's what you play then allowing your ideas to grow naturally without putting boundaries and stipulations around them in the beginning will get you closer to where you want to be. Just make sure that that is the kind of music that you're jamming and exploring.

I guess my only concern is that you're getting too worried about whether there will be enough time changes and whatnot. Music should evolve naturally as you play so if a time change is needed, then it happens (in my experience anyway), and if you don't need one you don't use one. You don't need to have certain elements of structure for music to be considered 'progressive' and you sure as hell shouldn't force that, especially if you haven't been writing for a long time.

Also, don't get discouraged when your music doesn't come out exactly how you want it. It happens to everyone in the beginning. It takes a lot of practice to be able to get down what you're hearing in your head accurately and even more difficult to get that down as a coherent piece of music that you'll be proud of. Keep writing and writing. You'll eventually get there.

EDIT: Also, listen to Br0C00ler. The last 3 songs I wrote, I spent more time pacing around my room playing out different melodic and rhythmic ideas in my head then I spent with a guitar in my hand.
Last edited by mjones1992 at Jan 14, 2014,
#11
I have very simple drum tracks in different times. They only have very simple embelishments to mark the ends of a phrase. I play guitar along with these, just noodling around until I get a part I like, which I then record. Then I listen carefully to where the accents are and where there should be holes, etc, and revise the drum track to match the guitar. I have an e-kit so I might do this by jamming around on the drums, or just program them in. I might even re-do the guitar again if the beat suggested changes or if it should groove a little differently.
The point is that I find it hard to write one track first, so I use scratch tracks and work a bit on one and then a bit on another.
#12
Quote by Four Thunders
I might have let myself talk way too much when I made the original post. To put it simply, I want to know what people find works better for them: Coming up with the beat first(drum track, time signature, tempo, etc) and then writing the guitar parts to match, or picking up their guitar and playing until they find something they like and forming a song based off that.


much of the music i write doesn't have guitars in it. some of the music i write has no percussion. sometimes i write music that doesn't have either.

my point is that you should focus on creating complete, coherent music that isn't limited to particular labels or instrumentation. start with a melody, sure. a drumbeat, okay. lyrics, fine. but you need to be able to turn an idea into a finished product. most anyone can come up with the idea - it's often a very intuitive thing. but the process of songwriting (by which i mean being able to reliably turn ideas into finished products) is far from intuitive. it's a learned skill -- and one that few musicians develop well.
Anfangen ist leicht, Beharren eine Kunst.
#13
It honestly really depends. I usually start writing a song based off of a riff I come up with and let it grow from there, but sometimes I'll be absentmindedly tapping a rhythm with my hands while driving or something and like that so much that I start building a song around that.

Though honestly, while structure is a very important thing in composing so your songs don't sound like riffs stapled together, you have to let the songs build themselves organically rather than trying to plan things out ahead of time. Especially for progressive music, bands like Tool don't write a song thinking "I want to write a song in 5/8 that goes into 5/4," that's just how they think. So sometimes you'll write a song with 20 different time signatures because that's just what the song is "asking" for, sometimes you'll write a song in straight 4/4 but it'll still be just as strong as the other song.

tl;dr: Just start off with a "seed" for a song: a riff, a chord progression, a rhythm, or what-not. Then just take it from there and see what comes out, you may surprise yourself! If you can't think of anything, no problem, just save that seed for later and see if you can make it grow later.
#14
if you really are a proggy, like me, you should try writing based off of different things. maybe one song is based around a recurring lead motif (octavarium by Dream theater), maybe one song is based on bass (schism by tool, or sting's cover of driven to tears), and maybe one song might be based on keys (these walls/pull me under by DT again)

my point is, prog is about trying now things, and about the "why not" rather than "why"

some of the best musical decisions ive made were things that originally scared me lol
Last edited by pAWNlol at Jan 16, 2014,
#15
I personally don't like an academic approach to songwriting. You should write what you hear. If you're not hearing anything, what you write is not likely to be interesting.
#16
Sometimes I come up with a guitar part first, sometimes a bass part, sometimes a drum part or whatever. It depends. But the thing is not to overthink. At least first come up with an idea and if you want it to be in a more "progressive" time signature, change it a bit. Add one eight note to it or something like that. But many times if you just decide to write a song in a time signature, it will sound forced. I wrote a song in 5/4 but it used an "even" rhythm pattern (2+2+3+3) and it didn't sound strange at all. https://soundcloud.com/jaakkorauhamaa/5-4-demo You can also hear some time signature changes in my song. They came naturally, I didn't force them.

Also, 4/4 isn't a boring time signature. You can make it sound strange if you want. It's not about the time signature. Music isn't time signatures and chord progressions. You can make whatever chord progression sound interesting, even the infamous "4 chord progression". But many times people just use them in a generic way which makes them sound generic.

http://www.youtube.com/watch?v=SdOEindANdY

In some parts (for example around 2:30) the drums accent different beats, not the basic 1, 2, 3, 4. But it's in 4/4 all the time. The time signature doesn't make it sound generic.

I would say don't try to write in a certain time signature. That usually makes it sound forced. Just come up with ideas and let the song write itself. That way it will have most flow. Yeah, it may not be the most progressive sounding song but so what? I don't like writing in genres.

Writing in odd time signatures is easier if you come up with a rhythm pattern. So maybe write the drum beat first and write a guitar part over it.

But yeah, when you write music, use your ears. Don't overthink.
Quote by AlanHB
Just remember that there are no boring scales, just boring players.

Gear

Bach Stradivarius 37G
Charvel So Cal
Fender Dimension Bass
Hartke HyDrive 210c
Ibanez BL70
Laney VC30
Tokai TB48
Yamaha FG720S-12
Yamaha P115
#17
Your favourite bands probably opt to change time signatures during song (although many on your list have a whole heap of songs that don't) because they feel the song would improve as a result.

But you are simply copying, chopping and changing out of some self-imposed standard, a need to complicate things when some songs don't improve at all as a result. And thats what your music sounds like.

Try writing simpler songs first and working from there. There's a lot of things that can vary the feel of a song other than the time signature.
And no, Guitar Hero will not help. Even on expert. Really.
Soundcloud
#18
Try to think in terms of rhythm rather than time signature. An interesting accent pattern will give you the rhythmic foundation you're looking for. Ask yourself what the kick drum and bassist are going to play and build from that.
#19
Quote by Four Thunders
I've spent many years studying guitar and teaching myself as much as I felt I needed to know about music theory, and I'm finally to a point where I feel I've learned enough to comfortably write music.

I'm very into progressive rock and metal. I take influences from Rush, Pink Floyd, Dream Theater, Tool, A Perfect Circle, and a few others. I wish it was a simple as writing along to a 4/4 beat, but that wouldn't be much fun now would it? :P

I'm at a crossroads now. I've spent some time writing the last few months, and I've only come up with a few song ideas. After taking the time to record parts separately, and placing them together and rearranging as I felt it was needed, I'm not happy with what I'm hearing. It doesn't sound like something I would proudly play. It doesn't sound like something I would actually like as a listener. It's close, but it's still not there.

I think my biggest struggle is deciding what aspect of a song to start designing first: the drum beat/time signature, or the guitar riffs/chorus/etc. Being as my music will be progressive, the time signature will change several times in most songs, but there will be a "main" time signature that will carry the majority of the song. So far when I've been playing around with arpeggiated chords and what not, I've been writing a lot of parts that match a 3/4 time signature, however, I formed an entire song around 3/4 and just didn't like the way it felt. I'm being very nit-picky about this, but I want it to be right.

I've tried looking up interviews with a lot of my favorite guitarists to see if they ever gave out some information about their songwriting process, but most of them don't, understandably. But that doesn't help out a new guy like myself. I would like to know what has worked for some people, so I can try their method myself and see if it helps.

So, when you guys sit down and start writing your compositions out, which do you do first? Do you decide what the time signature is going to be, as well as the tempo, ahead of time? Or do you get right into playing around on the guitar, recording/tabbing some bits, and finding out later what time signature/tempo to sync with it?


I personally think you may be over thinking it slightly. Most songs evolve and end up nothing like the composer thought they would. Don't be too rigid about the process, throw ideas about and then maybe leave them for a bit and go back to them and you may have some fresh ideas a few days later.

I would highly recommend reading "Life" by Keith Richards (funny that) not only is it a fascinating read but he also talks a lot about how the songs came together and how he and he and Jagger would compose, a lot of the time it was just playing a riff or a couple of chords to one of the band with maybe one line of lyrics and then seeing if anyone else had any input. Of course that's only one perspective but a pretty successful one.
#20
Also, bands like Dream Theater and Pink Floyd write a lot of stuff in 4/4. As cdgraves said, it's more about the rhythm, not the time signature. 4/4 isn't always 1, 2, 3, 4. As I said in my post, it could also be 3+3+3+3+2+2 which is pretty basic too. Most music is in 4/4. The time signature on its own doesn't make music sound simple. Time signature isn't all there is to music and it doesn't make music complex or simple. You could write a simple song in 7/4 time if you wanted. Also, what's wrong with simple music? Just write music.
Quote by AlanHB
Just remember that there are no boring scales, just boring players.

Gear

Bach Stradivarius 37G
Charvel So Cal
Fender Dimension Bass
Hartke HyDrive 210c
Ibanez BL70
Laney VC30
Tokai TB48
Yamaha FG720S-12
Yamaha P115