This is how i do it on LIVE

I start off by mere performance on midikeys – synth organs usually playing chords from various songs for inspiration. Next i add simple drum patterns, basically hh, snare, kick. Play again. Then i set up baseline to be in the range of low notes. Play around a bit then add arp and a pad. Tweak midi zones a bit. I also bind mute controls to computer keyboard digits to mute certain drum tracks. I add lead if it doesn’t sound like there’s a lead in already.

I have now all ensemble i can play simultaneously. So i play again and again. Until all parts I clear as to how to be performed. I always record. You never know when your best shot comes. I use one VST instrument for all tracks, drums too for better harmony. Like arturia minimoog. That’s it. That’s sort of LIVELY music i prefer to make.

The song under that link is my one example to base on the technique:
http://madimpactunit.com/audio/give_me_sometime_live_v2.mp3

My songs with vocals always start either from the hook or the first line of the song, and develop from there. Occasionally I have a melodic or chordal progression, but invariably it will be tied to some line or at least thought. The song then builds from there. All of the “Concert of Praise” songs on my soundclick site were developed that way.

I always write in chunks – the whole verse, the whole chorus, the whole bridge, etc. I write at the piano for pop songs and then flesh out the arrangement from there.

I have to revise and resubmit the first paragraph. If you listen to the song “Run the Race” on my soundclick site, it was built from the opening bass riff.

For purely instrumental songs, they can grow more organically, but i’ve really only written a few songs without words (arranged plenty of them, but only a few originals). The most significant one (and the one I’m working on right now) grew out of a musical theme that was running in my head. All the rest either grew from a theme or melody that was then developed.

I write instrumental songs at my notation software (Finale) with the appropriate instrumentation directly.

I start by recording an idea or two vocally. Then I usually convert the melody or beat into midinotes, choose an appropriate synth and vocode the two together. (Everything in my music speaks to you deeply, even if you don’t know it yet).

Once I have the basis for the song, then I start to fill the entire spectrum with instruments (minus the key vocal frequencies). I process and process until the song could stand on it’s own without vocals — this is so you can really get lost in my music. This usually means inventing all sorts of crazy ways for the timber of my instruments to change over time.

Then, I add the main vocals, tune em if I have to, follow it up with some backup vocals, fill any ‘dead space’ and master.

Tada! Future Music Now:
http://mikestop.com/future-music-now

I think for ages man has wondered if the creative process can be algorithmically determined to create truly great works. I’m sure there are volumes on the Beatles writing process alone!

Truly, everyone’s style is different and some things do not work for others, and it’s important to remember that you have to FEEL the music the way you want, even if the Beatles or the Stones did it differently.

Personally, I start with a great riff and then build an entire part off of that: adding drums, bass, and then decide if that is the chorus, verse, etc. From there, I mess with chord progressions to see what progressions fit with that initial riff, and that helps me form the rest of the song. From there I whip out the piano and start realizing the vocal progressions in contrast with musical elements to see how the vocals will work out. Sometimes I start with piano and see what vocal progressions I like, and work backwards. Melody is incredibly important to me, and I recently I’ve found myself starting with vocals first and instruments last.

Acoustic guitar and a small handheld recorder. Simple. Works fine. If the song is there, hard to make any productions.

Not that I dibble with pure productions as well that end up becoming tracks, mostly jam-centric, play and see what happens.

I like to write songs with no “chorus”, I mean songs with a chord progression and a melody that will repeat trough the entire song.
I know it can get boring so once I have a great melody I usually create alternative ideas for all the elements like bass lines, new melodies, several backing vocals, various guitar lines different drumbeats etc… that will fit in the harmony and rythm progression.
after that I play with it like a puzzle. removing or adding the layers. it’s a really creative thing to do, cause sometimes combinations that wasn’t expected happens and make the song more interesting.

sorry for my english

I love this topic.

Wow….there are so many different factors. I think it can vary so much even by genre. If you’re a rocker vs.more of a dance/tech/urban, your inspirations and ways that you start the creative process will probably vary greatly — a guitar riff vs. a beat, or a bass line.

I know for me personally I tend to “think” with my guitar when starting. But I’ve also had songs that grew from lyrics first, then music…or a melody that I started humming.

What’s interesting is how different the music can turn out when it starts creatively from different places. I know when I start things on piano, the songs tend to be a lot more “sappy” than when I start them on guitar, or if I do something more electronic it’s typically a much different personality (less “organic”, more mathematic type of feel). One of the reasons that I used to like some of the Thompson Twins’ stuff is that they seemed to have a way of doing “electronic” with an organic feel. (yikes, dating myself, here).

One thing that I try to be aware of is my own creative comfort zone, and trying to have a healthy appreciation for breaking out of it once in a while. Starting with my guitar is a great comfort zone, but sometimes our comfort zones can end up being creative shackles, too. Sometimes it’s good to stretch a bit and try new things. That’s when we grow…but it’s certainly not always “comfortable” pushing into new areas.

I like so many different styles of music. I just “went with” whatever I was feeling on my record, which made for some interesting differences across the album (lots of diversity). However, the flip side of that is that it may be harder for people to digest me as an artist (vs. an artist or album that is more cohesive creatively). I don’t really have one “sound” as a result. But, that’s okay…..I had fun creating it. )

But, at the end of the day, I would always take feel over formula.

Ironically, I was the anonymous reader that Joel refers to in the introduction. But I’ll share my current method (along with the drawbacks that I feel it has).

My main issue right now is that my two areas are disjoint. Lyrics tend to come to me and random times, so whenever I get the chance, I’ll sing them into my cell phone video recorder. It’s all pretty random and I don’t have a creative process for them.

However, instrumentals tend to be more embedded into my brain, so I’ll sit down and start with the drums that I have in mind. Then I’ll layer each instrument on top of the other (using Reason 4), and then slowly add variations until I have a finished instrumental.

The problem I feel with doing instrumentals this way is that, while I often start with a concept in my head, once I start adding the various layers, I’ll often end up in a completely different area than where I want to be. The more elements I add, the more the tempo seems to speed up (so I have to lower it), and then the whole song starts to change from there. It’s not always a bad thing, because it helps change up my patterns from what I had, but it’s also rather bothersome to try to create something that I have thought about, but end up with something completely different.

*goes back to studying for his exam…I promise*

There are 2 great bits of advice I received when I started – ‘don’t censor anything’, and ‘90% perspiration, 10% inspiration’. I’ve found that if you write 10 songs in a year, maybe one of them will be really good, a couple might be pretty good, and the rest filler. But if you write 100 songs a year, then you could end up with 10 that are really good, and maybe 20 or 30 that are pretty good. Doesn’t always follow, but the principle holds – the creative flow is more like a muscle I think, and the more you exercise it, the more it’ll work for you.

I would personally echo just about all of the above comments, i.e. I think a song can come in different ways, and whenever I’ve been stuck in a particular way of doing things, e.g. guitar riff first, then setting a challenge of starting with a drum loop or piano or bass can be a great way of attacking things from a different angle.

However, in general, I tend to get music first, and will record little snippets / riffs as ’sketchbook’ ideas that can be revisited later. Maybe one of them will catch my imagination and ’speak to me’, and then there’s energy to develop it musically and conceptually: depending on what the music ’says to me’, i.e. what vibe is there, that’s what determines the lyrics and what the song’s ‘about’. As to adding musical layers, that tends to be a jam-along addition of different parts, but these are honed to get the parts to fit well.

One example is a song called ‘Channel your Requiem’, which actually started as a rip-off of a chord sequence in one section of Mozart’s Requiem – I’ll be impressed if you identify which!

In the band I was in, we did a version we you can find here: http://www.myspace.com/onelifeleftuk. As you’ll hear, it’s a full histrionic rock epic, and that’s how it was originally written – the main ‘verse’ chord sequence came first, and the further sections were added as that main section developed. The lyrics and ‘concept’ came after, and the arrangement was honed afterwards. It’s not a radio edit, but it seems to want to be this long

As a production contrast, if you go to http://www.myspace.com/royalruin, you’ll find an acoustic version of the same song, except with a slightly ’swung’ feel. This was adapted to suit a solo (duo) acoustic gig that included old blues/gospel/spirituals standards, and it’s interesting to hear that the essence of the song still holds up despite being stripped of much of the rock shenanigans.

It all starts with a beat for me. I have kind of a crazy way of doing it. I might hear a song with a good beat then I go on the net and find a free midi of the song (there is a lot of GREAT free stuff out there), import it into Sonar strip out everything but the drums then route the drum part into a high quality drum VSTi.
Once I have the beat playing I strap on the strat and let the creative stuff come out. Building a song from there is easy given the ease that you can manipulate the midi. Once I have the basic chords laid out I usually lay down a organ track to hold it all together then add some bass etc…all my stuff is built one track at a time.
For me its all about the beat. I may come up with a lyric or two and days months or years later find a drum part for it. Most of my songs at http://www.soundclick.com/mothersbadson where written that way.

I generally begin with the chords / rhythm / riffs on the guitar, then move on to formulating a melody, and finally filling that meloday in with lyrics.

As for recording – I like to use Pro Tools and Reason hand-in-hand, combining recorded live instruments and samples.

To keep things interesting and keep the inspiration up, I just launched a new blog all about the craft of songwriting and producing. Check it out! http://www.servethesong.net

Yeah, this topic is great!

I usually work on the song in my mind for a long time, till I get a somewhat complete understanding of both arrangement and lyrical content. I never write down or record anything until I have this sort of whole picture. Some ideas get lost in the process. But I believe in evolution theory – survival of the ideas worth implementing. And it usually works – I end up with few songs but they all come out not too bad.

The exact process of programming/recording may vary. I usually do it in standart order: drums, bass, guitar and synth parts, then vocals, then the mix. During this I try to achieve the sound closest to what I had in my mind. Sometimes I start with a midi sketch before recording live parts.

I used Cubase but recently switched to Ableton although I haven’t yet done any solid works in it. May be this will change the process.

Melody is important to me. So I take a melody instrument, open the microphone, and just play for 10, 20, 30 minutes, whatever melodies I come up with. If I think there are enough good bits, then I start editing till I have between and 3 and 6 minutes worth. Then I figure out what chords fit under it, and depending on the character of the tune I pick the backing instruments, be that a single piano or 10 different synths.

I use my keyboard and play around unless i find a tune then I work towards the chorus,beas and the rhythm