Tuesday, July 7, 2009

The Process


Every musician has their own personal process for putting together a song. If you're like me, that process often evolves throughout time, sometimes almost as if it has a life all its own. I have always found it interesting to read about another artist's process, and since I happen to be in the last stages of finishing up an album as we speak, I thought maybe I'd share mine. Like I said, my process has changed with pretty much every album I've done, so this mainly applies to how the songs for this particular album came together.

THE IDEA

Even when I'm not formally working on an album, I keep an "Ideas" folder to sketch out ideas as they come to me. Inspiration strikes when it strikes, which may not always be the most opportune time to sit down and bang out a complete song, so keeping a folder where I can quickly record an interesting bassline, chord progression, melody, or rhythm ensures that when I am ready to begin work on an album, I already have a nice selection of potential "seeds" for new tracks.

SEPARATING THE WHEAT FROM THE CHAFF

Let's be honest, not all ideas are winners. So I further sub-divide my ideas folder into "Definite", "Maybe", and "Probably Not" folders. As time goes on, songs might migrate from one folder to another, but the important thing is that I save everything, even the stuff that I decide sounds like crap. Sometimes an idea that seems useless at first, might end up working perfectly in the context of another track, or you might be able to salvage it with a small revision that comes to you later. Regardless of the outcome, you're never going to regret hanging on to every idea you come up with, so you might as well do it, right? Once I have a good selection of potential songs for the album, I start work on transforming these raw ideas into proper songs. I like to be working on several tracks at once so that if I hit a dead end with one track, I can continue working on a different one. I don't know many musicians with extra time to spare, so anything you can do to increase your efficiency is a good thing.

BUILDING THE BASIC STRUCTURE

My ideas usually tend to come in small chunks... a chorus... a verse... maybe just a vocal melody. So the next stage is filling in the missing pieces. I usually try to build a single verse and chorus first, so that I have something that flows together well. I'm not talking about building the full arrangement at this point, I'm usually just working with a bassline and a basic drum beat or loop to play over. Once I have a verse and chorus structure that seems to be working, I arrange these into a full song-length structure, experimenting with where to have these portions repeat and change, where to put bridges or breakdowns, etc. This can take some time to work out, and can even change again later in the process (thank you, cut and paste!), but the sooner I can figure out the basic skeleton of a track, the sooner I can get down to the meat and potatoes of putting the song together.

LYRICS AND VOCALS

In the past, I always saved this part for last because, for me at least, it was the most difficult and least enjoyable part of putting a song together. This time, however, I tried to at least start this process fairly early on. There were several reasons for this. First, because I find it difficult and therefore the most like actual "work", getting it finished early on meant I would get the hard part out of the way and could approach finishing the song in a more enthusiastic and energetic way. Secondly, I wanted to concentrate more on improving my arrangements on this album. I've always liked to include lots of layers in my songs, but in the past, I feel sometimes my arrangements were a bit cluttered and muddy. By building the songs around a vocal, I was able to concentrate on getting everything to fit in its right place. Your mileage may vary, but I really found this helped me arrange the additional synth parts much more effectively. The new album is more layered than anything else I've done, but this time you can hear each part with a clarity that was missing before. Or at least I think so. I record several takes of each vocal track and once I have a few takes I am happy with, I build the composite. You can read more about that here.

BUILDING THE ARRANGEMENT

For me, this is the most fun part of making a song - fleshing out this basic drum beat, bassline, and vocal into a full, proper song. This usually involves looping sections of the song and jamming over it, trying different sounds and trying to find the melodies, chords, hooks, and riffs that will put some meat on the bones of my basic demo. At this stage I am not too concerned with finding the perfect sounds. I just want something that fills the basic role the final sound will fill. So if it's a string or pad part, I use something in that vein, if it's a lead, I use something that falls into that general category. What is most important here is that I am conscious of the frequency range each part occupies so that I don't have parts stepping on one another or competing for space in the mix. It is at this stage that I also program the drums. With the snare parts, at least, I like to play in the parts in a linear fashion, from start to finish. I think this makes the drums sound a bit more alive and natural and results in fills that fit the transitions better. On this album, I used drum pads for this purpose, which I found to be a lot more fun and spontaneous than tapping it out on a keyboard. I usually have to go back and do some edits for places where I screwed up or for fills that just don't work, but most of it is as it was played (with quantization, of course... I'm no drummer!)

CHOOSING THE SOUNDS

This is my other favorite part. Now that I have all the parts of the song recorded as MIDI, it's time to formally choose and commit to the sounds that will be used in the final song. This is also one of the most time-consuming parts. When I'm not working on songs, I'm usually sampling or programming synth sounds so that I have lots to choose from when it comes time for this stage. Even with this pre-built library of sounds, finding the right one can be really frustrating if, like me, you have a pretty large number of synths and softsynths. So try to figure out what each synth you own is best at ahead of time and you can speed up the auditioning of sounds significantly. Also ask yourself if you really need all those synths. The answer is probably "no". But, like me, you probably won't get rid of the ones you don't need either. It's a disease, I tell ya.

RECORDING THE SYNTH PARTS

At this stage, I bounce down each part to audio. I usually record everything dry to give myself the most flexibility at mixdown, so if a synth has built-in effects, I usually turn them off when recording. Chances are, I have a better-sounding dedicated effect plug-in I can use on it anyway. I did a lot of layering on the new album, so often at this stage, I'll copy MIDI tracks and assign different instruments to it to find what sounds the best to my ears. This is sort of like building an arrangement on a smaller level. You want to choose sounds that compliment each other, not step on each other. One layer might provide the low end balls, one might provide a nice, metallic bite, etc. This applies to drum sounds as well. I rarely ever use a single drum sample for a drum part anymore.

FILLING IN THE DETAILS

Once I have most of the song put together, I build any ambient effects in the intro, any edits for a breakdown, or any other 'small' special details that help bring the song up a notch.

MIXDOWN

Once I have everything pretty much where I want it, it's time to do the actual mix. I bring all my faders down and start bring them up one at a time with the kick. Then I bring in the other drum and percussion parts and get their levels sorted, compression, EQ, and other effects sounding right. Next comes the bassline, then the vocal, and then everything else. During this whole process, I get the effects set up properly, and usually go back and forth changing the volume levels until things sound right to me. The more you mix, the better your instinct gets for this sort of thing, but it never hurts to have a reference track by an artist you respect in your general style to compare your mix to.

INITIAL BOUNCE

When I have something I am reasonably happy with, I bounce the track down to a WAV. I'll burn it to disc and listen to it on my computer, my car stereo, the speakers on my television, my iPod, and whatever other playback systems are at my disposal. If you've never done this, you'll be surprised at how different the same mix can sound on different systems. You're never going to get a mix to sound perfect on every system you play it back on, so what you're listening for here are the big, obvious things you can tweak to make a better mix that "averages" to as wide a variety of systems as possible. I recommend you do at least one listen on headphones or earbuds as well, as this can often reveal mistakes or things that need tweaking that may not be as obvious on speakers.

TWEAKS

You may be a much more skilled mixer than I am, but I've never done this initial bounce and not heard things that needed adjustments. So, this is the time for those last minute adjustments. Once you've done them, re-bounce the parts and repeat the evaluation on different systems again to see if you got it right. If not, go back, tweak, lather, rinse, and repeat until it's right.

PUSH THAT MUTHA OUT

Once everything is as perfect as you can get it, it's time for the final bounce so you can send it off to be mastered. This is the hardest part for me, as I am always convinced there is something I've missed that will be obvious to everyone else and I'll be exposed as the fraud I am, my career will go down in flames, and I will die penny-less and insane. But, much like proud parents seeing their child off to college, at some point you just have to let go and have confidence that your song will blossom into everything you dreamed it would be once the listeners pop it in their stereo.

So there you go. That's my process. Do you follow a similar path, or do you do things a lot differently? Any tips that you've personally found useful in your own work? Let us know!

Sequential Circuits Prophet-10 on Ebay


Whether you want to pretend you're Vince Clarke (he had one back in the Yaz days) or just give yourself a hernia trying to lift it, this would definitely make you the coolest kid on your block!

Info at the listing...

Monday, July 6, 2009

Eric Beam Releases Free MFB-522 Drum Machine Samples

Via rekkerd.org

Eric Beam over at rhythminmind.net has released a high quality sample set of drum machine sounds sourced from the MFB-522, an analog drum machine. The release is donationware, so please give what you feel they're worth. Here is what he has to say about the set:


"
The Drumcomputer MFB-522 is an all analog boutique Drum Synth. It’s deceivingly small size produces HUGE sound. A call back to the TR-808. If you want the 808 vibe but are sick of the overused 808 samples available, this is for you.

9 analog instruments captured @ 24bit/96k & RNLA compression for a little added edge.

Five/Six individual velocity samples per instrument.

Format = Stereo 24/96 .wav"

D16 Releases AU Versions of Drum Machines


Polish developers the D16 Group have released updated versions of their Nepheton and Drumazon virtual drum machines, including the long-awaited availability in AU format for Mac users. These updates also unify interface and various other behaviors to correspond with that of their more recent releases.

Additionally, D16 has updated its recently-released TR-606 emulator, Nithonat to fix various bugs. Read my review of it here.

Cheap Trick Release New Album on 8-Track


Although I suspect most readers of this blog weren't around when 8-track cassette was a common format, I thought you all might get a kick out of this. According to Boingboing, rockers Cheap Trick have given a tip of the hat to the decade that spawned them by making their new album, The Latest, available on the long-obsolete 8-track format.

It's easy to kind of snicker at this, but I bet they do pretty well with them. Hardcore music fans love different formats, and Cheap Trick has a big enough fan base even now that will probably love the sly in-joke. I think this sort of thing represents one of the possible futures for the music industry - not 8-track, smart-ass, but doing smaller runs of specialized releases that cost more than the standard release for the die-hard fans. Making more from less, so to speak.


Up next, the complete works of Frank Sinatra on 78 RPM wax cylinder.

Roland SH-09 on Ebay

Excellent bass synth....

Info at the listing...

Friday, July 3, 2009

Steve Albini Student Lecture

Whether you're a fan of his music or production style or not, Albini has been at it for a long time and has a lot of valuable and often unconventional info to share. This is a 90-minute student lecture Albini gave last winter in NYC.

UPDATE: This apparently is an older lecture that was given in Tennessee, not NYC. Still a valuable view.

Enjoy.

Roland MC-202 on Ebay

http://rover.ebay.com/rover/1/711-53200-19255-0/1?icep_ff3=2&pub=5574751254&toolid=10001&campid=5336116974&customid=&icep_item=160346221706&ipn=psmain&icep_vectorid=229466&kwid=902099&mtid=824&kw=lg

Thursday, July 2, 2009

Beatbox Dub FX "Love Someone"



So this is what busking in the 21st century looks like! I love stuff like this. I think it proves nicely that you don't need a wall of gear to make music, just a little creativity and your mouth. (Actually, that's good relationship advice too... haha)

Roland System 100m on Ebay


Info at the listing...

Wednesday, July 1, 2009

Tori Amos Commercial

I guess that's what all that 'Corn Flake Girl' business was about... haha

Rob Papen Announces Algorithmic Reverb


Softsynth creator Rob Papen has announced his next product will be an algorithmic reverb effect called RP-Verb. It is expected to be released at the end of July and will cost $179.

Changing Your Sex The Easy Way

As I've mentioned a time or two recently, I'm in the final stretch of production for my next album. Today I was working on one of the final vocal tracks and was getting frustrated. The verses sounded fine, but the vocals on the chorus didn't punch through and 'lift' the way I generally like them to. One of the techniques I use a lot on choruses is to add another layer of vocals sung an octave above the main one. It adds some energy to the track and brings the chorus up a notch nicely. But with this song, the main vocal was already somewhat high, and an octave up was simply out of my vocal range. For the hell of it, I tried using a pitch shifter on a second layer of vocals, but, predictably, this lead to the dreaded "munchinization" effect and it sounded like I was singing a duet with the Chipmunks.

The reason this happens when you use a pitch shifter is because it not only shifts the pitch of the vocal, but also the formant and this doesn't happen in nature. Formants are a function of acoustic resonance. In layman's terms, part of what makes your voice sound the way it does is because your body and throat resonate with the frequencies of your voice, much in the way the body of an acoustic guitar does with the vibration of the string. So whether you sing a low note or a high note, the size of your body and throat remain the same and thus, the formant remains constant. Pitch shifting shifts the formant as well, so as you go higher, it sounds as if the voice is coming from a smaller body, and as it goes lower, it sounds like it's coming from a bigger one. (This is all gross over-simplification, but I'm trying not to put you to sleep...)


Enter Logic's Vocal Transformer effect which allows you to change pitch and formants independently. I pitched the vocals up an octave, which resulted in the expected helium-type effect, but when I also nudged up the formant (I used a value of +4), my vocal track suddenly bore a striking resemblance to a natural female voice singing the part. If I were female (apologies to anyone who just got that image in their head), I might pitch my vocal down an octave and nudge the formant downwards. As it exists now, the Vocal Transformer is not without some pretty significant artifacts that mean you can't use it on an isolated track without being obvious (although some might dig the unnatural timbre), but used as a harmony part, it came out pretty damn convincing. Convincing enough that I sent the track to my bandmates and they couldn't guess the source of the foxy female backing vocals.
(For you non-Logic users, there are undoubtedly other effects out there that do the same thing...)

So if you'd like to hear what you'd sound like as a member of the opposite sex, remember that shifting pitch and formants
both are necessary for convincing results. And that's a lot cheaper than a trip to Thailand.

(The album doesn't come out until October, so I'm not going to post a sample here, but for those curious enough to wait a few months to hear it, just look for the track called "Angels & Demons".)

Macbeth M5N on Ebay


Info at the listing...