Tuesday, April 5, 2011

Ode to music production circa 2003

Well, while I am waiting to begin work on the final master of yet another song, I decided to write a little something to pass the time.

A lot of people (fellow musicians and/or people wanting to produce) ask me what I use to make my music. I tell them: "A variety of things, but mostly FL Studio." Part of that's a lie. I don't use FL Studio...I use Fruity Loops V3.6. Why? Super old computer, that's why.

I bought my PC in October 2003. It has a 1.7 Ghz Athlon processor, 1 GB of DDR RAM, and an SB Audigy soundcard (the first Audigy with a front interface). I run Windows XP Pro. It came with an 80 GB drive which I am constantly trying to clear out to make way for new projects. I added another 320 GB drive which I am also constantly trying to clear out, and recently got a 1 TB external drive to move files to when I am trying to clear out my internal drives. For software, I run the aforementioned FL, Ableton Live 4, Wavelab 5, and Acid 5, as well as a variety of soft synths and FX plugins.

So here is my process (for the most part): I make my synth and bass loops and samples using Ableton, as loading up any sort of VSTi thru FL uses up most of my resources. I can generally only load 1 VSTi at once (two if I'm lucky) before the song begins to crackle and pop (I also can't use Ableton for anything else!) I render that at anywhere from 1-8 bars, and throw it into FL (although the longer the sample, the more it bogs down after a while). I do all the percussion and some FX in FL, so it's a real back-and-forth sort of procedure. Luckily, I am able to keep both open for the most part, and can switch back and forth.

So, make the sample, put it in FL, listen to it, maybe add some more percussion, render that file as a wave and bring it back into Ableton. Make another sample, render: FL, render: Ableton. Rinse and repeat. This process can take anywhere from several hours to several days, depending on how inspired I am with what I am doing. The timeframe is also dependent on how many FL plugins I'm using and how long the synth samples are, because there comes a time eventually when the whole thing gets bogged and begins crackling again.

Adding vocals is fun, but requires even more work. Luckily, not as much as I used to do since buying Wavelab so long ago (I used to record the vocals using Creative's recorder, then chop them up in Soundforge or Cool Edit/Adobe Audition and try to place them properly on the step sequencer in FL). Now I just record through Wavelab, and add FX via plugins there. If I want to utilize FL's awesome BPM delay (which I have yet to see work as well with any other program), I do chop up the lyrics and paste them in.

So now that the song is finished, I master it. This is my least favorite part of the entire process, because just to get to the mastering point is what I call a slow, tedious process. Why? Single processor, low RAM, and severe impatience on my part. :)

Mastering: First Stage

Most of my songs consist of anywhere from 25 to 40 separate instrument tracks, which I render as separate uncompressed wav files at about 75% volume level. Once upon a time, I used to render it all in FL, but I noticed FL tends to leave a certain 'taste' to the track, which I've never liked, and sometimes it'll squelch or duck parts of the song, no matter what kind of compression I use on it. I also like to add EQ, more FX, etc. to certain instruments, so it's best to do it all separately. Rendering each track for a 5 minute song basically takes about 1 1/2 minutes...longer depending on how many plugin FX I've included. Times that by 25, and add in time to make file names for each wav. To make things even more fun, sometimes I forget to turn the compression off and don't realize it until I get to the second stage, so I have to do it all over again. And to make things even MORE fun, I can't do anything else on my damn computer while I'm doing this, since it uses up all my processing power.

Mastering: Second Stage

Now that I've got all the tracks rendered and ready to go, I open up Acid and drag each one into it. It takes a while for them to load up there. In fact, this is what I'm waiting for as I type this. And I can't do them all at once, either. That would just be far too easy! It uses up most of my memory with about 10 tracks. After that, Acid will freeze up and I have to start the whole process over again. I can So I load 10 tracks, wait for them, save and close Acid, then reopen and do the next 10 tracks until they're all in there.

*side note: I just realized that by turning off the Acidize feature when rendering in FL, it saves considerably less memory. Wheee!

Mastering: Third Stage

So now that I've got all the track done and everything in its place, I begin adding FX to each track. Most of it is compression (done on each track individually). Now, most people would say that too much compression is overkill. And they're right...it is...when you are compressing an entire track as a whole. You know there's something wrong when you're finished a track, open it in a wav editing program, and the entire wav is one solid line, because you're not allowing the dynamics to come through fully. Myself, I only add a slight compression to each, and each is bussed thru...generally with 3 or 4 different compression settings for each instrument. As this lifts each track up, I also have to change the volume settings on each track accordingly. Synths and pads (and vocals) also generally have some reverb on them, and anything else that ends up sounding kinda cool. Again, I have to watch the amount of plugins I use...if I get too many in at once the track slows to half its BPM, complete with the fun crackling noises. I also use a frequency analyzer to determine if it's too bass-heavy or hi-heavy. This is normally something that should be done by ear, via monitors, but I'm a very visual person...and I don't have decent monitors, to boot.

When it finally sounds good to me, I save it. And THEN...

Mastering: Fourth Stage

...open it back up. While my songs sound decent enough after everything I've done, there's one final (and a half) step: select all the tracks and duplicate them so I can change the panning levels on each one. This is where I get a little more creative with the final stage. I generally keep my bass and kick around the middle, pan the hats, claps, drumloops, etc outwards to each channel (with varying levels...one channel can be at 20% with the other 25%), and pan synths, FX 'swooshes', etc nearer the end of the channel spectrum. This adds a bit of dimensionality to the song which is usually not apparent until I do this. Since duplicating each track will also change the volume levels, I have to readjust accordingly.

Mastering: Fifth Stage

This is my 'clean up' stage. I'll render the whole thing (takes about 2 min for a 5 min track), then listen to it a final time to see if there's any boo-boos I made, or if something doesn't quite sit right for me. I also like to take the song into my car to see how it sounds there. Sometimes I'll leave the song for a day or two, just because I've been working on it for probably close to a month straight by then, so it's good to take a break and then come back with fresh ears. If I'm lucky, no more needs to be done. If I'm not so lucky, I have to go back to any one of the points to change what needs to be changed, then render the whole thing again. THEN I breathe a sigh of relief, and hope that other people will like it once it's uploaded for the public's listening pleasure. :)