Adventures in home recording, Part 1
Back in the early days of the pandemic, I was unemployed, depressed, and locked inside. I was stuck. So I did what countless other people did, and picked up a pandemic project: I was finally going to learn how to use my recording interface. It’s a tiny one that I was given a year or two prior, just a little portable firewire thing. But that little thing was a big monkey on my back, because I knew opening that box opened a door to a world of challenges. So, like most intimidating projects, it sat unopened for a long time.
It came with its own DAW software, so it was really all I needed. I’d had some slight experience with Cakewalk in the past, but it barely scratched the surface. I had a backlog of songs from old bands and ideas that I’d stashed away – a sequence stored in a keyboard here, a live recording of an idea there, etc. With no obvious endpoint and no band to rehearse or create with, I set myself a path to learning basic home recording. I took the four song sketches I had and ordered them such that working on each one would require me to learn something new about ... everything. These weren’t to be final things, just high fidelity demos (I tell myself, still wanting to play and record with a band). I did set one big constraint: I wasn’t going to buy any other virtual instruments or effects; I was going to use only what was in the tin.
Of course, in retrospect, there are a bunch of things that I didn’t know I would need to learn about, things that I’d heard of for decades growing up but never needed to know myself, like effect chains, rendering, compression, etc. If I’d had an understanding of all of the things I would need to learn, I probably wouldn’t have been able to force myself to start. And if I were just writing (rather than recording) music and going to a studio, many of these things would be someone else’s problem; so learning these things is either useful for bootstrapping DIY recordings or ... making good-sounding demos? I guess?
Anyway, because I don’t know what else may happen (ever) with these, I decided to share them – and their backstories – here.
Track 1
Goal: The first song was a piano sketch I’d stashed in an on-board sequencer over a decade ago. This song would force me to learn basic DAW use, song setup, and – the biggest single hurdle – using a drum plugin to record a drum track.
Drums were my biggest fear. Despite playing with drummers for most of my life, I don’t play drum set, and I don’t hear drum patterns in my head. I had no idea what drums would even do in this song, let alone how I would “record” them (or if it would turn out well).
First, I set up a tempo and time signature that matched the sequence, and saved that in the DAW (Presonus Studio One, plug plug). Using a click track, I recorded the whole song as I envisioned it, as a piano reference track only. Then I recorded a bass track (using a synthesizer, as my attempt on a real bass guitar revealed that years of being next to a bassist does not make you a bassist), then a few synth tracks (strings, organ, etc). Then I re-recorded the piano parts and muted the reference track.
Next, I spent a few hours (ok, maybe days) setting up the virtual drums via MIDI (EZ Drummer, plug plug) and learning how to trigger and record them. I recorded one measure at a time until I had a pattern I liked, stitched bars together until whole sections were complete, and then edited the MIDI to do things like fills, flams, and rolls. When I was done, I noticed that it wasn’t as “tight” as I would’ve liked (this is a huge understatement: it was sloppy as hell, with beats not aligning across different instruments and editing artifacts all over the place), undoubtedly because I didn’t make sure any takes were precise to the click and I wasn’t paying enough attention to blending my edits. But doing the drums last didn’t solve the problem, which should have been something I already knew from recording with humans (you almost always record the timekeeping part first). Lesson learned: part of laying down the foundation needs to be a basic drum track or a religious adherence to the click. Maybe both.
And then there was guitar. I hadn’t planned any great shakes for guitar (I had in my head this idea of an e-bow for the main melody, but I didn’t own an e-bow), so this was approaching an unwritten part on an instrument I can barely play. I ended up using a mixture of synthesized guitar and real guitar for even the most basic of background power chords. I did record a melody track on guitar and tried to cut and paste parts of it, which ... kind of worked, but sounded terrible. I also broke the high E string on the guitar and ... left it off. I tried tracking an iPad instrument (my iPad is old, it was an early prototype for what would become Geo Synthesizer), and it turns out that you should practice any instrument you don’t play if you want to be able to do it well. Same deal as the bass. So. I ended up practicing this on guitar for a week, doing some damage to my wrists, and then was able to track it well enough to capture raw materials (and edit the rest).
This song wasn’t trying to be any particular thing. I just wanted to make it what it was from that initial idea, see if it turned into something that sounded decent, and get on to something more ambitious.
Listen to track 1, “altitudes” on soundcloud
Track 2
Goal: This song would require me to learn how to change tempos, use multiple busses and more automation, and – the next biggest hurdle – recording vocals. I am not a singer, and I don’t love singing, so this wasn’t about getting good vocals. It was just about having vocals present in the song and sounding ... passable.
I must’ve worked on the second track for at least a year. It was an unreleased/unrecorded idea from the late stages of an old band that I wanted to take in a different direction. Bolstered by my success on the first track, I decided to tackle this collage of a song. But this wasn’t an ordinary song, and I underestimated what I was getting into.
Years ago, when this was an idea for an old band, our guitarist told me “you know, if there was ever a night of songs in the vein of Pink Floyd, I’d want to do that song.” And this got me thinking that it would be interesting to write a Pink Floyd song.
Now, this was the height of the pandemic; no one was thinking straight. This is a ridiculous idea on many levels. For one, it’s almost impossible for someone to write a song that sounds like another artist’s style. It’s easy to steal bits and pieces, tones, lyrics, even moods, but to create a whole song that sounds like it could be another artist? Almost impossible. Secondly, I’m no singer and Pink Floyd has two quite distinct lead singers (Roger Waters and David Gilmour). Thirdly, it has David Gilmour, and it’s hard to conjure the spirit of Pink Floyd without that man’s guitar sounds, let alone his solos. Fourth, it has... well, it has Pink Floyd. Say what you will about them (and growing up, I certainly had no shortage of wrongheaded opinions about the merits of the players in that band), but they are like a band of Ringo Starrs: maybe they’re not in the top 10 of the most skilled players at their instrument, but they write the right part for the song. Every little synth sound Wright used, every voicing, every basic beat Mason played, these all contribute to the feel of the song, and to what the listener associates with the name “Pink Floyd.” Lastly, and certainly not least, which Pink Floyd? That band had several distinct eras, so what do you choose? Do you want to sound like the more psychedelic era (like Atom Heart Mother, Meddle, Obscured By Clouds)? Perhaps the height of concept era (like The Dark Side of the Moon, Wish You Were Here, Animals, The Wall)? Do you want to sound like the coda to a generation (like The Final Cut)? The bars are very, very high and, frankly, if you’re trying to make a song that sounds like this band, you have to go near a lot of these targets, but miss all of them. You need to, like they did, echo past and future material with what you do, which makes this task even harder: whatever you decide to do, you have to choose a point in time to target.
But of course you’re going to miss. That’s the whole point of a creative constraint anyway, to try and trace an impossible line in order to find a new shape. So it’s OK if you set an impossible task. And that’s exactly what I did.
This song starts with a basic beat in 5, while some tonally ambiguous chords wind their way through a two-bar progression. It’s a prelude that fits well in the post-psychedelic/pre-concept era, and it fades out with the sound of a car crash (a nod to “On The Run” but also setting the context for the song’s lyrics) as the next section of the song fades in.
The next section is the verse, and it was the last thing written for this song. I had written words for this years ago, but had no guiding light for a vocal melody. I wrote and rewrote this more times than I can remember. I notated vocal lines on paper. I reused material from the intro. I had attempts on multiple pieces of paper around the office for a year. I finally ended up with a reasonable verse, and then almost a year later added a bass line (this time with a real bass).
The chorus is the first thing I had written for this song. It sounds like an amalgam of a bunch of concept-era Floyd (a little The Wall here, a little Wish You Were Here there, and so on). It’s just three chords sewn through some quasi-chromatic runs, creating an endless loop.
The thing after the chorus (let’s call it a post-chorus) was the first new material I created for this song, and it is absolutely unapologetically concept-era Pink Floyd. It’s part “Dogs”, part “Sheep”, and part The Wall. It’s the seed of the rest of the song, and it has its roots in some of the clusters from the intro (and some of the stranger chords from The Dark Side Of The Moon). If this part nails the aesthetic the best, it’s because I have spent years listening to Animals, and “Dogs” is the best Pink Floyd song off of that, the best Pink Floyd album, and if you think otherwise on either point, you’re wrong. But I digress...
The bridge was unchanged from when this song was first written. It uses harmonic material from the intro, but is intended to be a completely different sound from the rest of the song, like an interlude from earlier Pink Floyd. It’s probably a little more progressive than they would really have gone, but it works.
After this is the end of the song, where I stopped trying to sound like Floyd. After the post-chorus returns (of course, slightly faster, so I couldn’t cut and paste it, ugh), it settles on a low E and starts sounding more like Tool than Pink Floyd. I miss both of these targets by a mile, but at this point it doesn’t really matter, because it’s all about riding this section to the outro (which is a return to the intro). In theory, this outro should have some sort of musical sound with it; but I’m no guitarist, and I couldn’t come up with (or articulate for someone else) the kind of melody (or even the basic melodic material) that a real guitarist would use, so I resorted to just sounds. The real star here, to balance the intro’s NASCAR crash, is sound bites of people who ... are public figures despite being liars.
Overall, I was pretty happy to just finish this. It’s not successful as a Pink Floyd song, but it does sound like it came from identifiable roots, so I’ll take that as a win.
Listen to track 2, “alive in this lie” on soundcloud
Track 3
Goal: This song would require a few time signature changes, but no vocals.
The third song was another idea from another old band that never got off the rehearsal floor. My old band’s drummer, Mike, wrote the A section as the theme for a song which he called “Darkness Visible” – Mike is a super-smart guy, but I couldn’t tell you if he based this off of Styron or Golding or both or what. But I wrote the bridge years later (and years ago), which changed the tone of the piece, so I wanted to reflect that in changing the title.
The A section is split into two halves, with the first half as originally written (in 13/8 with a pedal bass line), and using the same drum pattern as the original. The second half is in 14/8 with more harmonic changes under the same melody – my take on Mike’s original idea.
I wrote the B section as a way to get from 14/8 to 16/8 (the bridge is actually mapped as 8/4), to be something that almost feels like an odd meter without actually being an odd meter.
The bridge is something I had actually notated as something I wanted to bolt onto his song, years ago when I was in that band and pencil and paper were all the rage still. So... this is me doing that. Closing the loop, over a decade later.
(There’s a really fun thing after the bridge ends, a whole step down from the beginning of the song. It starts again, transposed down a whole step and then modulates back to the original key after the first half of the A section. Sometimes recording projects can be fun compositional exercises.)
This song is all synthesizer, because I lost all confidence in my ability for even rudimentary guitar when doing Track 2.
Listen to track 3, “ultraviolet” on soundcloud
Track 4:
Goal: This would require me to do a lot of time signature changes, vocals, and guitar. There was guitar involved in song 1 and 3, but not as predominantly as in this one.
The fourth song was the last song my old band wrote. We performed it live once, at our last show. It was more of the style of that band, so recording this song would require me to try and rewrite parts of it, forcing it into a new style. In a way, testing whether that old band could have evolved into something I wanted it to be.
But this song was so complicated in structure that to put it together (even before creating a map in the DAW), I had to write out and organize it on a giant sheet of paper I had taped to the wall (for years) to make sure it made sense. It had a fully-formed beginning and end, but I wanted the verse to get away from the “one chord per measure” treatment that that band did a lot (and which I think I ended up doing anyway). I also wanted the bridge to have more structure than whatever we did to it way back when. Being the last “Fast Eddie” song, I wanted to keep the ending true to its origins: with the band split into halves, one half playing in 6 and the other half playing in 7 until they meet again at the end. In the prog rock world, I believe this is called “rock in opposition,” and we did that a lot because we were young and show-offs.
What I hadn’t really thought through in advance was how much guitar there needed to be in this song. It was guitar-based. And I was no guitarist. So I recorded the guitar parts no more than one or two chords at a time, and edited them together. Probably eight different guitar tracks (still using only five strings and my Line 6 PodXT, plug plug). Sometimes a track was only two notes because I just couldn’t play them fast or clean enough to record them along with what came before and/or after. Eventually, I did research new guitar strings and how to put them on a guitar and replaced the remaining five 25-year-old strings with a new – and complete – set of strings (so I could play the verse part, a high arpeggio). Eventually, I muted all but three guitar tracks.
(One curse of this band is evident in this song: I would often supply words, with no melody line, and want a vocal line to sound like Adrian Belew in King Crimson – that’s a ridiculously tall order, for anyone! So when it came time to resolve those two things now, the only person in the room was me. So I had to figure out how to fit the words into the song, for a change. My apologies to my former bandmates for so many things, but certainly this.)
The time changes in this song are excessive, so it’s possible that there’s one final edit I’d make to it (making the last “in” supported by a full six-count measure before the final “time” is sung), but at this point that’s a problem to fix when/if a band goes into a studio and records this for real. This is just a demo, after all.
Listen to track 4, “black and gold” on soundcloud
Coda
Since starting this project, I have finished recording these four songs and also used my home recording rig to create tracks for three other records (sending them to engineers to be used alongside a reference track, instead of physically going to a recording studio). I am not claiming to be any kind of expert now, but I can get by.
It’s hard to be creative, and even more so when you have an unrelated fulltime job. I’m the kind of person who needs collaborators, so going it alone (even for demos) is not something that I would normally attempt. This took a lot of determination, but it also required support. I wouldn’t have gotten this far without the sounding board and encouragement of the multi-talented Mike Conley, who was game enough to start talking about balancing creative endeavors and work with a stranger at a work offsite years ago. If you don’t already know his band, you should check out The Johnson Report right now.