It’s been a while. I’ve been working on a whole bunch of songs together, which is great for giving one song space while I work on another, but also not great because overall progress is slow. But here’s the first of them, fresh from the oven.
Tell Me What You’re Looking For is on Spotify and YouTube (and probably anywhere else you like to stream your music).
The Story Behind It
The original version of this song was written between twenty and twenty-five years ago, and was called Head On Fire. It was on an cassette tape album that I made with a friend and musical collaborator1, using a little four-track cassette-tape recorder. We recorded it in my living room, and gave it away to friends and family. As far as I know, no copies have survived, which might be a good thing because it featured me singing.
Then at the end of 2023 I was reading a book (Rip It Up and Start Again by Simon Reynolds) on the New Wave/Post-Punk era in the UK from 1978-1984, and I decided to see if I could write a song that used some New Wave/Post-Punk ideas, such as:
A bass part that’s more forward and melodic
A solid rhythm section foundation with a Big Harmonic Space above it for vocals
A big ol’ chiming guitar (as in bands like Simple Minds)
Since I liked the words in the original song, and it had something of a post-punk feel, I decided to steal ideas from myself and update it.
Some of the guitar parts were written after listening to post-punk albums. One of my favourites from that era is PiL’s 9 (which isn’t on Spotify, but the track Disappointed is a good example) and it definitely influenced the melodic and harmonic ideas. A bass part that’s upfront and melodic is something you’d hear on Joy Division/New Order tracks, but I’m more influenced by The Cure. Maybe there are echoes of their track A Forest (Spotify, YouTube) in the rhythm guitar…
As I usually do with a mothershout track, I gave myself some creative constraints:
Follow the usual Rules of mothershout songs - only “real” instruments (bass, guitar, drums, electric piano, vocal), no loops, sound like a band playing the song all the way through, recorded together in a studio.
Use one chord. For someone who loves jazz harmony, this is a tough constraint.
Make the bass part prominent, and use a New Wave type of chorused bass sound (makes the bass stand out more).
When arranging the parts for the instruments, try to use lines (single notes) instead of chords.
Let the vocal notes define the chords - be minimalistic.
I think it worked out quite well.
What’s It About?
A fancy way to put it would be: it’s about the way that emotion or lust overcomes reason.
A less fancy description would be: when you’re turned on you don’t think about consequences. To have a head on fire means to be turned on, and you don’t think about consequences because thinking gets hard to do.
That was in the original song, but I added a new chorus to change the meaning. The singer’s tell me what you’re looking for puts her firmly in control of whatever’s happening. I was thinking of something like the dynamic between Lauren Bacall and Humphrey Bogart in To Have and Have Not, where she’s the one in control. That gave me a strong idea of how the vocal should sound, and led to lines like I don’t think you’re really sure. The words are not personal, in that I would never sing them. Though I’ve known a few women who might.
Hit Me With Your Rhythm Stick
Up to now, my drum parts have been generated by Logic’s built-in Drummer software, which will create basic drum track and fills. I’ve then taken the generated tracks and edited them to get the drum track I wanted. That works, but it’s difficult to get a good drum track that’s not too “busy”. Now I’m trying a new approach, and this is the first track where I’m playing the drums.
Not on an actual drum kit because (a) that would be really irritating for everyone else in the house and (b) I don’t have the necessary physical coordination to play a real kit. I played the track using a technique called finger drumming; the drum sounds are playd with a keyboard or a “pad controller”. It’s easier than playing a real kit, but that doesn’t mean it’s easy. I’ve been working through lessons from the excellent Quest For Groove website, and doing a lot of playing along with all sorts of music to get a deeper understanding of drumming.
This is what it looks like. On the soundtrack of this video, you’ll hear an additional loud “click track” (a cowbell on every quarter note) that played as I was drumming, to help me keep in time. The pads I’m using are colour-coded to the different parts of the drum kit. The kick drum is the bottom two blue pads, the snare is on the two purple pads above them, the hi-hat is the three light blue pads and the toms are the top row of red pads. The left and right cymbals are the two orange pads at the bottom left and right.
The drumming for this track isn’t very complicated, but it has to be played in time to create a solid foundation for the song, so it took a couple of weeks of practice to be able to play it well enough. It sounds like a drum loop or a drum machine, but I did genuinely play it all the way through from start to finish, several times. What you hear on the track is a combination of the best two takes (the take in the video was not good because I messed up the hi-hat pattern, so I didn’t use it).
The drum sounds for this track are from Native Instrument’s Abbey Road Drummer. All the sounds for these (really excellent) virtual drums were recorded in Abbey Road Studio 2, which has a great natural sound. This track uses the 1970s Tight Kit.
A production geek note: Abbey Road Drummer has the “room sound” of Studio 2, and I left it on the drums (you can hear it in the “bare” drum audio example below). I set up a reverb (an LX480) with as close a match to the room sound as I could get, so that everything sounds like it’s been recorded in the same studio.
Let’s Take It Apart
I’m going to break the track down into the component parts so you can hear how they work together.
Something you might notice as you listen to the audio examples below is that often an instrument played on its own will sound different to how it sounds in the full track, and sometimes it sounds worse. This is very normal in recorded music - what matters is how everything sounds together. For example, if there’s a bass instrument on a track, it’s common to take out some of the bass sound from the other instruments (especially guitars and pianos) so that they don’t compete with the bass in the lower frequencies. On their own they then sound weak, but when the bass is playing along with them, they sound fine.
Slave To The Rhythm
We’ll start with the drums, because they’re the foundation that holds the whole thing together. This is what they sound like on their own:
But you don’t hear the drums in that “bare” form on the full track, because there are some audio effects applied:
The cymbals and the hihat have a “flanger” effect on, to make the sound “sweep”.
The snare has an effect called a “ring modulator” on, which echoes the sound as well as changing it.
With those effects, it sounds like this:
There’s also a set of percussion instruments. The most important are the bongos (which I played like the drums, using a pad controller). Here’s what they sound like on their own:
You should be able to hear that they’re all on the left side of the stereo track (especially if you’re listening in headphones). To balance that out and add interest, I spread them out over both left and right sides by adding a delay effect (a delay is an echo whose timing can be precisely adjusted). This delay takes the bongo sound from the left side, delays it by exactly one quarter note, and plays it on the right side. And it sounds like this:
Notice how it’s now a more interesting sound - it has movement in it. Any hit on the left is repeated on the right one quarter-note later. In the full track you probably don’t consciously notice the bongos, but they add a lot to the “beat” of the song.
The other percussion elements are a shaker, a tambourine and a cowbell. The cowbell plays strict quarter notes in the choruses, and the others play off-beat patterns that match the timing of other instruments:
And if we put that percussion together with the drums, we get the basic rhythm track (this is from the start of the second chorus):
The percussion instruments used on this track are all from the Logic Pro factory library. I used the “Performance” versions of the instruments to get more variation in the sounds.
All About That Bass
The next layer of the track is the bass guitar. The bass has two jobs:
The notes played define the “root” of the harmony. Because this song is essentially in Bm, there are a lot of B notes.
The rhythm of the bassline notes accentuates the “pulse” of the music, and has to work with the drums, never against them.
This is the bass guitar on its own, playing the chorus bassline:
And if we put it together with the drums and percussion, we get the full rhythm section of the band. Listen to how the bass notes match up with the timing of the drum hits, especially the kick drum (the lowest drum). In a couple of places, the bass leaves a gap just as the snare drum hits, and in other places the bass and drums play exactly together, so that they “wrap around” each other.
This is played on a Squier Mini-P short-scale bass, through a Line 6 Helix amplifier simulator, with a touch of chorus added for more New Wave flavour. I usually play bass with my fingers, but this time I used a pick to get a more crisp sound.
Mr Guitar
Next, the rhythm guitar. The word rhythm is important here - this is not a flashy, look-at-me, Guitar God part. Its role is to build on top of the rhythm section of bass and drums by emphasising the rhythm, and to help define the chords of the song.
This is the rhythm guitar part on its own:
It does the job it has to and nothing more, keeping it simple. In the full track, I used a similar trick as with the bongos: a delay that bounces the sound from the left to the right. Delayed guitars were pretty common in the New Wave era, when digital delay guitar pedals had become cheap enough for most guitarists to afford, and players like The Edge from U2 and Charlie Burchill from Simple Minds made them an integral part of their sound.
The bongos use a quarter-note delay, but this guitar is delayed by three eighth-notes (a “dotted quarter note”) before you hear the echo on the right. The delay creates a much more interesting sound that has movement in it:
And when you put it all together with the rhythm section, you get this, the rhythmic basis of the track:
This guitar and the chorus guitar below are my Epiphone Dot, played through a Line 6 Helix amplifier simulator.
Sweet Harmony
Both the bass and rhythm guitars are only playing single notes. There’s an electric piano playing two-note chords in the background to fill out the harmony:
…and there are also some strings that play long notes very high, and quietly, through the whole track:
Adding those in to the rhythm guitar and rhythm section fills out the “space” above the solid bass-and-drums section.
Try comparing that to the audio above that doesn’t have the harmony (the one above the heading “Sweet Harmony”) to hear the difference it makes.
Leaving space in an arrangement can be challenging, especially in a real band where every musician wants something interesting to play, or an opportunity to show how good they are. When I first recorded a draft sketch of this song, the rhythm guitar and piano were playing full chords and there was much less space. I pared the parts down to the smallest number of notes needed, and the whole thing sounded much more open.
This electric piano is from Modartt’s Pianoteq set of electric pianos. It’s the Vintage Tines MkII (the Flower Power patch), which is really a Rhodes MkII (Pianoteq don’t use the name Rhodes, presumably because copyright and trademarks). The solo electric piano is the same sound with a ton of effects on it.
Let’s Groove
In the verses the focus is on the voice, but in the chorus there’s less vocal, and the chorus guitar riff helps changes the overall sound so that the choruses sound different to the verses. Here it is on its own:
It’s an electric guitar with a touch of delay, and a little bit of a “chorus” effect that mimics the sound of two guitars playing the same part, with very slight timing differences. This is a good example of what I wrote about above; an instrument that sounds worse on its own. Here’s how it sounds combined with everything else so far:
Now it’s beginning to sound a lot more like the full track, except that it’s missing the vocals.
The Voice
(Did you notice that all the subtitles in this section are song titles? This one is a reference to The Voice by Ultravox (Spotify, YouTube), which came out in 1981, in the middle of the New Wave/Post-Punk era I’m referencing)
This is the vocal line that starts the second chorus:
That’s two voices singing in harmony, with the higher notes quieter than the lower so that they blend together better. We tend to hear the higher notes as the more important melody, but I wanted both voices to be equally important, so I lowered the volume of those higher notes.
You should be able to hear that there’s a delay effect on this voice, echoing the words after the singer stops; in the full track it’s almost inaudible, but it helps the voice stand out amongst the rest of the instruments.
In the chorus of the track there’s yet another delay effect, which takes the last word of the vocal (in this case, “for”) and echoes it using a “tape delay”2. Tape delays were invented back in the 1950s. They use a loop of magnetic tape to record a sound and play it back after a delay. Because of how they work, every time the sound is repeated it’s a little more distorted. Here it is combined with the original vocal:
(Nearly all the delays used on this track are tape delays, because I like the way that the echoes are different to the original sounds).
I’m using the second chorus of the song for these examples because there’s one more vocal trick to show, and it’s only used in the chorus. It’s that lovely 1970s classic vocal effect, the vocoder. You’d recognize it from tracks like ELO’s Mr Blue Sky or Sweet Talkin’ Woman as the sound of a “singing synthesizer”.
A vocoder takes a “source” sound:
…and a “modulator” vocal:
…and combines the two, like this:
Can you hear the words “tell me what you’re looking for” in that last audio? The “tell me” is muffled, but “what you’re looking for” can be heard if you listen closely.
In track like Mr Blue Sky the vocoder “sings” on its own. On this track it plays at the same time as the voice so that it “sits under” the voice, like this:
By the way, that “modulator” vocal above is what the singing voice sounds like without any changes to the sound. I’m glossing over many technical details of what’s done to an instrument or a voice to get from the “raw” sound to what you hear on the track, but if you listen to that “modulator” example and the vocal example at the top of this section, you can get an idea of how much “polishing” has to be done.
The vocoder “source” is the first time I’ve used an actual synthesizer sound3 on a mothershout track, because generally I stick to the sounds of real instruments. I decided to allow it because the particular synthesizer sound I used is very 1980s, and a New Wave band would have had access to that type of sound.
The vocals are, as usual, generated by Dreamtonic’s Synthesizer V vocal synthesis plugin. The lead voice is Eclipsed Sound’s Solaria and the upper harmony is Dreamtonic’s Natalie. When I was arranging the song, I tried it in several different keys to find the one that brought out the soft, slightly husky tone in Solaria’s voice.
This track took a lot of voice work to get the timing, phrasing and sound of the vocals right. One neat example is on a line in the second verse: a little opinion and a lot of soul. Just to get really detailed for a moment (and for the other Synthesizer V users who read this stuff), this is what that line looks like in the Synthesizer V software:
The line at the bottom is displaying the gender setting - when it’s in the middle, Solaria’s voice is its standard female sound, but under the ou of soul the line temporarily rises to make the voice more male for just that syllable. The right hand side panel shows the different syllables (also known as phonemes) of the word soul: s ow l and ax. The last ax is an “uh” sound that emphasises the end of the word. Next to each syllable are sliders that adjust how long the syllable is (Duration) and how strongly it’s sung. You can see that the l is extra long and strong, so that the word is pronounced as soullllllll, lingering on that l, and that the ax or “uh” sound at the end is as short as possible so that it’s barely audible.
If you look closely at the syllables (shown above the notes), you might be able to tell that the actual words sung are more like uh liddluh pinyun an uh lodduh sowl. Programming vocals like this gives you a real insight into how humans really speak, and how much we change words (especially in English) to fit together.
The style of voice also changes through the track. At the beginning, both Solaria and Natalie are singing in their “soft” voice modes. From the start of the first chorus, more of the “clear” voice mode is blended in, increasing as the song progresses so that the sound of the voice helps build up the tension.
To create the vocals, I played the vocal melody using an electric guitar, singing along to get the phrasing right, and recorded the guitar sound. Then I used software to convert that guitar sound to MIDI - a digital record of the specific musical notes that I’d played. I imported that notes into the voice synthesis software and added words and syllables to each note, and then I carefully and painstakingly tweaked many, many parts of the vocal to get Solaria’s singing exactly right. Finally, I copied Solaria’s vocal and changed it to be sung by Natalie as the harmony vocal, figuring out the harmony intervals on a piano and adjusting the harmony vocal notes to match.
And if you think that sounds like a lot of detailed work, you’re right. It’s not necessarily easier to use a computer-synthesized voice instead of a human.
Come Together
To complete the chorus, we add the vocal and the vocoder to all the other parts:
And that’s it. Essentially, it’s two guitars, one electric piano, one bass, drums and percussion, and vocals. Sounds easy!
The Artwork
I have two challenges when it comes to artwork. The first is that I’m no graphic designer or illustrator. The second is that it’s difficult to come up with an idea for an image to represent a song. I’m better at audio than visual.
When I’m stuck for musical, arrangement or mixing ideas, I go off and listen to random(ish) selections of music to give my inspiration a kick. So I did the same thing, browsing sites like DeviantArt to get ideas, looking at anything and everything, in as many different styles as I could find. I took inspiration for the colouring from one image, the “vertical bars” pattern from another, and the female face is generated from a photo of a marble statue.
The final image started as a set of AI-generated pictures4. I combined and reworked them, and added the frame and the lettering. I like the way the colouring of the image subtly reappears as filling in the letters.
One of the edits I made was to change the eyes from open to closed. I can’t say why, but with eyes closed it looks more like the singer as I imagine her in this song.
The Music Theory Bit
One of my friends told me that she likes these writeups but skips over the musical details. I promised her I’d note when the really geeky stuff comes up. And here it is…
Harmony
One of the constraints I gave myself for this track was to use one chord. Just one. And this song pretty much does that - it’s B minor (Bm) all the way through. But dig a bit deeper and it gets more musically interesting. The song has a Bm “flavour” but there’s a fair amount of hidden spice.
This song uses the B Dorian scale; a natural minor scale with a major sixth (a G# instead of the natural minor’s G). That three-note guitar figure that you first hear in the intro ends on that major sixth G#, and it’s in there so that the listener hears the scale early on. Adding that G# changes the chord from Bm to Bm6.
The bassline uses mostly tones from a Bm7 chord; the minor third only appears as a passing note, and in the intro that note is hidden under the three-note guitar figure, so you don’t hear it. That keeps the sound more ambiguous - it’s not assertively major or minor.
In the intro, verses and chorus, there’s an electric piano way in the background playing an F# and a B (the fifth and the root notes of a Bm chord) for seven beats and then an E and a G# for nine beats. That E and the G# add an eleventh and a major sixth to a Bm chord, but because those notes are hidden in the background, they’re kind of subliminal. You don’t actively hear them, but your perception of the harmony is affected by them.
In the chorus, the guitar riff is playing figures that use the notes from Bm, A and E chords, all over that B-based bassline. You could think of those as Bm, Bm11 and Bm6(11).
Those same variations appear in the piano solo - it’s all E, A and D chords over a B bass note.
The harmony vocal sometimes moves to F, which changes the chord to a B diminished or B(♭5).
So it’s all Bm, but it comes in a wide range of flavours.
Melody
The basic vocal melody is pretty simple, but has an important feature - it’s mostly on D, the minor third note of a Bm chord. The rest of the instruments tend to stay away from that note to keep the sound ambiguous about whether the scale you’re hearing is major or minor, but the vocal melody tells the listener that we’re definitely hearing a minor scale.
From the second verse on, there’s a harmony vocal, and this is the voice that gets the fun, musically interesting part. While the original voice stays mostly on its same simple melody, the harmony adds interesting intervals on top. Sometimes they’re just thirds (a very basic style of harmony) but in other places they’re more complex, such as selling it all to get a little more (second verse) or whatever you want is true (third verse). Here there are minor thirds where a major would normally appear, perfect and diminished fifths and fourths. In a couple of places, I used parallel harmony, having the upper harmony stay a strict number of half-steps above the lead voice, ignoring the scale. Underneath, it’s all jazz…
Arrangement
The arrangement of this song uses a bunch of tricks/techniques to fit the different parts together:
There are gaps in the bassline left around specific drum hits. In the verses, the bassline is written so that key notes fit into the gaps between the vocals.
Although it’s difficult to hear, the background electric piano changes chord at exactly the same time as the bass plays key notes (such as the “pull-off” or “snap” sound that comes after the words head on fire in the first verse), so you almost never hear the chords change, only that they have changed.
The three-note guitar figure5 in the intro (D, E and G#) comes back once in the middle of the second verse, and then twice in the last verse, which helps the track change and build as it progresses. It fits in between vocal phrases, timed so that the last note fades in time to let the vocal be heard.
The bass plays that same three-note figure at the end of each two-bar section of the chorus, ending with a slide down from the G# back to B to restart the next two-bar section (I have a bit of a thing for sliding bass notes).
The piano solo was played with all the effects6 on (i.e. they weren’t added later), and written to let the effects stand out. For example, at the start of the solo there’s a gap after the second note, left there so the echo can be clearly heard.
In the first verse, the rhythm guitar is playing a different rhythm than the bass, but in the second and last verse, it plays most of the same rhythm, so the guitar and bass blend together to sound more like one instrument.
In the solo section, the bass part is simplified and the rhythm guitar plays fast muted 16th B notes, so that they form a background and let the solo piano stand out.
The drum part is deliberately simple, and only uses two drum fills. The first is played immediately before the intro starts (a double-kick, one snare and one tom), and then re-starts the drum part after it pauses before each chorus. The second fill is a four-hit fill on snare and toms that comes at the end of the choruses, without the drums pausing. Re-using the same fills helps tie the song together.
The drum and percussion patterns change subtly between the verse, chorus and solo sections of the track, to accentuate the bass and/or rhythm guitar.
All three verses start with the words loving is easy with a head on fire (with either if or yeah before), so that they link together.
All That To Say…
Does it work? Does it meet my goal? I think so. For me, it’s been a couple of weeks since it was finished (I have to upload it to my music distributor’s website in advance to allow time for it to be loaded onto all the various streaming services). Over that time, I’ve continued to listen to it, and I think it works. It’s New Wave-ish enough, and has the sort of soundscape I set out to build. I’m pleased with it.
But of course, you can make up your own mind.
And the next song doesn’t sound anything like this…
When I wrote this paragraph, I realized I’d never Googled to see what my friend Colin has been up to... Turns out that he’s still writing music, and is on Spotify.
Technically, it’s Logic Pro’s digital emulation of an actual magnetic-tape delay, because technology. And real tape delays are big, clunky devices that need too much maintenance and care.
Okay, so… technically, a lot of the sounds I used are generated by synthesizers, including the vocals. But they sound like real instruments, not like you’d expect a synthesizer to sound, and that’s the key thing. I avoid synthesizers because they’re too flexible - I could lose a week just playing with different sounds and not actually making any music.
I have mixed feelings about using AI in general, including generating images. But I spent a while looking at the portfolios of human designers who could produce a record cover within the kind of budget I have for doing this, and didn’t find anyone whose work looked right.
It’s mostly guitar, but there’s an electric piano playing the same notes in the background to fill out the sound.
A production geek note: the delay on the solo is a dynamic delay, which is ducked by the main piano sound so that it doesn’t clutter it up. Then there’s a chorus and autofilter that come after the delay, so that they change the sound of each repeat.