‘gazed and Confused: A Guide to Shoegaze in your DAW

This article was originally published back in 2015. As with other well-read posts on this site, I’m trying to revisit some to tidy up some of the writing and audio and generally make them a little more professional.

Shoegaze is the maligned term for a sub-genre of guitar music that started in the British Isles around the late eighties and culminated in the early nineties. Coined by an NME journalist as a derogatory reference to the guitarists’ motionless performance gazing at their pedalboards (or shoes). It encompasses the likes of My Bloody Valentine, Slowdive, Seefeel, The Jesus & Mary Chain and Ride. Taking inspiration from the washy dream pop of Cocteau Twins and the new-wave Goth sensibilities of The Cure, along with the psychedelic, hypnotic noise jams of Sonic Youth, somehow shoegaze came to be.

Largely based in Oxford, Reading, and London this small scene of indie bands rejected the label, not wanting to be compartmentalised by idle and apathetic print media. However shoegazing has come full circle, and it’s a badge of honor to be enshrined in the same auditory category as the luminaries who created it.

After a brief decline around the mid-nineties, the genre has had a resurgence with bands such as M83, Ulrich Schnauss, A Sunny Day In Glasgow, Pinkshinyultrablast, Amusement Parks on Fire and Deerhunter developing the idiom with modern production techniques and electronics. Many of the first wave of bands such as The Jesus and Mary Chain, My Bloody Valentine, Slowdive and Ride have also released new material within the last few years. Here’s an exhaustive playlist of some shoegaze from across the year, including some more modern stuff too. Suggestions always welcomed:

Part of its sonic makeup is the heavily processed sound achieved by using many layers of distortion, reverb and other effects. The sounds are saturated in warm fuzz and carefully caked in slow, undulating modulations, asynchronous delays and swirling washes resulting in an ethereal but heavy sound that is still sought after by bands and producers today.

Typically shoegaze pedalboards can be constructed with a combination of rare discontinued oddities, quirky digital multi-fx units and a plethora of vintage and boutique effects. The main focus of the sound usually comes from distortion, reverb and delays, and how they’re ordered in the chain. These sounds are achievable within modern DAWs, even with just what’s bundled in Logic and Ableton. I’m going to try and get some of the signature sounds with, where possible, just those plugins and some creative programming

Loveless

Kevin Shields is probably seen as the Lionel Messi of shoegaze and his band My Bloody Valentine are Pep’s Barcelona. Their album Loveless is a coffee-table classic and source of inspiration for guitarists since its release in 1991. The term game changer is probably bandied around far too easily but this album has certainly left an indelible mark, and its influence is still felt rippling through music. That makes it prime for looking at throughout this article.

Loveless was released on Creation Records and took four years to make due to engineers being hired and fired by the band, moving studios, ill health of drummer Colm, alleged budgeting problems and Kevin’s fabled perfectionism. That said there is a certain amount of mythology around the recording of the album: in Mike McGonigal’s book for the 33⅓ Series, Shields puts to rest rumours of nearly bankrupting Creation with their sustained delay of the record, citing other artists on the roster who were doing much the same.

Also around this time, a licensing deal had been brokered with Warner Brothers to the tune of around £70,000, which made a large down payment on studio and other production costs. The album was produced by Alan Moulder and you can read a really interesting write up about making it here, large swathes of which I will be quoting liberally.

Only Shallow

I’m going to start by trying to get close to the distortion sound on their opening track, Only Shallow. Since originally publishing this article back in 2015, Sound on Sound detailed Only Shallow in 2018, it doesn’t change my approach radically but it did shine a light on a few microphone techniques.

It would be a fools errand to try and replicate the exact combination of pedals and amp used by Kevin Shields and Bilinda Butcher, but we can make some educated guesses and we can approximate the fuzz/distortion sounds with stock plugins. I’m going to start by adding Logic’s Pedalboard and the RAWK Distortion. It’s not entirely clear what this is trying to emulate but it’s a good start. Adjust the Crunch to 10 o’clock and the Tone to just shy of midnight.

I’ve also added Logic’s Amp designer using the Small Brownface Amp and Vintage British 4×12 cabinet with a Ribbon 121 microphone. Loveless was recorded across a variety of studios so it’s not useful trying to second guess pre-amps so I’m using Waves G-Series throughout this article to sculpt the sound. There are two tracks of guitar here but both panned centrally. A great deal of Loveless is very mono.

I’ve added a third take with a much harsher, aggressive fuzz using FabFilter’s Saturn and the Monster Fuzz from Logic’s pedalboard. This can go much further down in the mix, just adding some buzz to the higher frequencies. Both this and the first guitars were tracked with a Gretsch G2420T Hollow Body, not the guitar used on Loveless but it’ll do!

The song has some characteristic string bends. These were quite difficult to do not being a great guitarist myself. I used a Fender Strat for these, double tracked into simultaneous copies of Guitar Rig (by Native Instruments) utilising an Auto Filter, Distortion, Pitch up effect and “Citrus” amp (based on Orange, of course). This is one of the less convincing sounds on this track:

To get that searing quality I added Logic’s EnVerb, this is a non-linear reverb capable of faking “reverse reverb” effects, something that can be found a lot on this album. I adjusted the attack to about 380ms to approximate this. Lastly I added an instance of Valhalla Shimmer (more on this later) to simulate some of the sparkly top end:

I was lucky enough to have my good pal Stuart Pringle record some drums for me years ago for a Shoegaze sample pack that never materialised (something you’d want? Let me know!). These were quick takes done not to directly emulate Loveless but just to get some grooves down. We used an RE-20 on the kick into an LA-610, an SM57 on the snare into a Chandler TG-2 and 1176 FET compressor, a mono Coles ST4080 overhead and a U67 and U47 for stereo image, all into Neve 1073s.

In Logic I grouped these together and used more SSL for gating, compression and EQ. I used some moderate Space Designer room reverb and another 1176 as a parallel compressor. To my ears, a lot of Loveless is quite close-mic heavy, so I had to keep an ear out for the cymbals/room not to be too exaggerated by the compression. Lastly I used Decimot 2 to emulate the samplers used to re-trigger Colm Ó Cíosóig’s drums, after he stepped back from recording due to health issues. They’re a little bright in isolation but work in the mix.

Lastly I recorded some p-bass. Debbie Googe used an Ampeg SVT with an 8×10 cab and a Vox Tone Bender pedal. To emulate this I used Amplitube’s Ampeg SVT clone and the Mudhoney pedal. It seems no one makes a Vox Tone Bender clone, but the Mudhoney Fuzz did what I thought it needed to, this was mic’d using a U87 or U47 (depending on the session). The clean DI and distorted amp sound were blended sound and mixed with some SSL. Please excuse my average bass playing!

Everything was summed together and ran into an SSL bus compressor and 2″ tape emulation before some gentle EQ and limiting.

I Only Said

Let’s have a look at another example, this time I Only Said.

It’s not clear how many takes comprise of the main guitar sound but it took me three lots of double tracks to approximate the enormity of the rhythm tracks here. Here’s Gtr 1, an adjusted version of the main guitar tone for Only Shallow:

To add a little more girth to it I tracked the same part with a different guitar (an SG) tuned differently, using a different amp.

It still wasn’t quite bright enough, or having the right warbling quality. Shields was famed for using Fender Jazzmasters and Jaguars and part of his playing style was strumming with a constant motion of the whammy bar, giving a fluttered old tape-like sound. This can be heard throughout not only My Bloody Valentine’s back catalog but also as a much-emulated effect by other guitarists, even those outside the shoegaze scene. Here’s the man himself giving a super-quick demonstration of what I’m talking about here:

To get this more undulating guitar sound without the correct whammy bar I tracked just the higher part with a bright Strat sound (cranking the pre-amp’s high shelf EQ up). I used the cheap and brilliant SketchCassette II to fake some tremolo-arm action:

The instrumental bit in the middle of this one took about 20 hrs to record and because we kept going over the same part of tape so much it kind of wore that part of the tape out a bit… …After getting over the shock we realised it was the perfect amount of sound change to allow the sampled part to come out properly.

Kevin Shields

Lastly in this song there’s the lead sound. This was tricky, and perhaps not as close as I’d like. I used Guitar Rig with an Auto Filter, Fast Compressor (really short attack to knock out the transient of the guitar), Big Fuzz, Marshall JTM45-style head, Vox AC30-style cabinet and delay. I double tracked it and ran into an 1176, SSL channel strip and a little Pultec EQ to soften it.

I seem to remember all the rhythm guitars went through a 1960s Marshall head with an old 4×12 cab that matched the amp, and also a Vox AC30. The signal was split between the two amps and they were close mic’ed with SM57s right on the cone. I always mic guitar amps right on the cone and usually in the middle.

Alan Moulder

On twitter, Kevin later clarified this:

The main guitar part (where there is no singing) was 2 valve amplifiers with tremolo facing each other, played a few times then sampled and played backwards and an octave higher along with the original part…

Kevin Shields

Here’s how the whole mix sounds with drums and bass:

Ok Loomer

Loomer is a great example of the wall-of-sound guitars sometimes heard on this album. During Tim’s Twitter Listening Party, Kevin described this song as “trying to make a sound like fire sounds, also train noises”.

This is of the few times in this article I hard panned the double tracked guitars, as I could hear some width on the record. For the first guitar layers I’m using the Happy Fuzz Face (emulation of an Arbiter Fuzz Face) and Candy Fuzz on the pedalboard plugin with the Clean Jazz Fat preset from Logic’s Amp Designer.

The above guitar was great for the impending-doom fuzzy low-end, but I could hear something more jangly, as I added a cleaner AC-30 guitar on top:

Like in I Only Said, Kevin’s glide guitar really makes this. I added a third guitar just strumming the chords and dipping my whammy-bar where I could hear the pitch drop significantly. This is a Rat-like distortion going into a Fender Twin amp:

There’s also a choir or synthesiser sound, I can’t tell. I approximated it using Logic’s ES2 (Choir Pad preset) running into plenty of reverse reverb and delay.

Here’s the final mix:

To Here Knows When

This is a good opportunity to look in more detail at Valhalla Shimmer.

The Alesis Midiverb II had a particular preset (Bloom, presets 45 through to 49) known for its use by Kevin Shields and others (although I believe this song to use a reverse reverb preset from the Yamaha SPX90 or Alesis Quadreverb).

Let’s start off with the dry sound. Using the same clean fuzz tone for Only Shallow, I’ve recorded something a bit like To Here Knows When. Here it is without Shimmer:

It’s okay, but it’s not perfect. I’ve inserted Shimmer across the guitar bus using a tweaked version of this preset (designed to emulate Bloom). Importantly the diffusion is set to 0.618 makes the reverse effect kick in and I’ve adjusted the Size to work better with my tempo.

I’ve added a fuzzier, brighter guitar underneath (essentially a Big Muff running into a Tube Screamer running into another distortion, all in Guitar Rig). Also there’s some sampled drums and tambourine loop off Splice:

Interesting Kevin noted on the recording of this song “this song was the one where everything went off the rails for awhile, 2 weeks trying to program a tambourine track which was then played by hand in 10 m[inutes]”.

Sometimes

Let’s have a look at Sometimes. This is another instance where I found it useful to pan the guitars to emulate the stereo field on the song. Kevin recalls “The acoustic guitar is actually 7 tracks each with a different sound panned like a fan left to right so it sounds like one track if that makes any sense”

First let’s deal with the acoustic guitar. On Sometimes I can hear double tracked steel string acoustic. I only have a 12-string guitar here, so the sound isn’t identical but it’s okay for demo purposes. This was mic’d with a U87 into an 1073.

I’ve added a few layers of distorted guitar, this first one is from my SG with a similar amp/pedal combo to Only Shallow. “The distorted guitar is the re-amped acoustic thru our trusty Vox fuzz, this pedal was pretty much the only fuzz or distortion used on loveless I think”

The above distortion is quite dark and bottom heavy. This guitar adds some sparkle to the top:

Here’s the final mix:

Here’s the outro, which I did in the original article back in 2015. It’s Logic’s pedal board routed to two tremolo pedals at different speeds in mono.

Synths

M83 is a French outfit comprising of Anthony Gonzalez and formally Nicolas Fromageau. Their sound falls somewhere between classic French pop, electronica, ambient music and of course shoegaze. In their earlier material, you can hear influences of everything from Moon Safari to Takk… in there. I Guess I’m Floating is taken from M83’s 2005 album Before the Dawn Heals Us. It’s a beatless, atmospheric number that demonstrates well a few of the production and composition techniques synonymous with Gonzales et al:

Without knowing exactly what’s going on, we can still get close to the sound. The composition is quite simple, a straight sixteenth-note-style bassline, two chords, and a repeated lead figure. Let’s start off looking at the bass. I’ve programmed the chord changes and used Ableton Live’s Arpeggiator MIDI effect to trigger new notes. This way we can change the gate length later on. The instrument itself is mostly the sound of the Arturia Modular V using the preloaded CE_BassClem4 preset without much tweaking at all. This is layered with some bass guitar.

The body of this sound is the Native Instruments’ Reaktor synth 2-OSC, using the Softpad patch as a springboard. I removed the resonance from both filters, synced LFO 1 and 2 to 1/4 and changed the phase of LFO 2 to 0.5, making it bounce off LFO 1. This is layered with kv331 Synthmaster 2 doing a slightly modulated wavetable sound. It’s important when layering to check the tuning of each layer, ensuring no additional notes are being added unnecessarily. The final part to this is an Ableton Core Library sound, M Tron Strings, reducing the Bright, Filter Reso and Overtone macros and adding a longer Attack and Release.

The lead is a simple three-note ostinato. Again this is a live performance, quantized, but I’ve left the velocity as is. This is comprised of three sounds: the excellent and free Sound Magic Piano One (be sure to tweak the Rel Vol); Ableton’s Tension instrument for some bright pluck; and Native Instruments’ FM8, using a tweaked version of the Electric Harp.

The final piece in the puzzle is the sound of some primary school kids playing, which I got from Freesound.org (thanks to klankbeeld). Because my clip is quite short and the sample was eleven minutes long, I overlapped various sections and panned them around slightly. Everything has a little reverb on a send (Live’s Reverb was fine for my application, using close to 6 seconds and run into a compressor) and there’s a touch of the Glue Compressor on the master:

If you’re interested in hearing more M83-style production, Anthony did this video for Arturia showing how he works with the MicroBrute and MiniBrute with a fairly hefty synthesizers.com modular. In addition, he has created a library for Arturia’s excellent CS-80V, which can be downloaded here.

Ulrich Schnauss is a London-based producer originating from Kiel, Germany. His style has been heavily linked to the shoegaze sounds, as well as more overt hat-tips to Tangerine Dream (whom he is now a member of), and the early Krautrock music scene. His music references everything from ambient to drum’n’bass.

Einfeld is taken from Schnauss’s 2007 album Goodbye (and it’s not on Spotify). It’s typical of his work: slowly evolving motifs, lots of ethereal textures and instrumentals despite having quite a song-y chord structure. The main idea kicks in around 0.32:

Let’s start off with the most striking element: the bells. Logic’s new addition Alchemy has a good starting place with the Chimera Bells patch:

The next sound is a sine wave lead. I’ve gotten close to this in Logic’s ES2 using two slightly detuned sine waves and some Analog detuning: give envelope 3 (hard-wired to amplitude) a tiny bit of attack, around 300ms of decay, just under 50% sustain and a healthy bit of release (around 600ms). Enabling the Soft Osc Start will ensure the oscillator’s phase starts at 0º each time. Finally I’ve added some of the built-in chorus and some Valhalla Shimmer:

At the end of each block of four bars is an ascending pattern caked in delay. Starting off with the Factory Default in Logic’s physical modelling synth, Sculpture, we’re close already. I’ve added Arturia’s CS-80V underneath using a preset called J.M_ThinDepth shortening the attack times of both oscillators. These are both run into TAL’s Dub III and Softube’s Saturation (both of which are free) followed by the basic Logic Compressor and some low-pass filtering coming from the Auto Filter:

The chordal part is probably the trickiest as it’s such an enveloping sound; there’s really no knowing what might have gone into the makeup of this sound. I’ve used a fair few sounds layered in a summing stack to try and replicate what’s going on. There’s Reaktor’s Titan, another CS-80V, some GarageBand synth called Hybrid Morph and Lennar Digital’s Sylenth, each with their own channel strip. It was close but not close enough.

The next step I took was to add a sampler in the background. I’ve taken a pad that has gone through many generations of resampling playing a low A through the chords. This adds just enough character that synths can’t replicate:

…and here it is in the mix with the other sounds and an Arturia Mini V playing the bass part:

In the above example, I added Eventide’s UltraChannel on the stack (utilizing its excellent H3000-esque Micro Pitch Shift as well as the Parametric EQ, Compressor and Stereo Delay). In addition, there’s an instance of Valhalla Shimmer, some compression, and EQ.

The loop was finished off with some wavetable synthesis courtesy of Native Instruments’ Massive, using two out of phase LFOs doing amplitude modulation and some ring modulation of OSC 3. I’ve also added in a very quiet kick drum (sine wave from ES2 with envelope pitch modulation) and a pitch-bent sine wave with some delayed vibrato, again from ES2 at the end of the loop. Some healthy compression and limiting on the master help glue everything together:

Well instrument wise what I’m really relying most heavily on is probably the Oberheim OB-8 synthesiser, which basically over the last 10 years has probably been my main instrument. My favourite synth is probably the Octave Plateau Voyetra 8, which is a great instrument, and in the digital world, probably the Prophet VS – I’m using that a lot as well. Effects wise, a couple of late eighties/early nineties Roland reverbs, the R-880 Digital Reverb is really nice and the whole SRV/STX series that Roland did in the early nineties is a nice reverb as well. Then I’m recording the stuff into the computer using Logic and plug-in-wise I’m using Pluggo mostly and Reaktor.

Ulrich Speaking to Barcode back in 2008.

A lot of these can be procured digitally. The Oberheim OB-8 is nicely modeled by the OBXD, and it’s free! He claims it’s one of his most used synth – a great starting point as it’s both warm and rich. The Waldorf Wave is something else synonymous with his sound. While many manufacturers offer wavetable possibilities (Synthmaster and Massive have already been mentioned) Waldorf’s own Waldorf Edition is probably the best starting place.

Other synths notoriously used by the man are the Yamaha DX7 (neatly reimagined by Native Instruments’ FM8), The Rhodes Chroma and MemoryMoog (both of which appear in Kontakt’s Retro Machines Mk2 library), the Prophet VS, which Arturia offers an emulation of and the Elka Synthex, which XLIS Lab has modeled. One of his synths not available in the VST market is the awesome, lush-sounding Octave Plateau Voyetra 8, which from the sounds of this demo is an absolutely fantastic piece of hardware:

If you want to learn more about Ulrich’s synths, effects, studio and composition techniques I’ve made a YouTube playlist of his various video interviews:

Equipboard also has a dedicated community of users detailing his equipment and Sounds for Synth has created a TAL U-NO library somewhat in the vein of Schnauss and Boards of Canada, which costs a mere £13.

Vocals

I’m not going to make any vocal examples as no singer am I, but we can briefly speak about some clichés attached to the shoegaze style. Vocals tend to be delivered “unconfidently”. I put this in quotation marks because that’s how it comes across, it’s not an assertion about specific vocalists.

As with ambient music part of what makes up shoegaze is the absence of the melody being the most dominant feature. In lots of pop music the vocals are supposed to be what’s pushed the furthest forward in the mix – it’s the bit designed to make you hum it for the rest of the day.

With shoegaze there are countless examples of the voice being used more like an instrument, contributing to a wider picture, an almost wall-of-sound texture dripping in artificial spacious effects. A good start would be recording a dB or two quieter than you might ordinarily. The use of slapback echo and reverse (or other nonlinear reverbs) helps, too.

They used an entire second mixing desk just for vocals, sometimes doubling the vocals up to 24 tracks – using a whispered voice and layering many times, the subtle nuances and slight pitch variations of each take would create that lovely phasey & bending vocal effect. They stated that they would pick the best vocal and have that slightly louder in the mix and give it some top end. The other layers would be quieter with a little top rolled off. The whole lot would be compressed to glue and voila. I’ve tried it and with the correct technique it sounds incredible.

Tape Op interview.

More can be read about what Kevin had to say here and a GearSpace thread about it here.

Huw Price, in-house engineer at The Garden during the early 90s and writer at Guitar.com, has this to say on the vocals on Loveless:

…He sang improvised oohs and aahs throughout one of the songs, and then, without listening to previous takes, he overdubbed several more tracks of the same. These vocals shifted arbitrarily between harmony and dissonance. When Kevin processed them with radical equalisation and reverb, he created something akin to whales performing Gregorian chant in a canyon. It’s among the most innovative and beautiful things I ever witnessed in a recording studio. My impression is that Kevin generally had a clear idea of what he wanted to achieve and when he eventually got started, he worked methodically.

Huw Price, Guitar.com

Drones

This is something from the original 2015 article that didn’t really belong anymore, but some of the audio results were quite good so I thought I’d keep for posterity!

Drones don’t have to just come from guitars and samplers. Here I’ve recorded the air conditioning coming from a cheap household fan through the in-built mic on my MacBook. Sounds pretty crappy:

However, let’s add Ableton’s Resonator (I’ve written about the interesting things you can do with Ableton’s Resonator here. I’ve tuned the root to C2 and the intervals +7 (G2), +14 (D3), -5 (G1) and -10 (D1). Set the Filter to Low-pass (the top button) and sweep out some of the hiss from the top. Finally, I’ve adjusted the gain. Now run this into your reverb of choice (Live’s native one will do) and voila!

Drums

With its origins in live music, it’s safe to assume most shoegaze music out there uses live drums. More recently, for example with A Sunny Day in Glasgow, the aforementioned Ulrich Schnauss, and M83 and other artists working within the (awfully-named) Nu Gaze genre, it’s less of a necessity and electronic drums and samplers have been used instead. Moulder remembers about the recording of Loveless:

All but two of the drum tracks are composed of samples performed by drummer Colm Ó Cíosóig. Because Ó Cíosóig was suffering from physical and personal problems during the album’s recording, samples of various drum patterns that he was able to perform in his condition were recorded.

Kevin was sub-compressing the drums pretty hard to get them to cut through… …It’s the age-old mix problem that we all have to fight with when there are really loud, overwhelming guitars. It is hard to get the whole drum pattern loud without it taking over, and if you are compressing it hard and you hit a cymbal, that’s all you hear, because the cymbals and hi-hats get through the compression.

I’ve written in more detail about parallel compression here. You can read more about faking sampled sounds here, too.

Other bands, either for stylistic or other reasons, resorted to cheap drum machines. The Casio RZ-1, Alesis HR-16, and Yamaha RX5 are examples of digital drums easy on the wallet. A lot of these have been sampled for EXS24 and Ableton’s Drum Rack; a cursory Google search should find you something suitable:

Finally, it’s not unheard of for artists to sample other records – this example is Chapterhouse’s Mesmerise, which samples the Hot Pants break (famously sampled in the Stone Roses’ Fools Gold):

The Hot Pants break:

Production/Mixing/Mastering

It’s said about most music when it comes to mixing but due to its complexity and thick textures, mixing shoegaze can really be a balancing act. Getting four or five layers of distorted guitar to gel whilst having a delicate delay and reverb algorithms sing over the top while leaving space for vocals, bass, and drums is no easy task.

I would advise keeping your channels low. It’s tempting to mix with each channel at 0dB and start there but I’ve found starting things off as little as -10dB means you’re likely to be left with more headroom. I’ll use something like FabFilter Pro-Q to scoop out unwanted resonances or notch out build ups of harsh frequencies, then using the parametric EQ on the SSL to more broadly shape the sound.

Anecdotally I’ve found hardware emulation channel strips much more forgiving when adding or subtracting larger bandwidth areas. With distorted guitars, using the high and low-pass filters on the SSL to box the sound it. Low rumble and necessary hiss at the top can be problematic when layering.

Another tip in this vein is to keep your soundcard volume high: this forces you to mix more quietly (although I wouldn’t doubt others would rebut this). Using buses and groups can help compartmentalise sounds, such as drums, groups of synths, backing vocals, etc. If you can get the relative volume of your drums settled then it’s a case of just moving one stereo fader rather than all of the channels together.

In terms of reverbs and delays: of course, the sound is anchored around their presence but choosing one or two really effective ones is going to give you a clearer mix than throwing four or five poorly-chosen ones into your mix-down.

Most of all getting things to sit well are about choosing the right part to play, playing it constantly with the rest of the backing track and choosing complementary sounds and/or designing sounds, paying attention to their context. For some additional shoegaze history, A Future in Noise has done a handy timeline while 3am Magazine has done something similar. Both well worth reading and very informative.