How To Make Your Own DIY MIDI Controller

Short on cash, got lots of time, and want/need a new piece of DJ gear custom to your needs? It’s time to go DIY! For about $100 you can build your own MIDI controller. Watch the full how to video from guest contributor Kyle Mohr and read the full construction guide inside.

DIY MIDI Controller Shopping List

First you’ll need to purchase all of your tools and materials. I’ve found the best prices and parts are usually sold on eBay from retailers based in China, but they tend to sellout fast and shipping to the States takes 2-3 weeks (this will, however, give you plenty of time to get your enclosure ready and board programmed). But, to make things easier I’ve provided many links for Amazon and US retailers.

Click the below sections to expand the lists:

Parts Enclosure Tools



This is my awful drawing on a post-it note, but it suffices as an ideal layout and sketch. I followed it up using Adobe Illustrator to make a to-scale blueprint. Download the PDF and EPS files here. This was designed for 4 potentiometers, 4 faders and 4 arcade buttons, but you can certainly swap them out for other components.

For the buttons, you’ll need a 1 inch or 24mm spade bit and you should aim for using a 5/16 bit for the potentiometers. Make sure to leave plenty of space between your components and do not overcrowd the layout. This will result in broken enclosures and tight areas for soldering. You don’t always have to be this exact, you can always just take a marker and ruler, measure distance between components and make sure to keep the distance the same, and mark up the back or inside of your enclosure with drill/cut spots.

Case Building

Once you’ve finalized your layout its time to “build” or make room for your components. You can use anything for an enclosure: an old VHS tape; plastic lunchbox; retro video game cartridge; a wooden box; or even 3D print your own. The main point of DIY besides saving some money and learning what is really inside your gear, is to customize it to be truly original.

Before you start drilling holes be sure to measure all components and their clearance! 

When drilling holes for components in the enclosure, especially with thin wood and plastic, to keep it from cracking when drilling, make sure to choose about 6-7 drill bits from very small to your final size.

For LEDs I always use this guide (look under “Making holes if needed”). This will help you slowly get to your desired size without putting too much tension on the material causing it to crack, chip or break. The key is to drill at slow speeds, and gradually go up to the size you need. If something starts to crack or chip, quickly stop and put your drill direction in reverse. Try using it in reverse for a bit to clean up the cut.


Once you’ve drilled, cut and punched your way through the enclosure its time to clean it up. Don’t worry, it always looks like a hack job. The key is to take your time. Try to drill and mark up the inside (spots people won’t see) and even place a spare board under your enclosure that you can drill into. This also helps to reduce tension on the enclosure.

Now if you used plastic, you’re in luck! You’ve just saved yourself days, seriously! For plastic you can simply use the hobby files to file off any jagged edges, or frayed pieces of plastic, which are blocking the holes.


Want to make an awesome wood enclosure?

If you used wood, get ready to go all Ron Swanson on your project. Most cheap wood like we’re using here looks, well, cheap! So how do we class up that piece of junk to look like a custom cut piece of expensive furniture?…Stain!

  • I highly recommend pre-stain. It will help the stain spread evenly, especially on cheap/thin wood and prevents spots.
  • After 5-15 minutes it’s time to add that stain. Just take a rag, dip in stain and apply to your enclosure. I like to use Dark Walnut. Apply and then immediately remove, and repeat this process a few times. This will give it a more rustic/steampunk look, but if you want something more solid, don’t wipe the stain away, just apply a lot and leave it to dry.
  • After a day of drying, apply polyurethane with a brush. Fully coat all sides to protect your enclosure from scratches, and protect, and even strengthen the wood. I suggest brushing it in the direction of the wood grain.
  • Let dry for 24hrs, then repeat to your liking. I recommend 3 coats, 24hrs between each coat.
  • After the final coat, let dry for 72hrs before adding component parts! This avoids putting my components into a sticky polyurethane mess.


Now that you’ve finished your DIY MIDI controller layout and enclosure, it’s time to add your components to the faceplate. Buttons usually just pop in, but some will require a fastener or nut to be placed on the back to secure it (if there is a little resistance, don’t force it, just take your hobby files to it a few times then try again).


Potentiometer knobs will then be placed in with washers and nuts. These should cover any jagged edges you may have from drilling. Tighten with pliers. For linear potentiometers (aka faders) make sure you have the number 1 (power) on the top. This may vary depending on make and model (feel free to test using a breadboard). 2 is usually your signal, and 3 should be ground. Then secure to the faceplate with your M2 screws.

To protect the components, add your Chroma Caps to faders and pots now (or any off-the-shelf caps as well, we just like those the best).

To add the LED, flip over the faceplate or top of enclosure and put your LED into the hole(s) you drilled for it. You may need to use your hobby file again for this to make sure the LED fits just right. Afterwards, push it through just enough to see the tip on the other side. Using a toothpick as an applicator line the circumference of the LED with Gorilla Glue. Once hardened it will ensure your LED doesn’t fall back into the enclosure. Don’t use hot glue, it may melt the LED’s plastic.

Soldering Your Components

Once the glue on the LED is dry, start the soldering process. This make all of the connections from your components to the circuit board, which will then relay the message to your computer or iOS device (oh yeah, this will also work with your iPhone/iPad)!

First you’ll need your Rosin-core solder (way safer than lead-based solder) – I use and recommend .050”-Diameter 63/37 Rosin-Core Solder.

If you haven’t soldered before there is a wealth of tutorials on YouTube. Here are a few of my favorites. Trust me, its not complicated at all, anyone can do it. Just take your time and pay close attention to what you’re doing.


Turn on your soldering iron and let it heat up for a few minutes. Grab your jumper wires. I use these instead of your average spools of wire because they have a single pin at the end instead of braided wire that is quite difficult to push through a hole. They are much more efficient since with normal wire you have to cut, strip, and heat shrink it every time. I also tend to pre-bend them all at 90 degree angles and trim about ¼ off of them to make sure they fit in any enclosure and the Teensy board.

Let’s start from the top and work our way down. First make sure each of your jumper wires will reach from each component to the next, and matching colors is always helpful. I like to use dark colors (black or blue) for the ground line and bright colors (red, orange, yellow) for power while using mild colors (green or white) for the signal line. For D-Shaft potentiometers you want to be certain your pins are on the bottom, this will ensure your knobs fit and that once wired they will be going in the proper direction.


For potentiometers you have 3 lines, ground, active (your analog signal which identifies where the position of your potentiometer is) and your power. They also usually go in that order when looking at the pins, a 4th pin is usually just a dummy pin and will not be used for our project.

Since we only have 1 power and 1 ground point to solder on the Teensy board we need to chain all of our grounds together then all of our power together. Once complete, we need to run that power and ground from the last pot to the first fader, which is soldered and treated as if it was a potentiometer. Follow the guide below.


Once that is complete, run the ground only from your last fader to your buttons. The other point on buttons is your active/signal line.


Solder the active line from each component, 1 single wire from each button, knob, and fader. The Teensy board has built in pull-up resistors that we will access using the Sketch we upload when programming the board so we do not need resistors soldered to the buttons active line.

Next, solder a 220 ohm resistor to the (+) powered pin of the LED, this is to ensure it doesn’t blow out from too much power. Then connect the other side of the resistor to a positive pin on the closest pot. We will then solder the short LED pin (ground) to the closest pots ground pin.


For the last bit of soldering, run the active line from every pot and slider to Analog pins A0 – A7. Where as the active line of each button will be soldered to the Digital pins on the other side of the board B0-B3.


Programming Your DIY MIDI Controller

Setting Up Teensyduino (Arduino + Teensy)

Teensyduino is an add-on for Arduino Uploading Software that enables the Teensy board to be used in the Arduino programming environment. Before we get started you will need to:

  • Download Arduino Software *Certain versions of Teensyduino are only compatible with certain versions of the Arduino Software. On the Teensyduino download page this is specified. As of August 2015, “Teensyduino 1.24 supports only Arduino version 1.0.6 and 1.6.1 and 1.6.3 and 1.6.4 and 1.6.5”.
  • Download Teensyduino + follow the install instructions on this page (be sure to install all the libraries!)


STEP 1: Plug your Teensy board into the USB port on your computer. After plugged in an orange light on the Teensy should be flashing on and off every second. This is called the ‘blink’ sketch – The Teensy board comes pre-loaded with it.

STEP 2: In the menubar, select Tools->Board->Teensy 2.0.

STEP 3: Also in the menubar, Tools->USB Type->MIDI

STEP 4: Open a sketch file (download mine here). This sketch is set to 8 analog pots (the faders are treated as a knob/pot when it comes to the code) and 4 digital buttons. If you have more knobs or faders, simply change the number here:


If you wish to add more buttons, there is a little more work, but it isn’t too complicated to fix.

Screen Shot 2015-08-25 at Aug 25.26.18 PM

STEP 5: Click the upload button (right arrow) to upload the new code to the Teensy. Since this is the first sketch you uploaded, Arduino will ask you to click the reset button on the Teensy. Click it (see image). After clicked, the sketch should immediately upload.


This code is based off a sketch I was given from fellow DIY MIDI engineer/Musician Otem Rellik. Click the button below to expand the code – it’s long!

The Code


Before we test our creation, we first need to clean up flux residue with a paper towel and/or Q-tip (for hard to reach areas) and rubbing alcohol. This will remove any materials which could overtime corrode your board and connections.

After that dries it’s time to cover your connections that could potentially touch another and cause a short. You have a few options here: simply tape up any loose connections of bare wire that could touch another or even a metal component part; cover them with hot glue so nothing moves or touches at all; use heat shrink tubing (which doesn’t always fit perfectly); or try liquid electrical tape which comes in a rubber cement style bottle and brush on (which makes covering tight areas a little easier). *If using liquid tape please do so in a well ventilated area.

When the bare wires are covered, plug in your fully-soldered and connected circuit board. The LED should light up and not burn out! Boot up your DAW of choice and make sure you see the Teensy recognized as a MIDI device. Try mapping each component to your DAW and see if a MIDI signal is sent, once you’ve verified, you know you have achieved success!


Lastly, it’s time to close up your MIDI controller so it looks less like Frankenstein’s monster. You can use many things to mount the board in your enclosure but I recommend a generous glob of hot glue on the bottom. After it is mounted, plug in your adapter which is mounted to the enclosure, close the lid and seal it up.

Final Product

Screen Shot 2015-08-25 at Aug 25.29.55 PM

You are now the proud creator of your very first DIY MIDI Controller.

Originally posted here:

Doomsday Survival Tips for Live Shows

Three live performance experts share their deepest fears and how to avoid onstage meltdowns.

Tutorial: Mark Davie

Formula One pit stops are pretty much where the race happens these days. Whether or not a team is gambling on a one-stop, two-stop or really trying to keep tyres fresh with a three-stop strategy is mostly what the BBC commentators bang on about for the entirety of the race.

It’s so predictable that the only differential between drivers is how many points of a second the pit crew faffs about in a pit stop. One second longer than normal is a total disaster. And if a tyre wrangler drops a nut, it’s race over and you may as well drive into the wall at the exit of pit lane.

Amazingly though, some musicians have no idea how long it would take them to get their show back on track if everything went pear-shaped. They’re gambling with their livelihoods every day. I’m talking about DJs, producers, bands, solo artists, anyone who uses a computer in their live setup. Even mix engineers to a certain extent, who are using computers to trigger events or add processing power to their rig. There’s a growing population of artists who run their entire show off one computer, completely unprepared for when it all goes bad.

So, let’s get redundant, baby!

To help save us from the inevitable doomsday scenarios, I talked to application specialist Jason ‘DJ Shine’ Spanu, a Canadian techno DJ who has toured the globe setting up systems and running playback for Nelly Furtado, Drake, Frank Ocean, Broken Bells, and many more artists; Lynden Gare, production manager for Flume’s Infinity Prism tour was also kind enough to fill us in on how Flume’s MIDI is managed; and Danny Harley, the one-man show that is The Kite String Tangle, walked me through his journey to figuring out how to get redundant. So if you’re walking the high-wire without a safety net, here’s some practical measure to land softly when things go to hell in a handbasket.


Technically speaking, Jason ‘DJ Shine’ Spanu is on the Broken Bells tour as we speak. Except he’s not. He’s at home, having sent a sub to cover for him on the last two weeks while he looked after a family member. He also got a call to set up an interactive iPad jamming station for a kids science event. It was a month’s worth of touring for a day’s work that came at exactly the right time.

Spanu’s life is anything but normal. Whether he’s on the road or not, he’s still the guy you call if you’ve got a problem, at any time of the night. As well as putting together and running the playback systems for touring bands, he also consults for others. “My phone rings at stupid times of the morning,” he explained. “And it’ll be the guys from Metric calling me or some other weirdo indie band asking, “How do you do the thing with the…?”

At which point he’s already put them on speakerphone and logged onto their desktop with TeamViewer. Which saves him yelling, ‘No, click that button!’ down the line thousands of times.

Spanu is a Toronto techno DJ who sort of fell into his current role as playback master. He moved in next door to where Nelly Furtado’s band rehearsed by fluke. The band got to know him as “the guy that used that f**ked up program Ableton Live,” and his roommate soon became their pot hookup. Spanu’s in with the band took a bit longer to eventuate — Furtado’s boyfriend also happened to be her DJ, thwarting any chance he had of cracking his way onstage with them.

So Spanu got on with making techno, and about three years later, they came back to him and asked him to show them that ‘f**ked up program’ again. Up to that point, they’d been using ProTools for playback, and the bass player/band leader was having trouble following Furtado when she wanted to stray outside of the bounds of a linear timeline. So the ability to time-stretch songs in one session, and loop and edit on the fly, had them in raptures. They were hooked, and asked him to join the circus at the height of her career in 2006. “On Day One we did every TV show on earth,” recalled Spanu. “All the different late night David Letterman-type shows. They had a variety show in Europe where they bring in all these international acts, and I realised I was the only one using Ableton, everyone else was using Digital Performer, ProTools or Logic. These days I don’t feel special anymore, they’re all playing Ableton.”

Spanu picked up Ableton at version five, and much of what he does stems from what he could do with markers back then. Depending on the job, Spanu switches between the Arrange mode and Session mode. If he’s DJ’ing, it’s Session mode, but if he’s just playing backing tracks, he sets up markers in the Arrange timeline at likely loop points, so he basically just has to hit a forward or back key at any time during the set. He uses the Arranger view because bands primarily know what note follows another, which part goes next, and the form generally stays the same. If he tried to get fancy by triggering sections in Session view, he’d likely stuff it up, or he’d have to program follow actions to progress to the next loop set… far too much hassle.



When Spanu first started out, the general consensus was that “computers were the devil, and you better hide that s**t. Don’t put it anywhere even near the stage!” Regardless of whether they made a dance album, and everyone knew where the sounds were really coming from, you wanted to look at the “hot chick, not the guy looking at a computer.” For some artists, like Frank Ocean, the whole band is under the stage, whether they’re tethered to a computer or not.

Artists tend to see Spanu more as ‘Wizard of Oz’ than DJ, so his ego has taken “more hits than it can handle”. It took years for him to finally make it up on stage with Furtado’s band. So when he got the opportunity, he carted his full DJ rig — two Novation Launchpads, an Ableton Push and two giant touchscreens — up onstage with him. Even then, for a show in Ibiza they still had a guy posing as the DJ onstage right next to him… At the end of the day, he gets paid the same whether he’s on stage or off.

For other bands, its clear what his role is. For instance, said Spanu, “Broken Bells came to me and said, ‘We need you to do this’. They’re not asking for ‘and be awesome’. They’ve got awesome covered.”


One of Spanu’s current clients is Frank Ocean, and the rig had to be scalable as required, but generally portable and fit for shipping. The rig comprises Ableton Live at its core, running on two MacBook Pros, with two portable MOTU Ultralite interfaces and a Radial SW8 eight-channel switcher between the two. He uses the seven-year old Ultralites because they fit in his hand, have 10 stereo line outputs — enough for any artist he’s working with — and are Firewire bus-powered. Which makes power supplies one less thing to worry about losing or going on the fritz.

When he first started buying gear for artists with a budget, he thought bigger would be better; more processing power would mean more stability. So he packed a MacPro tower into a foam-lined, touring rack so he could run 24 backing tracks. They went to play a show down in Mexico where the subs were plugged into the same generator supplying his computer. As soon as the sub kicked in he lost all power and his computer crashed. He soon realised it was dumb. “Who needs 24 tracks anyway,” said Spanu. “You might as well not show up.” Plus, when you’ve already got a 64-channel stage plot, no festival or TV show is going to give you 24 more lines to send to the computer guy.

From there it became, “What else can I erase from the equation?” Now the MacBook Pro runs itself and the interface from its internal battery. If everything else goes pear-shaped, he can still hit the next marker and get audio.

With eight tracks to play with, he usually divides his percussion up into frequency specific areas, so a stem of kicks, a stem of snares and other mid-heavy elements, and hi-hat loops on their own stem. Then it’s typically vocals and music separated onto their own stereo outputs, and two separate click tracks. One click goes to the drummer for the entire song, and the other one typically just gives the band a ‘1, 2, 3, 4’ count in. From there the drummer takes over, unless there’s a drum-less chorus breakdown or bridge.

Once he has the main markers in place, he then uses his experience to inform where other markers might be a good failsafe; places he knows he can loop if the artist decides to diverge into a 10-minute Kanye West-style monologue.

“Be aware that every time you put a marker down, you’re spending about a megabyte of pre-loading RAM,” said Spanu. “So when you’ve got 50 songs, averaging about 20 channels, and I put 10 markers into each song. Bang, I just spent some RAM. So I try to be a little sparing, but also not terrified about it, because it’s generally fine.”

Another thing he never does is upgrade straight away. “Live 8 came out and I didn’t even touch it,” said Spanu. “It was seven forever. When six came out I was like, five works, I’m not changing this.”




Once the audio interfaces are synced up, there’s the issue of syncing MIDI. Depending on the artist, he might set up an eight I/O MIDI patchbay. If a band has keyboard controllers all over the stage, he’ll build a MIDI web that does all the control changes for the different synths from his computer. Spanu: “As soon as I hit Play on a new song, the sounds are all ready. And in the middle you can take the keyboard player’s sound and do another solo with an arpeggiator, and it becomes this massive ballet of MIDI.

“The next level is adding in video. That’s usually a word that scares a lot of people, but it’s pretty simple. All you’ve got to do is run some SMPTE timecode. Live doesn’t natively produce SMPTE, but you can just record the audio file of it, run it as a separate output and shoot that to the lighting console so everything comes together.”


For Spanu, there’s no point in having redundancy if the changeover isn’t automated. Realistically, he says, if anything actually does go pear-shaped, it’s going to take him a second or two to notice, react and hit the ‘Oh s**t!’button. So he feeds one of the inputs to the Radial switcher with a separate VST test tone oscillator plug-in, set up in Ableton to continually emit tone whether he’s stopped a session or not. If the Radial stops seeing that tone at any time, it automatically switches to the inputs from the alternate audio interface.

The other issue is syncing his MIDI between the two sessions running on different computers. But rather than syncing MIDI timecode or having the computers talk to each other, he instead relies on control.

One of his associates, Matt Davis, who Spanu handed the Drake gig over to, programmed a mirrored Max 4 Live patch that allows one Launchpad to follow the other over MIDI. The Max patch waits to hear what button was pressed on the triggering Launchpad, tells the other Launchpad, and they both pass that information onto Live. It means that if the session stuffs up, the first Launchpad can still trigger cues to the second, allowing time for Spanu to shuffle over to the other system.

Spanu: “A simpler way of doing things is to get a MIDI splitter and a MIDI keyboard, map the names to certain notes, and by hitting that note it will split to both machines at the same time. You don’t want sync. Because sync relies on a relationship and if one of those things goes away, then what happens? There’s no point getting anal about phase, because at the time you have to switch between one and the other, whether or not it was on by a frog hair is not going to make any difference.

“I made a bunch of how-to videos with the Metric guys, and one of them was literally walking up to the computer and yanking out the Firewire cable. He didn’t even notice it switch, it was amazing.



Spanu also DJs on stage for Toronto R’n’B artist Julie Black. He has a lot of fun with, four iPads running a TouchOSC patch, Lemur and TouchAble. But his problem is always the network. “It’s the radiation soup of signals floating around us all the time,” reckons Spanu. “Cell phones, radioactive crap… Fukushima. Any of those things interfere with my ability to make a network a foot away, even via a simple connection — it’s stupid! I’ve got 100,000 people in front of me, and I can’t do any of it.

“The new version, TouchAble 3, will finally have a connection from the phone jack. So I intend to split the workload up between things I know are simple and stable, and things that are awesome and fun. TouchAble lets me adjust every parameter on the fly really quickly without messing with a mouse. But it’s all going through digital wonky wi-fi so it may go out at any moment. If I put a filter effect on each of the vocal tracks and control it from my iPad it better be a very stable connection, because if I pull it all down then go back up, and it doesn’t go back up… disaster!

“It happened to me the other day in a small theatre with 400 people. Everything was fine and all of a sudden the connection went bad. It’s all fresh gear, a new iPad Mini, but something is freaking it out. In-between sets I had to reprogram it all. Luckily, I always bring a hard-patched MIDI device with me — something with knobs. If I was hiding behind a curtain I wouldn’t care. Just hit spacebar and f**k off, but on stage I want to interact and not be the guy faking it. I’m of the generation that doesn’t do jumping jacks when I perform.

“I generate my own network, because generally you need a really high-powered transmitter to rise above everything. The problem is all those phones’ default channel is the same channel I’m using, so there’s all this crosstalk and inter-modulation frequencies shooting around. And if I go to a lower channel it doesn’t have enough power. The wi-fi world is a pain in the arse.

“Generally I’ll do things like mixing, sends, throws, and any automation that I want to have control over. I’m not doing too much, but if I want it, I know it’s there. I’ve simplified my interface because we’re all smart, but when something goes wrong on stage our IQs drop to single figures. you’ve got to build your system to be run by a baby who just got ran over.”


When Danny Harley (aka The Kite String Tangle) first thought about making his live system redundant, he really struggled to find any information on how other artists were going about it. Eventually he was given a leg up by Rufus, who in turn pinched their main ideas from Cut Copy’s stage setup.

“Laptops freak me out, I never liked performing with a laptop because I was scared it was going to screw up, that’s why I desperately wanted the redundancies. I moved completely over to hardware for a little while and then moved back again.” That’s right, to get around his fears of running a laptop, Harley carted around all his hardware synths and samplers. “It was a lot harder because you had to bounce everything over to the samplers and make instruments. It ended up taking too much time because if I wanted to tweak anything in my live set I had to re-bounce.”

After frustrating himself for too long, he eventually moved back to the laptop and vowed to figure out a redundant system. In the end he ended up with a similar system to Jason Spanu’s — two laptops, two MOTU Ultralight interfaces and a Radial SW8 switcher. He also uses a MOTU MIDI Splitter fed by the outs of a MIDI XT to connect his MIDI instruments to both computers, but steers clear of any auto-mapping features. “The stuff that throws me off the most is when a parameter will be auto-mapped and you spend an hour trying to undo something that’s supposed to be intuitive. Everything is mapped from scratch, it leaves less margin for error.” Also like Spanu, Harley doesn’t clock anything from the laptop, relying instead on the MIDI event triggers to keep both sessions in sync.

It was a bit of work for Harley to get his system together, but now he’s got a svelte, racked, flyable setup that was worth every penny. “It was expensive,” said Harley. “Because you have to double up on some stuff and your switcher but for me the peace of mind was worth it.”

No tangles here — Harley’s rig is controlled by a Livid Ohm controller, Akai MPK49 keyboard and MPD pad controller, and Edirol keyboard. His effects chain includes a Line6 DL4 delay, an Electro Harmonix Cathedral reverb, and a TC Helicon Voicelive.

No tangles here — Harley’s rig is controlled by a Livid Ohm controller, Akai MPK49 keyboard and MPD pad controller, and Edirol keyboard. His effects chain includes a Line6 DL4 delay, an Electro Harmonix Cathedral reverb, and a TC Helicon Voicelive.

 64-BIT, BUT NOT 24

When most people start using computers live, they’re petrified to do anything more than hit the space bar. “It’s shut up and hit Play, because everyone is still terrified and thinking we’re running it on a Commodore 64.”

But more than user error, the times Spanu has seen everything go wrong is when artists ask too much of a system. Like running 118 tracks, or insisting the tracks have to be in 24-bit. “I’m like dude, we’re not performing at a DVD authoring factory,” laughed Spanu. “We’re out in this gross s**thole and I’m one of 60 tracks on stage. It doesn’t matter if I don’t have the depth of a 24-bit dynamic range. Let’s pull back all those numbers and be cool.”

Other times, he just can’t explain what’s wrong. The Broken Bells session is pretty elaborate, they have four keyboards on stage all running internal Ableton soft synths in the one session, with lots of tracks playing as well. But they’re running it at 64-bit with 16GB of RAM, on solid state drives, using UA Apollo Quad interfaces with 16 outputs. The system is completely stable, but for some reason, one file is causing the whole session to “eat a dick in one spot that we can never fix no matter what we do.”

Other than the odd voodoo moment, it’s mostly “buffer overruns or disk overload messages where I’m trying to pull too many things at once, or fighting for RAM allocation with drivers,” said Spanu. “That’s why I tend to stick with what works. I don’t know what a Focusrite Saffire will add into my equation until I’ve had a couple of weeks personal experience running the thing.

The last time I did a massive upgrade, I stayed home for a week and just left everything running for 12 hours. I looped this massive two-hour show six times while another machine was recording what was happening. Once it passed that test, then it was alright to go out.”


Spanu always has the artists buy the kit needed to run the show, rather than operating on a loan-hire system. It’s part of the redundancy package; he doesn’t want the whole show to go down even if he decides to call it quits. But before he buys anything, he picks through what they already own, because “a lot of these bands have a courtyard worth of crap you could put together,” said Spanu. “Once you’ve picked up random cases and other bits and pieces, you budget it out and let them go to town.

“I like to buy three of everything in case one goes down, so it’s a redundant system. I try to roll like an Apple store as well and bring a couple of all the connectors you need. I also try to keep my spare systems small enough to carry them on. I don’t mind shipping the big rig, but if the plane crashes and I get out, I’m grabbing that bag and we’ve still got a show. I’m that stealth dude who ninjas around airports. I take up like nine bins, and put a 60-pound bag in the overhead — if that falls out it will kill someone.

“Everyone else can just get a guitar and plug it in, or a keyboard and find some stupid sound and play it and you’ve got a show. But if the computer guy has lost his Mr. Wizard box, then you’re crying.”


Lynden Gare has been Flume’s (Harley Streten) production manager ever since he took off a couple of years ago. Back then, Streten was touring the world, running Ableton Live out of his MacBook Pro through the measly 3.5mm minijack, into a DI. His sole backup was a spare cable in his Sportsgirl kit bag.

Since then, the production has moved through sitting a spare laptop under his main one, to having a second laptop networked with the first with the same show file running. They’ve got it down to about a 10-second changeover time, before Flume is back into a fully-featured show.

The reason the changeover isn’t instantaneous is sort of a redundancy issue in itself. With every piece of gear comes an extra point of failure, so a fully redundant system has to be replicated or triplicated. But Flume’s workbench isn’t off stage, it’s his trademark ‘Infinity Prism’, which isn’t really big enough to hold pigeon-pairs of every controller he uses to run his show. As well as the laptops and soundcards, he’s got two Novation Launchpads, a keyboard and an Akai APC40 running via USB, and a Roland SPD effects pad outputting MIDI to the soundcard. It’s a full deck. All the USB items are running into a server-grade USB KVM switch, which basically lets Gare switch all the USB inputs from one computer to the other.

Once the big red button is hit, Gare does an encoder pickup on the APC40 because the backup computer doesn’t know where the faders are in relation to their range, so you have to manually move them through their range so Ableton can pick them up. It’s not ideal in the sense of two systems talking to each other, but because most pieces are running over USB, Gare would have to convert the signal to MIDI, then MIDI back to USB on the other end to keep every controller in sync. And that would be a lot of MIDI flying around. Streten doesn’t have to sync up with anyone else, so the team figured simple is best.

It’s a choice related to the type of act Flume is. Flume is one guy, so it’s not about having him control absolutely everything at once in his session, he’s usually only playing or manipulating one instrument or effect at a time. The attention is on what he’s doing, not what he isn’t. “We didn’t want to try and replicate the full-featured show,” said Gare. “We just wanted a back up. The analogy I use is likewhen you’re in a motorboat and the motor cuts out, at least if you’ve got a sail you can still get to your destination. That’s what we’ve gone with, rather than try to have a secondary motor that’s going to add time, risk and other points of failure.”

The complete failsafe is returning back to Flume MkI — the 3.5mm jack into a DI already plumbed into front of house. To cover any gaps, Streten has strategically placed stretched-out transitions in his session that he can trigger at any time. At the end of the day, if he has a session on a working computer, people are still going to have a good time, and it buys the team enough time to get the main system up and running again.

There is a third audio laptop offstage, said Gare, “In case the roof opens up and 10 litres of water fall onto the stage and destroy both laptops.”

Like Spanu, any failures are likely going to be caused by heat. Gare: “In the sun, the MacBook reduces the amount of processing available to try and restrict damage to the computer itself. So we had a couple of shows where CPU usage shot through the roof and we got pretty close to using the backup. But we ditched it at the last minute and put a little sunshade over it.”

All in all, the show carries nine laser-cut, foam-lined Pelican cases with six MacBook Pros to run audio, video and the Infinity Prism. It all stems from the single session running on Streten’s main computer, with the others providing the processing grunt. Like everything else about the Flume show, the cases are a big upgrade from Streten’s ‘Sportsgirl backline’. Once, to get out to a show on an island off Corsica, the transport boat stopped 10 metres shy of the shore, forcing them to float the waterproof Pelican cases out to the boat. Needless to say, they paid for themselves right then.

While Gare has put a lot of effort into making sure the Flume show runs seamlessly, Ableton whiz Alex Alexander has now joined the team, and Gare reckons he’s been instrumental in ensuring the stability and security of the system, especially through the upgrade from Live 8 32-bit to Live 9 64-bit.



Spanu: “There’s been four times when I’ve had issues running Ableton and all four times were to do with one issue. I was on this big tour with Nelly and it was heat. Heat is a bitch. Computers do not like heat!

“We played in a hot steamy Parisian club and towards the end of the show the computer just went to grey screen. It was the first time I’d hit the ‘Oh s**t!’ button in two years.

“The next time was in New Jersey, again in a hot steamy club. I was like, “Yo Apple Store, fix my s**t, Nelly Furtado, let’s go right?’ And then the next time was in Denver. Same thing, hot sweaty club and by now someone else is like, ‘Dude, it’s the heat!’

“The last gig we had was in Las Vegas at 8pm, which means soundcheck at four in the blistering sun. We got industrial-strength fans and I duct-taped them to my computer at both ends and turned them to stun. Towards the end, one of them died. By then it was like, I’m just gonna use laptops and keep it simple.”

Why laptops? Not because they don’t get affected by heat, but because they’re mobile. You can move them to a cooler spot if you really need to. He’s seen people use computer gel packs to cool down their machines before, but Spanu puts beating the heat down to two things — circulation and decoupling. To get his computer off the deck, and keep air all around it, he wads up a few gaffer tape balls and sits his laptop on those.

The tape balls also help minimise vibrations getting at your hard drive (less important these days with SSDs) and your audio interface connections. He was DJ’ing for hip hop artist Socrates one day, and when the first beat dropped at full volume, the computer crashed and the audio interface got stuck on the last number it was making… at full volume.

Thinking about redundancy amounts mostly to common sense, before getting on stage. Because when you’re up there, if anything goes wrong, the only thing you’ll be able to do is hit the ‘Oh s**t’ button and hope for the best. As Spanu puts it: “You’ve really got to plan for the worst and expect the best. Somewhere in the middle you’ll have a good show.”


Originally published:

Desktop PC PROS

It’s flexible, easy to upgrade and can be adapted to your changing requirements — all hail the desktop PC!

Column: Martin Walker

Well the computing marketplace is once again being shaken up and thoroughly stirred, with Apple iPad sales expected to drop by 20% in 2015 as jumbo smartphones take over. New models are skinnier than your average ‘haughty couture’ stick figure, with PC maker Lenovo releasing its LaVie Z laptops, claimed to be ‘the world’s lightest 13.3-inch notebook PCs’. They have more computing power than Apple’s new Intel Core M MacBook — as well as two USB 3.0 ports, a HDMI video output and integrated SD card reader compared with the single USB-C port connectivity of the MacBook — yet match or undercut its price. However, as I discussed in Issue 108, choosing a PC notebook suitable for extensive audio use can be a minefield for the musician.

Old fashioned they may be, but the good old desktop PC is still the best bet for guaranteed high audio performance and longevity under fire. You can upgrade the majority of their internal parts as and when the need arises (no soldered-in RAM like the latest Mac Mini range that prevents their memory being increased after purchase). Desktop PCs also offer plenty of ports to plug in a wide range of peripherals simultaneously; including USB and Firewire audio interfaces, external drives for backing up your precious audio material, and even smaller noteboooks/tablets for use as touch-based MIDI controllers, MIDI synth editors/librarians, and the like. All these can be connected via low-cost generic cables that cost a few dollars, rather than the expensive custom white extras offered by Apple.

For more elderly card-based audio interfaces, desktop PC motherboards are still available with PCIe and even ancient PCI expansion slots, and if you favour one of the latest Thunderbolt-based audio interfaces, these can be plugged in via a PCIe expansion card as well. Flexibility has always been the strength of the humble desktop PC, which can adapt via upgrades to suit changing requirements. You’ll never get that sinking feeling if you find you need more ports, since cheap expansion cards are available to add a clutch more of whatever variety you need. And in the unlikely event that a component (graphics card, power supply, hard drive…) fails, you can nearly always replace it too.


Enough already I hear you cry — we get the picture! Ironically though, the picture is one area that can cause a few problems for the desktop musician. Ideally your loudspeakers should be perhaps four or five feet apart, and your ears a similar distance away from them. Even when you ditch the dinky notebook format, with its closeup screen, in favour of a larger screen placed a bit further away. It can still end up sitting in front of your loudspeakers, resulting in lots of acoustic early reflections that compromise your stereo image. In this scenario, you’ll notice a significant improvement in the acoustic detail of your mixes if you invest in an even larger graphic monitor that can be placed further back in between your loudspeakers rather than in front of them.

I faced this scenario myself a few months back — the last couple of years I’ve had a 23-inch widescreen monitor mounted on a heavy duty tilt/swivel stand that allowed me to pull it to about two feet away for detailed editing, and then push it back to just in front of my monitor speakers when mixing, but I knew I could do better. So I replaced it with a very affordable 27-inch widescreen monitor in a fixed position right back between my loudspeakers, and the difference was remarkable. When a client next visited my small studio he was convinced I’d done a lot of extra work on his mix, but it had always sounded that good — we just hadn’t been able to hear its unadulterated excellence while the monitor screen added its dose of sonic confusion.


The desktop PC is also very flexible when it comes to storage requirements. It pays to adopt at least one SSD for speedy booting and application launch times, and for those that can afford it, several larger SSDs if you want a totally silent computer. However, for those with smaller pockets, modern mechanical hard drives in conjunction with a decent acoustically-lined case are well nigh silent too, and there’s usually plenty of space inside the typical desktop case for at least four such drives. Many of us still rely on CD/DVD drives for audio CD burning and video duties, but Blu-Ray drives are also easily fitted for those with more expansive storage requirements. Once again, you can change your mind as you please, and still end up with a single case containing all the storage you need rather than a slimline notebook with a rat’s nest of cables emerging from it to cope with your peripheral requirements.

The one thing that undermines the ultimate longevity of the desktop PC in my opinion is Intel’s insistence that each new CPU family it releases must be accompanied by a fresh motherboard with a revised CPU socket. Although an experienced user can change the motherboard, it’s a major undertaking, and only really worth doing if you get at least an additional 50% of processing power in the bargain. Yet even here the PC musician is in a much better position nowadays than a few years ago, since the amount of computing power available in a modern machine is more than sufficient to run huge audio projects. Since it’s likely to be the single most expensive component in any PC by a margin of three or four, just make sure you choose your desktop CPU wisely within your available budget, and it ought to last you for at least several years (mine is coming up to three years old now with little or no sign of running out of steam).


So there we are. The humble desktop PC may seem a little passé to some, but for the musician it still does its job well, is incredibly versatile, adaptable, repairable, and can be upgraded in a host of ways at relatively low cost. I do feel sorry for those who end up having to buy a new computer to gain more processing power, only to find their perfectly good audio interface has to be abandoned in the process. There’s quite enough built-in obsolescence in the world today without discarding audio gear simply because this new machine lacks the relevant socket to plug it in. So stick with the desktop format, and carry on using what you already have!

Originally published:

HOW TO: Record Retro Drums

Tired of drums sounding big, wide and boring. Turn back the clock with us as we re-create drum sounds from the ’50s, ’60s and ’70s.

The diversity of our auditory preferences always amazes me. Sometimes we want to hear the highest of high-fidelity sounds. Hairs standing on end, ‘voice of the gods’ sort of stuff. So sonically pure that our brains immediately start humming in tune with the sensation of  ‘good sound’. Then there’ll be other times, when we hear a sound that is ‘effective’, but hardly what you would call hi-fi. It just sounds cool.

Often we hear old records and intuitively pick up on the ‘retro’ vibe. A big part of that is the way sounds were coloured — whether on purpose or not — by the recording process. Yet artists are often really thinking about drum sounds when they come into my studio requesting a retro approach. So let’s get retrospective and see if we can deconstruct, reconstruct and bring out the colour of drum sounds from bygone eras.


We’ve become so used to big, glossy contemporary drum production, that we tend to define ‘retro’ drums as being the antithesis to that; unpolished, or raw. Certainly, compared to the massaged, layered and controlled techniques of state-of-the-art modern records, this would be true. But when people come into the studio asking for a vintage approach, what they’re really asking for is something beyond simply lo-fi; they’re looking for character. It’s the same with photography. The first thing we do after snapping an image on our phones is apply a filter to it. The quality of the original image is usually good enough in itself, but a filter ‘colours’ that moment. Whether we fade it a tad, or add a vignette, it says something about us.

In a musical sense, the artist or producer is asserting a sense of personality on their production, because it says something more definitive about their creation than just a ‘nice’ sound.

As much as we’d all like to just fast forward to the bit where I tell you how to get that magical Mick Fleetwood sound. For us to approach this effectively we need some basic info on how the originals were recorded and what we’re listening for.

As a producer who is also a drummer, I’ve been a long-time student of drum sounds and record production. I’ve found that the more I know about the original approaches, the more authentic my results are when emulating them in a contemporary setting. It’s beyond the scope of this article to give you a complete history of Western Pop recording techniques, but I will break it down into a few very general time frames and fill in a few gaps about the general recording approaches for each era.


The most important parts of achieving any of these drum sounds are right at the source. The drum kit needs to be setup and tuned for the style, and the player needs to understand the concept of playing to the mics. For anyone trying to get in the ballpark for the first time, it’s worth spending time looking at images on the web of drummers (and their kits) of the era you’re interested in.

For sounds up to the late ’60s, it was at most a simple four-piece kit, usually with two full heads on the bass drum and almost no dampening, except some felt strips. The snares were almost always five-inches deep, and there were usually only two large crash/ride cymbals that weren’t hit very hard. If you’re using a kit with a big fat deep snare, a bass drum with a big pillow on it, five toms and six cymbals, you’re making your job a lot more difficult. And if your drummer is a basher, smashing out quarter notes on the hi-hats, with a weak snare backbeat and a zillion drum fills, you’re going to struggle to get an authentic Motown sound.

The point is that appropriateness will yield plenty of sonic options from any mic combination you use. And don’t be scared of how odd your drums might sound in the room. We have a bunch of drums that we keep tuned a particular way (including a kid’s toy kit) because we know under a certain combination of mics, they’ll sound a particular shade of incredible.

Throughout this article, there are some specific modifications to both the kit and to the drummer’s approach that will guide you towards the right combinations of approaches. Still, your best asset will always be your ears. Learn to listen through those old records for your best chance of figuring out why they’ve become so iconic.

In the ’50s, the idea was to position the single overhead mic so it captured the whole kit.


Key listening tracks:

My Baby Left Me – Elvis Presley
Lucille – Little Richard
Shake – Sam Cooke
Twist & Shout – The Beatles

It’s good to listen to the stereo version of Twist & Shout first, because one channel is actually the close mics, and the other channel is just the vocal spill. Only then listen to the mono version, and you’ll hear how much excitement is coming from the sound of the spill into the vocal mics.

The first stop in our way-back machine is just prior to 1960. Looking around the studio at that time, the first thing you would have noticed is that all the musicians were in the live room. No one was lounging in the control room with a DI’d guitar or MIDI keyboard. The complete ensemble was picked up live off the studio floor, balanced in the control room and printed to tape (usually single track, rarely two- or three-track multitrack) as one performance — no overdubs. There may have been only one mic on the whole kit, possibly two.

A big contributor to the drum sound on big band, jazz, and early rock ’n’ roll records, was spill. A lot of what you’re hearing is the band bleeding into each other’s mics, particularly into the vocalist’s.

Of course, there were no headphones at this point, so players were balancing their performances purely by the sound in the recording space. Crucially, the drummer had to learn to play to what the microphones were hearing — all of them, not just those over the drums. Whether that meant playing quietly, putting a wallet on the snare, or not hitting the crash cymbals. It was a case of deferring to the engineer’s instruction, which the good musicians quickly adapted to. The engineer would say something to the effect of, ‘if you play too hard, and hit the ride cymbal in the chorus, you’ll ruin the whole mix.’ Critical point number one: play to the mics.

Equally important in the ’50s is spill. In the distance, about eight metres away from this drummer’s view, you can catch one of the room mics that’s emulating the spill from a trumpeter, vocalist, you name it.

Here’s another ‘single mic-only’ drum position, effective for that ‘Motown’ sound…

…and one more mono position — from the top this time — with that floppy wallet snare dampening.

The next thing to remember, is that because of spill, room sound was inescapable in these early recordings. Mics were pointed in all sorts of directions around these reasonably large rooms. So drums were bleeding into the piano mics at the other end of the room, the upright bass mic, the vocalist, the string and horn sections. I’ve read accounts from engineers at the time where they didn’t use the mic that was on the drums because the drums were loud enough in every other mic when balancing down to mono.

Finally, it’s important to take into account the quality of the recording equipment at the time. It’s sometimes difficult to get contemporary recordists to comprehend that, once upon a time, recording engineers had very little equipment. And what they had wasn’t always particularly versatile.

In the ’50s, they were dealing with lovely new Neumanns (regularly the valve varieties), reasonably recent ribbon mics (15-20 years in use), and the introduction of dynamic mics. But the consoles were often purpose-built four or eight channel valve consoles, with either no EQ, or very simple bass and treble controls. More advanced EQ functions, like a sweepable mid frequency, were outboard pieces that were patched in when needed, and there were only a few modified radio-style compressors used to keep volume levels in check going to tape. Engineers were also at the mercy of all of the analogue tape process’s artefacts; alignment issues, tape hiss, as well as overloading the tape machine’s circuitry and the tape itself.

The expectation was that the sounds would be right in the room, then the mics would pick the sound up and send it through the console as transparently as all those highly-coloured components would allow. The hope was that at the end of the day, playback off the tape would resemble something close to the event. In a nutshell, nothing near the fidelity of a contemporary recording system.



Apart from making sure we use the absolute right kit, see sidebar Step 1: The Right Kit, some techniques to get close to this sound would include:

Don’t use many mics on the drums themselves, perhaps only one over the whole kit.

Remember, at this early stage of drum recording, no one considered the drums as individual elements — kick, snare, toms, etc. One mic was placed in a central position to pick up the sound of the whole kit. Ribbon mics were great for this, as the weight from the bottom end of the kit was emphasised, and the duller top end helped tame the cymbals. Listen for a mic position that gives you a balance of the whole kit.

Then scatter a number of mics of different varieties around the room, pointed in different directions; away from the kit at a distance, into a corner, aimed at the glass on your window, or pointing down at the floor. Imagine there’s an ensemble coming into your room, and you’re putting one mic up for each instrument. You may use a ribbon mic for where the horns would be. A condenser for the rear of the upright piano. Maybe a dynamic where the singer would be.

The idea here is to simulate spill — open mics around the room picking up the sound of the room, but not in a ‘drum-conscious’ way. So experiment with these and see which ones give you the greatest sense of ‘drums in a room’.

Get the drummer to play at the volume he’d need to play if he had a live singer in the room with no foldback. Tell him to back off on the fills and the cymbals, and make it all about the groove.

When listening back to the various mics, only listen in mono, perhaps putting a high-pass filter at 100Hz and a low pass at 8-10kHz on every mic. Don’t be scared to submix all the tracks to a mono auxiliary and treat the sound with light compression or even a saturation/distortion type plug in. Balance all the mics up until you get a sound that can be described as ‘character-filled’.

The early ’60s added another dynamic kick mic to go with the overhead mic position for a bit more bottom end. It was a closer sound than the previous era, but as you can see here, not too close.


Key listening tracks:

My Generation – The Who
Uptight – Stevie Wonder
I Want To Hold Your Hand – The Beatles
Mr Tambourine Man – The Byrds
Heatwave – Martha and The Vandelles


The ’60s ushered in some pretty significant changes. Though Les Paul and others experimented with sound-on-sound recording in the early ’50s (essentially, playing one mono tape into a mixer, and recording another part at the same time to another mono machine), true overdubbing facilities only started to appear on three-track machines in the very late ’50s. The initial benefit was that orchestras could be overdubbed. Tracking sessions soon involved having the ensemble balanced onto one track while the singer, also performing live, was recorded to a second track. The orchestra would then be overdubbed onto the third track — a massive change at the time. As the first four-track machines started to hit in the early ’60s, this practice remained intact. Ensembles, including singers, still performed live in one space, and occasionally the additional tracks were used for ‘sweetening’ — adding horns, strings or backing vocals. The pre-’60s practice of self-balancing remained.


The first major change ushered in by the rock ’n’ roll era was that singers could monitor themselves through ‘monitor speakers’ to compete with the louder guitar amps that had also popped up. Drummers started to play harder and used more cymbals to mirror the live experience, prompting engineers to begin looking at new ways to convey this live experience to tape. Basically, everyone got louder.

This manifested in some small but important changes. Large diaphragm condenser mics were replaced as the main ‘drum kit’ mic with either a ribbon mic to tame the cymbals, or a dynamic mic to protect the more fragile/expensive mics from flailing drum sticks. A kick drum mic became compulsory too. Though in this early stage the front head was still on the drum, and engineers, concerned with the amount of air pumped out of this bigger drum, placed the mic no closer than a metre from the head.

In the control room, things changed as well. The valve consoles had to be upgraded to handle the four-track recorder functionality. In many situations, a compressor was placed between the bussed output of the console and the tape machine, to protect the tape from overloading. It meant the live tracks were submixed down to one channel, sent through a compressor and then to one track of the tape, embedding those sounds and balance into the mix forever. So a lot of time was taken to get the balances and interaction between the elements and the compressor right before they hit tape.



Let the drums ring; take out or reduce the dampening in the kick drum, and use a full head on the front of the drum. Tune the drums up a bit higher than you’d expect, particularly the toms. Open up the hi-hats a bit more.

Play a bit harder.

Use two dynamic mics for the kit — one out the front of the kick and one at cymbal height pointing towards the centre of the kit, so that the cymbals are pointing at the side of the mic. Move your cymbal setup around so you only have a crash and a ride.

With this setup, you are again listening for drum kit coverage. By this point, engineers were looking for a little more ‘focus’ out of the mics, hence the addition of a dedicated kick mic. With these two mics, you’re getting a better representation of the close quality of the drums.

Place another mic or two — condensers or ribbons — a few metres back from the kit, and point them in the opposite direction.

With these mics, you’re strictly imitating the sound of open vocal mics. Pointed away from the drum kit, they’ll pick up drum spill and ambience. By themselves they should sound unfocused and ambient. Mixed in with the above close mics, you get that authentic crunch.

Do the high/low-pass trick mentioned above. Submix all the channels into a single auxiliary through a compressor with a moderate attack, release and threshold so the kit glues together rather than pumps. Readjust the balance once you hear the effects of the compressor, as this process will (and should) change your balance.



Key listening tracks:

Respect – Aretha Franklin
Honky Tonk Women – The Rolling Stones
Hello Goodbye – The Beatles
California Dreaming – The Mamas & the Papas


As pop music was taking over the world in the mid ’60s, there were also massive changes in recording technology. For one, by the end of the decade, stereo pipped mono for pop’s preferred format. As pop records and productions grew in creativity, so did the compulsion to innovate.

Producers started to use the extra room on four-track machines for more content. As they discovered reduction mixes (bouncing down a mix of four tracks to another four-track machine to free up more tracks for overdubbing) there became more demand for a clearer focus on the drums. Also, vocalists didn’t necessarily track live with the ensemble anymore, which reduced or almost completely eradicated the ambience and spill which had been such a big part of a record’s sound up to that point.

Combined with slightly more detailed EQ on consoles, more input channels and routing options, different compressor choices (the earliest 1176 incarnations arrived around this point, as well as The Beatles’ heavy use of Fairchild compression), and advances in tape formulations and machines, records just started to sound better. They became much more tonally focused, with a greater sense of the close mics rather than the ambience of single-track ensemble recording. Later, eight-track recording unshackled the engineers, often resulting in a whole tape track set aside for drums!


Add a few more mics for that late ’60s sound, and chuck a wallet full of cash on the snare for dampening.

Engineers still shied away from moving the mics as close as we do today — there were all sorts of concerns about overloading mics and consoles by getting too close to the drums. But they did start adding additional mics. Initially it was to represent the kit more faithfully, but became more creative as the decade rolled on. For example, engineers started to put mics somewhere in the vicinity of the toms. Often underneath the rack tom, and another somewhere vaguely over the floor tom. If they felt they weren’t getting the presence of the snare, they’d put a mic underneath the snare and add treble to increase the snap of the drum as it headed into its submix, before compression and prior to hitting tape.

Engineers blatantly broke the previous generation’s rules of miking up, and just put mics in key places, leading to plenty of experimentation with positioning. Remember, there was possibly a condenser on the overhead and snare bottom, but in every other position a standard dynamic mic was typically used.

As things developed, there were inevitably questions about drum setups and parts. Engineers were always looking for ways to get the maximum sound with the smallest amount of mics, which meant positioning became critical. Engineers experimented with pointing mics at the fulcrum of the kit between the snare and the kick drum, or at the side of the snare underneath the hi-hat, or indeed, a single mic on the floor pointing vaguely at the kick drum and bottom of the snare, often requiring the rack tom and all cymbals be removed.

Drummers were starting to control their drums a lot more — taking the front head off the bass drum and putting a pillow or blanket against the single head for a more focused sound. Placing a wallet or something heavy on the snare to reduce the ringing of the drum. In the latter part of the ’60s they even muffled the drums with tea towels or rags to totally control the sound.

In this vital and influential era, it was all about experimenting with both the drums and the techniques used to capture them. Yet undeniably, focus was in, and ambience was most definitely out.




Start using more mics closer to the kit, but not as close as you would for a contemporary kit. Think more about ‘coverage’ of the drum sounds. Possibly start with a kick mic (dynamic, right in the drum) a snare bottom mic quite close to the bottom snares (a dynamic or pencil condenser) and a condenser as an overhead. Place dynamic mics underneath the rack tom by about 30cm, and above the floor tom by about the same distance and pointed more towards the snare.

Take off the front head of the bass drum and deaden the sound. Drop the tuning of the snare a little more, but deaden the snare with a wallet or a folded up cloth. Tighten up the snares underneath, and possibly use masking tape (don’t use gaffer tape, it’ll tear the bottom head) on them to make the sound of the drum resonance free, with a short sharp snap to it. Tune the toms a little deeper, with maximum resonance. Use smaller, more ‘standard’ cymbal sizes.

Get your drummer to play at a moderate to quiet volume, laying off on the cymbals. For that Motown/Stax sound, don’t be scared to remove all toms and crash/ride cymbals, and just focus on the groove, using only one or two mics in unusual places to capture the whole kit.

Avoid ambience mics completely, and focus more on the closer mics, even keeping the overhead mic moderately low in the overall mix.

Submix all tracks to a mono auxiliary. Insert a compressor plug-in with a faster attack and release, and higher ratio. Adjust the release so the kick drum is pumping against the cymbals. Rebalance the close mics so the kick and snare are upfront and clear and a little crunchy. Add an EQ after the compressor to accentuate the low bottom end, and possibly highlight the high mids (around 4-6kHz) while low pass filtering around 10kHz. Don’t be scared to experiment with saturation/distortion plug-ins. Distortion and overdriving equipment was par for the course at this point, and is a key part of the sound of this era.

Mic positions for that tight close mic sound positioning, with kit dampeners in place.


Key listening tracks:

Rhiannon – Fleetwood Mac
Hotel California – The Eagles
Young Americans – David Bowie
Jive Talking – The Bee Gees
Close To You – The Carpenters


By the time we hit the ’70s, very quickly 16-track, then 24-track became the norm. Recording as an ensemble was no longer needed, as each instrument had its own tracks. The drums were spread out over four to six tracks, allowing for more control in the mixing process. Compressing to tape became less critical. Studios that were once big, open recording spaces started to be partitioned off for a more controlled sound, allowing for increased creative options at mix time. Valve consoles were replaced with transistor-based versions, with greater EQ and routing flexibility, allowing the drums to be sculpted tonally. Drums were made tonally dead, with bottom heads starting to be taken off toms and mics put inside the drums, or gated heavily for greater separation. Drummers devised all kinds of ways to muffle, mute and control every ambience and resonance from the kit, including copious amounts of gaffer tape, or pieces of felt that flopped up as the drum was struck and rested back on the head. Drums were tuned deeper for a fatter tone. Small diaphragm condensers became the norm for overheads and hi hats, and producers were looking for a more hi-fi approach to drum recording. It was all about a controlled, focused, deep, percussive tone that screamed ‘high quality’. Ambience was generally a thing of the past.



The deader you can tune and mute your drums, the better. Use much smaller cymbals for punctuation rather than explosive accents. Even put some light tape on the ride cymbal to make it a nice dry percussive hit without much wash. Don’t be scared to extend the hi-hat stand to its maximum height and drop the height of the snare to get as much separation as possible — or even move the hi-hat a bit further away from the snare. It may be less comfortable for the drummer, but will make the snare sound purer, and make it easier to work with.

Think massive drum kits — four or more toms were the rule rather than the exception. It wasn’t unusual to see four mics close to the cymbals, submixed to a stereo pair; a mic on each tom, submixed to a stereo pair; plus a kick and snare track, sometimes with two mics on each submixed.

It’s all about the close mics.

Once again, tell your drummer to play to the mics, which often means playing a bit quieter than normal. There are stories of legendary studio drummers who rarely played loudly, instead focusing on consistency of drum hits. Any sort of rimshots on the snare should be avoided — it’s all about consistent hits in the centre of the head. The whole effect is a drier, completely focused and controllable tone at all times.

When mixing the sounds, it’s all about clarity. Gate the snare and mute the silences on the toms. Sculpt the sounds of the kick and snare. Add high mids to the toms for extra definition. High-pass filter the cymbals and hi-hats aggressively, and keep them reasonably low in the mix. Insert a compressor on the kick and snare, and place one on a drum submix too, but with a low ratio and moderate attack/release. You’re just using it to tighten up things rather than change the tone.


Occasionally you’ll want to give contemporary-sounding kits or loops a retro overhaul. Here are a few simple tips to unsterilise your sounds and make them a bit more fun. When dealing with a kit’s individual tracks, start by submixing them into a stereo or preferably mono submix and treating the drums as one instrument.

One of the first, most simple tools is EQ. A lot of vintage gear was relatively lo-fi in comparison to today’s tools, so a few basic EQ tweaks can overhaul sounds completely. High- and low-pass filtering immediately reduces the fidelity. Finding the right gnarly frequency in the midrange, with a fairly wide Q, can give your tracks a certain ‘honk’ synonymous with vintage tone. Experiment and play around, keeping in mind that a lot of these sounds weren’t hi-fi to start with.

Next, apply compression. There were so many different levels of compression happening throughout the process. Preamps being pushed beyond their design. Compressors inserted into busses pre-tape. Tape compression itself was a big factor too, often squishing off the transients as records were tracked further into the red. So it’s worth experimenting with lots of different ratios and attack/release times, as well as both compression and limiting. For the earlier stuff, you’ll be looking at lower ratios, and longer attack and release times — you could use presets in your compressors similar to those you would use on a master bus.

For a mid ’60s flavour, set much more aggressive attack and release times to pump the cymbals. This works a treat, especially in combination with a reasonably aggressive limiter to squish the transients. You’ll hear a dramatic change to the sound of your tracks, but that’s the idea.

For late ’60s and ’70s, return to more transparent compression to glue the tracks together. Lower ratios and moderate-fast attack and release settings will tighten things up without squishing the sound too much. A little limiting to flatten out the transients will be handy too.

Last but not least is distortion. Everything from tape saturation, to overdriving compressors, to vinyl plug-ins, to distortion pedals. There’s a lot of distortion on these pre-’70s records, everything was being pushed too hard — mics, consoles and tape machines. If you listen to Motown records, for example, they’re sublimely overdriven. Having a blend control can be handy to dial in just the right amount of grit. But don’t be shy about it. This can be crucial for getting your grooves sounding spot on.


Of course, there are always exceptions to every genre. In the early ’60s, Phil Spector took to drum recording with a ’50s approach. In the ’70s, drummers like John Bonham and Roger Taylor tuned their kits to sound like Big Band drummers of the ’40s and ’50s. Nevertheless this should give you a general guideline to the way engineers captured the drums over a critical period in pop history.

Essentially, when approaching retro drums, you have to break down your own preconceived ideas of how drums should be played and recorded. Once you do that, you can get really creative with the way your drum sounds present in your records. It may just be the thing that separates you from everyone else.

Originally published:

Tutorial: Michael Carpenter

PC Laptop Investment

PC laptops and notebooks can be great for music making, but ensure you make a wise investment!

More and more musicians are abandoning desktop PCs in favour of more portable devices, both for live use and for a more compact setup at home or in the studio. Sadly though, while buying a desktop PC for audio purposes is a comparatively safe bet nowadays, buying a laptop/notebook is still a bit of a minefield. Modern PC laptops/notebooks can certainly provide more than enough processing power for most musicians, but the crucial measurement that will determine whether or not a particular model is suitable for a musician is DPC (Deferred Procedure Call) Latency.


Windows handles audio and video streams not in real time, but with periodic blocks of data, queued via an ‘interrupt call’. Whenever the next block of data is required to maintain a smooth stream, the appropriate device driver issues one of these DPC calls, which then gets placed by Windows in a queue, to be carried out when it’s finished dealing with whatever it’s currently doing. So far so good. However, as with most queues, calls are dealt with on a first come, first served basis, so if anything ahead of the audio request in the queue takes longer than expected, you’ll run into problems.

The raison d’être of any audio PC is to stream audio smoothly, without any dropouts that result in infuriating audio clicks, pops, and glitches. Any internal computer component whose driver hangs onto the processor for an excessive amount of time can cause these. It’s bad enough when they occur during playback, but if they happen during a recording you can be left with permanent interruptions in your audio that are very difficult to mask.

If you experience glitching audio on any PC you can try systematically disabling likely devices one at a time via Windows Device Manager to see if it cures the problem. Thankfully there are also a couple of free utilities that can help track down the most likely candidates. For a long time, DPC Latency Checker ( was the first port of call. It provides a graphic readout of DPC latency in real time — if all results remain in the green zone the machine is able to run audio streams without dropouts, but if any stray into the yellow area your machine might need its audio buffer size (and hence its latency) increased a little. Sadly, if results enter the red zone then you most definitely need to find and disable the culprit device. DPC Latency Checker is still useful if you have a Windows 7, Vista or XP-based PC, but it’s currently not suitable if you use Windows 8. A suitable utility that runs on both Windows 7 and 8 is LatencyMon from Resplendence Software ( The Home Edition is free, but there’s also a Pro version for business use. Their readouts are still fairly self-explanatory, and there’s also a mass of further stats on offer if you need to delve deeper into problem areas.


Now the fundamental difference between a desktop and laptop/notebook PC is that with the latter, you’re unlikely to be able to replace — and sometimes even disable — the drivers of various internal components that may cripple low latency audio performance. A few settings may be available to tweak, but even one ‘badly behaved’ device driver can bring a system to its knees audio-wise. It could be a cooling fan unexpectedly cutting in to stop your processor overheating, a Wi-Fi device being polled, network adaptors, an internal modem, or even an onboard soundchip. If you can buy a laptop/notebook on the high street, you can try taking in the above utilities on a USB stick and running them before you hand over any cash, but if you’re buying over the internet, your only option is to establish beforehand that if your chosen model isn’t well-behaved with low latency audio, you can return it for a full refund.

Sadly, a lot of people still think that specialist ‘pro audio notebooks/laptops’ are a con, but they can save you a lot of heartache. Laptops supplied by audio PC builders are simply models that have already been thoroughly tested across a wide range of audio hardware and software, and are guaranteed not to present the above problems. You also get informed support about high quality audio performance if and when you need it, rather than a blank look. Audio PC builders may discard many models before finding one that works well down to low audio latencies, while the larger audio suppliers may even have enough clout to ask the original manufacturer to provide a specially tweaked BIOS, offering options mainstream laptop equivalents don’t have. The specialists will charge you a little extra for all their time and effort, but I think it’s well worth this premium to save all the possible hassle.

If you want to go it alone, there are also a few good threads across several music forums where musicians detail specific mainstream laptop models that have been found to work well with audio. Although sometimes manufacturers end up changing one or more internal components and their drivers during a production run, perhaps because a particular part is no longer available, has been superseded, or the original spec gets ‘improved’. In such cases, audio performance may suffer, so no forum recommendation is foolproof. Sometimes even the specialists have to drop a previously good laptop from their range for the same reason.

Time after time I’ve seen musicians post on forums asking ‘Which is the best laptop to buy for audio from the following list?’ Rarely can people provide them with any specific advice, except on the odd occasion when someone has recently taken the gamble and bought that exact model, and therefore been able to test it out for themselves. Generally, model after model is suggested and discarded, before the poster either: buys a pro model and starts making music, gives up, or buys something blind in the hope that it will work well for them. The latter option is often the saddest thing of all because then the questions start. From the hopeful ‘What settings do I need to change to get acceptable performance?’ Through to, ‘Why doesn’t my BIOS have the settings you recommend altering?’ And the despair of, ‘I could never entirely eliminate clicks at any usable buffer size for playing VST instruments’. I’m still a firm supporter of desktop PCs for music, but when it comes to laptops and notebooks, all bets are off unless you stick with the professionals.


Originally published Column: Martin Walker