We are thousands, probably millions of “bedroom producers” in the world. What’s fascinating is that it represents as many setups, combinations of gears, softwares, FX, synths, decoration, furnitures… Since I put a lot of efforts iterating on my own ideal setup, I couldn’t resist to share some of my feedback here.
Disclaimer: I’m not advertising any of the products mentioned below. I’m happy to share my experience and justify my choices, but what is good for me might be different for you.
What do I want to achieve when building my studio? “Ergonomics” is probably the most recurring keyword I would use here. What should I put on my left? right in front of me? how can I optimize my most recurrent tasks? Everything is designed to help accomplish what I have in mind as effectively as possible. When it’s time to explore, I need to make sure everything is done to boost my creativity and don’t prevent me from going further. As you accumulate more and more hardware, and hopefully more and more knowledge, it’s pretty easy to get lost and end up using 10% of what you could actually do. Note that limiting your capabilities can be a somewhat creative process, but that’s another story.
This blog post will cover my workflow, and explain how I’m using those units:
- Presonus Faderport 8
- Vermona DRM1 MK3
- Jomox Alpha Base
- Elektron Analog Heat
- Acidlab Bassline 3
- Squarp Pyramid
- Moog Mother32
- Moog DFAM
- Modular synth
- Dave Smith Instruments Pro2
- Roland Studio Capture
- Teenage Instruments Pocket Operators
- Strymon Big Sky
- … and a few other FX :-)
Before diving deeper into the details, let me quickly describe the overall workflow so that you can better understand my approach of making music.
Current workflow in one line:
Sequencers → Machines → Patchbay → FX → audio interface → DAW
So what does it mean? First, everything is orchestrated by sequencers. When I press the “play” button on my main sequencer (Squarp Pyramid, will be described later), then the track starts. This main sequencer handles patterns, sequences, CC messages / automations, midi FX, and the clock signal. I’m also using embedded sequencers for some devices when I find them especially suitable for the task, but they will still follow the master clock signal.
All the machines are then routed to a patchbay with 2 rows of 24 patch points, which allows me to quickly re-route any signal to any input. Meaning it’s fairly easy to apply FX when needed without re-wiring the devices. An exhaustive description of the capabilities offered by the patchbay is provided later in this post.
Note that I’m not intentionally taking a dawless path and going full hardware. There’s even a DAW in my setup! My workflow is strongly focus on creativity, ergonomics, and getting things done. For the creativity part, sorry but I don’t find modern DAWs inspiring. They don’t encourage me to build tracks the right way, maybe because of the endless possibilities. For ergonomics and getting things done, I try to avoid diving into menus or spending hours programming devices in depth. This also eliminated DAWs as well as many menu-intensive gears (and yes, I mean Elektron…)
Sequencing all the stuff
To orchestrate everything, I deliberately made the choice to avoid spending hours in front of a screen. While software can do a good job at sequencing everything, I find it more intuitive, more creative, to deal with hardware sequencers designed for that purpose. Anything that inspires me is a great addition to my setup, no matter how “good or bad” it sounds, or how complete it could be.
My first hardware sequencer is actually pretty recent and was an Arturia Beatstep Pro. I realized at that point how creative and efficient it was. Turn knobs, press buttons, change pattern lengths, chain patterns, add a bit of randomess, that’s what you can do in a matter of seconds, really! And experimenting again and again is straightforward, and even encouraged by the interface. Admittedly, you have limited memory and cannot easily save hundreds of versions on your PC. But I learnt to appreciate the volatility, and I’m actually not that scared to loose some work, sometimes.
Anything that inspires me is a great addition to my setup, no matter how “good or bad” it sounds, or how complete it could be.
After this successful experiment, I decided to move forward and find a replacement that could handle all the new synths I’ve accumulated. So I needed to find a sequencer capable of generating polyphonic sequences over dozens of machines. My first choice was the awesome Sequentix Cirklon, but unfortunately, there’s a long list, hence a long wait, before you can get one (and second hand units are pretty rare or more expensive than the brand new units). Thus I opted for the Squarp Pyramid. Squarp is a French manufacturer, and their sequencer is considered as top-notch in term of feature and integration. I don’t want to list all the features here, but 64 polyphonic tracks, with unlimited number of notes & automation, CV/Gate connectivity and 2 MIDI outputs, that was the perfect candidate.
I can’t regret the Pyramid, and the price difference with the Beatstep Pro is definitely justified here. In term of effectiveness AND creativity, what I really love is their implementation of an euclidian sequencer. It’s pure madness, mostly for rhythms. You can build rich and complex grooves by turning only 2 knobs.
It’s a bit menu-intensive, but they’re limited and well designed. The implemented MIDI FX, for instance, are excellent. I primarily use the randomness and the MIDI delay. The delay can really do the trick when you want to add a bit of groove without any audio FX. The randomness applied to velocity, on hi-hats for instance, contributes to humanize repetitive patterns and surely changed my approach of building patterns.
To be honest, I think the Cirklon would still be my first choice though, because of the higher number of knobs and buttons provided by the front panel. This is closer to my ideal workflow, but the Pyramid is close enough. There’s a tradeoff often discussed. So if you consider buying a first-class hardware sequencer, I encourage you to carefully read the specs, and ideally try both. The key thing you should remember is that the sequencer is the central piece of my setup. In term of connectivity, I’m using both MIDI outputs. They’re both sending the clock signal, but only one is sending the start/stop signal, to trigger machines with an internal sequencer I want to use, as described later. The MIDI input is wired to the DSI Pro2, which is my only keyboard synth, and thus allows me to input notes with a keyboard to generate sequence, when it’s more relevant and easier than using the step-sequencer capabilities.
Practically speaking, I’ve designed a template project on the Pyramid that I can re-use for any new project. Since the configuration is saved per-project, this template has presets for my entire setup: device names, note names (for drum machines, e.g. “kick”; “snare”; “HH”; “tom1”…), MIDI channels, human-readable CC messages (e.g. “prog-change”; “flt1-cutoff”; “flt1-res”…). By the way, the current OS (PyraOS 3.0, beta) adds the capability to write such “definitions files” as regular text file that can be used by the Pyramid. Since the community is pretty active, it should be easy to find list of CC messages for your units, from this forum.
Add more drums!
Ok, so that everything is connected by MIDI cable to the sequencer, how do I actually generate sounds? Let’s start with the drum-related sounds! I feel lucky enough to own two awesome machines: a Vermona DRM1 MK3, and a Jomox Alpha Base. They both have individual outputs, but I choose to only use those from the Alpha Base. On the DRM1, the master output is connected to the Patchbay. I might switch the connectivity in the future, and the cool thing is that they’re close to each other, so that I can quickly change the setup without changing or moving the cables.
The Jomox has nice internal effects, reverb and delay, applicable on individual instruments. But the effect is actually routed to the master output only, and doesn’t affect the individual outputs. Same for the embedded FM synth, which doesn’t have a dedicated output. So the master output is also connected to the patchbay. Worth saying the instruments are actually removed from the master output when their individual output is connected.
Tons a knobs, a bunch of buttons, that’s exactly what I need to design a drum kit!
The Jomox and the DRM1 are both capable of extremely wide and original sounds. Although I really like the classical X0X sounds (I had a TR-8), I was looking for machines that would allow me to design my own kits, with a fine control of energy, attack or harmonics I want to put in my drum sounds. I don’t even know it makes a strong difference at the end, but I love the idea of proposing my own signature that way. Plus, it’s a super stimulating process. Tons a knobs, a bunch of buttons, that’s exactly what I need to design a drum kit! Moreover, thanks to the individual outputs, it’s fairly easy and creative to apply an audio FX on a specific instrument.
I’ve put these 2 machines on my left, so that I can turn knobs whenever I need. I found having drum machines separated from the rest pretty effective. Practically speaking, once the drums elements have been modeled and I’m happy with the sonic atmosphere, I barely need to touch them. So I can jump to the right side and work on the other parts, and the back and forth from left to right are somewhat limited, in practice. Depending on the track I’m building, drums parts come first or second. I usually start with the “main” element of my track, the one that should keep the listener busy. Sometimes it’s the bass line, or the lead, or the kick, snare and hi-hats to establish the foundations of the track. But when I don’t start with drums, they come in second position anyway. That’s not always a kick by the way, but any rhythmic element to immediately start forging an atmosphere.
Regarding sequencing, the DRM1 has no sequencer at all so needs to be controlled by the Pyramid. The Jomox has a pretty nice internal sequencer, with roll/flam effect you can enable per step, and parameter locks (p-locks) for the pitch and an extra parameter. At this point I don’t know if I will effectively use its sequencer, since I really love the euclidian approach offered by the Pyramid, which also makes it easy to automate parameter changes through the bunch of CC messages. Worth saying, the Jomox has a strong MIDI footprint. You will need 11 channels for the instruments, and an extra one to handle program changes… And no, it’s not configurable (yet?). No need to say I’m pretty happy to have a hardware sequencer meeting these requirements. Thankfully, if you cannot afford that many channels, there’s a channel that can translate notes for all the instruments at once. With this mapping, you have access to only 4 notes mapped to 4 different pitch values (while you can use C1 to C6 values to control the pitch when using dedicated channels). Still about sequencing, on the Pyramid I decided to use a dedicated track for each instrument. While it may sound overkill or tedious at first (I though so myself, I confess), it’s actually extremely flexible, and suitable for both production and live workflows. It’s convenient to mute/unmute a specific drum sound, or to add a MIDI FX, like a delay, randomizer or arpeggiator (provided the Jomox can process different pitch values).
So for now, I’ve extensively covered what is required to make some Boum! and Tchack! Enough for some kind of technos, but not for mine. I actually have many options to generate bass sounds. the Moog Mother32, Moog DFAM and the DSI Pro2 are all capable of awesome bass sounds. But the Bassline 3 from Acidlab becomes especially helpful when it comes to 303-like sounds. You’re probably aware there are many, many 303 clones on the market. I wasn’t looking for something as close as possible to the genuine Roland 303. First, I don’t make a lot of Acid tracks, and second, I don’t need to feel “close”. Well, that box is probably “close enough”, but I was way more excited to hear it had its own character and personality.
Except the pattern/track management that I found a bit weird at first try, sculpting the bass sound is straightforward. From round and heavy bass sounds to more acid tones, that’s the range you can expect, and you get what you expect. No menu, and an enjoyable sequencer well designed for jamming and building more complex sequences, to a degree. You get it: the Bassline 3 will respond to the start/stop signal sent by the sequencer, unlike the Jomox. Regarding this unit, I was really looking for something specific, dedicated to a somewhat unique kind of sound. I had a Roland TB-3, which made me pretty happy to be honest (don’t care it’s digital), but I was lost in the bunch of factory presets. I also disliked the sequencer interface, as well as the one to control the envelope. Once I managed to formalize what I disliked and what I actually needed, the choice of the Bassline was rather straightforward. Oh, and there’s an FM input to control the filter AND an audio input to process external signals, instead of the internal oscillators (2 waveforms). As you will see in the depiction of the Patchbay, that’s a perfect combo with other stuff!
It’s hot in here, isn’t it?
Believe it or not, in term of ergonomics, one of the thing I’m the most proud of is the placement of the Elektron Analog Heat (c.f. picture above). Immediately reachable at my right, perfectly inclined using a $1 support, it has become my favorite toolbox to give character to my sounds. One of the most obvious usage is to add distortion (especially cool on the kick, snare, hi-hats or bass) but I also often use it to simulate tape recorders and compressors on the master track, using smoother settings. Finally, it can also become even more creative when playing with the filter, EQ, LFO and envelope follower. I bought it a few days after the launch as a high-end distortion unit, and it’s now my best friend, always turned on. Studio One has a super-nice tool called “Pipeline”, which provides a convenient way to manage external FX and route the signal accordingly. In short, it’s a natural unification of software and hardware, even handling potential latency issues. After some months of extensive usages, I ended up with my own bank of presets which solidly contribute to my sound signature.
Ian Pooley has released an outstanding video comparing 4 similar units, and the Analog Heats honorably competes with more high-end units.
The synths and modular zone
When I jam or experiment to craft an atmosphere, I rarely start with drums or bass. Most often, inspiration comes from this zone. I switch one the 4 boxes you can see below and things start getting serious (Moog Mother 32, Moog DFAM, Modular synth and a Pro2 from Dave Smith).
They’re all packed together… because they interact absolutely well together! Mother 32 is a semi-modular, as well as the DFAM, and they both belong the same Moog family, so that’s obvious. Extending the capabilities with some other modules sounds natural, isn’t it? And the Pro2 has 4 CV in, 4 CV out, and 1 gate out. And its modulation matrix allows you to basically send or receive anything (LFO, pitch, envelopes, …). There’s so many things to tell about all of these synths, I’d prefer to give an thorough summary instead. They’re all inspiring (again), bring sounds that fit well in my tracks, and they’re versatile in a way that they encourage my creativity. I built the modular to provide utilities that add something new for the rest of the setup (a sample destroyer, aka Make Noise Morphagene, a percussive sound generator aka Elements from Mutable Instruments, the most-powerfule-CV-generator-ever aka Control Forge from Rossum, and the obvious and now classical Maths that can serve plenty of purposes for a single module). Patching and repatching them is pure joy, and thanks to the Pyramid, Mother 32, DFAM and Pro2, I have endless sequencing options.
Another similarity among these 4 friends: no need to dive into menus. I know I’m repeating myself, but I turn knob, flap switches and press buttons. I don’t care to loose my work (on the Mother32, for instance). This even encourages me to dig a sound as much as I can before experimenting something new. The Pro2 has menus with advanced options, but you can do almost everything from the front panel. The Control Forge might the most menu-friendly hardware that I have, but I can effectively launch a preset to generate CV curve, and tune the parameters (steps, length, magnitude…) by turning knobs and pressing buttons.
In retrospect, I’d say my process is rather iterative. Build a sound I like, sequence a simple loop, and make it live and organically evolve by adding envelopes, LFO’s, clock dividers and randomness until I get something interesting. Keeping these 4 synthesizers close to each other, alongside the central sequencer, is key in that regard. And again, since the drum machines are not there, I’m not tempted to change too many things at once. Instead, this layout helps me to stay focused and creative, doing one thing at a time.
The FX zone
I cannot thing about any official statistics, but I guess it’s safe to state 99% of the tracks are using effects at some points. So do I! I invested lately on hardware FX. I first thought I could do everything from the DAW. But as I thought more deeply about my approach of making music, how to make my workflow inspiring and engaging, I saw no reason to apply the same reasoning as the one used for synths. Thus I started accumulating a few boxes. The good news is that some of they can be really cheap, below the $100 or even $50 for some of them. As usual, I’m not seeking for something promising the perfect sound. Crispy or distorted sound is even welcome and part of the game. The Roland SDE-1000 delay (rackable FX in the middle) is a perfect example. For less than $50, it’s capable of unique grooves and turning the knobs and pressing the buttons encourages me to get something I would never get with software. It’s lazyness, I confess, and I could possibly achieve even more with bundled plugins, with cleaner sound, but practically speaking, using my mouse, sitting in front of a screen, is so boring that I couldn’t go beyond the provided presets.
The Strymon BigSky is as powerful as a synth. Just generate a long-tail shimmer, chorale or cloud reverb, and a any short and basic blop will send you to space. Then send the output to a ring modulator (Moog MF-102) or a Chorus (Boss CE-300, classic Roland chorus sound) and your FX laboratory will guide you to the path to unique atmosphere. I’m still building at improving this zone and constantly looking at second-hand pedals or racks.
Patchbay, the cables jungle
This has been mentioned many times throughout this article, it’s not time to get into the details. I wasn’t first convinced by this addition, but it happened to become the guarantee of modularity, flexibility and creativity.
By default (when additional cable is wired on the front panel), the routing is natural: each instrument is connected to an audio input. This way, I just to turn my hardware on to start playing, which is a strong requirement to feel productive. The 6 last input/outputs are dedicated to FX (2x stereo FX chains, and 2x mono FX chains). Audio inputs provided by the synths (Bassline3, Pro2, Alpha Base) are also connected so that they can process any external signal. Benefits from the bay:
- Avoid fatigue on the critical cables and connectivity (instruments, audio interface)
- Allow to virtually connect more instruments than available inputs (and then just re-patch when needed)
- Conveniently routing all the audio signals from a central piece
- Encourage complex paths, e.g. Instrument → FX1 → FX2 → Audio Input
And that’s a rather cheap investment. Another usage is to quickly plug some “random” sound boxes such as the Pocket Operators from Teenage Engineering.
Generating some stupid sequences from them are ridiculously fun and once processed with other stuff, they can integrate well in your track.
Laptop, DAW, controller
While I spend a vast majority of my time playing with the machines, everything is eventually routed to the audio interface, plugged to the Macbook, hosting Presonus Studio One, controlled via a Presonus Faderport 8. If you want to learn more about why I actually moved from Ableton Live to Studio One, I wrote another detailed blog post a while back, but basically it has been a matter of ergonomics.
The Studio One / Faderport combo is key in my post-production workflow. Since I do my best to record something as close as possible to the final version, the remaining steps intend to sanitize the signals (remote unwanted low frequencies…) and mix the tracks together (levels, panning, EQ) so that it sounds even better. And guess what, all of that is achievable via the Faderport controller! So I just need to load my custom template into Studio One, record, and play with knobs and faders to realize 95% of the job (a few minor edits are sometimes made afterwards). One complain about the Faderport is the poor readability of the scribble strips. You can adjust the contrast with the last firmwares, but slightly tilting it makes them perfectly readable, and also helps to quickly identify and access the buttons you need.
The audio interface is a Roland Studio Capture. What I needed here was a card with at least 16 inputs, 4 outputs, and autonomous, i.e. not requiring any computer to work properly. At that point I hesitated between modern mixing consoles offering audio interface capabilities (e.g. Soundcraft Signature 22 MTK) and more standard desktop interfaces. While the former option was appealing, I couldn’t find a solution satisfying all my concerns (good pre-amps, switchable pre/post-EQ and pre/post-fader routing…) and such consoles clearly require a lot of space. On the other hand the latter option is less flexible as you can’t mix straight from the interface (well, actually possible, but definitely not convenient), but save space, and often propose more advanced routing options. You got it, I opted for a rackable audio interface. The Roland Studio Capture had decent pre-amps for the price, capable of very-low latency, and gives me the inputs/outputs I needed. Plus, the frontend panel is quite convenient and you can control everything straight from the box (levels, amps, per-channel –digital– configurable compressors, high-pass filters, etc.) In term of placement, practically speaking I barely need to touch it except for (a) turning it on and off, and (b) adjusting monitors or headphones volume.
That was a long story, hopefully you enjoyed the reading and are now full of new inspiration for your own setup. In case you don’t know what you should remember, keep in mind that thinking diligently about your needs, workflow and approach of music is more important than the actual specification of the hardware you want to buy. Buying hardware you won’t gonna use, or getting high-end synthesizer that will inspire you nothing but the presets is probably not the best option. It takes time, and you’ll probably buy and sell many devices before being satisfied. I did “mistakes” and learnt a lot. Let me conclude this article with the list of hardware I bought and quickly sold (don’t mean they’re bad, actually I really loved some, but found better alternative as I made progress and increased my awareness):
- Akai MPD-32
- Akai EIE-Pro
- Arturia Beatstep Pro (actually, I still have it)
- Focusrite Saffire Pro14
- Korg Volca Beats
- Novation Launchpad
- Novation Mininova
- Roland System-1
- Roland TB-3
- Roland TR-8
- Soundcraft EPM-6