I had a cynical face on when I clicked this, but I found it a great article and I'd agree with all of the advice in it. It's pragmatic.
The one part I am on the fence on is on being able to play in notes, as opposed to just sequencing them on-screen. The reason I am on the fence is that, for me, there is an important question missing from the article and that is "What style, or genre, of music are you trying to create?". The reason that is important is that it can have a big influence on the tools you use to create your music - whether that is playing by keyboard then using a computer to tidy up your playing, or sitting with a blank piano-roll or step sequencer and creating loops ... I'll put that on a rough scale from Vangelis-style arrangements epics on the left, to synthwave in the middle, to electro and techno on the right, with the scale labelled "suitability of drawing patterns on screen"
Something I always agree with is that not doing it sitting at the computer is important. Not all the time, but it's nice to be able to use your gear and a sequencer without using the keyboard and mouse - if you are reading HN there's a fair chance you use a computer all day, and it's nice that when you are being creative you can put the computer to one side and treat it as a giant digital recording studio / tape recorder, as you create and make music on your synths/grooveboxes/drum machines
My perspective is written as a 50YO who has been making electronic music (acid/electro/techno) since I got my first Amiga at 17! I still can barely play the keys, but I am master of the sequencer. I have most of the classic Roland drum machines and analogue synths - I sequence the synths on Ableton mainly but program all the drums on the machines themselves as it's so much fun, using them to trigger sequences on the 101 and run the arp on my JP6 and Poly61. On the flip side, I sometimes wish I could play the keys as it would help with variations and chords for strings, so I understand the counter argument that I often hear from those classically trained
Another 50-something year old here, also a synthesizerist since I was a teenager.
I've found it very, very useful to have a piano in my studio, even though I haven't taken lessons - it has been great to engage in various Youtube channels and learn proper keys playing, fingering, chord construction and so on.
Also, Rick Beato is a god. If you're a synthesizerist but not a keyboardist, you need to spend some time with Rick.
An alternative is to get a good pad controller. You get a lot out of this beyond just playability (such as knobs — physical knobs are great for experimentation and fast, dynamic changes to things like filters), but it's still something that can be played like a piano in terms of expressiveness. But it's also something that will easily fit on your desk, and the grid layout lends itself to quickly memorizing chord patterns.
I would agree that regardless of the type of music you make (including techno and related genres, which are among my favourites), having a playable instrument will make a big difference to your production. A huge part of this is just happy accidents. The synth part at around 0:40 in this track [1] came about with just pad mashing. A standard part of my process is to record the audio output of whatever noodling I'm doing on a pad controller, pick out the good parts, and loop them. It's way more fun than drawing notes!
Agree with others that the Push is good; I have a Push 2 that I love. It's great for chords and melodies because the layout is massive (8x8). However, it's somewhat expensive (the 2, anyway) and the pads themselves could be better in terms of expressiveness/sensitivity. Also, it's only worth getting a Push if you have Ableton; as a controller for Ableton it really shines.
If you want really expressive pads, which foster dynamics and also allow you to do things like finger drumming (I really recommend questforgroove.com if that sounds interesting to you), the best pad controllers are the ones in the Maschine line-up, by Native Instruments. I have an MK3 but the more basic versions are also apparently very good. The layout is much smaller (4x4) but the pads are massive and super sensitive, they are a joy to play.
A lot of people also use Launch pads by Novation, I don't know much about them, but they are quite popular and they are, I believe, on the cheaper side. https://novationmusic.com/en/launch
Rumour has it that a Push 3 is not too far away, so keep that in mind when considering the Push lineup from Ableton.
> Also, it's only worth getting a Push if you have Ableton; as a controller for Ableton it really shines.
Quick addendum here: I've got a Push 1 that I cribbed for $150 on Ebay a few years ago (a steal at the time!) and absolutely loved my time with it/Ableton Live. When I switched my machines over to Linux though, Ableton didn't follow, so instead I picked up Bitwig and tried using that as a drop-in replacement.
Now, Bitwig is a pretty dang good DAW (I could write an entire essay's worth on why I love it), but what really knocked my socks off was DrivenByMoss' extentions[0]: this guy wrote an impressively expansive Bitwig plugin that lets you use the Push as something other than a generic MIDI controller. What's more is that I think I actually prefer the DrivenByMoss workflow to Ableton's. Partially because of Bitwig's modularity and partially because of the degree of control on display, you can do some outright stupid stuff with Bitwig+Push that isn't really possible on Live. It really is some of the most fun I've ever had making music.
Very cool! I didn’t know about that. I have been hearing a lot of positive stuff about Bitwig in the electronic music community. Personally I have invested too much time (and money) to switch from Ableton, but if I were starting now, I’d definitely consider it.
Ableton Push is absolutely fantastic - the original v1 device is pretty inexpensive as a used purchase, and it's still masses of fun. Very well built as well
Seconding this - the push was amazing for me without a big music background, as chord exploration became shape based, instead of needing piano knowledge.
Actually, just the NORNS itself is going to get you booted up with some ridiculously fantastic music theory, especially if you're the kind that reads the source ..
It requires Ableton (there might be unofficial integrations with other DAWs). I have an Akai Force which is completely standalone, not even a computer needed. But I'd honestly recommend Ableton+Push unless someone really needs it to be standalone.
I feel you should pick up and learn the synthesis architecture of the synths you you love the most.
This article/blog goes over VSTs and such, but theres nothing quite like sitting with a piece of hardware and learning that piece of hardware.
If I had to select a great piece of hardware to learn synthesis on, it world be probably the
*Elektron Digitone*.
- It's oscillator is a 4 op FM but the rest of the architecture is subtractive.
- Its very easy to get that 4 OP FM oscillator to do the classic triangle/saw square waves, for that more traditional subtractive feel.
- Its pretty popular and there are loads of high quality tutorials
- You can construct entire songs on it
- You can learn to synthesize drum hits, bass lines, etc.
- Its a 4 track midi sequencer! (4 track internal synth engine sequencer)
- Multitimbral 8 voices
- It's also a USB audio device!
Personally I'd recommend starting small with an Artruria Microbrute. It has one oscillator and no presets. Print some preset examples out from Brutal Patches and enter them on your own to learn. The manual that comes with has great synthesis explanations that aren't too complicated.
It has MIDI and a tiny patch bay. You can expand from there if you want. If you need a drum track, get yourself a Pocket Operator Rhythm and clock it from the Microbrute.
(Be careful, this is a gateway drug. I learned how to use a Microbrute and now I own 168hp of Eurorack modules, a 0-Coast, and an Arturia Matrixbrute.)
Microbrute is totally a good starter hardware synth, but I’d suggest people strongly consider the Korg Monologue. It has no patch bay, so if you’re headed towards modular it’s maybe the wrong purchase - but it’s built-in oscilloscope REALLY helps visualize what you’re doing in a way that I think a lot of people would find helpful.
Monologue and Minilogue XD are solid synths :). Those would be fantastic "first" synths. Tatsuya Takahashi, and by extension, Korg, did a really good job with those.
To each their own, but starting with FM synthesis over subtractive synthesis (particularly when it's an $800 box in question) seems to me like a good way to frustrate the average person. Now, the Digitone is no DX7 when it comes to complexity, but I've always felt like additive/FM synthesis just completely went over the layman's head. I love FM synthesis for the outright insane stuff that you can build with it, but subtractive synths have always filled the knob-twiddling hole in my heart. Once you know the basic components, you can dial in pretty much anything you can think of with a good degree of confidence.
My recommended beginners synth would probably be the Behringer Neutron [0]. You can often find it used/on sale for less than a third of the price of a Digitone, and it's a design that "grows" pretty well for the average learner. You can already get a ton of mileage out of default configuration, but then it also has a patch bay that would make your Mother-32 turn green with jealousy.
- Thats the trick though, it's a subtractive synth, you can completely ignore the FM part.
- You'll still need a keyboard or a sequencer to use the Neutron and some sort of mixer or euro-rack to headphone amp. Thats said, eurorack is a lot of fun, and patching cables is a lot of fun.
I would get a hardware synth. Much more engaging to program than looking at a screen.
I know people hate Behringer but the fact is, they are killing it with their clones of famous synthesizers. They sound pretty much like the originals and are inexpensive. Here are a few suggestions.
Hardware has the advantage of tactility and stability(both in terms of "boots the same every time, no software install or licensing" and "built to not crash"), and the disadvantages of physical maintenance, connectivity(a real studio setup needs physical cables and patch bays, so recalling a config is not so simple) and inflexible interfaces(some hardware is actually just bad to program because it's too complex to fit in a front panel). It's very personal and philosophical to decide on one or the other approach.
For me the right compromise has been to get a MIDI knob controller like the Korg Nanokontrol, use plugins to build up sounds, but bounce to samples frequently. When I want really fresh sounds, I go look for new presets or sample packs.
Working in samples using a powerful sampler/sequencer like Renoise, you've got a ton of flexibility to add production, since you can send it back through the synthesis path as many times as you want. The loss is primarily on the end of expressive performance in the raw source, which is where using a knob controller comes in to build up an expressive part when and where it's called for. Keyboarding skills can also do some of this, albeit in a different way: knob sweeps over an existing sequence is like conducting an orchestra, while little shifts in rhythm and velocity add individuality to a part.
And although extensive modulation is technically appealing, most uses of it will make a sound effect that doesn't seat itself well in an arrangement. The big stuff for performance is usually the same tried-and-true - sweep the filter, bend the pitch, vary amplitude. And this is stuff that is more satisfying to do with real time controls, for sure. But I am also a bit disillusioned of the idea that my synth noises will sound so much different from preset synth noises as to define my sound: it's like how digital artists will get overly focused on brushes, or journeyman programmers will have big ideas to solve localized code problems without really integrating the solution into the big picture. There's faster feedback when you tweak synth knobs or mess with brush styles or do a lot of edit-test-debug, and that's very engaging, fun to toy with, but the feedback won't drive you in a particular creative direction unless you already know the goal.
If you’re going to buy behringer, just buy second hand off craigslist/reverb. There’s so many of them flying around already because people buy them and then upgrade to something more powerful - no need to further saturate the secondary market.
Learned something new: Dynamo = a forceful energetic individual.
For me (being Dutch), a dynamo is an electricity generator, in particular powering the bicycle lights :)
I've been dabling in synthesizers for many years ad well, from software synths (c64 SID chip, Amiga) to hardware synths (Moog Grandmother, Korg Wavestate) and Eurorack. The Electrosmith Daisy module (based on STM32) has been a lot of fun learning low level DSP techniques , from reverb to loopers, ADSR envelopes and oscillators. The article seems to focus on software synths. The open source AudioKit Synth One for ipad is nice, as is VCV rack.
The term dynamo comes from metaphoric relationship to electric generators and the power they produce, not the notion of being "dynamic" (a modern usage of the word).
All of this is great advice to this 30+ year veteran of synthesis. One thing I'd add: instead of buying new synths, buy effects (pedals or rack or whatever). You'd be surprised how amazing the difference can be between a synth and a synth+foot pedals .. its like upgrading the signal flow by taking on another module.
Also, if you're getting into synths and haven't encountered VCVRack yet, you are in for a huge treat: http://vcvrack.com/
Get it, install it, log in and tick all the boxes for every free module in the list. There is nothing quite like starting a new VCVRack project and having close to a thousand new modules to play with and experiment around with.
Good call out - I was also going to make a comment about the importance of effects, but as the article is trying to keep things simple, I didn't want to complicate matters!
The Valhalla DSP effects are fantastic value, and SuperMassive, my absolute favourite, is also free as in beer. I suspect Sean reads HN https://valhalladsp.com/
If only Valhalla effects came in Linux VST version... I've been futzing around with linvst but it's a horrible kludge (and still doesn't work). I'd happily pay the regular buckaroos for a few professional plugins. Maybe these could get delivered through Steam?
I'm far from a pro producer, but I use tracktion's daw essentials for tasteful fx. Tal has some free old school chorus and cheesy(in a good way) reverbs, then of course there's Uhe.
I haven't tried there standalone fx, but if their built in synth fx are any indicator, you should be able to get some nice sounds from the standalone fx. I do feel like Uhe fx tend to be a little more opaque in terms of the names and values. E.g. having a filter that goes from 0 - 100 instead of just putting the cutoff frequency is a bit too old school. Having a parameter on a delay that's just called "wow" is not very helpful.
In any case all of these companies support Linux, and they all produce software worth taking a look at.
It's really a shame that more developers don't compile their VSTs for Linux. The good news is that if you look around there are plenty that do and give you everything you need to make music on Linux.
Got to say I sympathize with those that don't. All of mine including legacy stuff support LinuxVST (2.4) but it doesn't look like the KVR marketplace has me listed. Maybe they don't count free/OSS stuff?
There are whole little communities that do, but you have to dig for 'em. I recently joined Unfa's patreon (reviewer and popularizer of OSS and Linux music-making gear) and joined his discord server. Only to discover they had an emote of my face, due to my OSS plugin making and because they liked my catchphrase 'I'm Chris from Airwindows and hey… you can do this!'
And so you can. The stuff is out there, and if you look around it's absurd what you can do. My favorite discoveries lately are Bespoke, and the Surge Synth ecosystem (hugely impressive and helpful to other OSS projects of all types)
But you'll find it's a bit of a media freeze-out as far as people TELLING you any of this is happening. I can attest that if you do compile your stuff for Linux, you'll hear almost nothing about it, including from the folks you are supporting. I've got a Mediafire mirror for my stuff that tracks downloads across all my updates. Signed MacOSX bundle supporting M1 machines, 3023 downloads. LinuxVST, 325 downloads.
You do Linux support because you care about it as a concept, not because you're going to get any benefit from it. It doesn't pay, it costs (in dev time that won't be compensated, any more than if you were developing one of the little libraries upon which all the Internet depends). You have to have other work lined up to support you while you do the added Linux support.
Airwindows plugins are amazing! If you haven't tried Chris' plugins, you're missing out.
I buy music and other software to use on Linux all the time. A few individuals willing to play for software might not be significant but I just signed up with your Patreon so hopefully now you're at least partially incentivised to keep releasing your plugins on Linux.
Oh, I'm not doing that for money, or even thinking it will help me survive :)
The question is more whether I'll be compelled to stop work because I'm NOT surviving, but so long as I am, my incentive to release plugins on Linux is because it's the right thing to do. Also, I think it matters to have stuff out there where the barrier to entry is ability and education, not capital. I think basing everything off only how much capital you start with and are able to exploit, distorts things in very unpleasant ways.
Fortunately, so long as I'm alive and have internet I'm able to do something about it. Maybe, so can you! :D
Honestly there are times when I think the attitude is more useful than anything I do code-wise. And I still care a lot about some of the stuff I've come up with in code ;) but these days, more is needed.
It's worth thinking about why this is. I am very determined to stick with the kind of messaging and attitude that I like. I also have a strong intuitive feeling that it's not a good match to what platforms and competition and basically internet capitalism want to see out of me, and they've got enormous pressure to bring to bear, and many ways to direct me towards becoming what's useful to them, and indeed punish me for not doing so.
That would mean that if it's in any way hard to find 'that attitude' you're appreciating, it's because it is reserved for people who both have it and are willing to take a beating to keep it. As we think about stuff like that, think about what ways platforms would like you to be more, I dunno, Logan Paul or PewDiePie etc in nature, for engagement, and in what ways they're able to deliver pressure and advocate for those on their platform to take on that form.
There is this open source CloudSeed reverb, which has a C++ and C# version, I prefer the C++ one. With some knob fiddling it gets closer to infinite reverbs, like in Valhalla and Strymon Big Sky clouds.
I guess it can be compiled for Linux with some kung fu build skills. I ported it over to Electrosmith Daisy STM32, which is very cool too: https://www.electro-smith.com/daisy
Check out the Zynthian, it is a monster of an FX box and has a lot of quality stuff built in - plus, it has the ability to host VST's and in fact is set up to do that out of the box for the PianoTec VST's, so it may not be that difficult a stretch to get Valhalla running under Zynthian, too .. and once it is, you'll find yourself in a whole universe of power. ;)
Zynthian looks cool but it runs Linux on Raspberry Pi 4, so it probably won't be any easier trying to get Windows plugins to run it than on a Linux desktop system.
IMHO, adding an FX pedal to a synth rig is one of the easiest, simplest ways to upgrade that synth and give it a new echelon of dynamics.
I also have to say that the Zynthian is probably one of the best modules for this you can find, but that would definitely be a degree of complexity higher than the article tries to maintain. Still, an amazing FX box that really pushes the value of the F/OSS FX/DSP community (http://zynthian.org/)
Exactly! Great for the kids - my 9 year old son loves it. Even though he's got all the hardware here as well, he much prefers giggling away sitting in front of the computer, so much easier entry point!
100%. Doing a bunch of work in Syntorial is a huge shortcut to "getting it". And once you've done that, a synth is a synth is a synth. Like driving a car, becomes more, "oh yeah, this oscillator is a bit weird/interesting", rather than "wtf does this shit do"
> narrow your scope and focus on one synth... make sure it’s subtractive.
For my part I went with physical modeling first. It quickly gives results that are almost impossible with subtractive. Here's a track where I made all sounds using Apple's Sculpture (mallets, an acoustic guitar, and something like a bowed flute): https://www.youtube.com/watch?v=qkUs-BdxtN8
I wish there were more synths like this, where you could design sounds of imaginary physical objects. There's Sculpture, there's Mutable Instruments stuff, and then everything else is a notch below in sound or versatility. For awhile I was toying with the idea to make a new open source synth for bowed and wind type of sounds, but realistically I don't have the skills.
The advantage of subtractive synthesis is that each of its elements (oscillators, filters, amplifiers, modulation, effects) has an impact on the sound more or less independent of the others, and so their parameters can be understood and modified independently of the others.
In other techniques, notably FM and physical modeling, the parameters and elements are very tightly coupled and nonlinear: the particular effect of parameter X changes depending on the setting of parameter Y. As a result you can't study the pieces separately, but instead must understand the dynamics of the entire system as a global unit. This is what made FM nearly impossible for musicians to wrap their heads around: and it also plagues physical modeling.
The other problem with physical modeling is that there are many physical modeling methods and architectures, and learning one doesn't help all that much with regard to the others.
Why bother with learning a special "image editing" application ?
Why can't folks make good looking pictures from just code ?
Algorithms implemented in soft synths need to take certain things like aliasing into account and if every musicians had to bother with that, there would be much less EDM for you to enjoy.
But against a synthesist with a nice hardware synth or computer synths, there's no way coding can compete with the speed and dynamic interaction of the interface. I think this is a nice example that shows just how fast crafting a simple tune can be with a dedicated interface such as the Elektron Analog Four:
People probably could make good sounding music from code, but it's harder. When you're making a new patch on your synth, you don't want to think about complicated matters like code-as-text, you want to be able to turn a knob and hear a difference, with each change getting closer and closer to the wanted sound. Having to write even one line of code instead of sliding a slider up/down would instantly put me out of the flow when I'm creating new patches, no doubt. Changing values to some specific value instead of kind-of-here-sounds-good-to-me would slow down the entire flow as well. Switching between computer keyboard and my synth keyboard just to make a small change would do the same.
All in all, if you know exactly what you want you could probably do it in code, but part of the fun in creating a patch, and what helps with creativity, is experimentation, and I feel like that would easily get lost when coding with strict values and other "computer" syntax that fiddling with a real synth just flows together with the rest of the music making process.
There is no comparison really between the UX of sitting with a computer keyboard in front of a screen, writing/editing syntax to hear changes and hearing real-time sounds coming from something like Novation Summit together with a good pair of headphones, with zero screens involved.
You could do that, just like you could write a web app in assembly.
stuff like https://sonic-pi.net/#examples does exists, but what you start to realize is that, is even the "code first" approaches to making music, you are really just generating configuration on the fly that's being processed buy some other engine underneath, in this case it's Super collider.
Synthesizing audio, and by extention entire songs with code is really hard, try it some time. Sonic-PI seems like a good lib for it but its still really difficult - lots of threading needed to make layers of different "instruments"
I am a synth fan, and I agree with some of what's presented in the article. I think the author's approach overcomplicates things though. Subtractive synthesis is pretty intuitive. I started with a Korg Poly 800. The onboard battery was dead which meant no patch storage and no sound because everything was zeroed out by default. The guy at the music store showed me how to create a quick patch and get some sound out of it.
Other than this brief demo, I knew nothing about synthesis at the time and didn't even have a manual. Turns out, it wasn't really needed because the process was simple and the front panel clearly explained what everything did.
My advice would be yes, start with subtractive synthesis but don't feel guilty about using a preset. Pretty soon you'll know your way around the basics. You can always go to the manual or other patches for reference. I think there is little value in front-loading a lot of knowledge that you will soon forget.
One quibble with the article: If you really want a perfect playground for learning synthesis, U-he's Zebra 2 is much better than Diva (also by U-he). Zebra is fantastic when it comes to workflow and flexibility and has a UI that makes how the patches signal flow works much clearer.
I have a boatload of both eurorack and softsynths, but when I want to get some stuff done in a hurry, or on the move, or just prototype, I always wind up reaching for Zebra. Not surprising that even Hans Zimmer is a fan, and that guys physical collection is astounding.
If you told me I had to produce music with nothing but Zebra 2 and a sequencer, I could live with that. :-)
I started seriously playing around with synths a year ago, and for a while I convinced myself there was no reason to buy hardware with all the software that's available.
I would almost never recommend a beginner start with eurorack, no matter how passionate, unless you're talking about using VCV Rack or something. Physical eurorack is a money pit, and the people who own it are the first people to cop to it. Many modular systems I've seen are basically $10,000 worth of modules that add up to one well-built desktop synthesizer which you could purchase for $500-$1000.
Agreed that Eurorack or modular synthesis isn't something that I would recommend for a beginner. Also agree that a Eurorack system _could_ be an overly expensive version of an existing fixed-architecture synthesizer.
Eurorack's potential, however, is a way to explore synthesis after you've learned the basics. It can be a very open playground to try synthesis ideas that may not be possible on a fixed-architecture system.
To lower the costs of Eurorack, one could buy modules from the used market or the DIY method and build from kits if they have experience with electronics. The DIY path may also lead one to designing and building original Eurorack modules. Learn about synthesis and electronics at the same time!
As the above post mentioned, VCV Rack which is a software emulation of Eurorack is a free way to see if modular synthesis is something that appeals. Other free software near-alternatives are Pure Data (https://puredata.info/) (sibling of the commercially available Max/MSP), Bespoke (https://www.bespokesynth.com/), and SunVox (https://warmplace.ru/soft/sunvox/).
The one part I am on the fence on is on being able to play in notes, as opposed to just sequencing them on-screen. The reason I am on the fence is that, for me, there is an important question missing from the article and that is "What style, or genre, of music are you trying to create?". The reason that is important is that it can have a big influence on the tools you use to create your music - whether that is playing by keyboard then using a computer to tidy up your playing, or sitting with a blank piano-roll or step sequencer and creating loops ... I'll put that on a rough scale from Vangelis-style arrangements epics on the left, to synthwave in the middle, to electro and techno on the right, with the scale labelled "suitability of drawing patterns on screen"
Something I always agree with is that not doing it sitting at the computer is important. Not all the time, but it's nice to be able to use your gear and a sequencer without using the keyboard and mouse - if you are reading HN there's a fair chance you use a computer all day, and it's nice that when you are being creative you can put the computer to one side and treat it as a giant digital recording studio / tape recorder, as you create and make music on your synths/grooveboxes/drum machines
My perspective is written as a 50YO who has been making electronic music (acid/electro/techno) since I got my first Amiga at 17! I still can barely play the keys, but I am master of the sequencer. I have most of the classic Roland drum machines and analogue synths - I sequence the synths on Ableton mainly but program all the drums on the machines themselves as it's so much fun, using them to trigger sequences on the 101 and run the arp on my JP6 and Poly61. On the flip side, I sometimes wish I could play the keys as it would help with variations and chords for strings, so I understand the counter argument that I often hear from those classically trained