https://vcvrack.com (open source and wonderful)
https://www.youtube.com/watch?v=V35OhojjqDs <- your first tutorial
Happy patching :)
There's a bunch of other really interesting types of synthesis and you can explore them using the above software
- Frequency/phase modulation synthesis
- Vector synthesis
- physical modeling/Karplus strong
- Additive synthesis
- Eastcoast (subtractive)
- Westcoast (waveshaping/LP gates)If you want to actually learn subtractive synthesis minus the complexity use an all in one synth VST like Surge which is free and open source and you won't have to worry about tedious fundamentals that don't actually matter unless you're doing modular synthesis. Helm is another great VST.
Once you understand subtractive you can graduate to more complicated methods of synthesis like FM, vector, ETC.
Watch the video. It's 15 minutes. I wish I could learn assembly in 15 minutes! You build the synth one module and one connection at a time all connected to an oscilloscope.
If I gave someone a SH-101 with no context, and let them noodle around with it and then asked them, to explain the architecture, they wouldn't be able to.
Sure, they may make some cool noises buy they wouldn't understand, what is what, why is hooked up to what and how that might differ on some other fixed architecture synth.
If you gave someone an SH-101 and explained to them how it was architected it wouldn't really help them make that signature acid bass sound so ya they'd be a little more knowledgeable on how synthesizers technically work but they still wouldn't be able to make any music with it. Whereas if I showed them how to make that bass sound they could now go try it on every other subtractive synth they run into even if it doesn't sound the same. Besides that, they'd learn how a lot of it works in a musical sense which is way more important to using synths than any technical knowledge will give you if you actually want to write music.
Most tutorial videos are 5-10 minutes long, and completely reproducible and it's very clear the signal flow from one part to the next.
Your FM synth at home is 100x more time complicated than this patch you build but if you grok what's going on here, you should have a better idea of what going on in your other FM synths.
making my music hobby feel like my job sounds terrible lol
One of may favorite digital synths is the TX-81z. 4op FM. One of the first digital synths where the operators weren’t restricted to sine waves. (DX7 had 6op FM)
If you just look at the specs. And even play with the values you won’t understand why one synth could obtain sounds the other couldn’t.
That “source” is usually the synthesis method.(usually the complicated part outside of anything that’s not a traditional VCO)
If you have a filter, the resonance imparts a particular sound as well.
Digitone is 4op fm as the source then that’s funneled through a pretty standard east coast architecture.
That is the point of making music.
The rest is exhaust fumes.
You don’t have to know how something works to know how it sounds.
And nobody ever danced to a lecture on signal flow.
one could argue what made it great was that it only has 5 parameters that affects it's sound and it's sequencer, (and it was a total flop financially for Roland and could bought 2nd hand for like 50 bucks in Detroit. and dance music is better for it!)
and as per the subtractive synth that change music the most...I'm going to go with Moog or mini-Moog, without them there's no 303, 101, Juno, Jupiter, etc.
I love me some good acid-house though, and much rather have a 303 then a mini-moog. :)
Thanks to Uli, I can afford the both. And a 2600, etc.
With VCVRack and the right tutorial, a user will build a basic synth with an oscillator, filter, amp and envelope generator - which together make up the fundamental core of subtractive synthesis. The manual patching of modular is a great way to actually learn how these building blocks interact with one another to create sound.
The control-flow is obvious, the syntax is simple enough that novices shouldn't struggle with it, and writing directly to pins to control the peripherals gives immediate concrete feedback.
On the other hand, before vcv, seeing a vst synth just had me overwhelmed instead.
I'd recommend everyone reading this to get free vcv + the surge vcv library, and just play around with it.
Believe it or not, this is some people's preferred learning style. See also: Nand2Tetris, Linux From Scratch
We’re all uniquely different!, I promise; After a number of other paradigms had failed to teach me programming, assembly was what finally did it.
And with synthesis, it was FM first and then subtractive. The picture is more concise when looking at it from the perspective of frequency modulation because all oscillators can do everything, you know? –That’s the sense it made to me, privately and personally.
How many people will comment on a youtube video, course, strategy or book, and say something like - It's the best explanation. I tried all these different things and only this worked. The common denominator is previous failed attempts at the subject.
I think they are both acceptable places to start and learn.
For me, it was the microbrute which really taught me subtractive synthesis, simply because of how stripped back it was, I couldn't just add a new module to cover up my bad sound design. Though obviously that's far less accessible (£150 hardware synth versus free software you can load up right now), I'm sure there'd be a middle ground. I know iOS is really good for its synth ecosystem, maybe there's a nice subtractive synth there.
So much fun, in fact, that I bought a MatrixBrute not terribly long afterward. Now _that's_ a monosynth to last me a lifetime!
A few years earlier I also had a DX9 that I foolishly used to try and emulate analog sounds. Somehow I stumbled across an article on Fourier series and how infinite sinusoidal summations could be used to create the other types of fundamental waves. Programming a 4-op DX synth to emulate these and looking at waveforms in Audacity gave me a natural intuition for how time series waves relate to frequency and harmonic content.
If I had to do it all again I might get a Korg Minilogue since it can display waveforms on its LCD and is digitally-controlled.
Especially to put what it is actually called in parenthesis as if everyone calls subtractive synthesis "Eastcoast".
This is certainly something very specific to the path you took with synths.
East Coast synthesis, often associated with subtractive synthesis, emphasizes traditional keyboard performance, harmonic richness, and filters to shape sound. Meanwhile, West Coast synthesis (credited to Buchla) leans more experimental, focusing on waveshaping, FM synthesis, and unconventional control interfaces.
The terms themselves have been around for decades and have become shorthand to describe these philosophies of synthesis design and architecture of synths. You might not hear them as much outside modular or academic circles, but they’re far from obscure.
When I started using the modular, I was forced to understand the signal flow. And, the patch cables provided a visual cue of what was happening. I learned more about synths in a year with my eurorack system than I did in the previous 10 with hardwired synths.
After you learn those basic rules for patching a synth, then you get to break them. (-:
Recommending something like VCV rack seems like starting with Calculus before you can solve 2+2, but it's really not. The signal flow is right there for you to observe.
subtractive synthesis isn't synthesis. It's a transformation.
ADSR is subtractive even if you ignore the filter.
The (ideal) square wave contains the odd-integer harmonic frequencies, where the (ideal) sawtooth has all harmonic frequencies.
I think starting in the digital world may make this less clear?
You are subtracting overtones from a non-sinusoidal set, the sound synthesis in subtractive synths is the more like choosing digits to construct a representable number.
Additive synths are actually far more restricted...remember that the set of computable numbers is not quite as small as the cantor set, but is getting there.
- Two oscillators undergoing detune, sync, ring or amplitude modulation, or fm prior to getting fed into the filter?
- An LFO combined with an oscillator?
- An envelope (controlling the filter or amplifier) combined with an oscillator?
Perhaps these things might be considered combinations? I agree this is weak. You can blame the RCA Mark I and II for calling subtractive synthesizers "synthesizers".
Subtractive synthesis has a particular meaning in common use whether it’s right or wrong.
Purely logically, no, but that's not really the practical sense we're talking about. And by introducing a digital delay, you do induce a filter on the sound. So if you have a delay, then you have a filter.
> In particular, the type of effect known as a delay (e.g. a delay guitar pedal) isn’t a filter, certainly not in the musical sense, which is the relevant sense here.
I think it's best to reconsider my original comment. I was arguing that it isn't useful to suddenly rename subtractive synthesis as a "transformation", because you could just rename all of synthesis "filtering". But that's not useful, and synthesis just means building up sound from parts. I.e., it's best to work at a given level of the parts and think about things like waveforms, envelopes, LFOs, pulses, triggers, delays, reverbs, etc., most of the time.
Getting Started Making Sounds - https://news.ycombinator.com/item?id=31434208 - May 2022 (3 comments)
Abletone Learning Synth - https://news.ycombinator.com/item?id=31279526 - May 2022 (63 comments)
Synth Playground - https://news.ycombinator.com/item?id=26429207 - March 2021 (21 comments)
Learning Synths - https://news.ycombinator.com/item?id=20272346 - June 2019 (172 comments)
It's a good idea to show content before your page does anything that asks for scary permissions. (And, honestly, without knowing what the site does, its pretty scary to click on a link on hacker news and have a site ask for elevated permissions before it shows anything.)
A million years ago I had some analogue korg model and have been interested in playing around again. But I know myself well enough to know that my interest may very well be fleeting and I don't want to invest much in the way of $$ to go that route.
https://stgdownloads.novationmusic.com/novation/novation-sof...
If you’re an iOS user there are a ton of good soft synths on iOS that are criminally cheap. E.g. Moog has a version of the Mini that’s like $15.
Personally, I'm not into modular synths, so I don't recommend them as "Stop buying synths and start doing modular synths" but more like "If you're already into synths, but want to learn more about how they actually synthesize the sound, give VCVRack a try", merely as a learning tool.
If someone is going to buy a hardware synth to learn on, I think a well laid-out semi-modular that has a good default signal path is a good way to go. Just ignore those funny 3.5” jacks until you’re ready.
(but no, really not like this; https://www.meta.com/en-gb/experiences/synthvr/3748465338566... ; then I would just get a real synth)
Here's an example, although I'm not sure if it's good or not because I haven't tried it yet. https://www.uwyn.com/geco/
Check out the Ondes Martenot for a compromise between the two paradigms. https://en.wikipedia.org/wiki/Ondes_Martenot
https://lambda.cuesta.dev/ (repo: https://github.com/alvaro-cuesta/lambda-musika) -- check out the examples on the bottom toolbar's blue button.
The basic idea is you write a function `t => [l, r]` where `t` is time and `l`, `r` are output samples for the left and right channels in `[-1, 1]` range. You can think of it like ShaderToy but for sound synthesis.
It includes a small utility library but it's meant to be just a few helper functions instead of a full-fledged framework like SuperCollider, Sonic Pi, et al. I.e. it's still sample-oriented instead of module-oriented. E.g. in Sonic Pi you script modules, their parameters, and how they connect with each other, while Lambda Musika is all about outputting samples of a waveform.
It's very barebones -- I'd love to get some time to upgrade this to Monaco editor and add TS, intellisense, etc. -- and possibly buggy, but I still find myself coming from time to have some fun with it.
The links don't even work anymore on CMU.
Common Music might still work but I can't imagine bothering with Nyquist. https://ccrma.stanford.edu/software/clm/
There is just not much reason to not use SuperCollider or Csound instead of these though.
Edit: I did just find Nyquist has been rolled into Audacity scripting that sounds pretty cool https://audionyq.com/
I posted that one, and not others, because it is a lisp dialect, and this community is typically inclined to lisp in a more than average %.
>Behringer Model D >Novation Bass Station 2
Recommending VCV is horrible advice - unless your idea of learning synthesis is getting RSI...I think VCV is a great testing bed for trying modular ideas...once you understand synthesis...i think a huge draw back of VCV is the plethora of choice - it's just way way too much.
I learnt modular on a real life Doepfer modular...it was frustrating as hell until things starting clicking - i cant imagine the feedback loop on software being that good.
edit: side note making sound is one thing, making something actually worth listening to...
Have fun
It was successfully launched on Kickstarter a while ago, and is now available through a few retailers.
I relate to that: I was inflicted a couple years of piano lessons as a kid, and hated every minute of it - because I detested the music that my teacher and my family considered appropriate.
40 years later, I stumble on electronic musics and realize I would enjoy making some - and I joyfully (and incompetently) began on that path, lately becoming ripe for some dry theory because I now value it.
Each person is different though: my daughter always enjoyed the technical exercises for their own sake !
If instead of playing you're more interested in making cool sounds, I'd skip the keyboard completely and download a free DAW like Reaper, Ableton, or FL Studio.
Whichever route you take, the secret is practice. It always is...
Constraints can be a great thing.
Alternatively, an Akai MPK Mini MK3 costs $100 and gives you twice the range, no limit on simultaneous keypresses, velocity sensitivity, a mod wheel, analog knobs, and velocity sensitive drum pads.
You don't have that with a computer keyboard. Music is hard, don't set yourself for failure so early, for no good reason.
Many synthesizers (especially softsynths) map changes in sound to velocity, that is, how hard you hit the key. Hitting the key harder makes a different sound (e.g. layers more samples) than hitting it softly.
This isn't necessarily a problem. As you say, constraints can breed creativity. A good musician should still be able to play great music, but for somebody just learning it's a lot of unnecessary friction.
At the same time, if this is the best one can hope for on a computer keyboard, I feel comfortable resting my case.
But it can't help you learn piano, which is what the top-level comment asked for.
What you're asking about is impossible. Not just difficult. You might be able to learn something about moving your fingers independently with rhythm on a computer keyboard. But at about the time you get to one week of experience, you'll probably be doing at least as much harm as good in terms of learning to play piano music on a computer keyboard. It is too different.
Partial list of problems:
Size of keys
Position of keys
Number of keys
Travel distance of keys
Velocity sensitivity of keys
Sufficiently low latency of audio output (can be solved with pro audio hardware)
1. Choose a song you like and find out what chords are needed to play it.
2. Learn those chords and practice playing the song.
Use a keyboard with weighted keys for best results.
Refs: