Synthesizers (and other pro audio stuff)

[demagogue]

This is one of the better tracks that I've synthed together.
I think it's safe to say it's in the ambient genre.
https://soundcloud.com/user9513654/a...e-already-gone

[Aja]

Aja - I am really loving that track you linked above.

Thank you! I have more than an album’s worth of material pretty much ready to release, but I don’t know the best way to actually get people to hear it. Everyone and their dog makes ambient music these days, and I’m not sure how one stands out.

And I would definitely be down for a synth-off contest. For anyone without a synth, there are lots of free software synths you could compete with. VCVRack might be my first choice. There’s also Dexed, which is a pretty spot-on DX-7 emulation.

[faetal]

I use VCV rack. I'm going to be getting a disting ES-3 at some point so that I can use Ableton's CV tools with my semi & full modulars and I think being able to get VCV integrated would be a great way to test which modules I want to get next.

As for getting people to listen to music? Tough one.
I think if your music is good enough, it'll just happen organically - you'll play it to someone, they'll play it to someone else and so on.
I came to the conclusion a fair while back that my music just isn't good enough to get its own momentum.
That said, I'm probably being pessimistic, as larger issue is that there is just so much music out there, it's really difficult to stand out enough to pique people's sustained interest.

[Gray]

I know my place. I know I lack talent and the music I make isn't very good, so I see no point in promoting it. I just enjoy making it, and if there's the occasional track I feel pretty good about I might put it on SoundCloud, knowing full and well that there is lots and lots of better music out there made by more talented people. If someone stumbles upon it and likes it, fine, but I'm not promoting it.

[Aja]

I use VCV rack. I'm going to be getting a disting ES-3 at some point so that I can use Ableton's CV tools with my semi & full modulars and I think being able to get VCV integrated would be a great way to test which modules I want to get next.

As for getting people to listen to music? Tough one.
I think if your music is good enough, it'll just happen organically - you'll play it to someone, they'll play it to someone else and so on.
I came to the conclusion a fair while back that my music just isn't good enough to get its own momentum.
That said, I'm probably being pessimistic, as larger issue is that there is just so much music out there, it's really difficult to stand out enough to pique people's sustained interest.

Yeah, I worry about that lots too, but I guess all I can do is make the music I want to hear.

Getting an interface from modular to computer is a good idea. I wouldn’t mind doing it except my laptop is old and chokes up when running VCVRack.

[demagogue]

Well let me make a proposal for a contest here then. And people that are interested, please give your input and we can change things to whatever suits us best.

I think the best thing to make is probably a recording of a custom patch (although it could have multiple voices like a pad, drums, & bass), playing some kind of melody, played or sequenced or logic'd; something that people can listen to as music or as a piece. But not like a whole song with multiple parts--which could get complex, take a lot of time, and muddy up what we're listening for--just a nice excerpt that shows off the patch, and not just like a single voice playing a note (like "just make a guitar sound"), which would be boring to listen to. Then people post the recording here, and I'll collect them in the first post of a new thread for people to vote & comment. And I'd personally like to see a little description of how they made the patch or what's happening, although someone could fairly say they had no idea what they were really doing and this sound came out.

I guess to be a proper synth contest, there should be a rule that you synthesize every voice you use and not just play vanilla samples or factory patches, or at least not unmodified. (Maybe a drum loop would be okay, though, as long as the other voices are synthesized, but not encouraged?)

The main question I have is would we just be submitting one patch or more like three patches? I feel like if it were three, then people would be flexing their synthing skills into much different genres and techniques, which would be more interesting all around, but it would take longer. What do you think? Actually the idea I had was to have three categories, but people can submit to one, two, or all three of them as they have time and initiative, so it's no disqualification if you only submitted to one, but all the better if people submit to more than one. And then each one is graded individually.

Anyway, if we did have categories, it seems genre is the best way to separate them. Then we'd want genres that explore quite different sound spaces. The first three that come to mind for me are:
1. any type of EDM (house, techno, trance, acid, dub)
2. any type of synth-pop (New Wave & '80s synth-pop, Krautrock, whatever Chvrches is doing), and
3. anything ambient (chill, dark, environmental, drone, experimental, weird interesting noises that don't have to have anything to do with music).

I thought about industrial too, since I like the genre. But maybe that fits into synth-pop. Rather than have 4 categories, I'd rather just count it in that category.

Then for a deadline, a single patch is something that's good to put together kind of quickly and by inspiration, so could take just a weekend. If we were doing just one patch, then I'd think 3~4 weekends is more than enough. If we do 3 patches, then maybe more like 6~8 weekends. So ... Sept. 1 deadline maybe? But if people finish much earlier we can bump it up? I'm not sure, but does that sound workable?

Edit: Or we can just keep it really simple. One patch, anything you want to do, by Aug. 1. That might be more realistic.

Oh, and people can use whatever system they want, hardware or software. The more the merrier. As was mentioned above, there's a virtual modular synth system called VCV Rack that anyone can download for free (here). There are already 100s of modules for it that are free to use, and lots of tutorials on YouTube. (Omri Cohen's tutorials are the gold standard. Almost all of his videos are for individual patches, too, if you want to know what that means.) So anybody could participate using that if they wanted to join.

[Gray]

This is what I was talking about. The very first time I've put my actual voice on a track for over 30 years, albeit heavily vocoded. It's not great, I know. I made this on new year's day, hung over, pissed off about all the pointless nostalgia.
Ingenting Var Bättre Förr Men Allting Är Samre Nu (Nothing was better in the olden days but it's much worse now).

There are six verses of rambling lyrics not yet recorded, more or less just whining or shouting about how everything is wrong, and you don't need to hear all that crap. And this was before the pandemic, count yourself lucky I've not uploaded my lockdown songs yet, they're even worse.

[faetal]

I have a week off, so the timing is right.

[PigLick]

I am up for this, but it will purely be software, and as I have just moved have to go through the rigmarole of setting all my gear up.

[PigLick]

i guess this would be in the ambient category?
https://soundcloud.com/darryl-25/synth

[demagogue]

I'm thinking being too specific about categories just limits people's imaginations. So maybe it's better to say one more on the dance or rave end, one more on the ambient or weird or noise end, and one "normal" one on the pop or musical end (anything that isn't the other two), but mixing them up is okay too and one track could even count for multiple categories.

Really I think people can submit anything and others voting or commenting on them can rate it however seems best to them (as a weird sound it'd be xx, as a dance track it'd be xx, as a piece of pure synth music it'd be xx). Practically speaking there's probably one it's closer to than the others. I'd say that's on the weird end.

Edit: So I made a patch today. Here it is.



Here's a little explanation. It's in VCV Rack. On the bottom left there's a clock (Clocked) that's sending out triggers at 4 different tempos and the 4 modules to the right of it are all LFOs (low frequency oscillators) basically just going up and down in a sine wave at 4 different rates and modulating things.

On the top-left is Basal, that's the oscillator (the voice). It's a kind of roundish-squarish buzzy wave, you can see on the osciliscope. There's a slow LFO tweaking "Mod1" making the buzzy go up and down (changing from a round-shaped skinny sine wave to a square-shaped wave and back). And the pitch is being changed by Ions. The notes are just set in a circle, and every 8th beat it switches to the next note.

Then it's going through the Laika filter. Then you see the three Laika parameters are wiggling, cutoff (back and forth), saturation (how much it cuts off, so it slowly pulses in and out), and drive (a hard edge) which goes up and down even more slowly, all controlled by LFOs going different rates except the cutoff is by the clock, so it works like a pulse wave. Tomi is the boom boom. One of the slower LFOs raises and lowers the volume right on the mixer (the yellow arrow going up and down), so the boom boom comes in and out. Then we have the hi-hat, which has three parameters that I hooked up to different clock settings (closed high hat, open high hat, an accent ting, and FM to make the pitch go up and down). Then everything goes into the mixer, and the mixed audio goes into Plateau for some reverb, then it goes to the speakers.

My working title is usually the main modules and what goes on, so right now I just call it "Basal Laika Boom Boom".

[Aja]

I want to comment later on people’s tracks but for now I just wanted to say that I think we should narrow the focus and select a specific goal. Maybe something like “a steel guitar sound” is too specific (though personally I think that kind of thing would be fun), but at the very least I think we should all be working toward something that’s comparable to each other. So we could pick a specific mood, a specific technique (i.e. FM synthesis or, I dunno, self-oscillating filters) and stick to it. I think if we just say “post whatever synth tracks you want and stick them in a category” it might not be as interesting or educational.

So, that being said, here are some of my suggestions:

• a bell tone
• a monophonic drone
• a generative melody
• a krell patch (you’d have to look this one up)
• create airy sounds
• create cold sounds
• best impersonation of a piano
• synthesize a specific sound in the world — a car engine, a human scream, a bird song, the sound of water, etc.

[demagogue]

Those are really interesting ideas and I'm actually curious to try them out now that you mention them; and I see the issue with everyone just making random patches which make it hard to know what to even compare, if anything. Although our FM contests are like that, just larger scale art projects people judge in an overall way. So I don't think it's a major barrier either. I'm really not sure what's best, since they all sound like good ideas to me. Maybe other people can give their opinions too.

I wanted to post, speaking of synthesis techniques, because my mind was a little blown this week by what's called sine-wave speech. The idea is you can basically take a single sine wave, or sometimes 2 or 3, and just have it follow the envelope contour of a person talking and if you listen to it... Well you can hear for yourself if you watch this video about it. (The sine wave speech part starts from 1:24.) It's pretty crazy, and amazes me how our sense of hearing works and how synthesis can really play with it.

[PigLick]

I think the specific sound idea is a good one cos I just tend to twiddle knobs and such until I come up with something I like.

Just to add to Aja's list i think some more musical idea possibly
- best squirty saw wave 70's fusion lead tone
-bass tone of doom
-musical white noise

[Aja]

Okay, people seem to like the idea of synthesizing a specific sound, so let’s go with that. Why don’t we start off simple and go with the best bell sound? I’m open to other ideas, but a bell seems like something that’s simple to get a the basics down but difficult to nail.

[demagogue]

Well alright, let's get this party started then: https://soundcloud.com/user9513654/bell-sample

I'm hitting four consecutive notes (e5, f5, g5, and a5. The FM is really going crazy with the pitch. Outside of those notes it starts going haywire above and below, which this version shows after the first four notes: https://soundcloud.com/user9513654/b...le-other-notes. But those first few notes sounded pretty good so I stuck with it.)

Method (not entirely sure what's going on myself, but here's what I thought I was doing): I making this in Absynth. Two oscillators. I read that ringmod and FM are used to make metal clangs and bell like sounds, and I read that ringmods work like wave-packets, so I kind of ran with all three ideas to see what I came up with.

So the first oscillator is a wide sine wave-packet with only 3 smaller waves in the bigger one that I drew by hand (so kind of hand-made ringmod), transposed up 40 halfsteps. Then I turned ringmod on it which I believe mods it with the base sine wave for the note, which I guess double-ringmods it? Then the second oscillator is just a sine wave transposed up 22 halfsteps, but then I ran it through a filter that was set to an FM. (Vanilla FM mods a wave by another wave; I think the filter is letting the first wave pass by the 2nd wave. Anyway, the FM filter keeps a little more of the body of the original sine wave because I wanted some of that body resonance to come through and sound just a little clangy but not too much, and it did that job.)

And then each oscillator has its own envelope, both of which have attack at 100% right at the hit, and then the decay comes down concave, like a 1/x^2 skislope, sharp at the top but then leveling off a half-second in and coasting slowly down to zero after about 3 seconds. The decay curve is sharper for the higher pitched oscillator so it goes down pretty quickly, and it's flatter for the lower pitched oscillator so it rings out louder for a bit longer and falls off more gradually like a kind of faux resonance of the bell body.

[Aja]

That’s a decent sound! Nice and harmonically rich. I’d like to hear it with some reverb, too. I’ll try to patch one up tomorrow morning and post it.

[Aja]

Also, can you explain what a wave packet is? I’ve never heard that term.

[demagogue]

I tried reverb but then it sounded clearly artificial, and I wanted it to sound as realistic as I could. But I'll try to find a good reverb setting that sounds nice anyway.

A wave packet just means smaller waves inside a larger wave. When you do ringmod, the base wave is the wavelength of the larger guy, and the mod wave is the wavelength of the smaller guy. Then ringmod fits the smaller wave into the larger one as repeating wave packets.



Edit: "Wave packet" is the term they use in quantum physics for the probability amplitude wave of a particle that looks like this. It might not actually be the right term here; it's just what I wanted to call it for lack of a better term.

[PigLick]

so kinda like nested tuplets then?

[Aja]

Here's my entry.

https://soundcloud.com/dimedancing/j.../s-zcGBzcX3ujp

The Shapeshifter sends out a sine wave from its primary oscillator, which is then both frequency modulated and phase modulated by its second oscillator, also a sine wave (and which is set at a precise harmonic interval. Being able to quantize the second oscillator to harmonically useful FM intervals is one of the nice features of the Shapeshifter; it makes melodic FM much easier). That gets sent to a mixer along with five other sine waves of various related frequencies, and all of that gets sent into a low-pass gate, which is opened with an exponential-shaped envelope. It still sounded a bit harsh, so I sent the final mix into a low-pass filter, then on to the reverb and delay (for the second half of the clip), the latter of which is set to a very short frequency to add a little extra, I dunno, depth.

[Aja]

Edit: "Wave packet" is the term they use in quantum physics for the probability amplitude wave of a particle that looks like this. It might not actually be the right term here; it's just what I wanted to call it for lack of a better term.

Right, that makes sense. I don't fully understand how ring modulation works, but in practical terms I think of it as the sound that results from basically running a tremolo at audio rates (which is kind of what your diagram shows, a waveform varying in amplitude).

[Jason Moyer]

I rotate synths in and out constantly, so the only hardware I have out ATM to try this is my Poly D. This is just VCO's 1/2/3 on triangle waves at different pitches and VCO4 frequency modulating them at audio rates with a thin pulse wave.

https://soundcloud.com/jason_moyer/poly-d-bell

[Aja]

You sold your Voltage Research Lab?

[Jason Moyer]

Nah, I just currently live in a place where I can't keep most of my gear out. Partially because of space, but mostly because it's facing a major road and has a serious dust problem.