SSLS #2: Syntakt oddballs (sound design)

[Nils]

Hey fellow Syntaktians,

here’s a new Syntakt Science Lab Special. The focus of this thread is to highlight creative, unconventional and less obvious ways of doing sound design on the Syntakt. This topic came from a discussion in SSL #8 where @sbkboi posted a nice attempt on comb filtering.

I realize this is a very open-ended topic without a clear common denominator, and I’m not sure which direction this will take. But I’d like for it to be a source of sound design tips not related to particular machines (we’re covering them in the regular lab anyway). It may be about the FX track, the delay/reverb, tempo-related stuff, you name it.

The threshold for posting here is low, but please post the following:

• the general idea of your finding or tip
• a good description of how to achieve it
• Some reflections around its possible use and its limitations
• A recorded example always helps shed light and inspire

If this thread gets up to speed with lots of posts, I may update this first post with a table of contents to be able to find the main tips. As always, it’s OK (and encouraged) to discuss and pose questions.

Anything I should add? Let me know

[sdkboi]

I’ll allow myself to post a link to my attempt on comb filtering on Syntakt’s current firmware (1.30?) as mentioned by @Nils in the original post.
I’m very excited about this thread because I started to realize and enjoy the broader sonic potential of the ST, unfortuneately only years after I’ve bought the ST during its launch, with sincere thanks to all the awesome Syntakt Science Lab threads that help me “up my sound design game”.

Syntakt Science Lab #8 BD FM

My next BDFM contribution is less about the BDFM machine than it is about the project’s BPM and Delay FX settings. I was attempting to create a comb filter effect by having a high BPM setting in order to make the Delay FX “Delay Time” parameter get into comb filter territory (<20ms). I haven’t measured whether this is actually the case with my project, which uses a current setting of 234 BPM but the results sounded interesting to me ;p [image] The project is a single BDFM machine playing the …

[Nils]

Syntakt “wavetables”.

Well not really But by combining high bpm and high retrig speed, trigs are fired at audio rate, giving an impression of pitch. The sequencer acts as a master oscillator so to speak, and each trig acts as wave cycle. Changing the sound parameters gives the impression of scanning a wavetable. Each track can thus be thought of as a slave oscillator or divide-down oscillator.

The basic procedure:

• create a sequence running at 300 bpm with trigs on every step
• all trigs set to retrig on fastest division (1/80)
• add more trigs on more tracks, or vary trig speed in different sections of the pattern.
• Press play and start finding sweet spots by changing sound parameters.
• Add LFOs and other types of control for more variations

So the pitch can be controlled by varying the tempo, either manually or via song mode. Pitch is controlled by the retrig speed, with 1/80, 1/64, 1/40 etc giving different pitches. Several pitches can be added by usingmore than one track, but they will always consist of a very limited set of intervals (1/80, 1/64 etc).

The “wave” can be altered by applying LFOs to any variable sound parameter, or for example to amp attack (softening the edge). For example, LFO to pitch sounds a little like oscillator sync. It can also be altered manually (snd drastically) by scrolling through machines in the machine list.

It’s a lot of work sequencing something meaningful (it’s running at 300 bpm after all, and requires retrigs on all steps), but I guess if you’ve got a nice drone going you can simplify sequencing it by using the FX track as an overall amp envelope.

One more thing; by setting up preset trig speeds on the 4 trig modifiers, you can play these four buttons as a simple keyboard.

Here’s some random noodling, not very impressive but I think you get an idea of what it can do.

[Oxenholme]

@Nils

not very impressive

It sounds nothing like I’ve ever heard coming from the Syntakt, so that counts as impressive in my book.

Never mind the practicalities of it

[Nils]

I know what you mean It’s not very controllable, but it may be interesting either as sample fodder, or for further refinement within the Syntakt with the help of the FX track, reverb, delay, song mode (which I didn’t try) etc. My demo is pretty basic as I wanted to show the character of the basic technique.

[jeye_musik]

I can confirm the adventures in Syntakt wavetables also count as impressive in my book! It seems an ideal method for experimenting with less traditional forms. Musique concrete, or early Raster Noton, or Francisco Lopez. That sort of stuff, I’ll need to investigate that too.

[Oxenholme]

Sounds like a great way to create cool samples

[igors48]

Nice one @Nils !
Another idea - use microtimed TRIGs and tune the Quantization during play.

[Nils]

(I couldn’t sleep so here’s another one…)

Additive synthesis on the Syntakt

The idea is pretty simple: set up eight identical digital tracks tuned according to the natural harmonic series, and treat the eight tracks as a single multi-part oscillator. I used SY BITS for this, but it can be used with basically any machine. The advantage of using SY BITS is that it has 2 oscillators pr voice, giving a total of 16 harmonics.

The base frequency is set to 100Hz, and then the natural harmonics table follows with 200Hz, 300Hz, 400 Hz etc. I then had some help (source unknown ) calculating which notes to use for each oscillator, and how much cent deviation to achieve the correct frequencies.

In the demo below, the eight tracks are treated as a single sound at the beginning. This means using the same trig pattern for all tracks, and using Control-All to affect all tracks at the same time. At some point I start hitting the eight tracks differently, with different types of modulation and different trig patterns. I believe at this point the frequency intervals aren’t necessarily “correct” anymore, and this becomes a selection of random modulated sine waves. But it sounded pretty cool

Demo, starting off dry and boring, but I think it gets more interesting as modulation and effects are added.

EDIT Sorry for the wall of sound coming in at 2:45. Watch your ears…

Update. Here’s the table:

Harmonic #	Frequency (Hz)	Ratio	Nearest 12-TET note	Cent deviation (cents)	Syntakt track and oscillator
1	100.0	01:01	G2	+35.00	T1 - OSC1
2	200.0	02:01	G3	+35.00	T1 - OSC2
3	300.0	03:01	D4	+36.95	T2 - OSC1
4	400.0	04:01	G4	+35.00	T2 - OSC2
5	500.0	05:01	B4	+21.31	T3 - OSC1
6	600.0	06:01	D5	+36.95	T3 - OSC2
7	700.0	07:01	F5	+3.82	T4 - OSC1
8	800.0	08:01	G5	+35.00	T4 - OSC2
9	900.0	09:01	A5	+38.91	T5 - OSC1
10	1000.0	10:01	B5	+21.31	T5 - OSC2
11	1100.0	11:01	C#6	−13.69	T6 - OSC1
12	1200.0	12:01	D6	+36.95	T6 - OSC2
13	1300.0	13:01	E6	−24.48	T7 - OSC1
14	1400.0	14:01	F6	+3.82	T7 - OSC2
15	1500.0	15:01	F#6	+23.26	T8 - OSC1
16	1600.0	16:01	G6	+35.00	T8 - OSC2

Note: the “Cent deviation” column shows the exact offsets for each separate note value. In the SY BITS machine, the main TUNE parameter will offset both oscillators on that track, while DETUNE only affects oscillator 2. This means that for each oscillator 2 setting, I had to detract the TUNE value to get the right DETUNE value. If not, the main TUNE offset will be added to the second oscillator on that track twice.
Example: Harmonic #3: +36.95 cents (~37cents). Harmonic #4: 35-37 = -2 cents. Hope this makes sense, I’m not the best at explaining…

[sdkboi]

Waking up, finding and listening to another magical Syntakt creation by @Nils and it’ll be a great day!
Pure beauty! Who needs firmware updates if there is @Nils

edit: I hope you’ll have the chance to catch some sleep later!

[Nils]

@sdkboi That’s too kind! I do hope it can be of inspiration to some. I’m literally just scratching the surface here. There are a lot more things to explore. For example using other machines, shifting the base frequency, selecting only odd or only even harmonics, or just hand pick a set you find interesting and go from there. I also think both techniques have a great potential for ambient soundscapes. Imagine controlling the eight tracks with different slow moving amplitude envelopes, or filter envelopes.

Firmware upates...

Shhh, don’t say that out loud… we want firmware updates too!

[sdkboi]

@Nils

Shhh, don’t say that out loud… we want firmware updates too!

ok let me sound more demanding: long overdue!!

[Oxenholme]

Alloy Aviary.

Inspired by @Nils and his high speed ‘wavetables’ I started doing some random resonant filter and pitches at 250 bpm and 1/80 retrig with CY Alloy. Some delay time modulation as well. I’m slowing the pitch down right the end to 30 bpm so you can hear some of the ‘cymbal character’ come back in.

[sdkboi]

Excellent! I like these dripping/tripping birds

[jm2c]

Its a proverbial zoo. Nicely played sir

[Oxenholme]

Thanks! It reminded me of being in those big tropical greenhouses

[sdkboi]

Amp panning modulated at high frequencies introduces additional harmonics whose pitch is controlled with the LFO’s frequency and whose amplitude is controlled by the LFO’s depth. That’s what it sounded to me like at least It effectively created a second detuneable operator, which I used on a BD FM machine in order to create multiple drones. I bet there’s a name for it. I didn’t bother recording an example yet because I didn’t come up with something that sounded interesting for more than a few seconds.

[Oxenholme]

It’s actually a good trick for ‘stereoizing’ your hihats and/or giving them a bit of crunch depending on speed and depth

[sdkboi]

That’s how I’ve been using it too so far but never on tonal sounds at extreme settings, e.g. with a multiplicator of 2k yet. I’ll just record a few seconds later today with what I’ve got so far

[Nils]

@sdkboi If you’re modulating the VOL parameter it’s usually called audio-rate amplitude modulation. Applied to PAN I assume it’s the same, only applied across the stereo field. I agree it’s a nice and efficient way to introduce a stereo effect.

Here’s how amp mod works:
When the carrier and modulator both creep above 20Hz, two sidebands are introduced: the sum and the difference of the two original signals. It’s a simple but cool way of introducing more overtones.

An example:

• You have two sine waves at 200Hz and 600Hz.
• When you let one modulate the other, this will result in two sidebands at 400Hz (difference) and 800Hz (sum).

Amplitude modulation is different from ring modulation (aka balanced modulation) in the following way: Ring mod introduces the sum and difference as well, but cancels out the original signals. This means you will only hear the new sidebands, and the modulation effect is then usually more pronounced.