#733

Also a slew is recommended to “wash out” the stepping of the readings.

1 Like
#734

I thought I’d share a scene that I’ve been working on to create a performance system for improvising three part canons. The general idea is to create musical “gestures” with a turn of the teletype param dial and then create and modify a canon using input from the TXi and incoming triggers. Here’s an example of it in use:

https://www.instagram.com/p/CKHMxpjhit0/?utm_source=ig_web_copy_link

The scene detects when the param knob is moved and starts recording the gesture into the first pattern of the tracker. (So basically, you give it a quick wiggle and the notes you hear loop in a way to reflect that movement.) Each voice then steps through the pattern a note at a time at a rate determined by the incoming triggers (so the canon can move at different rates in different voices). The notes produced are quantised to one of a set of scales or chords stored in the second and third patterns. Each row is a different scale or chord, and they are stored as a binary number in pattern 2 and an offset in semitones in pattern 3. I used my quantiser haiku for this purpose.

Which quantiser row is used is controlled by the fourth knob on the TXi, so you can quickly change the key or chord that the whole canon is playing. The first three knobs on the TXi transpose the line being played by the corresponding voice in the canon (but keep the voice quantised properly).

The result is something that is remarkably playable with just 5 knobs and incoming triggers (e.g. from Kria). It is generative and complex, but there is no randomness involved at all. With a bit of experimentation you can really play the thing. Anyway, here is the scene if anyone wants to have a go. Remember to populate at least the second column of the tracker with scales. (In the example video I use 137 for a minor triad, 1193 which adds the fourth and seventh, and 1705 which gives it a dorian flavour.)

#1
TR.P 1
J % + J 1 8
X + P J TI.PRM 1; $ 8
CV 1 N X

#2
TR.P 2
J % + J 1 8
X + P J TI.PRM 2; $ 8
CV 2 N X

#3
TR.P 3
J % + J 1 8
X + P J TI.PRM 3; $ 8
CV 3 N X

#6
Z PARAM
PN 0 T Z

#7
B RSH B 1
IF NZ B: BRK
X - X 12; B 2048

#8
IF EZ A: A 1
B BSET 0 % X 12; K 0
W EZ & B A: $ 7
L 0 11: K ? BGET B I I K
X + * / X 12 12 K
X + X PN 2 TI.PRM 4

#M
T % + 1 T 8
IF NE Z PARAM: $ 6
A PN 1 TI.PRM 4

#I
PARAM.SCALE 0 48
TI.PRM.MAP 1 0 24
TI.PRM.MAP 2 0 24
TI.PRM.MAP 3 0 24
TI.PRM.MAP 4 0 8
P.N 0; M 100
20 Likes
#735

Really really cool!

If you want to free up script 8, QT.B should happily accept your existing bit scales :wink:

1 Like
#736

Cool idea, thanks for sharing!

1 Like
#737

Yes indeed! I’m really overdue updating my teletype to try your cool new ops… :slight_smile:

EDIT: In fact, this might give me space to slip in some Turing machine code in there for some random melodies too. Thanks for the prompt, @desolationjones!

1 Like
#738

and now I’ve been pushed over the edge - I’m buying a teletype!

3 Likes
#739

Don’t forget to ask if you have any questions. It’s a deep (but rewarding!) rabbit hole.

Oh, and remember you can mount it upside down and flip the screen with DEVICE.FLIP if you find the cables getting in the way of the screen. :slight_smile:

1 Like
#740

Nice code, I don’t stop to learn new things…
Here’s a video I’ve done today with your code. (I’ve used Just Friends in synthesis mode instead of using cv outputs.)

1 Like
#741

Really fantastic! I’m super happy you found a use for this approach :-). Just friends is great for this kind of polyphony, and it’s quite easy to modify the scene to do six part canons that way.

Next I think I’m going to try other ways of creating the initial looping melody - maybe with a complex LFO into the IN and gradual morphing of the stored loop of voltages using weighted averaging over time. Will share some code for doing that if it goes somewhere interesting!

2 Likes
#742

It sounds promising!

Thanks for sharing!

1 Like
#749

@ioflow how has your TT jitter-emulator come along?

Relatedly: has anyone peaked at the code and broken down what jitter does in more concrete terms across the full range of the knob? Interested in creating a pretty direct emulation for this Kria companion gridops witchery…

#750

Hello I am trying your scene and the melody stay static when I stop moving param knob.
Is it normal flavor ?

#751

When you move the PARAM knob it should record those movements into the first 8 cells of the first pattern at the rate set by the metro (so as it’s set up here it’ll take just under a second to fill the cells). It keeps looping round and refilling the cells as long as you keep wiggling the PARAM knob and then automatically stop recording into the pattern when you stop moving PARAM. The actual notes are played out of this recorded pattern at the rate of the incoming triggers.

So, basically, I give the knob a wiggle for a second or so, then leave it alone and listen to the melody created by sending triggers into the three trigger inputs. If I want to change the melody, I give the knob another quick spin and listen again. I think of it as a musical “gesture”. If you wiggle the knob fast and for the full range of motion, you get a melody that moves around a lot. If you move it less and - say - at the bottom of the range, you get a slower and lower melody and so on. But it’s important to realise that the rate its recording the “gesture” is very fast, but the rate of the melody is likely to be much slower (depending on the rate of triggers coming in).

I usually send three streams of triggers into the three voices at different rates to make a canon in which each voice moves through the melody at a different speed. Then vary the dials on the TXi to shift each voice up or down the scale to create different harmonies etc. It’s also important to have the scales set in the second column of the tracker (it’ll default to just octaves, but putting in different numbers will give you different scales based on binary, which I talk about in the other thread I linked to - 137 is a good starting point, giving you minor triads.)

Hope this helps/makes some kind of sense! I really hope this can be of use to some people. I’ve been surprised how varied the music is that I can get out of this scene!

TL;DR - wiggle the PARAM knob fast (less than a second or so), and then leave it alone, put a number like 137 at the top of the second column of the tracker, and send triggers into the first three inputs on the teletype. (I should probably do an instruction video, but I find doing them far too much like work!)

1 Like
#752

Ok thank you a lot @SimonKirby for your clarifications.
I have learn a lot with it…
So if I understand well, in the insta video, when we heard notes variations, you are turning the knob.
For always have variations, I replace the knob from Txi with fader from my SweetSixteen that accept CV, so it’s cool, I can plug LFO to the fader and it’s morph the melody … really interesting.
I replace the .MAP by .SCALE, for not to have dead zones on the faders.
Also I adapted it to send notes to my Er-301 and Disting.
No I will extend the number of voices…
Very fun play today with your Scene.
Thanks for the help :blush::+1:

#753

For that video, I was turning two knobs:

  • the first on the TXi, which was transposing the looping melody of one of the voices up and down slowly while the other two voices stayed where they were

  • the last on the TXi, which changed the scale being quantised. To start with it’s just using the minor triad, then it gradually adds more notes.

So, I don’t touch the PARAM knob for that video, all the voices are playing the same looping pattern throughout (which I “recorded” into the pattern using the PARAM knob before the video).

If you come up with something you want to share, I’d love to hear it :slight_smile:

1 Like
#754

Ok with all this infos I understand the trick.
I have to develop some fun works with it :blush:
If I have some good results, I will share it with pleasure
Thanks again @SimonKirby

1 Like