Torso T1 Algorithmic Sequencer

The Lounge Other Gear
1676

I understand your position. Maybe Syntakt would be a nice solution for your utilisation of T1. All in a box (drum and synth) with cool external sequencer.

Yes, when using T1 in conjunction with Elektron sequencers, it seems redondant. It is. I feel the same with AR and A4. Since i’m still more efficient for programming with the hands directly on Elektron, i sometimes asking me the same question.

But, i really see the point of keeping my T1.
I’m not keyboard player, and i really appreciate the very efficient and fast way of writing pattern variations with Torso. Lot of things to try and keep with Temp. Elektron sequencers can’t do that way as T1 for the research of evolutives chords or notes
Futhermore, being able to change lenght, number of trigs or division of all tracks or parts of them, so easely and fast, is a rare ability for sequencers. It give a fantastic and simple tool for live impro.
So t1 with AR and multitimbral synth is a very creative combo. Even t1, AR and A4 (that can give a 3 notes chords and a monosynth at the same time) is an nice and compact set.

T1 is also a powerfull scrapbook that can be recorded in other sequencers (even Elektron ones) to be developped and/or deepened.

I really think that Torso and Elektron sequencers is a productive association.

10 Likes
1677

anyone selling theirs yet? asking for a friend… :grin:

2 Likes
1678

T-1 people, help me see if I understand how the randomisation / variance works (from a perspective of not having a T-1 to try.)

Would it be somwhat accurate to say that the random modulation on each parameter is a randomly generated but initially FIXED 16 step sequence, and then the Random knob by itself turns up the degree or rate of slippage (or “variance”) for ALL parameters that have random modulation assigned?

So it’s like you can put a Turing Machine on each of the modulatable parameters (although the sequences are fixed at 16 steps?), and then they share one big knob (the Random knob) to introduce mutation.

Am I close?

1679

I’m not 100% sure but I believe it’s fixed for a few cycles, and then regenerated at an interval controlled by “rate” (alt function of the random knob)

1680

This sounds pretty close to me. To restate it:

  • Each parameter has a (hidden) sequence of values it can use to modulate itself.
  • The depth of this modulation sequence’s effect on the parameter is controlled per-parameter by holding [RANDOM] and turning that parameter’s knob. It defaults to 0, which of course means the parameter isn’t modulated by the sequence at all.
  • This modulation sequence of a parameter can be modified in a number of ways. It can have skew applied, it can be rotated, and it can be randomly mutated.
  • We do not have specific control over how the modulation sequences get mutated. All we can control is the chance of mutation happening. We do this globally by turning the [RANDOM] knob. At 0% there’s no chance of any mutation happening to any of the parameters’ modulation sequences. At 100% each step of each sequence will be mutated. In between, there’s that percent chance of any step of each modulation sequence being mutated.

Some things to note:

  • These modulation sequences of the parameters are mutated by [RANDOM]’s value. But they are not only useful for randomization. Though their values are hidden, with [RANDOM] set to 0% they are always the same and can be used for predictable effects. I seldom turn up [RANDOM], but modulate my parameters frequently.
  • The effects of modulating a parameter with its modulation sequence is different per-parameter. The list of effects can be found in section 7.10 of the manual. In particular, pay attention to how some parameters advance a step in their modulation sequence per-step, and some do it per-cycle.
  • Even when a step of a parameter’s modulation sequence is mutated, it’s not completely “rerolled” to some truly random value. Instead, like a Turing Machine, it kind of wanders; a little up or a little down.
14 Likes
1681

This is as good a description as I’ve heard, and closer than I’d get.

@ubiquitous, read that thoroughly, and you’ll get the gist. For an even more basic description, I think the idea of a ton of Turing machines all working on different parameters is applicable, but how they are controlled is somewhat global, if that makes sense.

The point, ultimately, is that it’s not just creating wildly random sequences and modulation all of the time, you can intentionally do that, but it’s not the default state. I would say that because of this, and what’s really my favorite feature of the T1, is that you have to listen to what’s happening, because you won’t always see it.

2 Likes
1682

Thanks, this has indeed helped to clarify and confirm; it seems the mental model I had was fairly close.

I see also as hinted at above that for some parameters a random sequence is stepped through on a Cycle basis and for others using a “Rate” which is a secondary function on the Random knob.

Do you ever feel like the random mod sequences being fixed at 16 steps is a bit of an arbitrary constraint given most other periods can be freely chosen, or does it not matter? I guess there are also Retriggers which will reset it.

1683

From my experimentations I think about it like:

Cycle-controlled parameters:
receive the -next- value of its modulation sequence at cycle start

Step-controlled parameters:
receive a -new- value of its modulation sequence at random-rate speed
So if the random-rate is 2x faster than the division-rate, the modulation sequence repeats 2x in one cycle
I guess that’s why by default the division-rate sets the random-rate accordingly when changed.

1 Like
1684

Now here is a nice setup “trick” which helped me to tweak ccs more immediately:
I use the bottom row as cc-tracks corresponding to each midi-track in the row above.
That way I do not need to switch between midi and cc on the same button.
Of course this is only an option if 8 tracks are enough for you.
Anyway, just wanted to share :wink:

20240506_0857221

12 Likes
1685

If anyone is using the T1 as a sequencer for instruments, etc. in Live, could someone confirm the following: the MIDI keyboard that is available when a Track is in FX mode only works when the Live transport is running (assuming the devices are synced).

It seems completely crazy to me, but it appears that the T1 will only send MIDI data from the keys when the transport is running.

Thanks in advance! :slight_smile:

Edit: Also, the default Velocity output value seems quite low.

1 Like
1686

Yes it has to be running for the keyboard to work. Maybe they will fix that. Problem is they already have promised so much more the s4 that I have no idea when that will happen.

1 Like
1687

Thanks for your reply, and confirming the crazy is not just in my head. (At least on this :smiley:).

I am really liking the t1 so far!

1 Like
1688

“The keyboard screen in FX Mode sending notes only while the sequencer is running” is a known bug that should be fixed in the upcoming firmware release (approx. June or so). Current latest 2.1.0-beta.7 includes the fix, if you wish to test it just use the T-1 Config Beta available on Torso website:

4 Likes
1689

Thank you very much! :slight_smile:

1 Like
1690

I was wondering if any of you t1 gurus have any tips for altering probability on a per step basis. I have a drum hit with some repeats and I cannot get it to play occasionally rather than all the time or not at all.

I thought that maybe since the keyboard issue I found was a bug, there might be one here also, but I am guessing it’s user error.

I am pressing Ctrl+ step to enter step edit mode, then Ctrl+Velocity to change the probability, but whatever settings I experiment with, whether orange or blue, left and right of the T, lead to either on or off, no occasional playing, as I would expect.

Thanks in advance for your help. Lots of helpful info in this thread.

1691

You could try using a few cycles of the same pattern and then slightly altering them based on the different kinds of steps you want to happen. Set modulation to run on the cycles parameter and it’ll randomly hop around through the different variations a. Not quite per-step probability, but I think the effect is close.

Othe roption is to use a midi filter of some sort (I use MidiHub or some software) and use a CC track assigned to some kind of probability module for that channel, and do it that way.

1 Like
1692

Probability is, like everything else on the T-1, deeply linked to concept of Random and the “hidden” sequences of modulation values it mutates (see my post above this one for a deeper explanation). In short, the probability of any given step playing won’t change unless the random sequence is changing. So TL;DR: turn Random up if you want to make your probability more random per step.

The more detailed explanation is something like this:

  • Imagine the probability parameter has a hidden sequence of 16 values between 0–100.
  • By holding ctrl+velocity you’re setting a threshold between 0–100. If the value of the random sequence at the current step is less than the threshold, the step is muted. If above, the note plays.

So let’s imagine we know what the contents of probability’s hidden sequence is. Imagine its 16 values are:

[30]  [70]  [38]  [42]  [21]  [86]  [80]  [54]
[99]  [40]  [39]  [76]  [72]  [22]  [29]  [18]

With a probability threshold of 30, steps 1, 5, 14, and 16 will be muted. If we turn the probability threshold up to 50, steps 1, 3, 4, 5, 10, 11, 14, 15, and 16 will be muted.

But, if that hidden sequence never changes, then those same steps will always be muted for a given probability threshold. It will, as you say, always be on or off, never changing. The thing that changes the hidden sequence is the random parameter. By default (all blue) the hidden sequences never change. In the specific case of probability, that will mean the same steps always play (or not).

With random turned all the way up (all pink), the hidden sequences shift their values per step and probability will behave something more like what you expect. For random values in between, each step has that much of a chance of changing each step.

19 Likes
1693

image
image500×701 99.7 KB

14 Likes
1694

I need a lie down. :laughing:

Eternal thanks for writing that explanation out. I read it again and some of it stuck; I will read it again and again and also experiment, so that I can learn it.

Again, thank you!!

2 Likes
1695

Still getting my head around these. A bit like Parts on the OT for me at this point: I’m ignoring them until I understand them.

(Although, hopefully not forever, like Parts on the OT, which I never ended up bothering with. :grimacing:)