Non-Cyclical LFO/Sequencer??

E

eric_

Member
I'm wondering is there a way to have a controller randomly run and modify a parameter? For example, let's say I want the depth of modulation on a delay to go from 0% to 10%. I can obviously assign an LFO to this but that LFO cycle will run constantly and cyclically. I can change the LFO wave shape to random but it will still be cyclical. If I want the modulation depth to remain unchanged for a certain longer period, and then change quickly to 10% and back (even as a sine wave), is there a way to do that? For example, let's say I want the modulation depth to remain unchanged for "x" number of beats followed by a quick change to 10% over just one beat. The other factor is, using that example, I would want "x" to also vary (maybe 4 beats, maybe 12 beats) to give it a sense of randomness.

The sequencer could be another option where I have the change occur on one or two steps leaving the rest as normal, however that will still run cyclically. Is there a genuinely random way to do this? Of course, I could manually manipulate this "randomly" however I'd rather not have control of it. Appreciate any ideas...
 
the random wave shape is cyclical? or are you saying you just don't want a defined "tempo" of change?

can you put another LFO on the Tempo parameter in the LFO settings? i'm not sure, just a thought.
 
chris said:
the random wave shape is cyclical? or are you saying you just don't want a defined "tempo" of change?

can you put another LFO on the Tempo parameter in the LFO settings? i'm not sure, just a thought.
Click to expand...
Sorry I should clarify. The random wave shape is not cyclical but it is either always running or it is not. It is not random in the sense of when it runs and when it stops. Does that make sense?

Yes - I suppose I do not want a defined tempo of change. More ideal would be that it randomly switches from no change to a random tempo of change. However even a set, defined tempo of change, and a predictable wave shape, would be fine if the trigger for the change was random and allowed for longer periods of no change.

Apologies if I'm not explaining this well!
 
why not use one random lfo that has it's depth modulated by another random lfo? you can use the steps parameter set to 2 on the second lfo to restrict it's behaviour to "max" and "min" only
 
simeon said:
why not use one random lfo that has it's depth modulated by another random lfo? you can use the steps parameter set to 2 on the second lfo to restrict it's behaviour to "max" and "min" only
Click to expand...
I'm just trying this now but can't seem to find the steps parameter? I can see "Min", "Max", "Start", "Slope" etc but no steps parameter?
 
sorry, it's "quantize", not steps. it's on this page...

attach lfo 1 to the parameter you want to change and then link lfo 2 to the depth control of lfo 1. set quantize for lfo 2 to 2

Screenshot 2020-04-17 at 12.37.27.jpg
 
I'd use the modifier curve set so only extreme values of the random LFO are expressed upon the depth parameter.
To set the "threshold" adjust the value for MID.

FM3-Edit 2020-04-17 08-03-04.png

For a softer "knee" blend between the settings above and those shown below.

FM3-Edit 2020-04-17 08-05-00.png
 
simeon said:
sorry, it's "quantize", not steps. it's on this page...

attach lfo 1 to the parameter you want to change and then link lfo 2 to the depth control of lfo 1. set quantize for lfo 2 to 2

View attachment 66428
Click to expand...
I've tried this and it's getting there but I'm pretty sure I'm misunderstanding something. I've spent the last while reading the manual on modifiers and controllers and I'm still not sure. I'll lay out a real example I'm try to achieve and that might make it clearer.

So one use for this would to be have the pitch of the guitar randomly shift between -50c and 0. I select the Detune effect type from the Pitch block, set Voice 1 to -50, reduce the level of Voice 2 to 0%, and turn the overall mix up to 100%. I can now hear my guitar detuned by -50 cents.

Now I add a modifier to the Voice 1 Detune, select source LFO1, start off with a sine wave, and select Min to -50, Max to 0. I can now hear the guitar signal uniformly shift from 0 down to -50 cents cyclically (at a given rate - let's say 0.800hz). If I then change the waveform from sine to random I can now hear the guitar signal randomly shift between detune values 0 and -50 (and anywhere in between) at the same given rate. If I set Quantize = 2 on this LFO1 (not LFO2), I can hear the guitar signal randomly shift between detune values of either 0 or -50 (but nothing in between). Given that the waveform is random the likelihood of the detune values being -50 or 0 at any given time are equal (in other words the pitch should detune by -50 half the time and in the half the time).

So getting close but not quite there. This is where the confusion comes in...
Let's say I leave the LFO1 Quantize parameter set to 2 and have LFO1 depth set to 100%. I'm still hearing the guitar signal randomly shift between detune values of either 0 or -50 (but nothing in between). Let's call this "Setting A". Now let's say I run LFO1 Quantize off, and add LFO2 as a controller for LFO1 depth. I have LFO2 set the same as LFO1 (random waveform, depth = 100% etc.). Let's call this "Setting B". Are Setting A and Setting B not essentially the exact same?

Thanks for putting the time in on this!
 
eric_ said:
Are Setting A and Setting B not essentially the exact same?
Click to expand...

do they sound exactly the same?

if lfo2 is controlling the depth of lfo1 and it's set to a different rate, the pitch shift will only occur when lfo2 is at max AND lfo1 is at -50.

it might help if you tell us what you're actually trying to achieve and we can tell you how to achieve it. it might not use this method.
 
Eric:

I think this conversation would be much better if you just described EXACTLY what you want to do.
Don't just describe the technique you're hoping to use -- describe the effect you hope to create and its role in the performance or song.

(Edit: Ha. Simeon just said the same thing while I was typing this!)
 
simeon said:
do they sound exactly the same?

if lfo2 is controlling the depth of lfo1 and it's set to a different rate, the pitch shift will only occur when lfo2 is at max AND lfo1 is at -50.

it might help if you tell us what you're actually trying to achieve and we can tell you how to achieve it. it might not use this method.
Click to expand...
Admin M@ said:
Eric:

I think this conversation would be much better if you just described EXACTLY what you want to do.
Don't just describe the technique you're hoping to use -- describe the effect you hope to create and its role in the performance or song.

(Edit: Ha. Simeon just said the same thing while I was typing this!)
Click to expand...

Thanks for the perseverance!
There are a couple of different applications. One is that I'd like to have, say, one in every ten delay repeats modulate, where nine have no modulation whatsoever. An additional element to this would be to vary this number somewhat randomly (i.e. maybe two in every ten repeats modulate, followed by none in ten, followed by one in ten etc.). I don't want it to be predictable. Or another way to think about it is having a 10% chance of modulation on any given delay repeat.

Likewise with the example above using detune; if the guitar could randomly detune by -50c for one in every ten beats but in an unpredictable way.
 
Can you explain these things in a musical context? Maybe record some audio and do it by hand so we can hear how it's supposed to sound?
 
i think what you're asking for is a specific type of modulation that's random, but within set limits. kind of like a euclidian lfo. there are eurorack modules that do this kind of stuff. i think what you're looking for is do-able, by using more than one modulation source at a time and using one to modulate another. imagine using a random lfo to run a short sequence, where one step is at max and all the others are at min, etc.
 
simeon said:
i think what you're asking for is a specific type of modulation that's random, but within set limits. kind of like a euclidian lfo. there are eurorack modules that do this kind of stuff.
Click to expand...
I've never heard of this. Will read up on it for sure.

simeon said:
i think what you're looking for is do-able, by using more than one modulation source at a time and using one to modulate another.
Click to expand...
Can you elaborate on this? Using one modulation source/controller to modulate another is something I'm still trying to get my head around.

simeon said:
imagine using a random lfo to run a short sequence, where one step is at max and all the others are at min, etc.
Click to expand...
Exactly. Is it possible to have the steps run randomly so it just doesn't run the same pattern over and over?
 
eric_ said:
Exactly. Is it possible to have the steps run randomly so it just doesn't run the same pattern over and over?
Click to expand...

no, but if you run it from a random lfo, then it will run randomly. not much different from a random lfo quantized to 2 steps, but the max step can be any length you want (depending on the rate of the sequencer). experiment!
 
simeon said:
no, but if you run it from a random lfo, then it will run randomly. not much different from a random lfo quantized to 2 steps, but the max step can be any length you want (depending on the rate of the sequencer). experiment!
Click to expand...
Perfect. I'll try it out and let you know. Thanks again.
 
Are you wanting only 2 possible values from the final target parameter in every case? It sounds like that above in post #9, but I'm wondering if you also want a setup that would allow various random values within some range for other things.

Does the duration of the "on" state (like detune at -50 cents) matter? If you want them to all be equal, that makes certain methods of generating this type of randomness unusable.
 
Bakerman said:
Are you wanting only 2 possible values from the final target parameter in every case? It sounds like that above in post #9, but I'm wondering if you also want a setup that would allow various random values within some range for other things.
Click to expand...

Not in every case no, although that is what I've been looking at so far. Allowing various random values within a range would be useful too.

Bakerman said:
Does the duration of the "on" state (like detune at -50 cents) matter? If you want them to all be equal, that makes certain methods of generating this type of randomness unusable.
Click to expand...

Yes I'd prefer to have longer periods of "off" states broken up by briefer periods of "on" states.
 
You must log in or register to reply here.
Back
Top Bottom