LFO waveform question

Another newb question but I’m sure this will be useful for others starting out.

The best way I can describe the unexpected behaviour is visually via the video below.

I’ve honestly tried everything to stop it, but it seems to be inherent behaviour.

Basically, I expect an LFO wave to behave like this, using a triangle wave as an example:

\ / \ / \ / \ / \ …

But instead it does this:

/\ ___ /\ ___ /\ ___ …

I have know idea why this would be helpful or wanted in any use of LFO? If you want to start at a random position of a long loop on each trigger, you’d constantly come back to the start for a period of time equal to the time you hear random notes, just as one example.

Maybe I’m doing something wrong. I’ve looked at start phases, modes, depths and speed and varying combinations thereof with no success.

Video explainer: In the video when I tap the screen I’m showing how it waits for 1 loop, before playing the wave again. Driving me crazy this one!

Image: shows the LFO settings I’m using.

Many thanks :raised_hands:t3:

DT has bipolar LFOs by default and the value adds and then subtracts from the value of that parameter.

What you hear as rests are when the minimum value for the modulation destination is reached. In your snapshot, for example, the DT cannot go before the first sample. So the thing to do is to move the start in a little until the modulation sounds right.

4 Likes

For the lfo to scan the whole sample, place the sample start in the centre (60) and set the lfo depth to 60. Because the sample is 120 and the lfo is bipolar, you’ll now see the full sweep with no ‘pauses’ where the lfo goes past the min and max values.

3 Likes

Yup. bi polar LFO depths and speeds.
Also, looks like you want to match the start of the duty cycle of the LFO to the trig. So you need to have the LFO in TRIG mode, not free.

Enjoy

2 Likes

fixed title. this is user misunderstanding as explained above

2 Likes

@avantronica all good with that thanks.

1 Like

this is probably not right, it would generally be 64 on other boxes (not DT owner) i.e. half Max range

if you use a monopolar LFO or a linear AD Envelope you can get things going with start point 0

2 Likes

Max range is 120 on DT

2 Likes

yes for sample positions, but the mod depth is 128 vals - so 128 is spread over 120, thus depth being 64

if it’s higher than 7bit, it’ll still relate to 64.00 imho, but I’ll look at AR for comparison one day

1 Like

FYI @garf And @avantronica The sample length in SRC is 120, but the LFO depth and speed is -64/+63, start phase 0-127. Kinda confusing, why is the SRC length at 120 but everything else 127?

Still trying things but so far no juice. I’ll report back.

use depth 32.00 then

1 Like

the Elektron guys explained the rationale for that with the AR release - it makes more sense as the number is more musically divisible

The depths are so on are usually the full MIDI range , other parameters will vary in terms of the numbering

Some depths link nicely to the parameter it addresses, but I suspect you need half the Max depth to reach sample positions at each end if you start in the middle

3 Likes

@garf and @avantronica thanks a lot for your help!

So it seems that setting depth to 30, not 32, does a true loop without a rest. If I set to 32 it hangs for a few moments on either side.

For others searching for this answer in the future, here’s the end result in video:

Interesting. As a web designer I love factors of 16 as well when it comes to pixel measurement. Guess the same goes for music, 4/4 etc. Nice to know the reason.

Interesting - they do often tie in depths so they become mapped, I am a bit surprised becasue it means there is dead depth here - but it’s good to know - I definitely want to see if it’s the same on the AR

keep in mind though that your loop has a length, so a depth that is too high will always eclipse the sample range -the logic I applied related to the position of start, not the outcome - if nothing else it has you thinking along teh right lines

Damn, now it doesn’t like depth at 30. The loop starts and ends short of the samples start and end when set to 30. If I tweak it to a number that it likes, as it plays through, the start and end gets shorter and shorter, until it reaches the center and is stuck on the 50% mark of the sample. Exactly like a swinging a pendulum that eventually find its centre of gravity actually. :thinking:

Things I did but then reverted to how I had it, could I have messed something up doing this, and even so I brought them back to the way it was in the above video so it should have solved that. Also haven’t changed the sample.

  • Swing from 60 to 80
  • Pattern scale from 1x to 2x
  • Tune of the SRC
  • Master length

I’m also finding that it gets stuck in the center, then I have to sketch LFO mode to another type, then back to HOLD, stop, play again and it works. But then does this weird shrinking start/end point thing.

What’s the logic That I’m missing here? Could this be a bug?

Update: cleared all parameter pages and went back to that setting and now it seems to play nice again. I’d just love to know if I caused this weird pendulum effect with the LFO or if it was some bug. If I figure it out I’ll post it here.

Could have been the FADE parameter? If you have it set even slightly off centre (the “X” graphic), when using the TRIG mode, it’ll fade the LFO depth out/in over time.

1 Like
1 Like

Interesting. Might have been that. Tend. It to use it yet as I often find it stops effects so been waiting to learn how it works.