Not sure if this is a bug or feature request … it seems like the Breath Controller and Mod Wheel CC can only be mapped to FX LFOs and not Voice LFOs
1 Like
May have already been reported, but if you change a bus from a midi machine to shaper, it carries on doing midi stuff in the background.
lfo fx bug ?
if i assign the lfo to a parameter on an fx and then bypass (remove) the fx , the fx lfo is sill assigned to the parameter when you return to the lfo page.
expected - it might go back to none ??
eg - assign lfo to time division on daily delay , assign lfo , then remove daisy delay.
1 Like
LFO shapes phase 0 is at a transition point.
To test, set LFO speed super slow like -1, TRIG mode, phase 0. SAW. Controlling a grainer POS (pos 60, depth 60) so you can visualise it on the grainer display.
Expected behaviour is that it would sweep up from the start of the sample to the end and then reset to the start.
However it start at the end of the sample and moves backwards before starting the intended forwards playing but it will of course be out of sync. I couldn’t find a good starting phase that starts at the minimum value. With a SAW wave I would expect it to never go backwards.
RMP Half can work as expected however it does not loop.
Even if you remove trigs ?
Midi lfos too ?
Can’t test atm but should be reported…
IIRC depth must be 64, not sure…
Range is 0-120, so starting at 60, ±60 will give the full range. But that’s not the issue. (Note: I’m wrong here!)
The issue is that SAW and some other waves like EXP do not start at their most positive or negative value. So if you want a forward looping sweep of the wave you’ll encounter it going backwards at the start.
With these settings I’d expect to see it start at the beginning of the sample and move forwards through the sample and then reset to the start.
Interestingly with depth negative and speed positive, the behaviour is a bit different and it skips the start of the sample and then reverses at the end.
Warble fx bug
Noise can still be heard when fx set to full dry
Muting buss has no impact on it
Expected - noise can’t be heard when full dry .
1 Like
Nope. I use -64 here :
Tonverk: Tips & Tricks - #443 by sezare56
Why is that?
60 - 64 is -4, is that intentional to start at POS=-4 instead of POS=0?
Won’t that result in the sync being off?
Or does grainer say the range is 120 but its actually 127?
Even with -64 you can see the wave play backwards at the loop point. So this isn’t a fix for the issue.
(But I agree it sounds better with -64 depth than -60 depth for timestretching, I just don’t know why that works)
This makes me more convinced there is a bug here.
Further Investigation:
@sezare56 you are totally correct:
It seems the 0-120 range for parameters like POS are mapped to 0-127 in the LFO so LFO depth of -64…64 is required to access the full range. This is super confusing and misleading, not sure if this is a bug or intended but what’s the point of those 0-120 numbers if the LFO depth doesn’t use them?
This is much easier to see in Wavefinder with A.POS=0, POS=60, depth=-60 does not reach the beginning of the wavetable, but -64 does.
IMO if a parameter’s numeric range is 0-120 the LFO depth needed to move it from the center (60) to the extents should be ±60 not ±64.
In any case the issue with SAW at slow speeds remains (worse with SPD < 0 since you get the interpolation at the start, SPD > 0 is fine until it loops).
There’s no way to get a slow looping LFO that moves in a continuous upwards or down direction immediately snapping back to the start.
I can only guess at exactly how it works but I think the LFO should be interpolated if required after sampling it instead of baked into the LFO shape.
That way at high LFO speeds if the previous and next sample at on either side of the SAW discontinuity it will be smoothed out, but at low speeds you’ll never get a weird value output.
If you don’t care about looping you can workaround it by using RMP+HALF mode instead which doesn’t have this issue.
This was definitely unexpected behaviour for me, I can see it making a bit of sense since you can just adjust the noise level as well, but it’s inconsistent with Dirtshaper whose mix adjusts the noise level.
Notes were coming out but not sure if I’d changed to shaper at that point and/or cleared the sequence. LFOs definitely kept running though.
Will report.
1 Like
As stated in the manual:
N.LEV Noise Level sets the amount of noise that is added to the effect. Please note that the noise level is not affected by the MIX setting.
That’s noise ring mode there, not just static noise as in Warble. A different thing.
1 Like
Follow up to investigations into SAW LFO weirdness on Tovnerk.
Here’s a 1 Octave sine oscillator pitch sweep via SAW LFO on both Tonverk wavefinder (top) and DN2 wavetone (bottom).
Ignore the noise on as DN2 is recorded via the TV’s AB input and TV is going via USB to the computer.
Identical LFO settings:
Speed: -8.00 x1 BPM
Wave: SAW
Phase: 0
Mode: TRIG
Depth: 12.00
You can see the TV starts at phase 0 with a sweep down, then rises gradually to the peak before sloping down again at the loop point.
DN2 starts at the lowest point, then rises gradually to the peak then jumping back down to the lowest point.
The DN2 behaviour is what I’d expect from any SAW LFO.
EXP has the same issue:
RMP:
SQR:
RND is fine though, I guess it’s the only one with a slew control.
TRI and SIN are fine of course.
4 Likes
TV’s LFOs apply a short smoothing between jumps. I don’t get it (trying to be more like analog modu behaviour…?), but it’s by design. Can lead to weird constantly falling/rising values at certain settings.
1 Like
I see it happen, anywhere you can reference that it’s by design?
Doesn’t seem very useful to me and the inconsistency is troubling
If there was a slew control then sure that’d be super useful, but as a default for things like SAW and EXP and SQR it seems problematic.
3 Likes
I had a ‘sample not found’ error pop up yesterday ( although I can’t recall deleting any samples )
I was surprised to get a ‘do you want to relocate sample’ ( or something like that ) prompt … I thought I’d seen a discussion about it not being there ? … maybe it was something else / related to a different machine type .
It was on a subtrack … I rebooted and it seemed ok afterwards .
Can be a VERY slow slew !
It depends on speed.
Example with a square, trig lfo, negative speed. The lfo should start from low value.
Definetely a problem. Good for kicks, though !
I didn’t face it before in my use case.
Yep if the smoothing was the same amount of time regardless of the LFO speed it would probably be fine. But since the LFO going slower drags it out longer and longer it’s quite an issue for slow LFOs.
I suggested in my bug report making FUNC+SPH control SLEW for all waveforms similar to Filter Env Delay, that way it’s user controllable instead of being stuck with it and would be generally useful when you want slewed LFOs.
1 Like
In a previous bug report about it