can you have free-running oscillators in DIVA?
voice 1
2 osc
transient mode = reset
same reset phase
voice 1
2 osc
transient mode = reset
different reset phase, 0 vs 33
voice 1
2 osc
transient mode = analog (?)
same note
same note
same note
etc.
in analog mode, reset phase should be ignored yet oscillators keep previous phase relation (which remains constant)
is it right?
DIVA free running oscillators
-
- KVRian
- 814 posts since 18 May, 2007 from Berlin
The screenshots seem to indicate that there's absolutely zero detune set up for the oscillators. That means nothing will move the phases apart.
Try the same with some detune between the oscillators please.
Try the same with some detune between the oscillators please.
-
- KVRist
- Topic Starter
- 42 posts since 24 Jan, 2019
can't 2 oscillators be exactly in tune and have different phases?Viktor [TUC] wrote: Wed Jun 05, 2024 10:46 am The screenshots seem to indicate that there's absolutely zero detune set up for the oscillators. That means nothing will move the phases apart.
Try the same with some detune between the oscillators please.
-
- KVRian
- 814 posts since 18 May, 2007 from Berlin
You mean different phase positions per note?
Your screenshots show they have different phase positions in a single note, and that consecutive notes retain these. My assumption, you tested the "osc reset" mode first, played several notes. These played voices have all been phase-reset to the values of the corresponding phase knobs in the "reset phase" panel, now all these voices share the exact same phase relationship between the oscs.
When you switch to the analog mode, phases aren't reset anymore, they can go where they want, so far so good.
Why then do we see the same phase relationship with all consecutively played notes in these screenshots?
That's your question, I assume?
Well, what changes phase relationships between oscillators?
Pitch differences, only that. No pitch difference, no change in phase relationship.
(Or you randomize them on each note trigger.)
In this case, you need to add some detune somewhere, either in the top oscillator panel, or in the Trimmers section where you define offsets for the different voices. With everything at zero, you won't see changes in the phase relationship of the oscillators in each voice.
Viktor
Your screenshots show they have different phase positions in a single note, and that consecutive notes retain these. My assumption, you tested the "osc reset" mode first, played several notes. These played voices have all been phase-reset to the values of the corresponding phase knobs in the "reset phase" panel, now all these voices share the exact same phase relationship between the oscs.
When you switch to the analog mode, phases aren't reset anymore, they can go where they want, so far so good.
Why then do we see the same phase relationship with all consecutively played notes in these screenshots?
That's your question, I assume?
Well, what changes phase relationships between oscillators?
Pitch differences, only that. No pitch difference, no change in phase relationship.
(Or you randomize them on each note trigger.)
In this case, you need to add some detune somewhere, either in the top oscillator panel, or in the Trimmers section where you define offsets for the different voices. With everything at zero, you won't see changes in the phase relationship of the oscillators in each voice.
Viktor
-
- KVRist
- Topic Starter
- 42 posts since 24 Jan, 2019
i understand what you're saying (thanks) but i still think there's something weird about how DIVA threats oscillators phase, especially across different voices...
i'll post about it later
(suspense music playing)
i'll post about it later
(suspense music playing)
-
- KVRist
- Topic Starter
- 42 posts since 24 Jan, 2019
polyViktor [TUC] wrote: Wed Jun 05, 2024 11:06 am You mean different phase positions per note?
Your screenshots show they have different phase positions in a single note, and that consecutive notes retain these. My assumption, you tested the "osc reset" mode first, played several notes. These played voices have all been phase-reset to the values of the corresponding phase knobs in the "reset phase" panel, now all these voices share the exact same phase relationship between the oscs.
When you switch to the analog mode, phases aren't reset anymore, they can go where they want, so far so good.
Why then do we see the same phase relationship with all consecutively played notes in these screenshots?
That's your question, I assume?
Well, what changes phase relationships between oscillators?
Pitch differences, only that. No pitch difference, no change in phase relationship.
(Or you randomize them on each note trigger.)
In this case, you need to add some detune somewhere, either in the top oscillator panel, or in the Trimmers section where you define offsets for the different voices. With everything at zero, you won't see changes in the phase relationship of the oscillators in each voice.
Viktor
2 oscillators exactly in tune
transient mode = analog
all trimmers = 0 (no voices or oscillators variations)
voice 1, 2, 3, 4 are identical
voice 5 is different (even if similar)
voice 6 is different
voice 7 is different
voice 8 is different
this looks/sounds strange and certainly not very analog-like
You do not have the required permissions to view the files attached to this post.
-
- KVRian
- 814 posts since 18 May, 2007 from Berlin
If voices 1-4 are identical, then you've played those 4 voices with 'osc reset' selected. If the other voices aren't identical, then you have not played these voices with 'osc reset' selected. Only when you play a voice, that's when the settings are applied.
What do you mean by "not very analog-like"? Can you define that? When you use phase reset and also remove every little bit of detuning from the oscillators, making them perfectly in sync, no drift whatsoever, what's the analog behaviour you expect, can you elaborate?
What do you mean by "not very analog-like"? Can you define that? When you use phase reset and also remove every little bit of detuning from the oscillators, making them perfectly in sync, no drift whatsoever, what's the analog behaviour you expect, can you elaborate?
-
- Banned
- 198 posts since 18 May, 2024
In my opinion the opposite is the case. It seems impossible to stop the free-running and make it do a reset.
You can hear it when playing bass notes with a simple pulse wave.
In my view detuning has nothing to do with free-running.
You can hear it when playing bass notes with a simple pulse wave.
In my view detuning has nothing to do with free-running.
-
- u-he
- 30206 posts since 8 Aug, 2002 from Berlin
I gave it a quick try and could not find anything odd.
Osc Reset works on each oscillator.
After oscillators are reset, if no detune is applied, all oscillators remain in same phase relation. Once detuned, phase relation at note start depends on the beating of the oscillators, and the timing of note starts in relation to the beating.
Osc Reset works on each oscillator.
After oscillators are reset, if no detune is applied, all oscillators remain in same phase relation. Once detuned, phase relation at note start depends on the beating of the oscillators, and the timing of note starts in relation to the beating.
-
- KVRian
- 814 posts since 18 May, 2007 from Berlin
Detuning the oscillators in the trimmer panel helps making the per-note differences obvious as it causes the oscillator phases to drift to drift apart, not remain in the same position relative to each other.
I'm getting perfect resets here if I so desire.Nicolau wrote: Wed Jun 05, 2024 2:17 pm In my opinion the opposite is the case. It seems impossible to stop the free-running and make it do a reset.
Reset all voice detune trimmer pots to zero to avoid differences between voices, use 'osc reset' mode. When I do that, all notes sound the same here. Or if I want to keep the voice detune settings, I use not 'poly' but 'mono' mode so that I trigger the same voice each time. Result here, each note sounds the same.
-
- KVRist
- Topic Starter
- 42 posts since 24 Jan, 2019
polyUrs wrote: Wed Jun 05, 2024 3:07 pm I gave it a quick try and could not find anything odd.
Osc Reset works on each oscillator.
After oscillators are reset, if no detune is applied, all oscillators remain in same phase relation. Once detuned, phase relation at note start depends on the beating of the oscillators, and the timing of note starts in relation to the beating.
2 osc exactly in tune
transient mode = osc reset
osc 1 reset phase = 0, osc 2 reset phase = 33
all trimmers = 0
voice 1 ONLY is played
TRANSIENT MODE IS CHANGED TO ANALOG
voice 2 is identical to voice 1
voice 3 is different
voice 4 is different
voice 5 is different
voice 6 is different
voice 7 is different
voice 8 is different
looks like all voices are affected by voice 1 initial phase reset parameters, even if they've all been played in analog mode only; shouldn't all voices waveforms be identical (and identical to voice 1 waveform)?
You do not have the required permissions to view the files attached to this post.
-
- KVRian
- 814 posts since 18 May, 2007 from Berlin
Maybe it's late and I can't follow, I'll have another look tomorrow.
But it does not look like all voices are affected, above screenshots 3 to 8 it says "different" too. But affected would mean they'd be the same, yet they're clearly not, or do you mean something else?
When I repeat the test here, it behaves as I'd expect. The one played voice gets the osc reset settings, and switching back to analog mode, all subsequent voices retain their previous phase settings.
In your above example, voice 2 looks the same as voice 1, but there's actually a chance that voice 2 already had a similar phase setting to the 'reset phase' values. Did you test that before, in analog mode? You can cycle through the voices once, check their phase relation, and after playing voice 8, switch to reset mode, play one voice (that would be voice 1), then back to analog mode.
In any case, I've done that experiment several times now, and each time, only the played voice or voices (in osc reset mode) received the phase reset values.
But it does not look like all voices are affected, above screenshots 3 to 8 it says "different" too. But affected would mean they'd be the same, yet they're clearly not, or do you mean something else?
When I repeat the test here, it behaves as I'd expect. The one played voice gets the osc reset settings, and switching back to analog mode, all subsequent voices retain their previous phase settings.
In your above example, voice 2 looks the same as voice 1, but there's actually a chance that voice 2 already had a similar phase setting to the 'reset phase' values. Did you test that before, in analog mode? You can cycle through the voices once, check their phase relation, and after playing voice 8, switch to reset mode, play one voice (that would be voice 1), then back to analog mode.
In any case, I've done that experiment several times now, and each time, only the played voice or voices (in osc reset mode) received the phase reset values.
-
- u-he
- 30206 posts since 8 Aug, 2002 from Berlin
I have no idea what you're measuring exactly, but the graph on the right looks different between Voice 1 and Voice 2. Similar, but different. If you put them on top of each other in Photoshop and use difference, it doesn't zero.
-
- u-he
- 30206 posts since 8 Aug, 2002 from Berlin
they are not. Phase reset affects voices when they are triggered. It does not affect voices that are not triggered during the time the setting is active.mentooool wrote: Wed Jun 05, 2024 8:49 pm looks like all voices are affected by voice 1 initial phase reset parameters,
At not time and in no setting does Voice 1 have an effect on any other voice.even if they've all been played in analog mode only;
I do not understand where that assumption comes from. Did I say something that could be interpreted that way (since you quoted me)?shouldn't all voices waveforms be identical (and identical to voice 1 waveform)?