Nord User Forum

Hi - Is it just me, or are the NS3 oscillators in synth modes 5-8 (MixSin, MixTri, etc) slightly out of tune with the main oscillator? This is particularly noticeable when the Osc Ctrl is at about 50% with the sub oscillator 1 octave below the main oscillator - I get about 6dB of phase cancellation every 2-3 seconds. Makes the synth pretty much unusable for consistent bass sounds unless you make your own single cycle waveforms, which is quite a faff. Am I missing something to be able to make them perfectly in tune, or is this a 'feature'?

If they were out of phase, you probably wouldn't hear a regular beating, so it must be, like you say, one of the oscillators that is out of tune. Have you tried intialising the synth section before trying the mixing? That would make sure that there are no other modulations at play.
According to your comment they should be 0.5 to 0.33 Hz off. I wonder if that can be picked up by a spectral analyzer or a tuner.

Yes, in synth mode 5 to 8 they are slightly out of tune with the main oscillator. I think this is on purpose, to create a more "rich" sound, and I kind of prefer it that way... It's like with a real analog synth, where two oscillator probably would not be completely in tune either or -- worse -- have random phase shifts.

You mention the 2nd osc being one octave lower and getting phase cancellation. It is true that you get a subtle slow variation of the sound, and not just a static tone; it sort of varies between the normal and the sub oscillator -- but I don't get what sounds like 6 dB attenuation here (as expected). If you want "proper" phase cancellation, just set the 2nd osc to the same tone and put the mix at 50% and then you can hear it clearly fade in and out. This is of course only the case when the two waveforms are identical.

However, I would like to ask what you're trying to achieve?

I think the main idea with the xMix modes is to use the LFO or envelope to modulate the degree of mixing, and thus create some variation in the sound anyway. Setting it to a static mix with no modulation would also not be audible at all with two completely identical oscillators (when at the same frequency).

Hence, if you're trying to mix in a different waveform, it's maybe less of an issue, and the slight detuning creates a more rich sound.

If you want to mix two identical sounds like in your example above (a sine with a sine one octave lower), you could use mode 4 instead (detune) and just set it to sub (or any other interval you like) and put osc control (fine) to 0 and no modulation. This way you can mix two completely in-tune oscillators (of the same kind). But maybe you want to do something else also that is not possible in this mode?

If you end up liking slightly out-of-tone oscillators, use the unison mode instead to add more of that

Just some ideas to maybe help you create the sound you want

I don´t think this has to do with the crossfade. If you just blend A and B i bet it´s still there.

Thanks for confirming that it's not an issue with my board

I was trying to use modes 5-8 to create bass patches comprised of an oscillator + sub oscillator. It's true that you can use mode 4 to do that, but then you are restricted in two ways - you cannot freely select an independent waveshape for each oscillator (they have to be the same), and also (critically for bass patches) you cannot adjust the blend between oscillators.

According to my tuner, osc 2 is set about 2 cents higher than osc 1. I get that this is similar to how a real analogue synth would respond, and respect the design decision, but it would be nice to have the option to keep the oscillators perfectly in tune (and reap some benefits of being in the digital domain). I guess this is a consequence of the synth architecture - i.e. having separate modes for detune, sync, mix, etc.

I have worked around the issue by using softsynths to create a single cycle waveforms with the blend of waveshapes that I want, but would much rather to be able to use the interface of modes 5-8.

NordPH wrote:I don´t think this has to do with the crossfade. If you just blend A and B i bet it´s still there.

Correct, A and B oscillators are also slightly out of tune with each other -- for the same good reasons. If they were perfectly in tune but not phase aligned, you would risk having no sound at all, so it's better the way it is now. How much they are out of tune varies; just try to set up two identical sine waves on A and B and hit a key repeatedly and listen for the fading in and out (beating). I've had beating as fast as maybe 0.5 - 1 Hz up to 0.0x Hz (didn't want to wait for it).

Just like the analog equivalent

Anyway, I didn't mean to say that the xFade modes are something special; I was mainly comparing modes 4 vs 5-8 -- maybe it's more like mode 4 is something special, where 0 detuning really is 0.

BTW: In mode 4, you can even get a situation where the two oscillators are perfectly in tune and has destructive interference -- the superimposed wave disappears (i.e. you cannot hear it play) -- but I've only seen this if I go from e.g. +1 semitone to 0 semitones with the step dial while a tone is running.

cupachump wrote:I was trying to use modes 5-8 to create bass patches comprised of an oscillator + sub oscillator. It's true that you can use mode 4 to do that, but then you are restricted in two ways - you cannot freely select an independent waveshape for each oscillator (they have to be the same), and also (critically for bass patches) you cannot adjust the blend between oscillators.

...

According to my tuner, osc 2 is set about 2 cents higher than osc 1.

Good to hear you found a working solution! I agree that there are some things you cannot do in mode 4 -- adjusting the mix is one obvious issue, of course!

As for the detune in modes 5-8, I think it depends on what tone you play; possibly higher pitched notes have a faster beating. I'm not sure if it is the same from time to time; between A and B they are somewhat random, I think.