Evolver Ring Mod

[Photosynthesis]

Hey does anyone know exactly what is going on behind the scenes with the evolver ring mod?    I have noticed that ring mod 4>3 and Ring mod 3>4 don’t always produce the same sound. 

When you have both digital oscillators set to the same wave and the same frequency they do or might sound the same.    However, whenever you have them set to different waves and especially when you have them set to different frequencies, maybe with an octave or more difference between, the ring mod sounds different between the 2.   I always make sure to have the filter split turned off as this would influence the sound coming out of the left and right channels.

I’m not an expert on circuits but the way I understand a ring mod it takes two signals and multiplies them together. a times b should be equivalent to b times a. So why do the 2 ring mods sound different at times?   

Before I thought maybe my evolver was broken or needed service.  This week I just got a 2nd evolver and it exhibits the same behavior sometimes. 

Could anyone confirm for me that their evolver also does this? I could upload a patch or provide settings later for someone to test if needed.  But it’s not hard to recreate if you have the digi oscs set to different waves and a different frequency.  Again no other modulation or filter split that might affect the output.

My only guess or explanation is that behind the scenes it’s not actually taking the signal from osc 3 and osc 4 and multiplying them together.   I was thinking maybe there is an internal carrier or modulator wave that is actually ring mod-ing one of the digi oscs.  Let’s just say hypothetically it’s a sine wave at 100hz.   This could  explain why when you do ring mod 4>3 it might sound different than ring mod 3>4.   
(Because 100 x C1 frequency would be different than 100 x D2 frequency.)

This is something that has puzzled me for a while.  I have tried contacting sequential but the lady i spoke with didn’t actually know herself.   She seemed surprised like she had never heard this before or had ran into it.

Also, I just want to say that it’s sometimes it’s not just  a slight difference either.  Sometimes we’re talking 2 completely different tones.  I know that the manual says there could be slight differences between the left and right channels because of the 2 independent analog filter and there potentially being slight variances in tone or output,  even without engaging filter split. 

[Soundquest]

I'm not an expert on ring mod, but just taking a stab.  The way I understood it worked was not necessarily the same as two integers multiplied....  3x6=18 or 6x3=18, rather the order of the application matters if other variables exist. 

  Lets say osc 4 wave amplitude is greater than osc 3 and its a sinewave on 4, with a saw wave on 3, that will be seen in oscilloscope as some iteration of the two.    When doing reverse setting (making 3 the sine wave and 4 the saw) keeping same amplitudes,  the resulting oscilloscope form will look different before the ring is even applied,  so maybe its just the amount of ring variance applied you are seeing. 

[chysn]

I haven't had an Evolver for a few years. But with an analog ring modulator (and presumably a digital one), it doesn't matter which oscillator is the carrier and which one is the modulator, the sound winds up the same. Ring mod, like multiplication, should be commutative, like you said.

Are you sure the oscillator levels are at 0, so that you're only hearing the ring modulated signal? If either Osc 3 or Osc 4 are sneaking in there, that could explain your difference.

[Soundquest]

I got a chance to play with this- I used a sine and a saw wave, one offset on frequency slightly.  As you indicated chysn, with both levels adjusted carefully the same, the order of ringing 3 to 4 or 4 to 3 made no difference that I could tell.  Of course, there was all sorts of differences when one, or both, of the feed levels were changed so that levels would be at relatively different amounts.  Then, the inevitable PEK progression happened.....I couldn't turn the darn thing off for the next 3 hours

[Photosynthesis]

Sorry for responding so late. I double checked and triple checked to make sure I didn’t have the other digital oscillator turned up.  No filter split either.

 Like I think I mentioned in my initial post I have 2 desktop evolvers and they both exhibit this behavior. So I can eliminate the possibility that my evolver is malfunctioning or out of spec I think.     

You should try testing it with the digital oscillators set to different octaves.   For example try  oscillator 3 set to the lowest octave and oscillator 4 set to the highest octave. Then just try it with some different waveshapes or frequency combinations. It is most noticeable when there are big differences.  Unless it is different in the polyevolver, then you will see that what I’m saying is most certainly true.

When it is also  apparent is when you have some sort of modulation applied to one or both of the oscillators.   For example envelope 3 routed to oscillator 3 frequency or oscillator 4 > oscillator 3 FM.   Even tho it’s not explicitly stated in the manual I believe,  the Ring mod is applied/calculated after FM.   But even if you have FM to one or even both oscillators in theory the ring mod should be the same between the 2 channels  because it comes back to an A x B = B x A thing.   



I can upload some sysex files later. 

[Photosynthesis]

I should add that this isn’t always a bad thing. In fact I think it’s kinda cool at times and is one of those “it’s not a bug it’s a feature” things.  I was only asking here because I was hoping someone might know more as to why or what’s going on behind the scenes so to speak