Ring Modulation...and more!

Discuss John Bowen Synths - Solaris
John Bowen
Site Admin
Posts: 1877
Joined: Sat Apr 07, 2007 3:00 am

Ring Modulation...and more!

Post by John Bowen »

With the news of either a re-release of the Elka Synthex, or an updated digital version called Synthex 2, I thought it was interesting that one of the features talked about was 'extended cross modulation’, including 2 Ring Modulators. It made me think that no one really knows about the two Solaris AM sections, which can be 2 Ring Modulators, or several other types of audio cross modulation algorithms, including Shift, Clip, and Absolute. (See page 41 of the Solaris User Guide viewtopic.php?f=11&t=889)

christeck
Posts: 83
Joined: Sun Sep 21, 2014 9:13 am

Re: Ring Modulation...and more!

Post by christeck »

I played a bit with the ring modulator, resulting in a couple of sounds, including »Poor Man's Wurly«. I didn't manage to create really outstanding stuff with it, though. In case someone knows of some tutorial resource (including audio material) posting I'm glad for any pointers.

ce

GraigDev
Posts: 1
Joined: Wed Jul 26, 2017 12:58 am

Re: Ring Modulation...and more!

Post by GraigDev »

christeck wrote:
Thu May 14, 2015 9:20 am
I played a bit with the ring modulator, resulting in a couple of sounds, »Poor Man's Wurly«. I didn't manage to create really outstanding stuff with it, though. In case someone knows of some tutorial resource (including audio material) posting I'm glad for any pointers.

ce
I know what you mean, I can make some basic stuff with this but I can't seem to make any that special.

stefanovic
Posts: 8
Joined: Thu Mar 01, 2018 1:49 pm

Re: Ring Modulation...and more!

Post by stefanovic »

Hello everyone, on the same topic(s) (Elka Synthex and Ring Mod):

as part of my self-taught sound synthesis class, I came across the description of patch 46 of the Elka Synthex ("Ring Mod", aka as the Laser Harp), by Paul Wiffen, borrowed here (https://www.youtube.com/watch?v=zaYkgY5_toE) and there (http://forum.vintagesynth.com/viewtopic ... 6&start=15):
LFO - Triangle Waveform routed to PW2. Depth 3. Frequency 3. Delay 0.
Tuning - OSC2 Sync On.
OSC 1 - 8' Octave. Transpose 0. Pulse Waveform. Pulse Width -1 (of -5 to +5). Volume 7.
OSC 2 - 2' Octave. Transpose 7 (of 0 to 12). Waveform OSC1 PWM + Ring Mod. Pulse Width -1. Volume 7.
Glide - OSC2 Glide. Speed 10. Amount 2.
Filter - Low Pass. Resonance 0. Frequency 5. Keyboard 10. Envelope 5.
Chorus setting - 1 (Lowest of three).
Filter Envelope - Attack 0. Decay 2. Sustain 2. Release 0.
Amp Envelope - Attack 0. Decay 5. Sustain 5. Release 3.
Release envelope ON.
I did obtain something (somewhat) close, but not exactly using the structure above, and I'd like to understand why. In particular, my osc 1 is not tuned to 8', and I'm not sure of the final mix (and whether both oscs should ever be part of the mix).

The manual of the Elka Synthex is here: http://www.synthmanuals.com/manuals/elk ... manual.pdf
With my understanding, here is my translation of this patch to Solaris:

OSC1 - pulse - 40% - coarse: -24
OSC2 - pulse - 42% - fine: +7 - sync: OSC1 (no glide, see below)
OSC2 mod1: OSC1 - 52% - shape (supposedly imitate 'Waveform OSC1 PWM')
OSC2 mod2: EG2 - with a ~1s decay and ~80 sustain (depending on the filter's current cutoff freq)
OSC3 mod3: LFO1 to shape, cannot remember the values
Mixer 1: well, I tried several combos, all including AM1, like OSC1 + AM1, OSC2 + AM1
with AM1: carrier: OSC2 - modulation: OSC1

OSC2 mod2 is supposed to imitate the glide control of the pitch on the Synthex: it seems to me that we can't use Solaris' glide since the glide reaches its next position relatively to the previous one, while Synthex's glide reaches the next position from a positive or negative pitch position, set by the 'amount' parameter.

As is, this patch sounds very dull and does not resemble the synthex at all. I can only obtain a closer result when OSC1 coarse in set to 0, which is odd. Also, The ring modulation does not produce that much of a different sound compared to bare OSC1 and OSC2, so I think I did not understand Ring Mod... The thing is, OSC2 is sync'ed to OSC1, plus it's also synced with the 'Waveform OSC1 PWM', and they are ring mod'ed, it seems to me that there are many (redundant?) correlation/dependencies between OSC1 and OSC2...

Best, Stéphane

(PS: and I know that this sound is not for the taste of everyone :-) https://www.gearslutz.com/board/the-moa ... hread.html)

John Bowen
Site Admin
Posts: 1877
Joined: Sat Apr 07, 2007 3:00 am

Re: Ring Modulation...and more!

Post by John Bowen »

The question I would have is, what is the Synthex’s Ring Mod section using for carrier and modulator, and what amounts? Also, what does it mean to say Osc2’s waveform is 'Osc1 PCM + Ring Mod'? What are the levels, and balances? You are showing an estimate for the Solaris of 52% modulation of the PW, but this is really an unknown.

I would assume, however, that the structure is very basic and simple. This would mean that the ring mod is taking the 2 oscillators with the waveform just as they are being processed, so this might mean the source for Osc2 into the ring mod is already the signal with Osc1 modulating the PW, however, it could also be prior to that happening. In the video of Jarre using it, I really don’t hear much of anything that is ‘ring mod’, but just mostly sync, and since Osc2 is sync’ed, I wouldn’t expect much would come from Ring Mod anyway. The manual doesn’t tell much about the way the signal paths are set.

Also, what is the Synthex Lowpass filter? 12 dB or 24dB? With gain compensation or not?

Overall, I’d say you are making good guesses as to what the settings might be for the Solaris, but again, there could be something different going on with internal settings, pulse widths, etc.

stefanovic
Posts: 8
Joined: Thu Mar 01, 2018 1:49 pm

Re: Ring Modulation...and more!

Post by stefanovic »

The question I would have is, what is the Synthex’s Ring Mod section using for carrier and modulator, and what amounts? Also, what does it mean to say Osc2’s waveform is 'Osc1 PCM + Ring Mod'? What are the levels, and balances? You are showing an estimate for the Solaris of 52% modulation of the PW, but this is really an unknown.
I don't know, that's part of the game :wink: I am not the author of the partial specification of the sound, and at first sight, the user manual and the service manual is not very informative on this. I have the exact same questions as yours. Actually, answering those questions would be very valuable for preservation purposes. At least I have confirmation that the basic structure I envisioned makes sense wrt to the description. I'm tweaking things around, but to no satisfying results so far (except when Osc1 is not shifted toward the low end).
I really don’t hear much of anything that is ‘ring mod’, but just mostly sync, and since Osc2 is sync’ed, I wouldn’t expect much would come from Ring Mod anyway.
Yes, that's what I suspected (or did not understand), that those multiple modulations might (might!) be redundant. The synthex is supposedly CEM-based, and multimode, and I have no information as to the amount of attenuation used for this patch.

Unfortunately I don't not have means to record the sound I came up with...

stefanovic
Posts: 8
Joined: Thu Mar 01, 2018 1:49 pm

Re: Ring Modulation...and more!

Post by stefanovic »

a quick note on my latest investigations:
- when I swap the octave of OSC1 and OSC2 (OSC1:2' i.e coarse:0 and OSC2:8' i.e. coarse:-24), the result is much closer to the original. Hypothesis: the description above might have erroneously swapped the OSC octaves?
- final mix: OSC1 and AM1, but AM1 sounds like a muffled version of OSC2, and does not add much to the sound compared to bare OSC2.

Of course, the best way to settle the contribution of Ring Modulation would be to disable it on a real Synthex and hear the result, I'm curious to learn it.
Anyway, I'm satisfied with the result so far, I'll come back to it later when my design and analysis skills have improved (if ever), and change everything lol.

Last tips:
- filter MM1, LP4 24dB, freq:27, res:0 (the Synthex is multimode, but the LP is 24dB only)
- EG2 decay: ~2s and linear

It's attached, for the curious.
Attachments
p15.Harp Sync Sweep.pre.zip
(1.18 KiB) Downloaded 16 times

polybonk
Posts: 15
Joined: Sun Jul 21, 2013 1:11 am

Re: Ring Modulation...and more!

Post by polybonk »

Hey John.

Just wanted to ask. What would I need to do to mimic the structure of the Cross Mod in my Jupiter 6?
Its stunningly different sounding from any other cross modulation I have heard. When done right you get endless worlds of shifting harmonic and textural content coming in and out.

Any idea on what it is doing that I could mimic with Solaris?

Thank you.

John Bowen
Site Admin
Posts: 1877
Joined: Sat Apr 07, 2007 3:00 am

Re: Ring Modulation...and more!

Post by John Bowen »

It depends on what and how the Jupiter 6 is processing what they call Cross Mod. I know in a number of earlier synths, this was just audio rate modulation of the frequency, but it could also be modulating the VCA, or even the Pulse Width.
Since the Solaris has several ways to modulate the amplitude (in addition to Ring, there’s several others), you should try using the mod Source of 1 osc to Linear FM destination of the 2nd osc, or try Exponential FM (Pitch in the Solaris mod paths) to see if either of those come close to what you want.

John Bowen
Site Admin
Posts: 1877
Joined: Sat Apr 07, 2007 3:00 am

Re: Ring Modulation...and more!

Post by John Bowen »

I also found this online:

"In the early VCO-based instruments, Roland used “Cross Mod” to mean frequency modulation (FM) of one oscillator by the other. The term is used this way in the Jupiter 8, Jupiter 6, and the MKS-80 module. These are the only synths that have this option."

polybonk
Posts: 15
Joined: Sun Jul 21, 2013 1:11 am

Re: Ring Modulation...and more!

Post by polybonk »

Hmm thanks for the info.
Jupiter 8 is definitely ordinary sounding FM. JP6 is utterly unique tho.
I did find online once someone said that JP6 cross mod was actually some kind of AM mod when I went asking in a forum years ago.

Its absolutely crazy what it can do. Like a wall of constantly evolving modular. I was hoping it was possible with the AM modulators in the Solaris.
I will email you an example of what I am talking about so you can hear how unusual it is. If anything could do that stuff it would be Solaris.

On most of my patches I mod the pitch and FM of osc 1 to 2 and 2 to 1 at the same time to create rich string like overtones.
Im often finding that the resolution of Solaris pitch destination is way too crude tho. Its like I need a lot more steps of resolution between 0 and 0.10

Is there a control solution that could use as an offset to get better resolution? Damp down the % somewhat?

polybonk
Posts: 15
Joined: Sun Jul 21, 2013 1:11 am

Re: Ring Modulation...and more!

Post by polybonk »

Actually I will use soundcloud to keep the question relevant to anyone reading.

Cross Modulation on a single Jupiter 6. Split upper and lower with keys in hold mode. Some delay used that creates the panning. Ping pong on the VS 880 HD recorder.

https://soundcloud.com/polybonk/in-the- ... sleep-edit

Everything that you can hear happening is all 2 JP6 parts deep in Crossmod with the keys set to hold a dual layer drone of 2 chords.

Nothing is being touched on the synth. Its playing itself.

John Bowen
Site Admin
Posts: 1877
Joined: Sat Apr 07, 2007 3:00 am

Re: Ring Modulation...and more!

Post by John Bowen »

Well, I’d need more information. What are the Splits made of? What waveshapes? To start it, you play 2 chords...are they one chord for each side of the split? Are they the same chords for each? Can you just isolate one note without any effects so I can hear what it sounds like?
I am completely unfamiliar with the Roland gear.

John Bowen
Site Admin
Posts: 1877
Joined: Sat Apr 07, 2007 3:00 am

Re: Ring Modulation...and more!

Post by John Bowen »

A quick lookup of the Jupiter 6 states that the splits limit one side to 4 voices and the other side to 2 voices. That means one of the sides isn’t playing a ‘chord’, but in any case, it would be good to get a single dry sound example to analyze. Perhaps you can record without a split a single sound using cross mod, and to play it up and down the keyboard. List the waveshapes involved, tuning ranges, etc., so we can better try to figure out things.

Another reference states: "Note that this FM is fed to the 1V/Oct CV input of the oscillator, so it is exponential FM, not linear FM.”

So, what we need to know:

1) what are the 2 waveshapes involved?
2) what is the tuning of both oscillators?
3) what is the depth of Exponential FM (what is the mod amount)?
4) what notes are you playing?

And I would like to know these points for both of the patches used in the split configuration.

polybonk
Posts: 15
Joined: Sun Jul 21, 2013 1:11 am

Re: Ring Modulation...and more!

Post by polybonk »

Thanks for looking into that.
This recording is from 2002 and just meant to convey how much evolution/harmonic tonal content that continues to come in and out.

Would have been 2 notes on upper and 4 on lower.

This might help
https://soundcloud.com/polybonk/jp6-xmo ... osc-2-sine

Here is middle c 2 saw waves and I turn up the crossmod. Then a second go Then Osc 1 Square Several Octaves Down Osc 2 Sine.
This really doesn't convey just how deep it gets when you use multiple notes and multiple waveforms all at the same time.
Then you can get worlds of shifting harmonic content.

Post Reply