#21

I haven’t tried Odessa… I wasn’t as compelled by demos of it. So, probably? :slight_smile:

1 Like
#22

Ensemble and Odessa have a disconcerting number of parallelisms in interface and occasionally even sound, yet I think the similarities are ultimately superficial, and these modules have very little to do with each other. But it’s weird.

1 Like
#23

Yeah, the concept seems similar, but the sounds from what I’ve heard are quite different. Get both, I guess.

#24

Or neither. You‘ve got to want what they do. They are each utterly unique, which is exciting, but not necessarily equally constructive for everyone.

Odessa is a true additive synthesis oscillator that emits a sine wave root and then some (potentially huge) number of harmonic partials above (always above) the root. The interface is dedicated to FM and a few macro controls that change the distribution and levels of the partials. Alone and droning, it produces some signature sounds (that can get real old, real fast), but it’s envisioned as a ‘basic oscillator’ to be used with filters and other waveshapers, not as a standalone voice. I’ve spent time with Odessa, and a friend has one and likes it; I respect it, but it’s probably not for me, at least not right now.

Ensemble Oscillator, as the name implies, is a virtual oscillator bank with a clever interface scheme for managing and manipulating the oscillators relative to each other. Like Odessa, sine waves are at the core, but the EnOsc is not capable of the harmonic partials of Odessa (a divergent infinite series); rather, EnOsc distributes its voices over a repeating pattern (not divergent or convergent), with hefty quantization features to make its behavior more useful. EnOsc can and should be used as a ‘basic oscillator’, too, but more than Odessa, has significant potential (IMO) as a somewhat standalone (droning) voice. For example, an animated droning EnOsc might be a very useful transitional element between two different pieces in a live performance: it’s interesting enough to fill that role in 16HP.

6 Likes
#25

Thank you. That is the clearest description of their differences that I have come across.

1 Like
#26

After some more time with ENOSC, I came to the realization that it’s really not an additive synthesis engine (though in very limited ways it can be used as such) – it’s a bank of waveshaping oscillators.

I suspect that each individual oscillator is a phase accumulator (rising ramp), which undergoes phase distortion (Twist) and phase modulation (Cross FM), a sine lookup, and then wavefolding (Warp). It’s then mixed with other oscillators – but if you keep the Balance knob fully CCW or fully CW, you can isolate one and work it in a fairly conventional way. It offers a wide variety of timbres with just one audible oscillator. (That’s what I was doing in my demo above.)

The real complexity comes when you’re hearing a whole bank of oscillators, and the myriad of FM relationships between them. For instance, if you’re using a major chord and Cross FM in “up” mode, you’ll a blend of integer FM ratios (on the octave, octave+fifth, etc.) and non-integer FM ratios (the thirds), and those ratios are different for every oscillator. Those 16 oscillators could be producing hundreds of sidebands. In “all” mode, each oscillator FMs the next highest – so with that major chord there are only 3 different FM ratios in play, but all of them are non-integer ratios and there are still several of them across the spectrum.

But if you choose the circle of fifths or Bohlen-Pierce major scales for instance, every ratio from one oscillator to the next is equal, so Cross FM in “all” mode produces sets of sidebands that align, while “up” or “down” modes produce a tangle of inharmonic complexity.

And adding to all this, there’s the doubling and crossfading of oscillators that happens when their frequency falls between quantized values on the scale. This effect can cause interesting shifts to happen with the FM as Spread or Root are modulated.

…

Aside: for what it’s worth, I believe (but can’t prove) that the FM modulation is always a sine, and “all” mode works from the top down rather than involving a serial chain of modulations. (The phase accumulator for the modulator probably runs through a separate sine lookup, without phase distortion or phase modulation.)

E.g. every FM mode is {1 unmodulated oscillator + 15x 2OP FM]. That’s complex enough… the alternative would be that All mode is [1 osc + 2OP + 3OP + 4OP + 5OP + 6OP + 7OP + 8OP + 9OP + 10OP + 11OP + 12OP + 13OP + 14OP + 15OP] and you might as well just use a white noise generator :laughing: And this is assuming no crossfading is happening…!

…

TLDR: set Balance to either extreme and ENOSC is a “simple” waveshaping oscillator with a diverse range of timbres. Set Balance somewhere in between but leave Cross FM at zero, and it’s a whole stack of waveshaping oscillators. But as soon as you involve Cross FM and multiple oscillators, things can get complicated in there… yet it still feels easy to turn knobs and get great results.

6 Likes
#27

I was very drawn to the ENOSC when it was announced, pre-ordered as soon as I could and it became the first dedicated oscillator in my system, but truthfully I didn’t immediately connect with it. I made this patch with the specific intention of getting used to it, and seeing if I could make something groove-oriented with it. I do love the sound of it and am excited to explore it further. (This is also my first time sharing any of my music here on Lines - so, hello!)

11 Likes
#28

I don’t know if anyone else has tried this, but ENOSC is an amazing modulating oscillator. I’ve been experimenting with Through Zero stuff the last month or so and I’m loving it paired with either SSF Zero Point or the Joranalogue oscillator.
Here’s a very rough video I made to illustrate this -

6 Likes
#29

I made an expanded version with better quality audio.

Here’s another example of ENOSC as a modulating oscillator. This time without Thru Zero. Still sounds great to my ears.

3 Likes
#30

Hello! I just got the Ensemble Oscillator last night and in my first patch there was magic straight out of the box, even before I read anything from the manual. Such an inspiring module! For me the best thing that 4ms have made so far. I hope you enjoy the video.

This is the finished track I’ve made with this patch: https://thanosfotiadis.bandcamp.com/track/uspicious-oon

5 Likes
#31

Here’s another video to show the Ensemble’s strength as a Modulating oscillator. Here it’s modulating the Dannysound through zero oscillator.

4 Likes
#32

After some rack shuffling I ended up with one of these and it has ended up being one of my favorite modules. My sweet spot is backing off the Balance to about 25% and then modulating Twist in Ramp mode. CS80 vibes when I plumb that in stereo into a Mimeophon. Those two together is the most fun I’ve had in a long time!

image

5 Likes
#33

I was playing with it again tonight and was astounded by some of the textures that can come out of it.

6 Likes
#34

It really is very fun, it just sounds lovely (although I am still figuring out how to incorporate it with other voices etc… although it is nice just by itself0.

1 Like
#35

When I first saw this module I was pretty blown away, but figured it had enough overlap with my Qubit Chord V2 that it wasn’t really worth looking into further (a strategy I use a lot to avoid GAS haha).

Red Means Recording posted some videos on this a few weeks ago that I’m just now getting around to watching and I’ve got to say I’m smitten again. I don’t have the funds or space to have both so… does anyone have direct experience with the Ensemble Oscillator vs the Chord V2?

Curious if it’s worth swapping out. For the record I LOVE the chord V2 for all the cool music theory tricks and interesting quantizing options. There’s so many interesting compositional things you can do with it just with some pretty simple modulation. But I don’t love the stock wavetables. I haven’t swapped them out though because I don’t really have access to any banks that I think I’d like better. They built in wavetables are solid, but I’m just not always feeling more than “okay” about the sounds.

#36

Is it possible to modulate the sequence/order/series of the harmonics? Am I correct in understanding the “Root” offset controls the ordering position of a harmonic degree, with modulation resulting in various inversions and permutations of the “scale” degrees? Or is this affected by the “Pitch” offset (if achievable at all)? The manual terminology is a bit conflatory. Hope this makes sense.

#37

The Pitch parameter transposes/tunes everything without quantization of any kind.

Root will set the lowest oscillator’s pitch within the scale (not the root of a scale, so it’s kind of misleading), and other oscillators follow from that according to Spread. So yes, it does inversions, more or less. It’s also subject to Balance and crossfading.

Of course if the “scale” is all equal intervals, there’s no real concept of inversion anyway.

1 Like
#38

Thanks. For clarification, the linear order of the interval series is fixed. Root will only modulate the interval from which the scale is ‘arpeggiated’. e.g. Let [0,1,3,4,6,7] be a scale. With modulation applied to Root, the series might proceed: [3,4,6,7,0,1] or [7,0,1,3,4,6] or [4,6,7,0,1,3]—rather than eliding, via permutation, order relations within the series (e.g. [4,1,7,3,0,6]). Yes?

#39

With a given scale, the order is fixed. I haven’t tried programming custom scales on it, but I assume they are always sorted, so that higher values of the Root parameter always ascend in frequency.

You could program a set of custom scales that repeats every 2 or 3 octaves rather than one, skip different steps in each, and then sequence or modulate the Scale parameter.

It seems like a lot of people try to think of it as a fancy quantizer, or as a sequencer or arpeggiator, or a polyphonic chord synth, and then struggle to figure out how to use it. I tried to think of it as a harmonic oscillator at first, but it’s not quite that either.

#40

Yeah, I’m understanding that construing it as any of the above would be far too reductive and detrimental to its ‘possible’ compositional applications. I don’t think they lean into the ‘complex waveshaper’ aspect of its design enough, mainly because it’s literally not polyphonic (from which much of the confusion stems, maybe). My question about permutating the series is merely out of concern for accessing diverse harmonic structures that aren’t too qualitatively redundant once Root/Pitch parameters are modulated, independent of other timbral modulations (whether internal or externally input).