S-Poly

[huggermugger]

I'm not 100% clear on the difference between S-Poly and 'vanilla' Poly. But I'm pretty sure that the only developer working with S-Poly is Adroit. I own Adroit's Chord module, a fine CV generator for predefined chords. It's nice to have a wide variety of chords available with a simple click. However, without Adroit's various S-Poly converters and splitters, it seems impossible to tap into Chord's CV values in the usual way because it outputs S-Poly, not regular Poly. I can accomplish my goal with other chord-related modules such as Aarnville, but I was hoping to use Adroit Chord. However, it's only available in a bundle with a ton of other modules I don't want. Is any other developer working with S-Poly?

[UrbanCyborg]

S-Poly is Adroit's own development, amounting to a sub-system running on top of VM. Very clever chunk of coding, I suspect, and probably pretty complex. So, no, you wouldn't be able to coopt it for other purposes; a module would have to be custom built to use it, and that's Adroit's balliwick.

More generally, I think what you're asking for is a specialized data poly. We've all asked for that, but it looks like CA isn't interested in going in that direction. So you'll just have to live with using the single-voltage-per-channel poly connections that are there. That's what I did for my Boolean Truth BLink connectors.

Reid

[ColinP]

The difference between S-Poly and the regular poly connections is that the regular ones are limited to the number of channels set by the user in the NUMBER OF VOICES part of VM's control panel.

Additionally the polyphonic system in VM requires both pitch and gate connections because of the way that voice allocation works. So if I used the regular poly setup then users would need to use two cables instead of one and ensure that NUMBER OF VOICES was always set high enough to include the number of notes used in chords or scales.

Also S-Poly is used to transfer up to 16 different control signals between various Adroit modules, so if vanilla poly was used instead then things like N-Step would break if NUMBER OF VOICES was set to anything other than 16.

More details here...

https://adroitsynthesis.com/s-poly-connections/

I did ask CA to remove the restriction of their poly signals a few years ago but they didn't respond. If other developers want to adopt S-Poly I'm happy to license the technology for free, but in an ideal world CA would remove the restriction as the NUMBER OF VOICES parameter is available in the API so there doesn't seem to be any sensible reason why the restriction is hard-wired into the cable code.

[ColinP]

However, it's only available in a bundle with a ton of other modules I don't want.

I'm curious as to what's putting you off LSSP XL as it's a very popular bundle with a user base size second only to the Granular Synth.

In the current sale it's ridiculously cheap at $35 for 54 modules - so 65 cents per module and some of the modules are very sophisticated. Also there's a 30 day free trial and it's supported by extensive documentation and high-quality tutorials.

I suppose I could make the S-Poly to Mono module available separately to LSSP but the technique of spilitting a poly signal into parts is fairly rudimentary stuff. I'd like people to explore the higher-level harmonic manipulations available in LSSP. Getting completely familiar with diatonic triads, standard scales, chord progressions and inversions Then to experiment with key modulation, modal scales and automatically generating chords from scales and such like. Using the S-Poly signals to control Pitch Adjuster, Chord Player, Melody Sequencer plus non-LSSP products such as N-Step and the Granular Synth.

Also the use of V/Bar and time-splitting in LSSP is something that I'm passionate about promoting as using voltage as an analog of time is far superior to using complicated, crude and fragile clock-based sequencing. I understand that it's a difficult paradigm shift but now that I'm used to V/Bar it's almost painful to go back to using old-fashioned clocks. You can do things with the Groove and Time Flow Changer modules that are close to impossible with a clock-based system for instance and stateless sequencing makes editing large-scale work a breeze as it's so easy to jump backwards in time.

As an aside, I decided to make N-Step clock-based so that people wouldn't have to get their heads around V/Bar but it made the project far more complicated and the end result far more restricted than it would have been if I'd used voltage = time.