A Poly-Moog Synth in the future ?

[ColorForm2113]

For a polyphonic to be "done right," it would have to be fully polyphonic, allow individual tuning, voltage control, waveform, filtering,and articulation per-note.

But that's not what you lot want.

I would love that! Minimoog per note! Or at least micromoog

[EMwhite]

The expense comes in the form of having the "rest of the voice"; meaning Env gens, filters, vca(s) and all of the modulation options. So divide down as you like, even the 12 voices that Dave Smith threw into the P12 are going to cost if done with discreet electronics (and he uses ICs but it still sounds good... not AS good in my opinion but still good).

Bring as much as you can under microcontroller digital control (Moog is close to mastering this: started with the Phatty Tribute then to Taurus to Minitaur and now Sub) and you can reduce cost and improve capabilities but if the purist wants discreet electronics and purity of analog tone, there is only so much you can bring into the digital paradigm. Anything that Moog produces which exhibits the hands-on merits of the Sub with perhaps more voices is going to grab folks attention now that they have folks attention. I've got no inside information but the serial numbers that I'm seeing posted here and there indicate that they are selling well.

But back on topic (Poly). My Opus 3 has 49 keys of polyphony and I recently impressed a guitar player friend of mine whose knowledge of synths stops at the Arp or Model D. Well, he was impressed by so much 'poly' until I explained that the Filter sounds wonderful but it (Singular) is only useful for fat chords, monophonic leads or playing legato with no re-trigger.

That's the dilemma. Opus 3 Organ section, which I'm not thrilled with (mainly because I have a Hammond) works 100% as it should, but strings start to get funny due to limited VCA control and brass, as mentioned above... limited by single filter.

On a bit of a rant here but the balance between discreet 2nd stage audio (post OSC) and all of the modulation capabilities that folks demand is going to cost a pretty penny.

Some of you may have read MC's excellent OBX blog and some may be following Scott Rider's ambitious OBX voice card DIY effort on muffs: http://www.muffwiggler.com/forum/viewto ... 63&start=0 (inspired by and with help from MC); between development of projects like these and some of the early work invested in by Olivier and the community of folks at Mutable (http://mutable-instruments.net/), there is lots of Analog and Digital-Analog hybrid going around.

It's a good time to be involved in electronic music. I'd argue that it's the renaissance of analog as many have pointed out. Really looking fwd to what is next!

[megavoice]

But the problem is VCOs must be heated up to keep them stable and in greater numbers this has an effect to the price. This is not the case for DCOs.

Not all VCO designs need to be heated.
Many (if not most) VCO designs are not.
But it is true that they must have some sort of autotune or feedback circuit to keep them stable when used in a polyphonic structure.

DCOs don't need such circuits or techniques as above, but have their own share of problems.
Some use a master clock which renders all DCOs perfectly in phase.
Most also don't allow "easy" modulation.
That is, one can't just "plug in" a modulation source (env, lfo, noise, etc) and have the DCO follow along.
Those sources must either be generated or digitally encoded first, then applied to the digital stream controlling the DCO.
Modulation frequencies are therefore limited to the speed and resolution of the digital processing.
So a design that uses VCOs could add a modulation control with something as simple as a pot and resistor, whether originally manufactured that way or modded.
On a DCO design, it could require considerable software changes.
This is one of the big reasons you don't see modular synths that use DCOs.

As for Moog making one, give them time. One day they might.

Oh Great ! That's what I like so much. Qualified statements.

And for the Poly, yes, It will take some time. Im feeling this. The customers pressures are too high.

[mico]

For a polyphonic to be "done right," it would have to be fully polyphonic, allow individual tuning, voltage control, waveform, filtering,and articulation per-note.

I'd slaughter many innocents as an offering to Baphomet for exactly that.

[MC]

Wow I didn't realize the progress that Scott made on that OBX board. I had been in contact with him through FB but my internet access got cut off due to payment dispute and FB is temporarily blocked at work. Not a member of MW (yet).

[MC]

For a polyphonic to be "done right," it would have to be fully polyphonic, allow individual tuning, voltage control, waveform, filtering,and articulation per-note.

Well, I had a different thinking of "done right"... I was refering to reliable design.

No amount of features is going to sell an analog polyphonic if it isn't reliable, and the vintage market is littered with unreliable ones.

A very common achilles heel is connectors. If the contacts are not gold-plated then they will not be reliable for carrying power rails and critical CVs, which translates to tuning problems. Connectors with gold-plated contacts aren't cheap though, thus many vintage polyphonic designs settled for tin-plated contacts. The problem with tin plating is it will oxidize to the point that they can't be cleaned at all, and you lose voltage over that ozidation layer. OK for logic levels but bad for CVs and power rails. Those were issues I identified and corrected in my Memorymoog and am now applying the same corrections to my OB-X.

So if the design decision for a new analog polyphonic was made to go with tin-plated connectors, then the connectors should carry ONLY logic level signals. This necessitates putting the DACs and S&H buffers on the voicecards to prevent analog CVs from being propogated via tin-plated contacts. So many hybrid analogs got this wrong by piling logic and analog signals onto tin-plated contacts. You can get around gold-plated contacts for power rails by using local voltage regulators - an approach favored by Oberheim, but the gain was erased with analog CVs on tin-plated contacts and no thermocouple compound between the tempco resistors and the CA3086 array performing the expo conversion (insert facepalm here).

Local DAC/S&Hs does carry a higher price due to redundant circuitry on each voicecard, but it does permit multitimbrality and voice modulation. Global modulation sources such as pitch bend and LFOs can be done in software and can modulate via the digital domain, which has been the approach since the 1990s.

As I said, there's a right way and a wrong way to design a polyphonic.

[museslave]

Well, reliability problems are design flaws. Everyone should assume that a synth put out will arrive without reliability problems.
Analog polyphonic synths are only good if the design is good.
But what the hell good is a solid design if it isn't an expressive and creative tool?