For You Fellow Synth Programmer Geeks...

[Markyboard]

I was just thinking about all these cool synths with 2, 3 even 4 oscillators and...they all go through the same filter(s) controlled by the same envelope or other modulation source. The Cwejman S1 has 2 independent filters and allows routing each oscillator to either or both filters (LP and SV). There's also 2 envelopes that can control either filter as well as the VCA. This is really cool and opens up many more sound possibilities.

 

Besides modulars I'm curious what other synths have similar control? Matrix Brute? Prophet 12?

[timwat]

What comes to mind:

 

Matrixbrute, and (effectively) any multitimbral synth with a multimode filter (e.g., Nord Lead, Modal 008).

[Doerfler]

The Cwejman S1 ...

 

hey Mark, how do you pronounce that?

[timwat]

I'm guessing Schway-mahn or Tsvay-mon or something similar, I think. The etymology is Polish, isn't it?

[Markyboard]

What comes to mind:

 

Matrixbrute, and (effectively) any multitimbral synth with a multimode filter (e.g., Nord Lead, Modal 008).

 

Yeah, I didn't get back in time to edit and specify analogs. But yeah, multi-timbral synths are certainly a valid answer.

 

Matrixbrute:

 

 

6. Voice Modes:

 

Multiple keyboard modes for mono, paraphonic and duo-split modes. Split oscillators for paraphonic mode or use the duo split mode to play a sequence while playing notes over the top of it.

 

Not sure if this is the same thing? Is this really the equivalent of multitimbral mode, allowing 2 different independent "sounds"?

 

 

Modal 8 sounds like a good one Tim.

[Markyboard]

The Cwejman S1 ...

 

hey Mark, how do you pronounce that?

 

For you Dave... I pronounce it with one finger

[aronnelson]

I don't know, maybe the Casio PX-5S. 6 oscillator - really almost like 6 Roland tones with settings for each. Becomes a nightmare to program on the LCD screen.

[Nathanael_I]

The John Bowen Solaris does this. You have four filters. Each can be 6/12/18/24db HP or LP, or an Oberheim filter, a Moog filter, the Curtis filter, or a comb filter! Plenty of options. Because there are also four mixers in it, you can set up whatever filter chain suits your fancy, including feedback loops.

[Markyboard]

Good one Nathaniel. From the User Guide:

 

The signal for each filter is sent to its own dedicated VCA,

where its pan position and level can be set. Each VCA can be

controlled by a different envelope, but the EG6 (VCA) has the

final say for the overall output. Using separate envelopes, you can create articulated shapes for each filters output; almost a multi-timbre approach to the sound, enhanced by the fact that each can also have its own envelope and pan position.

 

As opposed to separate programs in a split/dual/multi approach the Solaris allows assigning each oscillator to one of 4 filters and use one of 6 envelopes to modulate each filter. Very cool.

[Al Coda]

Good one Nathaniel. From the User Guide:

 

The signal for each filter is sent to its own dedicated VCA,

where its pan position and level can be set. Each VCA can be

controlled by a different envelope, but the EG6 (VCA) has the

final say for the overall output. Using separate envelopes, you can create articulated shapes for each filters output; almost a multi-timbre approach to the sound, enhanced by the fact that each can also have its own envelope and pan position.

 

As opposed to separate programs in a split/dual/multi approach the Solaris allows assigning each oscillator to one of 4 filters and use one of 6 envelopes to modulate each filter. Very cool.

 

Well, Solaris is a top notch digital synth originally designed w/ (Creamware) Sonic Core SDK.

I describe it that way because it´s SOFTWARE.

 

You can create the same mentioned above in Kurzweil VAST using the VA OSCs in single layers,- w/ the limitation of the few non-aliasing waveforms,- then cascade these layers into other layers used for the KURZ´s moogish emulation filter p.ex..

Have in mind each layer already offers a "VCA", 3 ENVs, 2LFOs, 2 ASRs and it´s easy to use 2-Pole (and other) filters in the SAME layer,- except the MOOG filter emulation consumating the most DSP blocks, thus using a single layer itself.

 

For sure it sounds different than Solaris, but the flexibility of VAST is great too.

 

A.C.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[The Real MC]

Chroma

[mate stubb]

Chroma had VCO VCF VCA chain, up to 4 on each voice. Then you could rearrange the patch algorithms to double up filters to get 4 pole or HP LP combinations.

[Markyboard]

I suspected Chroma was a candidate even though I know nothing about it's architecture. I'd have to put this and the CS-80 at the top of my list for cool synths I missed along the way.

 

Regarding the Kurzweil, I've got it sitting in front of me and perhaps I should finally give programming a patch from scratch a go. Maybe it'll prove more fun then I think.

[Al Coda]

 

Regarding the Kurzweil, I've got it sitting in front of me and perhaps I should finally give programming a patch from scratch a go. Maybe it'll prove more fun then I think.

 

The trick is editing the factory DSP algorithms for different signal routings and save ´em under new names as "user objects".

SpotlightKid made some custom algos in the sense of Minimoog D architecture, typically 3 or 4 layers, incl. 2 OSCs using different VA DSP waveforms (not sampled waveforms, so the keymaps are set to "none" or "silent" !!!), some kind of mixer plus the filter.

Doesn´t sound like a MOOG when you load that stuff but you have the source, routing and shaping tools.

Tinkering w/ p.ex. keytracking is essential and one very important trick is using the AMP envelope for the MOOG filter emulation occupying a single layer for itself, so the out-level of that layer is set to -96dB, AMP ENV not affecting volume and being free modulating cutoff (and resonance).

The AMP ENV in each layer is the only one which can be modulated in the ENVCTRL page, ENV2 and 3 cannot.

So use AMPENV whereever you can.

For slow attack filter sounds (PADs p.ex.) the ENVs 2 and 3 are good enough, but not for the snappier stuff.

With FUNs you can make the AMPENV "analog",- logarithmic instead of linear.

Scroll down to Demystifying FUNs_random controllers_analog envelopes .

Amplifier is always present in VAST even invisible at 1st glance.

I´ve got really cool sounds out of my PC3, not using the samples.

Take care using the few non-aliasing waveforms consumation 3 DSP blocks when programming from scratch.

It´s a limitation but sounds best.

 

Many factory synth patches I don´t use at all because they alias like hell,- my own, incl. sync patches,- don´t .

 

A.C.

[Shamanczarek]

Most Yamaha CS series are two complete single voice Synths in one box. The CS70M is the only model that has Key Split that allows a layer on each side of the split. Some CS Synths have a choice of LP, HP, or BP filters.

 

The Oberheim Xpander/Matrix 12 has 15 filter types plus each voice can have a different patch when set to separate MIDI channels. The Matrix 12 can layer at least 6 voices each with a different patch.

[AnotherScott]

The Maxi-Korg was two-voice duophonic, where each voice had entirely separate controls.

 

There's also the Oberheim Two Voice (or however many voices, if you just connect more SEM modules), and I think you can stack Slim Phatty modules and still control their filters and envelopes individually. These aren't quite modular systems, but maybe a kind of hybrid.... the stackable units are in a sense modular, but each individual unit is not.

[Markyboard]

What you really have with the Obie 2 voice is 2 independent synths in a common box that allows cross connectivity through patching. Sounds like the Yamaha CS series. This is really crossing into modular territory - and is certainly the case for 2 slim Phattys (or any 2 synths that have external filter inputs. Not that I'm objecting to the versatility you get with this approach.

 

Not familiar with the Korg but the Matrix 12/Expander is another great example, one I hadn't thought of until someone brought up the Solaris.

 

OK, back to my PC3x - I just needed to come up for air

[aronnelson]

I'm pretty sure a number of Roland synthesizers had "tones" which are basically a single synthesizer chain X 4 or 6 etc... Not that a single tone sounded fat or anything.

Same with the PX-5S HEX layer.

 

 

[Markyboard]

I'm pretty sure a number of Roland synthesizers had "tones" which are basically a single synthesizer chain X 4 or 6 etc... Not that a single tone sounded fat or anything.

Same with the PX-5S HEX layer.

 

 

Yeah, further up I re-qualified my original question with ANALOG synths of which a number of good ones have been mentioned. Seems many digitals including PX5 and Rolands have this capability- all too obvious to me now.

 

The AMP ENV in each layer is the only one which can be modulated in the ENVCTRL page, ENV2 and 3 cannot.

 

Experimenting with this today I developed a 2 layer cascaded program. Couple of head scratchers for a while but got through it. Not really trying to develop the Moog-like patch but using most of your concepts - this is pretty cool. Thanks AC!

 

 

 

[Al Coda]

The Oberheim Xpander/Matrix 12 has 15 filter types plus each voice can have a different patch when set to separate MIDI channels.

 

Not only that,- we´re talking "Multi Patch" Mode here ...

In fact, you can "group" any number of voices, give ´em a "Zone", group another number of voices and create a 2nd "Zone", then play w/ the key-assignment options and MIDI channels (or in the Xpander only,- w/ CV/GATE INs).

Because that can be done w/ up to 3 Zones, a result could be a 3-voice, a 2-voice and a monophonic synth all three playing different patches,- or a 4-voice and 2 monophonic ones which is the scenario I prefer.

This in addition to what you mentioned above,- playing 6 mono-synths w/ different patches on individual MIDI channnels.

Finally there´s the option playing all 3 Zones on the same MIDI channel but w/ different patches and different key assignment modes.

Means, you´re able to set the mono-synths´s key trigger to "RESET" and the polyphonic zone to "ROTATE" p.ex.

 

At the end of the day, it´s still much more flexible than J.B. Solaris which lacks true (MIDI-) multi timbrality.

But Solaris also has other strength making it one of the best, if not the very best, digital synth(s) out there.

 

The Matrix 12 can layer at least 6 voices each with a different patch.

 

That´s what it has in addition to the Xpander.

Technically it layers it´s 2 internal voiceboards (1 voiceboard = 1 Xpander) and offering a "Detune Page",- which doesn´t exist in the Xpander.

 

A.C.

[Al Coda]

 

Experimenting with this today I developed a 2 layer cascaded program. Couple of head scratchers for a while but got through it. Not really trying to develop the Moog-like patch but using most of your concepts - this is pretty cool. Thanks AC!

 

I just only mentioned the "MOOG-like" stuff demonstrating THAT filter emulation consumates a single layer alone, forcing you to set the amp-output of THAT layer to -96dB to cascade into the next layer.

You´ll also recognize that ONLY Kurzweil factory algorithm (IIRC) numbers 100+ offer additional "cascade" inputs and probably don´t allow insertion of every DSP block you imagine for a new patch.

So, you´ll come soon to the point requireing custom algorithms.

That "Minimoog architecture" algorithm offers something you won´t find in any of the Kurzweil factory algorithms even they offer the required cascade input.

An algorithm is empty always, you benefit from routing though.

If you use a MOOG filter emu to make it sounding like a MOOG or any other filter, shaper or combo of "what´s possible in that algo", is up to you.

When you reverse engineer most synth patches in the PC3, you´ll recognize most aren´t done w/ the VA,- they are done the old K2500/K2600 way instead,- mostly using the sampled waveforms, each in a separate layer, every layer using the same filter w/ modifiers identical set up just only to minik a simple monophonic analog patch.

It´s a waste of polyphony since every single layer being active simultaneously cost DSP horsepower which might vary depending on which DSP blocks are been loaded into each layer and so on.

You can get a lot of sound out of a PC3 when dealing cleverly w/ the overall DSP load.

 

A.C.

[Shamanczarek]

When layering sounds on most Synths the polyphony is halved. With something like an OB8 or Jupiter 8 it becomes a four-voice when layered. The Matrix 12 becomes a six-voice with one layer and then a three-voice if layered again. The advantage with dual oscillator eight-voice Synths is that a layer has up to four oscillators and up to six or more in the case of the Matrix 12. The Yamaha CS scheme doesn't reduce polyphony but a layered sound has only two oscillators per voice.

[Markyboard]

You can get a lot of sound out of a PC3 when dealing cleverly w/ the overall DSP load.

 

A.C.

 

Almost 8 years in I'm finally taking a stab at VAST programming; not too bad really. 10 hours or so and I think I have a pretty good handle on the concepts but it is somewhat tedious. I used to have a whole lot more stamina for this kind of thing and to make matters worse I wasted way too much time yesterday trying to use the Soundtower editor .

 

Back to using the PC3 interface. I find myself really wanting to turn some knobs but think I'll stick with it for a bit longer. I'm now creating custom algorithms which is pretty cool.

[Al Coda]

Almost 8 years in I'm finally taking a stab at VAST programming; not too bad really.

 

About 6 years here,- PC3 was my 1st Kurzweil after the Micropiano module,- and it was a long way since being used to knobby synth-surfaces and lots of global parameters, I had to overcome my inhibitions 1st before diving deeper into VAST.

 

... but it is somewhat tedious.

 

Yup !

Nonetheless I appreciate I can do things in VAST I cannot do w/ the other hardware-synths I own.

OTOH, there is also stuff you cannot achieve w/ that design regardless how much you tinker w/ it.

The DSPs in PC3 series are simply not fast enough to get the best out of LFOs, ASRs, ENVs,- and many layers stacked too.

There´s some latency between stacked layers which seems to be most obvious when using samples/keymaps and play chords w/ higher velocities.

 

... I wasted way too much time yesterday trying to use the Soundtower editor .

 

OMG !!!

The editor, even for free, is the worst tool to try w/ the PC3.

I found out pretty fast since it was I tried 1st after I got the instument new and in the hope to avoid all that menue diving when using the PC3´s UI alone.

Meanwhile I wont´ even think about using an editor programm at all and now I thank Soundtower for making my decision, using the PC3´s surface controls, so easy.

 

I'm now creating custom algorithms which is pretty cool.

 

 

A.C.

[MoodyBluesKeys]

My experience is that the Soundtower editor is not harmful ONLY if one never uses it to write anything, just to get a different view (lots of stuff on one screen).

Having said that, it has been several years since I've even opened the program while connected to any of my PC3s.

 

I don't really get into VAST editing that deeply, but I got tired quickly of having to reload everything including the OS, after ST puked in my system. I do always have an "everything" backup.

 

[Mike Martin]

I was just thinking about all these cool synths with 2, 3 even 4 oscillators and...they all go through the same filter(s) controlled by the same envelope or other modulation source. The Cwejman S1 has 2 independent filters and allows routing each oscillator to either or both filters (LP and SV). There's also 2 envelopes that can control either filter as well as the VCA. This is really cool and opens up many more sound possibilities.

 

Besides modulars I'm curious what other synths have similar control? Matrix Brute? Prophet 12?

 

Actually the solo synth in the XW-P1 is unique in this regard -

5 Internal oscillators (plus one external input) - each have their own pitch envelope, amp envelope, filter and filter envelope and then they all get summed and go through a "Total filter" with its own envelope.

[AnotherScott]

Actually the solo synth in the XW-P1 is unique in this regard -

5 Internal oscillators (plus one external input) - each have their own pitch envelope, amp envelope, filter and filter envelope and then they all get summed and go through a "Total filter" with its own envelope.

Same with the XW-G1?

[Mike Martin]

Yes the same on the XW-G1.

 

I'll add that it is possible to also route any of the other "polyphonic" sounds on any of the other MIDI channels through the solo synth's "Total filter", like a Paraphonic synth.

 

 

[Markyboard]

Hey cool Mike, that solo synth although digital is exactly the architecture I was talking about; everything under 1 patch as opposed to a multi layered approach although effectively this can yield the same result. I watched your demo video - amazing what this synth does for the price- wonder if any owners actually mess with this stuff?

 

I'm exploring (with forum assist) the Kurz PC3x' seemingly unique multi- layered approach within a single program where each layer is it's own sample and/or modeled oscillator (or 2). Each under control of independent envelopes or whatever. You can then stack up to 32 programs - this is plain sick!

[Mike Martin]

Hey cool Mike, that solo synth although digital is exactly the architecture I was talking about; everything under 1 patch as opposed to a multi layered approach although effectively this can yield the same result. I watched your demo video - amazing what this synth does for the price- wonder if any owners actually mess with this stuff?

 

No, I don't think many users do although there are a few that have gone pretty deep. There is one user that used the mod matrix to essentially fool the engine into becoming duophonic. The XW series has its quirks but it is very unique sounding and one of my favorite products.