Moog Voyager XL

[soundwave106]

It still could be discrete and pretty compact and neat looking. There exists such a thing called an "IC array" which has that nice chip look, but still is analog components (arrays of resistors, capacitors, diodes, and the like). We'd actually have to see *what* those chips are to know what is up.

 

A poly is in many ways more possible these days, in a more compact form, compared to the old days. Not only are there IC arrays, but there are some very small components (SMD resistors are *very* tiny for instance, other SMD parts are quite small). Depending on the design you can also get a higher part count with larger components (open a Waldorf Pulse for an example of this, it uses larger parts than the typical SMD but is packed very tightly on the board, simply because with today's machines this is possible).

 

Mind you, polys still can be a large part count, and still be expensive. But size wise... much less of a problem these days.

[mate_stubb]

Yes, you can go to tiny surface mount components - they do not have the expansive sound of boards with big traces and thru hole components. Cram most of the components of an oscillator or filter onto a chip, and bad things often happen.

 

That was my original point - Moog will have to give up sound to bring an affordable poly to market.

[brandonm]

Tom Oberheim acheived a 16 voice analog in a 1u rackmount enclosure with the MSR-2 back in 1994. The DCO's Tom used were a custom CEM IC. The DCO's used in the DSI products are an updated version of these same DCO's. I got an MSR-2 16 voice a few months ago and opened it up to check out the boards. Each voice board was filled with components, but we are talking 8 voices per board in a 1u rackmount surface area. The envelopes and LFO's are software.

[mate_stubb]

Perhaps you are not aware that Tom Curtis died, and there are no more custom Curtis chips being designed.

 

Ironically, the state of the art in chip making has passed on to ever smaller and smaller dies, and there are no machines around these days which could be made to remanufacture those old Curtis designs which sounded good.

 

And for every Curtis design which DID sound good, there were at least two that didn't sound so hot.

 

It is ever harder to design and manufacture great sounding integrated analog synth chips, because all the money is in digital these days. And the machines to make these chips can cost hundreds of thousands of dollars.

 

I used to be a manufacturing engineer, and so am coming at this pessimistically because I know what is involved in designing an analog poly synth.

 

There are 3 sides of the triangle in product design and manufacturing - high quality (great sound), features (high polyphony, flexible voice architecture), and cost.

 

Now pick any two.

[brandonm]

Perhaps you are not aware that Tom Curtis died, and there are no more custom Curtis chips being designed.

 

 

I wasn't stating that the DCO used by DSI is a new CEM design. Just that it is a derivative of a CEM-designed DCO that Tom used. That is what Dave Smith stated in an interview. Sorry for the misunderstanding.

 

As for the quality of this DCO, it's OK. Nothing like a CEM3340 though.

[mate_stubb]

Agreed. I was just saying that Curtis is not around to design any new chips for Moog.

 

I don't know if they have any chip designers on staff, or if they would have to bring on some talent to come up with their own dedicated synth chips. It's an expensive proposition all around.

 

I don't remember the figures Alesis spent designing the custom ASICs for Andromeda, but it was over a million dollars. And I don't know if that figure even included the setup to fab them.

 

Moog is a SMALL company - they have to be very very careful on large outlays. One misstep could ruin them. None of us want that!

[soundwave106]

Yes, you can go to tiny surface mount components - they do not have the expansive sound of boards with big traces and thru hole components. Cram most of the components of an oscillator or filter onto a chip, and bad things often happen. That was my original point - Moog will have to give up sound to bring an affordable poly to market.

Affordable and high quality can be a contrast, yes. I was more commenting on size, particularly the electronics. Studio Electronics have managed to cram an 8 voice into a 4U rack (15" depth). It's a voice card system; you can see some pictures of the insides of the Omega here. I cannot identify all of the chips (save for one DAC) but it's pretty compact (not a whole lot of small surface mount, either). And I also have heard real good things about the Omega's sound. To be honest, from what I know of electronics, a bigger challenge is mechanicals, which have not had the same price decrease for the quality that pure parts have.

Perhaps the Moog Taurus experiment showed that there is a market for expensive, niche synthesizers after all. If Moog brought a poly to the market, I kind of think they won't make an *affordable* one at first. Instead, they'd release a limited edition sort of thing, perhaps something akin to a poly Little Phatty (they seem to be keen on recycling designs lately). It may be $3000-$4000 for say a 6-8 voice Phatty (I assume there will be some scale benefits) but some people would buy it.

[ChristianRock]

One of you guys (not me, I'm not an EE) should read Keith Barr's book and build the next VCO analog chip

 

http://www.amazon.com/ASIC-Design-Silicon-Sandbox-Mixed-Signal/dp/0071481613/ref=sr_1_1?ie=UTF8&s=books&qid=1284481031&sr=8-1

[willi]

Not to be pedantic or anything, but if it's in a chip at all, it's not discrete. "IC" = integrated circuit; e.g. NOT discrete. Every chip is analog inside, when you get down to it. Try looking up something as 'digital' as a flip-flop, and you will see that it is constructed from entirely analog components inside. When you get down to it, all electronics - analog, digital, and everything in between - are made from the same basic (analog!) parts. In fact the first flip flop was built with tubes.

 

I digress...

[carbon111]

Plonk!

[soundwave106]

Technically, Willi is correct, but realistically, I'm not sure if there is going to be any difference between a "fully discrete" design versus using a few simple IC arrays. Perhaps we could use the term "electrically discrete" to describe this type of design.

 

To be honest, it would be interesting to read what, if any, technical reasons there would be for a well-designed analog IC to behave differently than an all-discrete version.

[Bernard]

One of you guys (not me, I'm not an EE) should read Keith Barr's book and build the next VCO analog chip http://www.amazon.com/ASIC-Design-Silicon-Sandbox-Mixed-Signal/dp/0071481613/ref=sr_1_1?ie=UTF8&s=books&qid=1284481031&sr=8-1

This pixel has it's eye on you



Resistance is futile

[brandonm]

Technically, Willi is correct, but realistically, I'm not sure if there is going to be any difference between a "fully discrete" design versus using a few simple IC arrays. Perhaps we could use the term "electrically discrete" to describe this type of design. To be honest, it would be interesting to read what, if any, technical reasons there would be for a well-designed analog IC to behave differently than an all-discrete version.

I would have to believe it simply comes down to the build quality of the IC versus the discrete components. An IC may have a number of transistors inside, which creates heat and we all know that heat can be an enemy to electronics. Heat is not so much an issue with a single discrete transistor, but it may take many of these components to make up what a single IC can do. If an IC is made well and disperses heat very well, I personally don't see how there could be much a difference, but what do I know.

[mate_stubb]

The more you pack into a chip, the smaller the traces, the thinner the layers, the bigger the heat dissipation problem becomes. You may have electron migration problems, interference between layers, etc.

 

Certain people claim the the big fat PN junctions in discrete transistors, along with wide traces on sparsely populated circuit boards, produce a more open sound. I personally can't say.

[carbon111]

It's too bad that so much of electronics is mostly magic. Sad that we'll never be able to actually measure things like resistance, impedance, capacitance, crosstalk, signal-to-noise, ect.

I doubt we'll ever be able to ever build high fidelity chip-based circuits...all those horrible op amps and transistor arrays twisting the sound beyond recognition. The horror.

I'm off to go get some $1000 wooden knobs for my stereo now to warm up the sound...

This reminds me of a conversation I had with people who thought different DAWs sounded different from each other.
Either it sums or it's broken. Period.

[soundwave106]

I doubt we'll ever be able to ever build high fidelity chip-based circuits...all those horrible op amps and transistor arrays twisting the sound beyond recognition. The horror.

 

 

Just Googling around, I can only find two advantages to an all-discrete design.

 

(*) ICs are more commonly mass-produced; it may be hard to find a good IC with the type of tolerance / quality / etc. a high end device might need. A boutique shop probably cannot afford to get a design they want on an analog IC.

 

(*) For *really* precise instruments, it is easier to swap out transistors / resistors / capacitors with a discrete design when you really need matched specs etc..

 

#2 is probably a thing of the past for the most part. Example: The Minimoog had hand matched transistors by necessity (some of the "sound variation" between models is because later models were only partially hand matched to save cost). The Voyager has an IC with precise enough transistors to not need matching, as I understand it.

 

#1 might be very relevant, but not for theoretical custom IC designs.

[carbon111]

Can I take my tongue out of my cheek yet?

 

Either an IC can perform exactly to specs/tolerances or not. If "not", use something else.

 

I'm an electronics freak...I breathe flux and eat solder...thank god for RoHS.

[willi]

All chips sound digital! And all tubes look best when lit by orange LEDs!

 

There's no tubes in this! Nor springs! (nor delay...doh!*)

 

Anyway, Voyager XL = want.

 

I own a CS-50 and personally came to the conclusion that polysynths aren't all they are cracked up to be... but then again I don't play a lot of pads or backing sounds right now. I do play a LOT of chords**, but what I personally found was that a lead sound should be fine with a monosynth format, and that the additional oscillators and 24db/octave LPF would have benefitted the bass sounds in that particular synth immensely. A Moog polysynth with nice fat filters would be great, but the whole experience left me wanting a Voyager more than ever...

 

It probably highlighted the necessity of patch storage, too, in performance situations.

 

XL looks like a nice synth. The price is a pain; so much that it is an investment, and I guess my biggest concern would be that the additional complexities might make it harder to maintain 30 years into the future, compared to a (already 30 years old) model D!

 

 

* Yes, yes, I have a MF-104z.

** I also use a lot of sustain, which can be more challenging to manage with synths compared to electromechanical pianos like the CP's and Wurli's, etc.

[Spleencage]

Seeing all of the cv jacks have been removed from the top edge, its same to assume those are the 251 and 351 installed with an addition of ribbon controller.
I am more concerned with what is said at their website as they keep comparing it to a model- D. Have they gone back and "dirtied" up this model?

As far as affordability goes. I think anything under 10K would be reasonable for a patchable 8- voice poly, when you consider polys for the late 70 and early 80's were for the most part priced between 5-10k, and that was 30 years ago.

[Buuzer]

I'm buying this as a nice controller/addition to a huge dot com modular.

As soon as I win the lottery...

[triton76]

I hardly recognize this place.

http://www.bigbriar.com/voyager/?section=product&product_id=21396

[triton76]

yeah it's still mono. 3 osc is great but 4 splittable with be Moogalicious

disclaimer: I've only dreamt of owning a Moog

[Tomkeen]

What, no glasses?

[k2500x]

I made this exact same post last week.

I guess there is just very little buzz about Moog synths.

 

Funny, when Roland releases a new VA, the forum is buzzing for months and it's STILL on the front page. We're all hypocrites!!!

[Tomkeen]

At least most people can afford the Roland VA. $5000 is a LOT of money for most people. I could never justify buying anything like this moog, however cool it may be. And is.