How do different VCA modules *sound* compared to each other?

[BlinkyLights]

I have not laid hands and ears upon many, so I just don't know.

Do some inherently (all other things being equal) sound darker, or brighter, cleaner or dirtier, do they respond differently to envelopes, or is sonic performance pretty much equal and do they all respind to envelopes pretty much the same (seems unlikely, intuitively, but...)?

Are there (forgive the term) high-end ones that purportedly sound better? Are any somehow more "colorful" or characterful than others, noticeably? (Obviously the tube ones sound different). What's the general consensus, if anything?

I need a couple of additional VCAs, and do I get to pick different sonic characteristics as well as features? I don't see much talk about VCAs from a How They Sound perspective. Do they mostly sound the same? Where do I learn what's knowable about this?

Thanks.

[hawkfuzz]

Most sound the same. The parts that are different are when you increase signal for clipping. Look up some videos and see if you can tell a difference.

LPGs are inherently different as far as sound goes due to the low pass part, but they are a VCA/Low Pass Filter combo.

[mdoudoroff]

In my experience, decent VCAs all sound about the same (i.e., transparent) until you start talking about overdrive and distortion. Some VCAs might (deliberately) distort in ways you like more than others. Some VCAs have special features specifically for this purpose. Two recent examples of the latter include the Koma Dual Discrete VCA and the XAOC Tallin.

[fluxsignal]

Typically discrete VCAs will have tasty overdrive and distortion. Another key difference is if the VCAs are 4- quadrant or 2-Quadrant. This is relevant for when you use audio-rate modulation to open the VCA.

[luketeaford]

Most of the VCAs I have are characterful, but the most obvious difference is whether or not it's DC coupled because if it isn't it makes a very weird (and different in each VCA) slew effect.

Sure, they often don't distort until you push them very hard (with the exception of moddemix which is more distorted at low voltages and therefore has a really interesting tail), but it has to do with how much voltage is used to open them. I think 8V is fairly standard, but some could be 10V or 5V and even if it's 8V it might begin to saturate before that-- working backwards from a non-electronics perspective to manuals that say panel controls normalled to offsets start to drive the VCA after a certain position.

[addendum]

LPGs don't just differ in that they respond slowly and feature filtering, which can be switched off with some, but the optocouplers in their signal path also eliminate the thumping that analog VCAs exhibit more or less when opened with a fast Attack.

Regarding coloration, there are several factors that determine which harmonics are added (odd, even and odd, nonlinear). Some harmonics are always added, more or less, others only appear upon hitting the roof (which can result in hard or soft clipping or various forms of folding). Tube VCAs color very strongly and interestingly.

[kwaidan]

With audio, I prefer VCAs with an adjustable response curve, even with envelopes that can be changed, for example, from linear to exponential to logarithmic. To me, the feature affects the sound, sometimes dramatically.

[cptnal]

With audio, I prefer VCAs with an adjustable response curve, even with envelopes that can be changed, for example, from linear to exponential to logarithmic. To me, the feature affects the sound, sometimes dramatically.

I wonder whether this differs from the effect you'd get from passing an adjustable envelope to a linear VCA. Are their shapes different on the scope, or does it depend on the character of the VCA - whether it's being overdriven during a longer part of its curve or something like that...

[SB-SIX]

I think some (most? all?) vca's will increase gain alot when you adjust the curve from linear to more exponential. Which can lead to (more) distortion/overdrive. On EG's like maths, the envelope range is not changed when dialling in more exponential curves.

[addendum]

An envelope's range won't change much if at all when the slope shape is bent by feedback.

Regarding VCAs, speaking for the Wiard Mixolator and I believe also the Malekko XMIX, the range doesn't change when the response is changed from lin to log, probably because they are wired as cross-panners based on phase cancellation.

Practically speaking (i. e. due to settings, not physical changes), the noise floor, signal to noise ratio and (due to its own attack time) the threshold of a noise gate used later in the signal chain may change because one tends to set different initial volumes (on VCAs that allow to do so) depending on the VCA's response curve and the envelope curves.

These characteristics can't be underestimated when creating sounds, especially imitative ones. With the right contours on loudness and (micro-)pitch, even the initial waveforms can become secondary with regards to recognizability of a sound.

[damase]

consider, how something distorts or saturates has a huge effect on its sound... how it handles sharp transients, low end, audio rate modulation, and very low volumes; you have a bunch of these throughout your synth before and after other processing it really can change the character of a synth in my opinion.

listen to demos with a very critical ear... do you like the distortion? or is it something you would avoid? in my experience if you dont like the distortion then you end up building your patches fighting against it. I think the tone of distortion can relate to the range for tone of clean too... how does it sound when pushed almost to distortion vs well below?

another thing i listen for with vca demos is “punchiness”. do the sounds seem 3d and pop out at you or are they 2d and far back? theres more to do with this than just the vca sometimes though.

i really like the WMD/SSF amplitude... this demo here...

not euro but i also really like the vcas in jomox alphabase

[Rex Coil 7]

... meh, just stick one of these after the output ....

[video][/video]

[BlinkyLights]

So, imagine a patch, any interesting patch, with just one VCA in a critical role at the end of the chain, letting the rest of the patch pass through as/when the VCA is opened up..

In this scenario let's say that this VCA is not set to overdrive like, say, you'd do with a CP3/DTM thing. Instead the goal is a basic, clean, good sounding outcome.

And if we were all in a room with this setup, and someone swapped out that lone VCA every few minutes as the patch went on and on, swapping from all of the majors and favorites, an Intellijel, a Malekko, a Coldmac, a crazy expensive Cjewman, an Erica, and on it goes, say 20 different ones, how different do you all think it will sound from VCA to VCA, in this scenario? Could we even tell?

And understand - I seek to steer clear of audiophile arguments, I don't mean some Magic Ears level of difference that only 1 in 100 claims to hear, I mean - obvious differences that (subjective evaluation of the difference's merits aside) most would hear it as indeed different somehow...? Not whether or not you can hear a 70s mylar cap over a 90s one.

Cjewman and others brag about the chips inside being special high end ones. Can you hear that difference, for instance? From a fancy L1 or Cjewman to a garden variety $50 VCA?

Because - in a sense - everything that audio passes through has "an EQ" of its own. In many cases its all but imperceptible, passing though audio unchanged, in others there's a bump somewhere, or a (these terms are suspect but they're out there) warmpth or sparkle, etc. Like, a Yamaha receiver just sounds different from a Poineer, or whatever, even if just a little bit. Part of what I am curious about is whether or not VCAs are similar.

Do (insert brand X) seem to sound more clean/hifi than (insert brand Y) who tend to be warmer, more colored, or does one have more pronounced mids, or less bass, or some thing like that?

If not, right on. Either way, I'm just new and dont know...

Thanks.

[sko87pro]

Part of what I am curious about is whether or not VCAs are similar.

Well, technically they are amplifiers - they should just amplify, as determined by the control voltage you give them. So that's consistent among VCAs: they just multiply two signals.

But there's more to the VCA than the chips they use. Some circuits are noisier than others; some have a different frequency response curve; some will clip before others. Whether these are benefits or shortcomings is a matter of viewpoint.

Applying changing amplification to an incoming signal is a form of modulation (AM), which changes harmonic content. So when we audition a VCA, we are reviewing how they respond to different signals. Ring mods are a type of VCA and they are known for "how they sound": uMod2 sounds clean, moddemix is characterful. (I don't have either.)

So I guess we can say "VCAs have their own character".

[kwaidan]

With audio, I prefer VCAs with an adjustable response curve, even with envelopes that can be changed, for example, from linear to exponential to logarithmic. To me, the feature affects the sound, sometimes dramatically.

I wonder whether this differs from the effect you'd get from passing an adjustable envelope to a linear VCA. Are their shapes different on the scope, or does it depend on the character of the VCA - whether it's being overdriven during a longer part of its curve or something like that...

I don’t know. I never tried the envelope with a linear VCA, though I have a few. The envelope is the Quad Modcan, and it has four different curves plus CV control. However, I seem to have more control adjusting both the envelope and VCA, more so than when I use a Doepfer A-140 or Peaks. Of course, it could all be in my head, but I like having control at both ends!

[BlinkyLights]

Part of what I am curious about is whether or not VCAs are similar.

Well, technically they are amplifiers - they should just amplify, as determined by the control voltage you give them. So that's consistent among VCAs: they just multiply two signals.

But there's more to the VCA than the chips they use. Some circuits are noisier than others; some have a different frequency response curve; some will clip before others. Whether these are benefits or shortcomings is a matter of viewpoint.

Applying changing amplification to an incoming signal is a form of modulation (AM), which changes harmonic content. So when we audition a VCA, we are reviewing how they respond to different signals. Ring mods are a type of VCA and they are known for "how they sound": uMod2 sounds clean, moddemix is characterful. (I don't have either.)

So I guess we can say "VCAs have their own character".

Thanks. I thought as much, but aside from mentions of overdrive here and tbere you hardly ever see people talking about much less choosing their VCAs for the sonic characteristics.

Or have I misssed those juicy discussions and comparisons?

Seems like obvious and important criteria.

[WaveRider]

well I have a Malekko vca (old) it distorts a lot, STG very transparent, and a crazy vactrol vca witch is handy with envelopes that have some voltage at rest... and has its own sound. They all have their own response to incoming cv...

[addendum]

Any VCA that doesn't go beyond a gain factor of 1 is NOT an amplifier, but an attenuator only. This gets mixed up frequently, but as long as one doesn't understand the difference, discussing the reasons for sonic characteristics of VCAs is a bit like comparing gears to brakes on a bike.

Some instances of distortion are only introduced above a certain gain factor, when the signal gets clipped, rounded or folded. Think of this as the "brakes".
Secondly, other forms of distortion may take place even below unity gain (because no VCA is perfectly linear even at unity gain, though some modern ones come extremely close. Actually, technically speaking, even mere attenuation is a form of distortion, which becomes practically relevant in mixing and compression because it lowers the signal to noise ratio... but let's stick to nonlinear distortion here).
Thirdly, the filtering in lopass gates technically is a form distortion also, but let's disregard this for now, too.
Fourth, there's bleedthrough of the modulating signal. Since any DC offset in the VCA's output will influence how that signal will be sound when put through a dedicated distortion device (waveshaper etc), this is can be an important part of the sound of a patch even though the bleedthrough may not be very audible if the VCA is at the end of the signal chain.

The second type is the hardest to discern and to discuss because it can be subtile and there are many factors at play, but this is the type you've been trying to discuss.

To me, the electrical details of what makes something like an old Moog VCA sound more colorful than a more linear modern VCA with a wider dynamic range are beyond my technical knowledge, so I can only approach it the way a music consumer judges their records versus CDs or originals versus remasters. But the distinction described above will serve to least know which functional level of a module one is trying to discuss.