From Brian Willoughby Sent Fri, Nov 30th 2018, 06:55
The explanation hinges on the distinction between adding and multiplying = modulation sources. You can combine as many as 6 modulations onto a single parameter. Those = all add up to a single value, so you cannot reduce the effect of one = modulator unless another modulator actually produces negative values. However, with two virtual VCA controls in series, the sum of the first = set is multiplied by the sum of the second set. Thus, if either one goes = to zero, then the final volume is zero. Even if neither values actually = fully reaches zero, either one can reduce volume proportionally. This is why, as Dan points out, Velocity can control volume on one VCA = and an Envelope can control volume on the other VCA. Either one can = silence the voice, or make it quieter, even if neither modulator = produces negative values. Note that there is still only one analog VCA at that stage of the voice. = The Voice Processor calculates the product of the sums and sends the = final CV to the VCA input of the 3372. You could literally patch 12 = sources to the CV that controls that VCA, and get very flexible and = powerful results. Compare the Minimoog, where I understand there are two analog VCA units = in series. I don=E2=80=99t actually own a Minimoog, so I don=E2=80=99t = quite understand how the vintage keyboard without velocity would make = use of both. I really only noticed the peculiar (to me) architecture by = reading about the AJH Synth recreation of the Minimoog VCA. Brian On Nov 29, 2018, at 4:27 PM, Dan Goldstein <xxxxxxxx@xxx.xxx> wrote: > I=E2=80=99ve always thought there were two VCAs so they you could use = the final one to (easily) do velocity-to-volume for any patch. The = first VCA can be controlled by envelope(s), LFO modulation, etc., and = the second VCA still controls the overall volume. So, for example, you = can have a tremolo patch that still responds to velocity and doesn=E2=80=99= t play when notes aren=E2=80=99t held. > =20 > Dan > =20 > From: xxxx xxxxx [mailto:xx_xxxxxx@xxxxxxx.xxx]=20 > Sent: Thursday, November 29, 2018 3:25 PM > =20 > Funny you should mention this... was talking with another AH-er today = about their newly acquired Xpander and the strange inclusion of the two = final VCAs. I have never come across any sort of explanation of why = there are two. You could make the last one drone just as easily as the = first. Anyone know the logic behind this?=20 > =20 > Also, that's an interested idea for a pseudo-sequencer. If you ran an = LFO through a tracking generator and then quantized the output to pitch, = you could get something musical. Will have to try! > From: Andrew Scheidler <xxxxxxxx@xxxxx.xxx> > Sent: Thursday, November 29, 2018 3:15 PM > =20 > Yes, you could have a lot of fun just opening up the Xpander's VCAs = and sending quantized modulations to the VCOs. Instant sequences = without any other gear, not even a keyboard :-) > =20 > Drew (missing his Xpander spaceship) > =20 > On Thu, Nov 29, 2018 at 3:08 PM Brian Willoughby = <xxxxxx@xxxxxxxxxxxx.xxx> wrote: >> I recently noticed that the Xpander / Matrix-12 has the option to = quantize individual modulation =E2=80=9Cpatch cables.=E2=80=9D I think = that the CV values are 14-bit resolution, and I=E2=80=99m not precisely = sure what amount of quantization is enforced when Quantize is enabled, = at least not in bit depth. >>=20 >> Reading the Owner=E2=80=99s Manual, it seems that Quantize is applied = on the final output, rather than the source. The result is that = quantized modulations of pitch end up jumping between semi-tones. >>=20 >> The default routings, such as Vibrato, do not offer a Quantize = option, so you=E2=80=99d have to use one of the 20 modulations to apply = quantized LFO (Vibrato) to Pitch. >>=20 >> Brian >>=20 >>=20 >> On Nov 25, 2018, at 7:18 PM, Glenn Gregory <xxxxxxxxxx@xxxxxxx.xxx> = wrote: >> > It actually would be kind of a cool option to be able to run a = modern digital LFO with 8 bit (or lower) stair-stepping, just for fun. I = can=E2=80=99t recall ever seeing that as an option. >=20