Muff's Modules & More

[fac]

[BKehew]

[drip.feed]

[robotmakers]

I have all the respect in the world for Mr. Peake, who has contributed tirelessly to synth discourse for a heck of a long time. I always look for his posts as among the best thought out and most interesting. I almost always agree with him. He has always been a champion of this particular issue, and has been quite consistent in his stance over the years. I have to offer a different point of view, however, based purely on my own experience and with a strong desire to NOT lessen anyone else's opinion. (I really hope that was enough of a caveat).

I have one 901 and four 921's. If I listen to a static saw from the 901 (alone) and a static saw from a 921 (alone), I personally don't hear a difference. My 901 tracks extremely poorly, so obviously there's a huge detuning impact over a range of notes, but I never got the big thrill of the 901 "sound" at a single pitch. I really wish I did!

Your mileage will DEFINITELY vary,
Roger

[drip.feed]

Maybe it isn't the raw waves that sound different? Maybe the 901s come into their own when modulated or something? Do the 901 and 921 sound identical when PWM or FM is introduced, for example?

Or maybe it's the combination of 901 + the equally vintage filters and amps that 901s are routed through that all adds up to making a difference?

[ericD13]

[Peake]

Look at them. They're actually serious.

[drip.feed]

[JohnLRice]

You may have seen these vids before. I listened to them a lot when I was first getting back into modulars! VCO's are all 901's I believe.

[robotmakers]

Here's a snippet of text from the late great Juergen Haible's page on his Living VCO design:

"Early Moog Modular VCOs: The have a linear (Hz/V) oscillator core, driven by an exponential converter. An offset voltage at the point where the two parts interface will cause what I've called "linear detuning" - the possibility of being slightly out of tune that increases towards the bass range, where offset voltages have more influence than on the upper range. (Later Moog Modulars don't have this: They have an exponential VCO core, i.e. the linear ramping of the integration capacitor is controlled by an exponential current source. Offset voltages don't matter in that context.)"

This would tend to support a theory that a bank of 901's will sound fat and detuned in the bass while being relatively more in tune in the upper octaves. JH designed his Living VCOs to incorporate this behavior. This could be ONE way that the 901 and 921 VCOs (in banks) could sound different, while a single VCO of each (alone) might sound similar to some ears (e.g. mine).

Interestingly, it appears that this could explain one reason why Taurus I pedals are prized over Taurus II pedals. Again, from JH's excellent website:

"Moog Taurus Pedals also have a linear (Hz/V) oscillator core, driven by an exponentially scalled keyboard divider. Offset voltages are a source of possible "linear detuning", just like in the early Moog Modular VCOs. (Taurus II doesn't have this - it has plain exponential V/Oct VCOs.)"

I hope this moves the discussion forward in a polite and useful manner.

Best wishes to all,
Roger

[CZ Rider]

[drip.feed]

[drip.feed]

CZ Rider, how did you make that patch? The closest I could get with two DotCom Q106's was this:

[s]http://soundcloud.com/dripfeed/q106-test-1[/s]

Two Q106's, one octave apart, detuned slightly against each other. I also had to filter it through a Q150 transistor ladder (no resonance) to get it sounding as mellow as yours.

[CZ Rider]

[BKehew]

[drip.feed]

[robotmakers]

Interesting pieces of information. Glad the discussion is moving forward. Three points came to mind:

(1) I believe the "Linear detuning" described by J Haible refers to a characteristic of a VCO with a linear core (vs. an exponential core). As a result, a VCO with a linear core will respond differently to offset voltages. According to JH, these offset voltages, which may creep in for who knows what reason, will tend to have more influence on the tuning in the lower ranges, and therefore give rise to a detuning effect in the bass and NOT give rise to a detuning effect in the higher registers. I believe this would occur WITHIN a bank of 901s. This is all quite a different rationale than a patch where there is a constant difference (measured in Hz) between two VCOs, which is what CZ Rider had described. JH wrote that what he described as "linear detuning" is not possible with exponential core VCOs. He attributed the source of the magic sauce of the 901s to their linear core.

(2) Differences in the amount of sharp cancellations as a result of the destructive interference of the waveforms in multiple VCOs will be due to two things: (1) possible differences in the shapes of the waveforms across VCOs within a bank (not too likely) and (2) the distortion characteristics of the mixer, filter and VCA chain following the VCOs. The old CP3 Moog modular mixers are famous for their distortion, and will likely have a big impact in this, rather than something to do with the choice of VCOs.

(3) As for BKehew hearing the difference between the 901 and the 921, that could be down to the calibration of the waveforms, or the output levels, or a lot of things and I don't doubt for a moment he could hear a difference. My personal opinion (FWIW) is that inherent instability of the frequency or the randomness in the waveshape in the 901 probably plays only a small role in any difference. I agree with J Haible, who on his Living VCOs page wrote:

"A good part of that special sound of early Moog and EMS oscillators is not because of any "randomness", "unstability", "instability" or "noisyness", as so often is said."

In any case, wouldn't it be great if someone produced a 5U version of the JH Living VCOs? It would seem as if that might provide a measure of the magic mojo of the 901s, while being easy to calibrate and offering modern features such as PWM.

You mileage will DEFINITELY vary,
Roger

[BKehew]

[spinach_pizza]

[megaohm]