The discussion is pretty close, but not exactly right.
A full parametric EQ lets you control the center frequency of a tone control, the bandwidth or group of frequencies that the filter effects, and the Q factor or the sharpness of the filter's effect. The type of a filter is also a factor - these parameters define a bandpass filter that lets a range of frequencies through and attenuate anything higher or lower than the passband; shelving filters (a high- or low-pass filter) affect only frequencies below the passband or above.
The Alembic filters with Q-switch are semi-parametric. The knobs and switches let you select the center frequency and the Q-factor, but not the bandwidth which was preset at the factory. The filter is also configured as a low-pass filter.
The effect of the filter is like there's a resonant peak at a certain frequency. Everything below the resonant frequency passes through to the output. As you turn the tone knob, this resonant frequency is moving around. When the Q-switch is set at 0db, the circuit acts pretty much like a passive tone control When you flip the Q-switch to the +9db position, there's a bump introduced into the frequency response of the bass - lower frequencies are unchanged, notes at the center frequency are boosted by 9db (a lot, that's 8x the voltage!), and notes higher than the passband are filtered out. There's no knob to tweak the bandwidth, but the Q switch is making a peak of 0, +9, +15, or whatever db in this range. This radically effects the way you hear the sound. This is also why the effect of the Q-switch is expressed in dB - it's the ratio of output levels in the affected area, and that's what dB are for.
If you flip the Q-switch to the high Q position and turn the tone knob you'll hear a sweep like a wah-wah pedal. That's the resonant peak sweeping through the frequency spectrum of the instrument.
Obviously you can build a full parametric EQ. On the instrument it probably didn't make sense to do that - it's hard enough to grok the effects of the controls that are there, much less add additional ones that widen the tonal effects. If there had been Q- and bandwidth-switches, it would likely have caused a lot of confusion among users. The circuit also is only set up to have a resonant peak. It could easily have a notch instead, sucking out the sound in the passband, but again, it's getting really complex and not very user friendly.
There's other more obscure effects as well - for instance, active tone controls can cause a phase shift that depends on the frequency. Even beyond the parametric characteristics, this is one of those things that most people can easily hear but probably can't describe in any meaningful terms.
David Fung
