Bombaywalla, Thank you for the link, very interesting. The only problem is that what they describe:
The digital control circuitry constantly monitors the pulsating waves from the regulator and the rail voltages. It will then make a decision to turn the coupling transistors on or off at each zero point to add as many or few pulses as required to hold the voltage constant.
is switching regulator. Any time something is completely on or off it is switching. Idea of controlling number of full wave cycles is not new - it used to be called "Group Regulation" and was even used in such applications as quiet light dimmers. The problem is, that power supply is still "linear" and in order to respond fast it needs low inductance capacitors. The best you can get is "Split Foil" type, but it is expensive and still poor. Adding parallel low inductance capacitor helps, but also creates resonant circuit with inductance of large main caps.
Their regulator is a step in right direction (switching), but why not use switchers? Linear power supply switches anyway at 120Hz at max voltage. Yes, not many companies use SMPS with class A or AB amps but some do (Rowland, Benchmark, Linn etc). I don't understand term used in their paper "Linear Amplifier". I've never heard it before. I assume, they mean "not class D", but class D is still linear (and not digital) in all respects.
