I thought I would add my $.02 to the conversation, so here goes.

First of all "Digital Amplification" is a misnomer, it is actually Class D amplification that is being discussed here. Although the "Digital Amp" name is becoming common place to refer to Class D it is misleading. Granted they often contain a great deal of "digital" circuitry as part of the design. However the thought that you are amplifying a digital signal just makes no sense. So what do you get bigger 0s and 1s! Furthermore, the input signal is an analog signal as is the output signal and nowhere in the path is there and ADC or DAC circuitry.

No in regards to the efficiency of a Class D amplifier (now this is really stretching my memory back to my college studies taking Electrical Engineering) in theory they are 100% efficient. But like most things theory and reality are often not in line with each other so as a practical matter good Class D amplification is easily over 90% efficient and some of the better ones are claiming in the mid 90s. The reason they are so efficient is a simple matter of electrical physics. A Class D amp (also known as a switching amp) is nothing more then a very high speed switch. The frequency at which it turns on and off is generally 10 to 20 times the highest audio frequency it will reproduce. In order to remove this very high frequency signal the output will have a purely reactive (i.e. no DC current due to a resistive load, it's a purely capacitive and inductive load) low pass filter designed to allow only the audio signal to pass. The majority of the time output circuit is either on or off. When it is off, there is high voltage across the output, but since the circuit is effectively open no current can flow. When it is on, it is a zero resistance path across the output so infinite current could flow, but since there is no voltage (you can't develop voltage across 0 resistance)then no current actually flows. In either case, no power is consumed since Power = Voltage x Current.

Now this begs the obvious question, how does power get transferred to the speaker? Remember the reactive load across the output? Not only is it used to filter out the very high frequency signal created by the rapid switching, but it is used to effectively transfer the audio signal to the speakers. Capacitors and inductors are both devices that store energy. They are either being charged or discharged depending on the switch state (on or off). Again in theory, capacitors and inductors hold there charge indefinitely. They require a resistive path in order to discharge. Well conveniently, the speaker attached to the output provides that resistive path. And I think we all know the rest of the story.

The only other question that hasn't been answered; how is the audio signal used to modulate the switching circuitry so that the final output is an amplified version of the input signal? Without going to deep into that whole process, and there are certainly many different ways to accomplish the task, I will just give you a simple explanation. The basic concept is called Pulse Wave Modulation(PWM). This might be where the idea of "digital amp" comes from in that the PWM circuitry could certainly be called a digital circuit. So without going into the nitty gritty of PWM lets just say it is this PWM circuits that turn the output switch on and off. This does not occur at a steady frequency. It is the varying times that the switch is in either the on or off state the creates the audio frequencies that drive the speakers. The longer that the output switch is in either the on or off state the greater the voltage swing seen at the output, i.e. louder. The difference in the on pules times vs. the off pulse times controls the output signal going in a positive or negative direction. The rate at which these pulse difference are generated is what determines the audio frequencies being output.

OK, now that I took a little trip down memory lane from my EE days, let me just say I am all for the growing acceptance of Class D amplification. After all, it's been around for a long time. Hell, it's been used by a large number of Audiophiles, though they might not know it. Most of today's powered sub-woofers are using Class D amps for driving those big ole speakers! But seriously Class D has a lot going for it:
1) Over all smaller design for same power
2) Less expensive components, therefor same price should get you better quality
3) Far more efficient, so it produces less heat
4) Less heat means small (or no) heat sinks which leads to 1 and 2 above
5) Due to its inherent power capabilities they should handle low impedance loads (2 ohm speakers) as easily as conventional speakers

Therefore, I would love to see Outlaw come out with a Class D based amp!