Watts and power

The difference in acoustic power delivery between amplifiers regardless of design class eventually comes down to the maximum instantaneous peak voltage the power supply can deliver relative to its RMS average power rating. 
Music can demand peak to average power ratios as high as 10 times or more the average output.
Class D amps are limited to the maximum rail voltage of the power supply.  A more traditional power supply with large transformers and very high capacity filter caps can potentially deliver several times the rail voltage on peak demand as the caps discharge before hitting clipping.So an amplifier with deep voltage and current reserves beyond it's nominal rating can deliver higher average power in relation to peak demands than one with a lesser power supply of the same rating.

Class D and linear amps have the exact same rail voltage limitations. They often have the same linear power supply front ends. Certainly audiophile Class D amps. They also get more oomph out of the same sized capacitor bank for real music.


It's torque at the wheels that moves a car, not engine torque. I can always convert more horsepower to more wheel torque with gearing. 


Many consumer brand receivers have poor ability to drive difficult loads and poor damping factor. The more audiophile brands even though rated lower watts at 8 Ohm tend to be much better. 


Everything is designed to a price point. Not doubling in power into 2 ohms (continuous) is often just a factor of the limitation of the transformer. It can only transfer so much power. A nice capacitor bank can give you much of the benefit for real music (not simple sine wave power tests).

Power is the product of the signal voltage * current * the cos(phase angle between them).
That cos term and the ability to act as an ideal source as the load impedance drops, is what makes most of the difference when it comes to balls.
The cos term varies between +1 and 0 as a factor depending on whether the load is completely inductive or capacitive, or something in between, at a given frequency. A purely resistive load will always be a factor of 1.
The impedance, usually simplistically thought of as 8 or 4 ohms, can often vary as low as 1 ohm or as high as 20 ohm, as a composite of resistance, inductance and capacitance.
A good indicator of the balls of an amp is if it can double its’ output power down to loads as small as 1 ohm. Most can’t.
Then there’s the concept of bridging or strapping an amp to make it a mono-block. In the case of bridging an amp, the two outputs are connected in series, out of phase, potentially doubling the voltage swing, but the same current capability, across the two pos terminals. This can result in a potential V**2/R power increase of 4x. Strapping places the two outputs in parallel, resulting in the same output voltage swing but with a potential doubling of the current capability. You might think that this could also result in an I**2*R 4x power increase but it doesn’t because the required voltage level can’t be met.
What this means is that in my opinion, the strapped output results in a much ballsier amp, able to easily drive difficult loads, than the bridging case. The BAT VK tube amps, and Accuphase A200 or A250 comes to mind. If you have a relatively easy load, the bridged case will likely provide greater volume or power levels but will peter out (current starved) if trying for the same levels with a difficult load loudspeaker.

The impedance curve of a speaker takes into account the reactance already and hence the current. Few speakers dip below 2 Ohm. Some do, but they are rare. If your speaker doesn’t dip below 2 ohms than having an amp that doubles in power into 1 ohms is likely paying for something you are never going to use. Their is more to doubling into 1 ohm than just beefing up the supply, and there is no guarantee that the trade-offs , like higher feedback, are going to sonically pleasing. This is why "rules of thumb", like doubling of power into 1/2 the resistance is "better", should be taken with a grain of salt.


If your speaker does not dip to low ohm, then a so called beefy amp will not sound any beefier, but one with higher voltage rails and more capacitance will, even though it can't double into 2 ohms or 1 ohm or whatever artificial metric you want to use that may not apply to your system.

As a very general rule of thumb: I think it safe to suggest that one double the speaker manufacturers minimum power recommendation and to consider amps that can double down that power to that speakers minimum impedance. Again, just a general rule of thumb!