The world of power requirements for audio reproduction can be a bit confusing. I think the main reason for this is the use of the continuous RMS power output as the benchmark measurement (not that I necessarily have a better one mind you). But for audio what is important is the peak power available for the dynamics not the continuous RMS output. The dynamic power available is determined by three factors; the continuous RMS output capability, the size of the power supply, and the amount of capacitance storage available. With a big power supply and lots of capacitance storage you can safely use a factor of 3 times the continuous RMS output as a good number available for dynamic output (this multiplier could be higher as the impedance of the speaker rises above 4 ohms). Keep in mind that for virtually all receivers and even some big power amps the multiplier for dynamic power is likely only 1 or just slightly more than 1. Consider you want 15dB of available head room for dynamics. This means you will need 32 times the continuous power to achieve it. So it we use the ADA1500 as an example we would have around 2,000 watts available for dynamics (650 x 3). Dividing this by 32 gives us 62.5 watts continuous. So we don’t need the 650 continuous watts available in the ADA1500 for our continuous power requirement, but to have a high enough peak power to allow for listening at 62.5 watts continuous. The last piece of the puzzle is how many channels the power supply can drive to full power at the same time. Once you cross this line both your continuous and peak power will start to drop per channel if all channels require the max all at the same time.
