I have a Kill A Watt AC power meter and this thing is such an amazing instrument for the price.
Picture a load like this:
a motor, a TV and a floor lamp using a half wave rectifier to dim. I was surprised that the Kill A Watt was able to correctly measure a combination of load like this that present difficulty even for a true RMS meter. The motor draws a current with cos phi shift, the TV with harmonics and the half wave rectified floor lamp pulls a DC bias.
Standard AC true RMS meter will not accurately measure this, because it ignores the DC contents introduced by the half wave rectifier load. To accurately measure a load like this, you have to measure the true RMS AC current, then the DC current and calculate the total RMS(total RMS amp= DC equivalent ampere to make the wiring heat up the same) current by sqrt (ACA^2 + DCA ^2) or use a meter with "AC + DC true RMS" capability. The total RMS "AC + DC RMS" is only found on very high end multimeters and is can only be found on hall-effect type clamp meter (the current transformer type transduce the DC content). My Kill A Watt appears to take the total RMS into account and reads within in a few percents of AC + DC RMS reading on a Fluke 189(which can't even measure power, yet cost about ten times as much as this gadget)
Now, to be a power meter, it has to be able to measure power. The custom DSP combines the voltage and current to calculate the true power and gives a direct reading of watts and power factor.
Although the accuracy is likely not as good as an analog integrator/multipler type power analyzer that cost a grand or two and I don't know this gadget's sampling rate or bandwidth but the functions are more or less identical.
Anyways, I find it amazing how they can build an instrument with this level of functionality for thirty dollars using a custom DSP.