A sticker on my microwave oven states its output effect to be $750\ \mathrm W$, which is $180$ calories per second. This means that heating $250\ \mathrm g$ of water by one degree Celsius would take $250/180 = 1.4\ \mathrm s$.
Now, my cup of $250\ \mathrm{cl}$ coffee had cooled to $35$ degrees, so I figured heating it to $45$ degrees would take $14$ seconds. But the actual temperature after that time was only $40$ degrees.
I did not measure the coffee volume exactly, and the thermometer probably has an error margin, but still the $50\%$ deviation makes me think that I've failed to take something into account. The porcelain cup should not have been heated much, and $14\, \mathrm{s}$ is probably too short for much heat to transfer from the coffee to the cup.
Best Answer
Your microwave creates a standing electromagnetic wave inside itself but it doesn't consume much energy to create this wave, so if the oven is empty it will only consume a small amount of power - a perfect microwave oven would consume no power when empty. Power is only consumed when you put something in the oven that absorbs energy from the standing wave. The oven needs to take energy from the mains to replace the energy absorbed by whatever's inside it.
The rate at which something inside the oven absorbs energy depends on what you put in. For example if you put in a cup of cooking oil it will absorb energy slowly and heat up slowly (I wouldn't try this at home as traces of water in the oil can boil explosively!!).
The 750W rating of the oven doesn't mean it pours 750W into whatever is inside it; it means that's the maximum amount of power it can pour in. The actual power absorbed will typically be less than this and possibly much less.
It would be an interesting experiment to try heating two (or more) mugs of coffee at the same time. I would bet you'll find the total energy absorbed increases as you put more mugs in, up to the 750W limit.
Later:
Prompted by Anna I have done the experiment. I used two identical coffee mugs containing 400g of water each. I first heated just one mug and measured the temperature rise every ten seconds from about 10C up to about 25C - I didn't want to go higher because you have to start worrying about heat loss to the mug and air.
I then replaced the water and heated the two mugs together, spaced as widely apart as possible, and again graphed the temperature as a function of time again up to about 25C. The results? Well my oven is rated at 600W and with one mug I measured the rate of temperature rise to be 481W (plus or minus a few percent). With two mugs the rate of temperature rise was 530W.
Now I'll just post the results up to the arxiv :-)