Now I know with speakers.. the lower the Ohm #, the "lower" the impedence/more dificult load/less efficient/less sensitive.
So how come 300-600 ohm headphones seem to require more power than lets say... 32 ohm grados?
Shouldn't the higher ohm headphones need less power?
The MKIII is an OTL tube design, so it puts out more power into higher impedance loads. This is somewhat of an exception among amps -- most other designs put out less power into higher impedance. I think the perception that high impedance headphones require more power has to do with the fact that many amps (pretty much all solid state amps, for example), put less power into high impedance headphones. So an amp with plenty of power for low impedance headphones may have a harder time driving high impedance headphones. It's the other way around for the MKIII however -- the MKIII is best suited to high impedance headphones. It can still work with some low-impedance headphones headphones of course, but it may not be ideal. Grados have such high sensitivity that they should be fine, but I can't say for sure.
The impedance of a headphone doesn't tell you much by itself. You also need to look at sensitivity. If you have two headphones with the same sensitivity but different impedances, both would require the same power to reach a given volume. However, the high impedance headphone would take more voltage and less current, and the low impedance headphone would take less voltage and more current.
There is also the issue of damping factor. The output impedance of an amp forms a voltage divider with the impedance of the headphones. Since dynamic headphones typically have changing impedance at different frequencies (that 32 ohm figure was probably measured at either 500hz or 1000hz), they will get different amounts of voltage at different frequencies. To keep this voltage fluctuation low enough that it doesn't affect frequency response by more than 1 dB or so, it's recommended that the output impedance of an amp be 1/8 or less of the nominal headphone impedance. I asked David at Little Dot about the output impedance of the MKIII and he only said that it's "less than 10 ohms." This isn't exactly confidence-inspiring. That ratio I mentioned could be anywhere from ~1/3 to well under 1/8 for your 32 ohm Grados. Perhaps Frank has a more precise measurement that would be helpful...
EDIT: The output impedance changes at different gain settings. See post #10 here http://www.head-fi.org/t/272595/question-about-little-dot-mkiii-gain-settings
Edited by manbear - 1/6/14 at 10:16am