As PurpleAngel suggested, a hybrid should be sure-fire which is why I did order the Little Dot 1+ and "tubey" tubes. I'm still new as hell to this but one aspect I was focusing on before buying the 1+ was it's an amp that gives closer to 1W of power for lower ohm headphones, with that power going lower as impedance goes higher. Which seems to be the opposite with OTL tube amps as the output is greater with higher impedance.
OTLs have higher voltage swing/output impedance, generally speaking, and that makes them more efficient/better suited to higher impedance loads. Their current output is generally limited, which is why you see less power at low impedance (this all tracks with Ohm's Law).
1W of output power is just nutters for the vast majority of headphones (the headphones it isn't nutters for can probably be counted on one hand). And I do mean "you can damage your headphones and more importantly destroy your ears" not just "its over specified." Of course that doesn't mean you're drawing all that power out all the time, just that it isn't needed.
Okay, that does make sense. Basically, the higher ohm headphones would "trim" some of the frequency response if I follow you correctly?
Output impedance will interact with the reactivity of the load. So if there is a very reactive load (Sennheiser HD 600 has been mentioned, and is a good example) its frequency response will be impacted by changes in output impedance (this is measurable and modelable), here's an example, from PersonalAudio.ru:
I wish the scale was a little finer, but they're showing around a 2 dB rise with the 300 ohm output. This doesn't matter if the output impedance is the result of being an OTL, or just high value resistors on the end of a solid state amplifier.
If the load is not reactive, however, the frequency response largely doesn't care and all higher output impedance will do is diminish efficiency (its basically "getting in the way"). Here's an example of a non-reactive load, a planar magnetic;
** For the peanut gallery: I know I'm using the word "efficiency" wrong here, and it should be "power transfer" instead, but "efficiency" probably makes more sense for most people.
A couple of years ago now, some very vocal folks latched onto this idea that "output impedance needs to be 0 because anything else represents error" based specifically on how the Sennheiser headphones (as reactive loads) respond to changes in output impedance, and that idea seems to have taken root on a large scale. The problem is that argument is a slippery slope: who gets to arbitrate what the "truth" should be here? It's an observable change, but that's where objectivity stops - the event can be seen and described, and has been. How you feel about it or whether or not you like it or desire it is something else entirely. "Damping factor" really should have no place in this discussion (or any discussion if I'm being entirely honest) - its a marketing spec borne out of a bygone era that's been appropriated to push a specific agenda.
To your original question: I don't actually have an answer, and its something I've wondered about myself as well. My guess is that if the amplifier has enough current delivery and the headphones are sensitive enough, it can probably get by well enough, but I'd be nervous about asking too much current from the amplifier depending on the cans (without more information about the 336SE's output specifications this really can be written off as idle speculation, and that's really what it is: I don't own the 336SE, and I'm speaking in generalities - maybe it works great, and at least one response to this thread seems to indicate that is the case).
Oh, and you don't need to say "higher ohm" or "lower ohm" or "ohm rating" - you can just say "impedance."
