Nope, didn't create a filter from the ear-enabled measurements. Probably wouldn't have been the exact same given that the earless filter didn't filter the eared measurements to quite a complete match. Thing is though, when I measure without the ear, the microphone is looking down at the center of the baffle, whereas with the ear it's moved slightly down off center (due to the anatomy of this particular ear). That could account for the channel difference in the treble in the filtered results - but even with that slight offset in mic placement the filter gets it almost right, which is pretty good. Possibly it means that this channel balance fix is independent of ear anatomy, unless one has asymmetrical ears to some large degree or if one of the drivers has some serious issues. But it may also be that this only applies to half-open over-ear phones, which are what I'm testing.
I've also done some experiments on matching frequency responses on two phones based on only taking the difference in response from earless measurements, i.e. in a way that is independent of ear anatomy. It works to some degree, though it's always a bit off. Need to experiment on positioning, etc.
Edited by vid - 10/21/12 at 11:25am