Here (not really a dip but lack of sub bass):
My graph begins at about 88 dB and rises up to 91 dB by 35 Hz.
Seems like this pattern is typical for units manufactured prior to Fazor production.
Nevertheless, the lack of sub bass is not as bad as with the Fazor models?
My graph for a september 2013 model, which was assembled in june 2013 (anyone here with fall/winter 2013 models, pre-fazor, mostly had theirs made around summer time) goes all the way flat to the lowest sub-bass, but where your treble areas cut a bit lower, like the 10k spike, mine cut higher, only by that small 3dB +/-
I compared my graph to the ones posted after your post, and I noticed that at the end of the day it doesnt matter, because where one graph does better in one frequency range, +/- 3dB or so, another does better in frequencies the first graph doesnt. it ultimately means the headphones will sound pretty much the same to us, because they compensate, to end up making the same sound. so these little differences dont actually mean much. in regards to a difference between pairs.
the fazor drop off at the sub bass is more visible and consistant, thats simply what it does to the frequency. Tyll claims to hear a slight change in the overall sound, because of the fazors, claiming it is slightly less audeze house sound, pre-fazor, and closer to lcd X neutrality, but without being as efficient of course. I would figure we shouldn't care about more, or, less bass, just that the overall sound is slightly changed, SLIGHTLY lol. Still I'm glad I got a pre-fazor model since I prefer the older sound, vs more neutral sound, and these models just months prior to fazors, would be the best version of the lcd 2 that they could do without using fazors, and technically will forever be this way, if they always include fazors in every future revision or model. So thats cool lol.