Quote:
Originally Posted by
fuzzy1969 
Ah ok, frequency response graphs rely on frequency sweeps and measuring the sound pressure levels at the other end, which is fine for say dynamic drivers which have no crossover but with multi BA which have crossovers which filter the bass,mid,treble to appropriate BA it doesn't mean a lot, also the crossover is the main reason why multi BA's can change impedance at given frequency's and music contains a multitude of different frequency's something else frequency sweeps don't allow for. edit sorry for not being specfic but Ive just woken up.
Well, as to the first part of your answer, there's something else that measures sound pressure levels "at the other end" -- the human ear. Some people would say that dynamic drivers are indeed fine for this
Some of the problems with multi BA and crossover set-ups will indeed show up on a graph, just as they can be heard. For example, a flaw in the crossover tuning of the C4 leads to a dip or suckout in the frequency response that the reviewer LFF heard and this was later seen in the graph. Therefore, a frequency response graph is far from meaningless.
As for the second part, I've just shown that a frequency response graph can convey meaningful information if done right. That it may not be right to measure all aspects of an earphone's sound production doesn't make it meaningless. You may be aware that there are several different types of ways to plot an earphone's sound. You can see purrin's website for more information.