I am glad Aleksander does not put complete stock into frequency response measurements. On another forum there is a guy who is determined to denigrate a particular brand of headphones and uses FR graphs solely in an attempt to prove his point. Aleksander points out the folly of using such graphs as the only measure of a headphone's sonic character. A good tool, yes, but not completely definitive. After all, we do not call up our friends and say "you must come over and see my new headphones. You cannot believe how well they measure!"
As a tool for proving the product quality, FR is pretty much useless, until you engage other measurements that explain what FR shows and why it shows it like that.
Simply, you can have a ruler-flat FR response that sounds utterly dead. Yeah, great tone balance, but dead, so now what? Good for playing music?
Simply, there are headphones that will probably measure better FR (not many, I think) and many more that will measure worse, but in the end, it's not about that.
Here's a very recent example:
Simone's Immanis measures the same before and after he got the new Lampizator DAC.
And I'm pretty sure that Lampi DAC does not have tilted up low-end or dropped down HF, or anything like that in it's FR.
However, now he doesn't need to engage EQ to get what he's looking for.
And we're talking 3-6 dB both up and down that he had to use in EQ to get what he needed.
So, the fact that Immanis' FR has small enough variations doesn't really matter for him. It wasn't until the whole character of the sound was changed by the DAC, before he could accept a pretty flat response.
What I did with FR was a certain manipulation, but within a tasteful measure. Nothing overdone or undercooked, I don't want to ruin things, just to get to point where inherently flawed kind of listening, with ears surrounded with donut-on-a-cup-like contraptions, starts to sound more realistic than usual for such contraptions.
The rest of the SQ, which is much more, is within the inherent SQ of the drivers that don't show on FR.