Mics and Frequency Response Charts

[ithoughticouldrelate]

So what, if any, value do these actually have in making EQ decisions?

I was puzzling over this: you could theoretically (assume we have really good and precise EQs) EQ an SM-57 to have the same frequency response curve as a U-87, yes? But I have a feeling that if you A-Bed the two to most people they'd pick the 87 on a vocal over this EQed 57. Or would they? Frankly, I can't see how in terms of frequency response the mics would be any different once you EQed them the same (again so long as the EQ is sufficiently transparent, and given that most frequency responses would only need a cut or boost of a few dB this would be somewhat feasible).

But then why pay $1000+ extra? So that leads me to believe there's something far more than frequency response at play.

From what I could find out those response curves are measured with pink noise in an anechoic chamber. But (AFAIK) pink noise is not variant to the degree that a vocal (different timbres of voice, vocal styles, vocalist technique) or any waveform you'd ever actually be recording. So doesn't the frequency response graph leave out everything that's actually "useful" quantitatively about a mic (such as its behavior when faced with a more "musical" waveform)? I mean, different sized capsules would have different masses/characteristics it seems and thus more or less "inertia" and so better/worse // faster/slower transient response, right?

So is the "magic" in a certain mic a function of a given engineer's experience and intuition (and this intuition is what you DON'T see on a frequency response curve)? Because when I read about people doing EQ I never hear them say (maybe I just haven't been listening enough, though!) "Well, I just pull up this frequency response chart and level everything out and go from there."

Any comments/setting me straights/advice/etc. would be really useful here! I hope this isn't a stupid topic/question. I'm still very new to all this and have not acquired much gear yet, so I'm doing my best to get a good technical understanding of the numbers sides of things and what actually is/is not possible so that I don't get hoodwinked by a marketing guru. I know that in the end it comes down to "trust your ears," but what can I do in other situations when I don't have that option (besides, er, trusting OTHER people's ears!! )

[Snarl 12/8]

Pink noise doesn't have transients (attack). Transients are where most of the "information" in a sound is. The way you can tell a piano note from a violin note, etc. Transients are very hard (impossible?) to capture exactly. And who cares, the sounds we've come to know, love, covet and emulate are full of imperfectly reproduced transients. Every single mic (even within the same model) is going to handle transients differently. Whether you find one mic more euphonic than another is going to be subject to your taste, expectations, experience, etc. There's way more to sound than frequencies. It all makes a difference, or doesn't. I believe that the place where you really get the most "magic" in a signal chain are in the transducers. Pickups, mics, speakers, transformers, record/playback heads, etc. Where one form of energy is transformed into another. People use different different mics the way they use different pickups, speakers, transformers, etc. To get different forms of magic (whether it's "pristine clarity" or "dirty raunchy fucked-upped-ness") into their sound. Pristine clarity and fucked up distortion (and everything in between) are both magical, highly subjective and elusive in their own way.

[Scodiddly]

There's also phase over the frequency range. Condensor and dynamic mics show significantly different curves. EQ changes the phase curve as well.

But all that aside, it's a matter of trial and error. Even two mics that look very similar on a frequency response graph will likely treat an instrument rather differently. Off-axis frequency response, etc.

[dfuruta]

Besides the transient response and the phase curve, polar patterns will vary (even for two cardiod mics, it won't be the same), and distortion characteristics (how much and what kind) will be different. It's also worth keeping in mind that the published frequency response curves are generally heavily smoothed & averaged, if not totally fictional.

[joelpatterson]

... if not totally fictional.

If that's the case, I'd like to find a mic that was somewhere between William Faulkner and Tom Wolfe.

[vvv]

... if not totally fictional.

If that's the case, I'd like to find a mic that was somewhere between William Faulkner and Tom Wolfe.

Well, of course, those that go to GS know that would be the SM7b.

[Rob Coates]

... if not totally fictional.

If that's the case, I'd like to find a mic that was somewhere between William Faulkner and Tom Wolfe.

I want the Lord Of The Rings Mic. One mic to rule them all, one mic to find them...

[Snarl 12/8]

I was thinking about this thread earlier today. Wouldn't you guys say it would be easier to get something like an earthworks sdc or some other legendarily fast and flat mic to sound like an SM57 rather than vice-versa? I'm thinking using something like transient designer and eq you could get an expensive mic to sound like a cheap one, but not the alchemy of going the other direction, which seems like the only point in doing such an experiment. Has anyone really done that though?

[dfuruta]

I was thinking about this thread earlier today. Wouldn't you guys say it would be easier to get something like an earthworks sdc or some other legendarily fast and flat mic to sound like an SM57 rather than vice-versa? I'm thinking using something like transient designer and eq you could get an expensive mic to sound like a cheap one, but not the alchemy of going the other direction, which seems like the only point in doing such an experiment. Has anyone really done that though?

Wouldn't it be really difficult to adjust the off-axis characteristics after the fact? Seems like one would need to identify where each recorded sound was relative to the mic to do that, unless I'm missing something.

[Magnetic Services]

Also keep in mind that the polar pattern of a given mic varies by frequency. I'm guessing they usually stick around 1k, but I have seen polar pattern diagrams with a thick line representing a midrange frequency in addition to secondary lines (dotted, dashed, etc) representing higher or lower frequencies. This is much of what makes microphones with the same published frequency response different.