Microphone Science

[nopster]

This may or may not be the best section for posting this article, please feel free to move it if needed.

I guess when it comes to selecting and/or classifying microphones, there seems to be lots of vague (to me anyway) terminology that floats around in all of these discussion groups. (Terminology such as warm, bright, good low-end, etc.) While this sort of terminology and similar is possibly helpful to those with lots of experience with microphones, it stills seems to me to be a vague sort of rating system, even if there is no better ways of expressing such things at this time.

I've got a Shure SM58. I researched before I bought it and found out that it was classified as "common, durable, relatively inexpensive, but more suited to live use than serious recording applications."

So then I started wondering what is a typical, common "recording quality" microphone and how does it differ? From reading various posts, if I'm not mistaken, the Neumann U87Ai microphone seems to fit that bill for a common, serious recording studio-style mic, or something similar anyway.

So, next I'm wondering, how do these two microphones differ performance wise ? For one thing (as the more experienced know) the frequency response curve of the U87 is very much flatter over a wider frequency range, which allows more faithful reproduction, with less EQ required latter for "correction".

Is it just that simple a difference? If so, it would seem that you could place an EQ after the preamp section, but before any other section, and EQ compensate so that an inexpensive SM58 would have a flat response curve similar to the higher end microphones. However, my guesstimation is that the large diaphrams of the higher end microphones are more adept at picking up whispers and other subtleties due to their design.

As a result of my curiosity, I started looking at large diaphram microphones and comparing their frequency curve versus their price. I really wanted something close to a U87 in frequency response, but without having to take up prostitution to afford it. ;-) After looking at several models, and offerings on ebay, and potential mic modification kits (Pelso Microphone Lab), I decided to get the AKG C414B-ULS. (Discontinued model.) At $500 used, it seems to come fairly close, frequency response-wise, but at a lower average cost, apparently. Picking a mic based solely on a relatively flat frequency response characteristics may not be the best move, but it seems reasonable to me at my current stage of development. (Besides I can unload it at about what I paid, without having to put out the "red light" to compensate.)

I haven't yet been able to try out the microphone but I am curious to find out how much closer the SM58 can come to competing to higher-end microphones(U87, etc) simply by flattening out its frequency response curve by using post-preamp EQ. Obviously, having the larger diaghram is probably going to make a difference, but how much of a difference?

Anyone attempted similar tests? Or have subjective or not so subjective opinions about how well the SM58, or similar can be poked, prodded, or otherwise tweeked for greatness? I'm not sure I'll have the time or patience for extensive, carefully-controlled comparisons so any feedback is probably helpful.

Once my RNP8380 Pre shows up, it and the C414 will go head-to-head with the SM58 & ART Tube MP V3 Pre. It might be a close race though. ;-)

Thanks!

ns

 

[sleepy]

If only eq could solve all our problems.

Let's take Barry White. Now Barry White had that very nice and deep voice. We could all make lower-pitched voices but most of us would hardly come close to Barry White's voice.

With microphones, we can take the SM58 and eq it as we wish but it'd never have that U87 sound.

Have you ever seen how sometimes mics are sold on pairs? And they speak of testing the mics to make sure that their response is equal, etc... This is because two mics of the same model from the same company will not sound exactly the same either. The materials used to build the mic and very subtle differences will very much affect the sound of the mic.

If you eq, you'd be throwing a log into the fire. You'll most likely find yourself trying to eq out the other mic signal to get that same signal that you are getting from the first mic that you eq'd. Then there are the eq's adding their own signature to the sound.

SM58:

U87:

http://www.neumann.com/infopool/mics/produkte.php?ProdID=u87ai
("diagrams" on the top for freq. res.)

As you can see there. The SM58 cannot capture the lower frequencies that the U87 can capture in this case. So no matter how much you eq you will never be able to make something stand out more that was never there in the first place.

Mics will capture sound differently obviously, just that some will capture more of one thing than others. EQ'ing will just bring out more of what the mic did capture.

Not to say that one needs an U87 to capture nice vocals. The SM58 is definitely capable when using a good pre-amp. It's good to keep in mind that no matter what mic you are using you may still choose to eq anyways. It all depends on what you need. The SM58 will not always work and then again the U87 will hardly fit any situation.

 

[nopster]

@sleepy, Thanks for taking the time for a detailed reply.

In the SM58 frequency response curve you posted, the response appears to drop off by about 10 dB at 50 Hz. (It's a little less than 10 but, let's call it 10 for purposes of this discussion.) I guess I was envisioning post-preamp EQ that would flatten out out the whole response curve along the -10 dB line, if you wanted a flat frequency response from 50 to just over 10kHz. To do this it might take a 16-band, or greater, EQ or similar. Obviously the more transparent the EQ (less signature) the better. So what this has accomplished is an apparent flat frequency response from an inexpensive microphone from 50 to slightly over 10khz.

The obvious drawback here is that this flat response has come at the price of dropping the whole signal by about 10 dB, which then requires 10 dB of gain to compensate, thereby raising the noise floor by 10 dB, if I'm not mistaken.

However, if you only needed a freq response from about 100 to 10khz (in the case of higher pitched singers), you could probably get away with a flat line response around the -3dB level.

I guess the tough part here would be setting up the EQ. It would probably take high quality studio monitors sweeping from 50 Hz to 10KHz placed in front of the microphone, and some sort of FFT spectral analysis to plot the results of the EQ adjustments.

I guess you could even have VST plugins with radio buttons for SM85 modeling, and similar, that gave you presets designed to flatten out the freq response of popular low-end mics. (Which could then be tweeked to taste.) However, I can envision problems with attempting to "mic EQ correct" after any compression, so that's why I'm thinking post-pre, pre-comp EQ would probably be the ideal.

I know the "easy" route is to just get better mics, but I'm not sure how easy this is for many that are on a tight budget.

If I'm not considering something obvious, please anyone feel free to comment.

Thanks!

 

[sleepy]

Antares mic modeler uses the same idea that you have.

In case you never heard of it it's a vst plugin. It allows you to select your source mic from the available list and then you choose which mic you want it to sound as. It also has a saturation option and a few other options.

Of course, there's a good reason why people still purchase mics as they need instead of one and using mic modeler.

So mic modeler already does all the eq work for you. The problem is that pre-amps are not taken into account. And it just isn't as good regardless.

However, if you only needed a freq response from about 100 to 10khz

You're forgetting to take harmonics into account. 10khz will work but you do miss out a lot of subtle detail without those 10-20khz.

Ideally you also want to eq after compression. The signal chain goes in that direction. When using a compressor as an insert on a mixing console, the compressor is also before the eq. You can check it out if you have the signal patch diagram for a console.

The best thing you can do to make a mic sound like another is this. Get an SM58, then if you need an SM57 remove the SM58 capsule and ta-da! (Yes, really actually) DD

So all the eq in the world will not make an SM58 sound like a neumann. But with some tweaking it will sound like a very nice-sounding SM58. This is what I am trying to get accross, you don't have to have a Neumann to have a good recording.

There are Rode's out there which many people like. Shure makes some condensers which many people enjoy also. Well shure makes a lot of industry-standard mics. They're not Neumann's but you will literally see them all over the place in Live setups and many many studios. Some people will sound better on an SM58 than on a $1000+ mic. This is taking into account a good pre-amp but still.

Here's a link that favors mic modeler:

http://www.proaudioreview.com/par/jan01/ANTARES-AMM1-web.shtml

(hardware and software versions available)

There was another that I read a while back that reads that MM doesn't make one mic sound like another. But it does help give it a different characteristic. This can be favorable in many cases. use EQ as a tool. EQ is a very good tool when used properly. You have the concept in mind. Forget about Neumanns and go for what sounds good to you.

Edit: tags