Sorry, I realize that seems cryptic, like all of the information surrounding these mics. I was just trying to clear the waters, because I actually know someone that owns two of the mics. No bullshit, and I am in no way related to Mr. Simpson or his mics. I've never even heard them. I was just trying to get across some information from an objective standpoint.

Uncle Ovipositor wrote:What, exactly, is a "reverse equalizer"?

The mics' frequency response is very non-linear. In the design, he decided to sacrifice frequency response for "time coherence," or something like that. Basically, they're extremely time accurate, and not especially frequency accurate. So, to compensate, you equalize the sound after the recording (or maybe you can eq before - I'm just going on what I was told).

In short - on a two microphone recording of an orchestra, when the timpani hits, it usually sounds like a slow-attack thud. With these mics, even at a distance, the timpani strike was clean and crisp, like it is in real life.

I hope this clears up any confusion.

And Uncle O., just because something is unclear to you, you should not decide that it is bullshit. That kind of attitude is unbecoming.

danielruder wrote:

Uncle Ovipositor wrote:What, exactly, is a "reverse equalizer"?

The mics' frequency response is very non-linear. In the design, he decided to sacrifice frequency response for "time coherence," or something like that. Basically, they're extremely time accurate, and not especially frequency accurate. So, to compensate, you equalize the sound after the recording (or maybe you can eq before - I'm just going on what I was told).

So, by "reverse equalizer" you meant "equalizer". Got it.

In short - on a two microphone recording of an orchestra, when the timpani hits, it usually sounds like a slow-attack thud. With these mics, even at a distance, the timpani strike was clean and crisp, like it is in real life.

I'm not sure I've ever felt like a recording of a timpani lacks attack or dynamic range, but I get your point. That this is being discussed as a result of "time domain performance" is interesting to me. It's something I don't know anything about, and would be curious to hear. How acoustic impedance comes into that is also beyond my pay grade.

And Uncle O., just because something is unclear to you, you should not decide that it is bullshit. That kind of attitude is unbecoming.

Please re-read my post. My issue is not with the mics themselves, about which I admittedly know nothing (in spite of my curiosity) and with which I have no direct experience with. My issue is with the lack of information supplied about them and the dubious way it's dished out. The fact that I've learned more from you (someone who's never actually been in the room with these mics but knows someone who owns a pair) than from the manufacturer is a bit of a head scratcher. And the use of quasi-technical terms makes it all the more suspect.

When I come up with my new type of microphone, I'll be quite up front with technical information. I will go out of my way to explain how it's different and why you should buy it, and I'll do it in the clearest language I can.

And really, Daniel - unbecoming? Watch the condescension, bud. There's nothing becoming about me to start with and implying otherwise is downright rude.

I don't really understand this... Doesn't equalization introduce phase distortion, hence defeating the time-domain accuracy?

Acoustic impedance matching, as i understand it, means that a material (usually some sort of soundproofing) will reflect as little of the incident sound wave as possible.
At least, that's what I vaguely remember from an acoustics course I took last year. And how it can be applied to transducer design, I have no idea.

It doesn't surprise me that heavy equalisation is needed for these things.
Time accuracy and frequency accuracy are sort of a trade off. When you improve one, the other suffers.

I'm taking a shot into the dark...

when you match something, you essentially match some sort of mode (be it vibration, electrical, whatever) that a circuit or object has intrinsically.. With "acoustic impedance matching", I think they're saying that this thing essentially reacts the exact same for all audio frequencies, with a near-perfect transfer of energy at the proper time to an object.

from what it sounds like, the microphones can detect all frequencies at the same time, however the amplitudes of those recorded frequencies aren't recorded truely, just the fact that you know exactly when they happened. By understanding how you lose the signal, you do the opposite to restore it.

I think Juice is closest so far....

In the simplest terms, matching acoustic impedance between the diaphragm and the air gives a greater transfer of power from the air to the diaphragm, which amounts to simple acoustic gain.

However, this gain is important because it is relative to the physical properties of the microphone, which means that relative to this gain, the general constraints & related artefacts of the diaphragm have gone 'down' by some 15-20dB.

Or, more simply, by raising the impedance at the microphone diaphragm we more effectively restrain errors/distortion such as diaphragm resonance.

Or, you can consider that the mass of the diaphragm (as constrained by sprung restraints) is relatively reduced by the same factor as the acoustic gain is raised, as is the restraining force of the diaphragm restraints. So the diaphragm is (relatively) more free to move against the restraints.

I usually describe this as enhanced damping.

Anyway, whoever said that time domain & frequency domain are traded off against eachother is on the right lines but not quite correct.

As was also mentioned, this microphone is designed to be 'post-equalised' or callibrated. Of course, the acoustic horn does not provide a flat frequency response but what it does is ensure a time-response that is so free of resonance that the post-equalisation is essentially transparent and brings the microphone to flat.

If you take any of my recorded samples and sweep an eq accross them, you will find a very different response to the eq than that of normal recordings. This is due to the lack of interaction between diaphragm resonance and equalisation. Something like boosting a 12k shelf on a great ribbon mic but even more so.

I hope that's not more confusing than it is enlightening! The samples on my site illustrate better the advantages of this time-domain improvement (dynamics/clarity/material resonance signature differentiation/etc).

Andy

PS, there is a little more on these things over at my forum:

http://oswaldsmillaudio.com/forum2/index.php?board=10.0

How about some charts and graphs? I like pictures.

jason smith wrote:I don't really understand this... Doesn't equalization introduce phase distortion, hence defeating the time-domain accuracy?

Actually, although in this context phase-shift (ie. group delay) is not something that the ear is particularly sensitive to, this is a good question.

In a standard application the EQ process does introduce phase-shift, such as is present in many very popular microphones (eg. u47).

However, results can be quite impressive with linear-phase equalisation - which represents something of a watershed for digital technology, since it can't be done in hardware.

Recently, I have been working with Refined Audiometrics Lab on a linear-phase solution for the Model A.

The first linear-phase clips of which you can hear on my site (from a recent location session):

Tabla 96k/24bit

Sitar 96k/24bit

Ensemble 96k/24bit

Andy

You seem to be taking a stab at doing something new with recording technology, and those clips sound very nice.

salut

Andy, what is in the signal path in those recordings? I'd like to hear it compared maybe to a Royer R-122 or some other ribbon mics. For a single mic, ~1000 EUR is not outlandish. Are there any diagrams of its internals available?