Oh dear.
Ben, you seem to have forgotten all about Physics. Air weighs 1.2kg per cubic metre, or 1.2g per litre. Saying something is heavy compared to air is just crazy. How much air?
How much distortion do you expect when you try to accelerate something weighing a few tonnes?
Like a car, or maybe a newspaper printing press. To accelerate anything, you need force. Heavier things need more force (Newton's 2nd). So long as the driver's motor is enough to control a heavy cone, what's the problem?
It's interesting that you say they're dinosaurs. Can you show me an old-fashioned light-coned driver that'll do what this can? Data-Bass
The moving mass is a kilo, by-the-way, but the output density (noise per cubic foot) is like nothing else. It'll also sink a couple of kilowatts with 1dB of compression.
Distortion measurements are there, too, and I suspect any of your preferred drivers would've turned themselves inside-out before putting up the SPLs that driver can in a sealed box.
Chris
We sold our mic business so I have no one to ask anymore. We did a 31 element array and I assume it was compact enough to still be smaller than the popular 1" mics. True story (as I was told) - Sales arranged a test along with one of the large European mic manufacturers at a recording session with Rihanna. She asked to keep the only prototype because she thought it was "perfect" for her voice and they were too embarrassed to say no.
I'm perhaps too curious but, could you name your company please ?
I'm also often victim of this strong cognitive bias, my preference goes on prototypes because this is an adventure.
PS : 5 years of apple "think different" ads on the TV... our poor brains.
All arrays will alias, it's just at what frequency. A small array of an inch or so won't alias below about 20 kHz. It's not likely to be a problem.
scott wurcer seems clearly suggest that they can't compete with large membranes microphones ?
Why ?
Last edited:
I would say that performance wise that would be true, but cost wise, not even close. I can't imagine how low cost MEMs have gotten to be, but it can't be much based on the volumes and the forecasts.
I used to argue that MEMs mics could never reach the performance of a larger electret (which is clearly true.) But then I was thinking only of a single one-v-one application. I did not envision the low cost that they could reach in very high volumes such as today and allowing the use of an array of them.
I used to argue that MEMs mics could never reach the performance of a larger electret (which is clearly true.) But then I was thinking only of a single one-v-one application. I did not envision the low cost that they could reach in very high volumes such as today and allowing the use of an array of them.
She certainly has a "Presence".
Part of which includes,
"The body of Venus/Athena/Diana,
With arms ..."
I disagree with the first part. My main experience with sound is live audio. When mixing on low distortion subs the difference is immediately apparent, if you know what to listen to. First off, the bass notes are more readily noticeable one to the next. Each note the bass player plays is distinct. But more importantly are the harmonics. Since a subwoofer has higher sensitivity in it's higher range, this means the harmonics are in the higher sensitivity area. Subwoofer harmonics can thus gum up the vocal range. Low distortion cones, like BMS with shorting rings will lead to a cleaner mid range. Also band pass lows will be do the same, as the harmonics will get *passed* out, since the box will not reproduce them.
To the second part, this is unfortunately true in live audio, to the uninitiated. We did sound for a country rock band one time and the sound man did not like the sound of the low distortion subs. We added a small front loaded powered speaker to the sub output, with only hundreds of watts, compared to the thousands of watts in the main sub horns. The powered speaker was set to go into clipping every time the drummer hit the kick drum. This gave the sound man the *guts* he was looking for, and he was happy.
I think this right here is key. Making, and the fact that it is not linear, but a slope. As long as the distortion is under the slope, it is masked, or un-noticed. The distortion in the second example does not fall under the slope, but rises above it, and is noticeable.
I would quibble with this based on crossover points. Yes, distortion would be more noticeable the higher in frequency one goes, but, the harmonics from the midrange might fall into the highly noticeable range, and the harmonics from the tweeter might fall into the above human hearing range, and thus, not lethal.
For a single MEMS element it's just physics of scale the mechanical impedances i.e. loss mechanisms scale just like any loss (resistive for instance) while the generator capacitance falls (as area). The net result is that given real materials that you can actually fabricate something out of there is roughly a square root relationship between noise and size (1/2 the area 3dB worse noise).
Distortion is more complicated, early MEMS mics used a MOSFET in deep subthreshold and biased the input with back to back diodes. This means there is a hard max limit on input voltage from the generator, we got out of the business before completing some experiments on true charge amplifiers at the input. IMO the MEMS guys only had to target the cheap Panasonic type capsules and when they equaled the performance they were done and many of the conventional capsules have gone by the wayside.
I think we are mixing in the enormous market for voice quality mics with recording mics. There is little value added by increasing the performance in this market which is absurdly huge, every device there is has a mic these days. The same is true of the cameras in your car vs a top DSLR. The car-cams probably are well under $1 and produced by the 100's of millions.
EDIT - I'm retired from ADI, and we used to make MEMS mics but no more. I was never in that group just consulted internally on occasion.
Last edited:
It's a Beyerdynamic Classis RM 31 Q, the data sheet says classic electret elements, but the smarts here is clearly to make huge meetings more intelligible. The SNR and dynamic ranges are fairly ordinary.
THIS!
I think one thing that confuses people is that they are used to thinking of everything in sound as separate, when everything is combined. The ear only hears one signal. The brain separates it into different components.
If a piccolo and a bass drum are being played on a stage, there is not two separate sounds that meet the ear, there is one complex sound, and the brain separates them, and identifies for itself what portion is the piccolo, and what input the bass drum has.
These instruments also have the advantage of zero distortion, in the fact that they are the source. If you cut a slash in the bass drum, and it rattled, it would still not be distortion, but source.
The speaker system is the exact opposite, in that it has to get everything exactly the same, or it is distortion. The sound quality, by definition, can only degrade, it cannot improve. Any *improvement* is itself distortion.
I think there are lots of non-linearities that can be measured, that are chalked up to doppler, but are in fact other forms of distortion.
If you were to look at the instance of the piccolo and the bass drum, there would be huge wiggles in the signal when the bass drum was pounded, and the piccolo notes would look to change, when in fact if the piccolo pitch were measured, it would not change, it would just *ride* on the bass drum poundings. The same happens when amplified. Same sound, only louder. The measured distortions are not doppler, but other factors.
Even in a leslie, the magnet is fixed, there is no pitch shift, compression or expansion, only vibrato. No doppler.
There was a reference early in this discussion to alnico, and it's timbre. This is an underdiscussed portion of this thread. There is a big difference in magnets. Not only in flux density, but in variations in flux throughout the power band. Alnico is known for having very stable flux density.
Also the main attraction for field coils is the fact that they have very stable flux density. There is a discussion going on about microphones, and I am surprised that this has not been compared to speakers. It is well known that generally speaking, dynamic mics do not have as much precision in movement, (not frequency response), as much definition, as condenser mics. This is the same factor. The condenser mic has a 48V (typically) rail, that is applies to a coil in the microphone. This is a very steady field for the diaphragm of the mic to work against.
The best sounding speakers I have ever heard were field coil speakers, and they were not the flattest frequency, nor the widest frequency response speakers I have heard. The biggest downside to field coil speakers is heat dissipation.