Let’s recap this all: the posts above don’t prove wrong the assumption that this mike is doing pretty well in some ways. And it seems to do level measurements from 20Hz to 30kHz quite OK. That is something and could (like in other places) be investigated further.
let's review the reason why we would need calibration for a small capsule omni electret.
absolute sensitivity influences:
diaphragm to backplate distance variation
fet drain source capacitance variation, fet forward transconductance variation when in common source configuration, fet drain gate voltage modulation (cause for distortion and gain variation)
high frequency response is also determined by diaphragm stiffness, and mic front grill design.
I have designed a panasonic capsule source follower with stabilized and bootstrapped drainvoltage such that electronics do not play a role in the frequency and gain response. Then even a panasonic 61 capsule showed a class1 behaviour .
absolute sensitivity influences:
diaphragm to backplate distance variation
fet drain source capacitance variation, fet forward transconductance variation when in common source configuration, fet drain gate voltage modulation (cause for distortion and gain variation)
high frequency response is also determined by diaphragm stiffness, and mic front grill design.
I have designed a panasonic capsule source follower with stabilized and bootstrapped drainvoltage such that electronics do not play a role in the frequency and gain response. Then even a panasonic 61 capsule showed a class1 behaviour .
The beheringer curve looks good. The difference are small and having a cal file is only a starting point if you want absolute accuracy. There are a number of factors that will increase the uncertanty of the measurements. Everything from ambient temperature, humidity and pressure affect the accuracy as does the superposition accuracy. And even then if the mike casings are not identical they will affect the measurements at higher frequencies. NIST level measurements of class zerouncertanties are in the .3 to .5 dB range in ideal conditions. In practice .5 dB is very good. And above 10 KHz don't expect 1 dB from measurement to measurement with the same stuff if you move the hardware even slightly.
For distortion keep the level constant and move the mike to 2X distance and see if the distortion falls. If it does then you are looking at mike distortion. If not its driver distortion.
For distortion keep the level constant and move the mike to 2X distance and see if the distortion falls. If it does then you are looking at mike distortion. If not its driver distortion.
Someone here has done a comparison of 7 ECM Pro samples: https://www.diymobileaudio.com/thre...ics-7-behringer-ecm-8000-vs-7-ecm-pro.473008/ Looks fairly consistent.
Looks way better as the old one - as expected but great to see.
Seems to be a useful & cheap measurement mic.
Seems to be a useful & cheap measurement mic.
I haven't seen any mention of the minDSP UMIK-1 and UMIK-2 mics. Are they not "audiophile" enough?
not yet, i did modify the panasonic capsule into a source follower version to achieve that., will try to find the design files, it has been a while, wanted to sel, the design to neutrik, but they did not believe the class1 behaviour... I did test the sensitivity using a piston calibrator at -20degr , 25 degr and 50 degr. could not to a freq response test, sadly.
No, that’s not what I said.
For context, statistical analysis and research were a very big part of my field while studying applied physics.
So I think I am safe to say that I have quite a good oversight how this works.
In fact, I have done experiments and projects for institutions that specialize in these kind of things.
From that background, I can say with confidence: the data here isn’t objective in a meaningful sense.
With a sample size of one or just a few, we can’t (and shouldn't!) draw any statistically significant conclusions.
So if we’re concerned about misinformation, it’s important to keep that in mind.
There’s no formal hypothesis, and I don’t believe I’ve made any unsupported claims.
If we want to get really nitpicky, the data you're showing is only valid for that one specific product, under those exact conditions, and in that particular situation.
IF, and only if, a company has a solid track record of delivering consistently high product quality, we might treat it as a meaningful indication.
But let’s be honest, I don’t think anyone here would argue that Behringer fits into that category, especially at that price point.
If you would show me some kind of certified GRAS microphone, sure.
That said, I’d much rather have a constructive discussion than let things turn personal.
Those kinds of exchanges rarely help anyone move forward.
So I am looking forward for something constructive, otherwise I will leave it at that.
I think people can draw there own conclusions.
Well, we would need statistic proof on that, wouldn’t we? 😉 But: back on topic!