Beta Test: Room Correction for Android (Feature now publicly available)

[slartibartfast]

Never said anything of the sort.
It's a fact that true measurement microphones that measure flat--and have traceable certificates to prove it--do NOT employ DSP. They are expensive because it's actually quite difficult to get flat response 20-20k.
I'd just like to know how Dayton & MiniDSP do it for $39 & $79.
I assume they use cheap DSP chip to correct the major flaws, and supply a deviation sheet so the end user can correct the remaining error.

You two assume their mics are made of fairy dust and provide $1000 performance for $79.
I'm sorry but I'm a realist and suspect things aren't as simple as you assume.

Ultimately it doesn't matter to most. They're certainly more flat than most uncorrected rooms.

Where would these DSP chips get their power supply from? The iMM6 only has four connections. Mic, Left and Right audio out and ground.

 

[guy48065]

I don't assume anything. I'm an EE and i know.

I'm a Metrologist. Good to meet ya.

 

[canard]

Never said anything of the sort.
It's a fact that true measurement microphones that measure flat--and have traceable certificates to prove it--do NOT employ DSP. They are expensive because it's actually quite difficult to get flat response 20-20k.
I'd just like to know how Dayton & MiniDSP do it for $39 & $79.
I assume they use cheap DSP chip to correct the major flaws, and supply a deviation sheet so the end user can correct the remaining error.

You two assume their mics are made of fairy dust and provide $1000 performance for $79.
I'm sorry but I'm a realist and suspect things aren't as simple as you assume.

Ultimately it doesn't matter to most. They're certainly more flat than most uncorrected rooms.

these small capsules are relatively linear, and dayton systematizes a passage on a dedicated, standardized measurement bench, and comes out with an individual calibration file....
it is a very classic approach...but just not common for such inexpensive products

 

[harkpabst]

I'm a stubborn Metrologist. Good to meet ya.

FIFY

 

[guy48065]

FIFY

... Don't forget skeptical!

About that...
So I've been suspicious these cheap measurement mics had some hidden trickery going on in order to be +/- .5dB (per the cal report) over much of the audible spectrum, as has been claimed in this thread.
This morning I downloaded & graphed the reports for my 2 cheap mics (iMM-6 & UMIK-1) and found they are far from flat so I was wrong to conjecture that their errors were corrected by DSP. They would make a very poor choice for any application where the correction file could not be employed.

Edit: swapped jpgs for the pdfs--for those who don't have a pdf viewer app

 

[harkpabst]

... Don't forget skeptical!

About that...
So I've been suspicious these cheap measurement mics had some hidden trickery going on in order to be +/- .5dB (per the cal report) over much of the audible spectrum, as has been claimed in this thread.
This morning I downloaded & graphed the reports for my 2 cheap mics (iMM-6 & UMIK-1) and found they are far from flat so I was wrong to conjecture that their errors were corrected by DSP. They would make a very poor choice for any application where the correction file could not be employed.

I'm very sorry that you got such bad microphones (if you got all that Excel graph trickery right).

My UMIK-1 happens to be within +/- 0.5 dB from 28 Hz up to 2.43 kHz. Perfectly suitable for RC with no calibration file at all.



This is really not so special for cheapish electret condenser microphones. Cheap dynamic microphones cannot cope with such figures, of course.

 

[canard]

.....
anyway...the calibration is there to compensate in fr...that's all...

 

[guy48065]

That "Excel graph trickery" is simply the oem cal report. The actual errors measured by their reference standard.
What is yours?

 

[harkpabst]

That "Excel graph trickery" is simply the oem cal report. The actual errors measured by their reference standard.
What is yours?

MiniDSP do not provide an "OEM cal report". You can download a text file containing calibration data. Do we agree on that?

I don't know what your PDFs (which you now exchanged with JPG files) are created from, but they look like Excel graphs.

My curve is very simply the original calibration text file imported into REW. If you want me to I can upload the pure text file as well. It's amazing how fast you changed your sceptical opinion from "these mics are too good to be true and surely need DSP correction" to "these mics are unusable crap anyway".

 

[canard]

......
This file integrated into our software allows fairly precise compensation, sufficient for our use....
that's all....

 

[guy48065]

MiniDSP do not provide an "OEM cal report". You can download a text file containing calibration data. Do we agree on that?

I don't know what your PDFs (which you now exchanged with JPG files) are created from, but they look like Excel graphs.

My curve is very simply the original calibration text file imported into REW. If you want me to I can upload the pure text file as well. It's amazing how fast you changed your sceptical opinion from "these mics are too good to be true and surely need DSP correction" to "these mics are unusable crap anyway".

Whoa there mr Gaslight. My chart is taken from the same data source as yours and shows the deviations from flat on the oem cal bench.
Is mine atypical?
Is yours?
Looks like the huge peak centered around 7k is the same for both.

 

[slartibartfast]

... Don't forget skeptical!

About that...
So I've been suspicious these cheap measurement mics had some hidden trickery going on in order to be +/- .5dB (per the cal report) over much of the audible spectrum, as has been claimed in this thread.
This morning I downloaded & graphed the reports for my 2 cheap mics (iMM-6 & UMIK-1) and found they are far from flat so I was wrong to conjecture that their errors were corrected by DSP. They would make a very poor choice for any application where the correction file could not be employed.

Edit: swapped jpgs for the pdfs--for those who don't have a pdf viewer app

+/-1dB from 30-3500Hz is fine if you limit your correction within that range. A lot of people only correct to 400-500Hz.

 

[guy48065]

I agree.
At the end of the day don't we all want WiiM to allow use of the oem correction file so we can be more flexible with these less-than-perfect microphones?

 

[harkpabst]

Whoa there mr Gaslight.

Careful with your accusations. You might fall flat on your face. Again.

My chart is taken from the same data as yours and shows the deviations from flat on the oem cal bench.

You picked the best possible way to hide that. Did you use Excel to create those graphs? Come on, give a sound and precise answer. You did, right? That just means that there's one more potential source of error. Nothing, really nothing to do with gaslighting.

Is mine atypical?
Is yours?

I did not pretend to know. Do you? Hint 1: You can download my calibration file by entering my UMIK-1's serial number. Hint 2: You can download the calibration file of any UMIK-1 ever made by just guessing the serial number.

Looks like the huge peak centered around 7k is the same for both.

That "huge peak" is 2.8 dB in the case of my UMIK-1. And at 7 kHz it's completely irrelevant.

 

[harkpabst]

I agree.
At the end of the day don't we all want WiiM to allow use of the oem correction file so we can be more flexible with these less-than-perfect microphones?

I never disagreed with that.

 

[guy48065]

Careful with your accusations. You might fall flat on your face. Again.


You picked the best possible way to hide that. Did you use Excel to create those graphs? Come on, give a sound and precise answer. You did, right? That just means that there's one more potential source of error. Nothing, really nothing to do with gaslighting.
...

Yes if course it's Excel. Doesn't everybody?
It doesn't create errors. It doesn't hide anything. What the hell are you trying to imply?

 

[harkpabst]

Yes if course it's Excel. Doesn't everybody?

My graph should answer your question.

It doesn't create errors. It doesn't hide anything. What the hell are you trying to imply?

Depends completely on what the user does. Don't try to make a straw man out of it. We are still discussing how good or bad these cheap microphones are. I say they're good and provide a graph to proof it. You say they're bad and show some print outs with handwriting on it.

 

[guy48065]

My graph should answer your question.


Depends completely on what the user does. Don't try to make a straw man out of it. We are still discussing how good or bad these cheap microphones are. I say they're good and provide a graph to proof it. You say they're bad and show some print outs with handwriting on it.

Never thought I'd see anyone trust some 3rd party app over MS Excel.
Nevertheless I'm sure if you fed the same data into each they would produce identical graphs.

 

[guy48065]

27 lab mics + misc.

 

[rcon7]

show is on ...popcorn time