RoomFit Moving Mic Measurement (beta) - improvements

[dominikz]

I wonder if it affects how RoomFit chooses the target level.

That it might, but given that the difference in response is 1-2dB at most (and mainly above 6kHz) I expect the difference in target level should be much less than 1dB.

 

[canard]

a dayton "cross spectrum" calibration ( your ""6khz"" down), no ?? ;-)

That it might, but given that the difference in response is 1-2dB at most (and mainly above 6kHz) I expect the difference in target level should be much less than 1dB.

examples to illustrate, a little, the classic situation:
cal "cross" ( in rew ) ;-)

(end of cross spectrum, for those uses, general public, it was a really great offer, what a shame. :-( )
,
(on Earthworks , which normally does not provide 90-degree calibration, the drop is less pronounced but starts lower)

(The Umiks and the little Dayton USB remain great bargains suited to these uses... just be aware of their limitations depending on how you use them... ;-) )

 

[dominikz]

a dayton "cross spectrum" calibration ( your ""6khz"" down), no ?? ;-)

examples to illustrate, a little, the classic situation:
cal "cross" ( in rew ) ;-)

(end of cross spectrum, for those uses, general public, it was a really great offer, what a shame. :-( )
,
(on Earthworks , which normally does not provide 90-degree calibration, the drop is less pronounced but starts lower)

(The Umiks and the little Dayton USB remain great bargains suited to these uses... just be aware of their limitations depending on how you use them... ;-) )

Not sure I understand what you were meaning to say, but the reason for the difference in measured in-room response above 6kHz we're discussing now is not principally due to microphone calibration file, it is due to microphone orientation and the way reflections are registered by the microphone (which are not really 100% omnidirectional at all frequencies). I've explained the effect previously here.

By the way, I own two measurement microphones: a Cross-Spectrum Labs calibrated Dayton Audio EMM-6, and factory-calibrated miniDSP UMIK-1.
Their calibration files look pretty different:

But if we normalize the 90° cal file to the 0° cal file we can see the difference between the two cal files is actually identical for both mics:

This is not surprising given that the physical design of both mics is very similar, so their directional characteristic must be similar as well.

As a result, in-room measurements taken with either microphone look almost identical under the same conditions:



We should note that there are of course some take-to-take variations, especially since MMM is susceptible to environment noise (whereas sweep measurement is much less sensitive to noise - but more sensitive to placement).

So as I mentioned before, microphone orientation is not that important - it is however important to use the appropriate calibration file for the chosen microphone orientation:

• Use 0° calibration file if you point the microphone to the speakers (i.e. horizontal), and
• Use 90° calibration file if you point the microphone to the ceiling (i.e. vertical)

For any pencil-shaped omnidirectional measurement microphone the measurement with vertical orientation and 90° calibration file will have about 1-2dB more level above about 6kHz, compared to a measurement with horizontal orientation and 0° calibration file due to the effects I've explained previously here, with examples provided here and here.

Of the two, vertical mic orientation (with 90° calibration file loaded) gives slightly more accurate results at the highest frequencies (compared to anechoic PIR), so if pressed that's what I'd recommend to use.

Hope this is helpful!

 

[canard]

My point was essentially to present the calibrations of the same microphone (made quite seriously) at 0 degrees and 90 degrees...
Just to shed some light on the subject ;-)

(Under 1kHz using 0-degree calibration in 90degres use.. should be fine with omnidirectional mic)

but it wasn't addressed to you...
hihi...
;-)

 

[slartibartfast]

My point was essentially to present the calibrations of the same microphone (made quite seriously) at 0 degrees and 90 degrees...
Just to shed some light on the subject ;-)

(Under 1kHz using 0-degree calibration should be fine with omnidirectional mic)

but it wasn't addressed to you...
hihi...
;-)

Who was it addressed to then? You literally quoted @dominikz

 

[canard]

Who was it addressed to then? You literally quoted @dominikz

The first sentence was addressed to him...hence the quote that follows... ;-)

 

[dominikz]

A Roomfit vs rew ( in "variable") ???
(I think it will be interesting in the future to compare them with those who can implement both approaches, mmm )
;-)

As expected, there's not much difference if similar configuration is used in both REW and RoomFit:

 

[canard]

As expected, there's not much difference if similar configuration is used in both REW and RoomFit:
View attachment 31792

Well, that's rather good news ;-)
Corrections<300hz ?

And
Thank you
;-)

 

[canard]

(off topic..
Although for my part, I am more interested in a subject (I don't use Roomfit) which was brought to my attention by an ASR member concerning DSP etc., which will certainly be clarified with our Wiim (especially if there are corrections in the bass...)

DSP Measurements and Rising Noise Floor

In several threads related to DSP platforms I've seen requests for measurements with filters engaged. I thought it unlikely that implementing filters in a DSP would have measurable effect on noise / distortion other than the obvious change in the fundamental, however @Sokel recently mentioned...

www.audiosciencereview.com

I imagine we'll have to observe it in detail on our Wiims.

;-)
)

 

[Fletcher Munson]

As expected, there's not much difference if similar configuration is used in both REW and RoomFit:
View attachment 31792

So the engine behind roomfit is REW ?

 

[dominikz]

Well, that's rather good news ;-)
Corrections<300hz ?

And
Thank you
;-)

These examples are with corrections <500Hz.

(Although for my part, I am more interested in a subject (I don't use Roomfit) which was brought to my attention by an ASR member concerning DSP etc., which will certainly be clarified with our Wiim (especially if there are corrections in the bass...)

DSP Measurements and Rising Noise Floor

In several threads related to DSP platforms I've seen requests for measurements with filters engaged. I thought it unlikely that implementing filters in a DSP would have measurable effect on noise / distortion other than the obvious change in the fundamental, however @Sokel recently mentioned...

www.audiosciencereview.com

I imagine we'll have to observe it in detail on our Wiims.

;-)
)

Not really on topic here, but I haven't seen any negative effect of EQ/RoomFit on N+D, here's for the WiiM Mini:

So the engine behind roomfit is REW ?

No, I doubt that.
It is just that all of these room correction tools work in a very similar way under the hood - they measure a response and then use PEQ to match the response as close as possible to a desired target.
So with similar configuration (e.g. similar correction range, target, smoothing and filter parameters) they all should produce a similar response.

REW is more powerful than RoomFit (there's much more things you can measure and configure in REW), but also more complicated to use and learn. But RoomFit allows you to get similar results in many cases, assuming appropriate configuration, while being significantly simpler to use.

 

[slartibartfast]

So the engine behind roomfit is REW ?

I don't think so. He is only talking about measurements with the same microphone not the filters generated from those measurements.

 

[Fletcher Munson]

I don't think so. He is only talking about measurements with the same microphone not the filters generated from those measurements.

ah ok