MiniDSP Dirac Live optimized for UMIK-1 or UMIK-2?

[skyfly]

I compared the difference between 0 degree and 90 degree calibration files of UMIK-2.

The drop in sound level at 90 degrees from that at 0 degree is much more when compared to Beyerdynamic MM1. The UMIK-2 is less omnidirectional than MM1.

Will the use of a mic other than UMIK1 or UMIK-2 noticeably influence the Dirac Live in miniDSP?

 

[Keith_W]

Provided that the correct cal file is loaded, the answer is no.

 

[skyfly]

At the listening position, we are also measuring indirect sounds from off axis.

UMIK-2 is at least advertised as omnidirectional mic. As far as on-axis response is reasonably flat (with calibration file, if needed), can we also use uni directional mic (shotgun mic would be an extreme case), or polar mic?

 

[DVDdoug]

can we also use uni directional mic,(shotgun mic would be an extreme case), or polar mic?

I think no. You need to measure on-axis response outdoors (or in an anechoic chamber) or with a "trick setup" and special software like Klippel. Directional mics aren't "perfect" and they are less-perfect off-axis.

Your ears are essentially omnidirectional so if you're trying to measure what you're hearing in the room, an omni mic is used.

But Floyd Toole says:

(Automatic room correction) can yield improvements at low frequencies for a single listener, but above the transition frequency to claim that a smoothed steady-state room curve derived from an omnidirectional microphone is an adequate substitute for the timbral and spatial perceptions of two ears and a brain is absurd.

 

[skyfly]

Your ears are essentially omnidirectional . .

Yes, and no. We can easily experiment with a sound source (e.g. a loudspeaker) and an ear. Depending on the direction to the ear, the sound level noticeably changes.

 

[skyfly]

From my UMIK-2 calibration files

	0 degree calibration file	90 degree calibration file	Different (90deg. – 0 deg)
1002.579	-0.0358	-0.0358	0
2004.513	-0.0418	-0.4166	-0.3748
4007.738	-0.8607	-2.0638	-1.2031
8012.901	-0.9835	-4.1497	-3.1662
16020.654	-0.8590	-8.2405	-7.3815

Now MM1.

 

[skyfly]

I think no. You need to measure on-axis response outdoors (or in an anechoic chamber) or with a "trick setup" and special software like Klippel. Directional mics aren't "perfect" and they are less-perfect off-axis.

My concern is not whether an off axis frequency response curve is "perfect" or not.

I do Dirac Live correction for in-room response correction. The "omni" mic receives sound from various directions. Depending on the level of omni-ness of the mic, different sound is recorded. So, even with the same target curve, the Direc Live correction result will be different.

For example, suppose I use the measurement mic 90 degree. Compared to Beyerdynamic MM1, MiniDSP UMIK-2 will receive more high frequency sound from the ceiling. See the chart with orange dots above. Are you sure this does not affect the Dirac Live correction filter?

I think the level of omni-ness should be specified by Dirac, because different omni-marketed mics have different omni-ness.

Another question: In the Dirac Live software when used with MiniDSP products, (not merely the mic calibration file) the specific omni-ness of MiniDSP mike (UMIK-1 or UMIK-2) was already considered, and thus aother brand's omni-marketed measurement mic (even with calibration file) will give unreliable result?

 

[kyuu]

For example, suppose I use the measurement mic 90 degree. Compared to Beyerdynamic MM1, MiniDSP UMIK-2 will receive more high frequency sound from the ceiling. See the chart with orange dots above. Are you sure this does not affect the Dirac Live correction filter?

Two suggestions:

1) Don't correct above Schroeder.

2) If you want to know for sure whether the use of the Beyerdynamic mic will be an issue, your best resource would be Dirac support rather than random forumites. I would guess that even doing full-range correction, the difference between the mics will be immaterial. But I'm just a random forumite.

 

[No. 5]

My concern is not whether an off axis frequency response curve is "perfect" or not

If you wish to have a omni mic that is closer to omni at high frequency, use one with a smaller capsule/body. A 1/2" capsule like the UMIK-2 will be kinda directional at high frequency owing to its large size relative to the wavelengths in question, a 1/4" mic (Earthworks M30 for example) will be less directional, and an 1/8" even better still. Although I don't know of an 1/8" mic that could be called "affordable", the Behringer ECMPRO has a 4mm capsule, slim body, and is more omni up a high frequency than a 1/2" mic.

 

[Keith_W]

All microphones have a polar response, including so-called "omnidirectional" microphones. As you can see from cal files, omni mics are less sensitive to sounds arriving from behind the mic and at 90deg incidence, especially with short wavelengths. It's just physics. You can get around it somewhat with a smaller capsule and smaller mic body which pushes omnidirectionality higher up the freq range, or you can EQ the mic to have an omnidirectional response.

 

[Miniyouuuu]

I haven't been able to test it in Dirac yet, but I have tested the difference between the Umik1 and Umik2 in REW.
Both in exactly the same position, and using their respective calibration files. The first thing that stands out is that with the same settings, the Umik2 produces more volume, and in general the frequency responses differ quite a bit.

The delay value also differs. I've tried to place the microphones in the exact center between both speakers and the Umik1 gives offset values of 11mm from L to R that should be 0. With the Umik2, that value is closer to 0mm.

And another surprising thing is that the mathematical sum of channels A+B from two measurements with the Umik1 doesn't give a value similar to the measurement of channels L+R. However, with the Umik2, the sum is exactly equal to the actual measurement.

I assume from all of the above that the Umik2 is more accurate and therefore will give more precision to Dirac.
Tomorrow I'll update my impressions with Dirac—up to now I haven't been very happy with the results obtained with the Umik1.