UMIK-1 vs UMIK-2

[Mike-48]

Here is a measurement comparing the two mics. Not much difference above 20Hz, and below 20Hx I don't care.

So the between-mic results are close, but between-calibration results are not. It's frustrating that mic calibration is so variable. A third independent calibration might give a hit of what's going on (not that I'm suggesting you find one).

I owned a TacT 2.2X DSP preamp almost 20 years ago, which came with a mic and cal file. At my request, TacT recalibrated the mic and sent a new cal file. Then, I bought a third-party mic with its own calibration file. The TacT's automated DSP software gave quite different-sounding results across the three possibilities. I picked the one that sounded flattest and most natural to me when I played back acoustic piano runs and such. Today, I tweak bass by ear, 1/8 of a dB at a time, on my Anthem preamp, whose proprietary mic comes with an encoded calibration file that will frustrate any user who wants to know what really is going on. So not much has changed! The scientist in me is a bit perturbed by all that, though it's usually easy enough to put out of my mind and enjoy the music.

 

[thothsong]

So the between-mic results are close, but between-calibration results are not.

That chart is comparing UMIK-1 vs UMIK-2, both using the CSL cal files. It's not comparing a single mic using miniDSP vs CSL cal files. So it's not addressing your original point.

 

[Mike-48]

That chart is comparing UMIK-1 vs UMIK-2, both using the CSL cal files. It's not comparing a single mic using miniDSP vs CSL cal files. So it's not addressing your original point.

Yes, of course. My comments were based on both that post and the previous post, which together show that the mics give the same result, but the two calibration curves for each mic differ quite significantly. This is a big area of uncertainty when we use such mics in audio. It is helpful to know that in the range 300 Hz to 4 kHz it didn't make a difference, at least with this pair of calibration files. It's still troubling that the user can't have much faith in these mic calibrations. Like @AustinJerry, I'm more inclined to trust the CSL curve, but that inclination is not based on any data at all. It's mainly a hunch, though I suppose the demonstrated agreement between mics with the CSL curves are used suggests that they are likely to be accurate (or inaccurate in the same way).

 

[thothsong]

It's been a week now, and miniDSP hasn't responded to my support ticket about the UMIK-1 vs UMIK-2 measurement differences. Disappointing. I may look into having CSL calibrate both.

 

[Chuck Gerlach]

Would I be correct in assuming, based upon the above, that the UMIK-2 mic provides no real benefit over the UMIK-1 AND that the CSL calibrated mic provides no real improvement over the miniDSP provided calibration files? I'm more than willing to spend the money if it really buys me anything.

 

[AustinJerry]

Would I be correct in assuming, based upon the above, that the UMIK-2 mic provides no real benefit over the UMIK-1 AND that the CSL calibrated mic provides no real improvement over the miniDSP provided calibration files? I'm more than willing to spend the money if it really buys me anything.

I would say "marginal benefit", unlikely to be worth it for casual REW measurements that most of us are engaged in.

 

[thothsong]

miniDSP support finally contacted me and asked for an mdat file, which I sent on 26 Jan. I've heard nothing back from them since. I inquired via CSL's web site if they would cal both of my mics, but they never responded back. So right now I don't really have a reason to trust either mic.

 

[3dbinCanada]

I watching on the sidelines seeing how this plays out as I own and use a Umik1. I'd be interested to know which mic/cal file is the most accurate, The problem with the comparions I'm seeing in this thread are relative to one another and these measurements are missing a benchmark or standard to compare too. Does this make sense?

 

[thothsong]

Sure, but who's got a reference quality method to evaluate them? That's why we're buying calibrated mics in the first place. I'd love to pay CSL to calibrate my mics, so I can at least get another view on accuracy, but such is life.

 

[3dbinCanada]

Sure, but who's got a reference quality method to evaluate them? That's why we're buying calibrated mics in the first place. I'd love to pay CSL to calibrate my mics, so I can at least get another view on accuracy, but such is life.

I hear you. I guess the best outcome one can hope for is that the measurements move towards being the same.

 

[AustinJerry]

Sure, but who's got a reference quality method to evaluate them? That's why we're buying calibrated mics in the first place. I'd love to pay CSL to calibrate my mics, so I can at least get another view on accuracy, but such is life.

Did you read my earlier post (number 15). My comparison was using thenUMIK-1mand the UMIK-2, both with custom calibration files from Cross Spectrum Labs. Why is this not accurate enough for you?

 

[thothsong]

Sure I read your earlier post. It's exactly why I want CSL to cal my mics, to see if (and hope) they then match each other. And if they do, then compare miniDSP measurements with the CSL measurements, to see if either mic using a miniDSP cal file comes close to matching CSL.

 

[3dbinCanada]

Did you read my earlier post (number 15). My comparison was using thenUMIK-1mand the UMIK-2, both with custom calibration files from Cross Spectrum Labs. Why is this not accurate enough for you?

Yes I saw it but I misread the text underneath. I thought you were talking about your measurements being amiss, not another poster's. Those measurements are close enough for me. How much did it cost to get the cal files from CSL and how long did it take?

 

[AustinJerry]

Yes I saw it but I misread the text underneath. I thought you were talking about your measurements being amiss, not another poster's. Those measurements are close enough for me. How much did it cost to get the cal files from CSL and how long did it take?

Well, I have made a number of comparisons using both the MiniDSP web site cal files and the CSL custom cal files, and I have never seen a significant difference for my mics. Of course, you may be dealing with a defective mic, which would give different results.

 

[3dbinCanada]

Well, I have made a number of comparisons using both the MiniDSP web site cal files and the CSL custom cal files, and I have never seen a significant difference for my mics. Of course, you may be dealing with a defective mic, which would give different results.

Nice to know that the cal files from MiniDSP are accurate enough. I only have the Umik1 but I am curious.

 

[thothsong]

His may be, but mine certainly aren't. If it's happened to me, how do you know it hasn't happened to you?

 

[AustinJerry]

FYI, shortly after I received my UMIK-2 from MiniDSP, I sent it to CSL for calibration. The turnaround time and communication were not very good. It was almost six weeks before I finally had the UMIK-2 back with its custom calibration file. Expecting business to be normal during the pandemic is probably not realistic.

 

[Andrew Slater]

So, I just got a cross spectrum umik2

and it measures the same as my umik1s

but sounds way better on Dirac….

what exactly is the differences in mics besides a much better sound card and 1/2” condenser.

maybe it’s the cross spectrum cal that I’m hearing… but I suspect it’s not being I have 4 other mics that also measured the same.

and why (with Dirac) am I getting different sound all together by just using a different mic?

im starting to believe that comparing frequency response is not valid argument. If the umik1 and umik2 produce the same frequency response measured, why do they produce totally different sounding filters in Dirac (umik2 being way better)

there’s more to it. And I think measuring frequency response is boring and not the whole story..

can anyone give any actual useful info? Having so many different mics I’m just not going to entertain this frequency response make it the same argument any more….

any helpful information is so much appreciated. I’m thirsty for knowledge:-)

 

[JStewart]

So, I just got a cross spectrum umik2

and it measures the same as my umik1s

but sounds way better on Dirac….

what exactly is the differences in mics besides a much better sound card and 1/2” condenser.

maybe it’s the cross spectrum cal that I’m hearing… but I suspect it’s not being I have 4 other mics that also measured the same.

and why (with Dirac) am I getting different sound all together by just using a different mic?

im starting to believe that comparing frequency response is not valid argument. If the umik1 and umik2 produce the same frequency response measured, why do they produce totally different sounding filters in Dirac (umik2 being way better)

there’s more to it. And I think measuring frequency response is boring and not the whole story..

can anyone give any actual useful info? Having so many different mics I’m just not going to entertain this frequency response make it the same argument any more….

any helpful information is so much appreciated. I’m thirsty for knowledge:-)

This a question for @Flak from Dirac.
You might also start a support ticket for which you’ll need to create a support account which is different from your Dirac account.

https://jira.dirac.services/servicedesk/customer/user/login?destination=portals

 

[thothsong]

So, I just got a cross spectrum umik2

and it measures the same as my umik1s

but sounds way better on Dirac….

Have you tried using RTA to measure what you're listening to, using both mics, to see if there's any difference that way?

 

[AustinJerry]

So, I just got a cross spectrum umik2

and it measures the same as my umik1s

but sounds way better on Dirac….

what exactly is the differences in mics besides a much better sound card and 1/2” condenser.

maybe it’s the cross spectrum cal that I’m hearing… but I suspect it’s not being I have 4 other mics that also measured the same.

and why (with Dirac) am I getting different sound all together by just using a different mic?

im starting to believe that comparing frequency response is not valid argument. If the umik1 and umik2 produce the same frequency response measured, why do they produce totally different sounding filters in Dirac (umik2 being way better)

there’s more to it. And I think measuring frequency response is boring and not the whole story..

can anyone give any actual useful info? Having so many different mics I’m just not going to entertain this frequency response make it the same argument any more….

any helpful information is so much appreciated. I’m thirsty for knowledge:-)

Expectation bias?

I have both mics amd have performed Dirac calibrations with both. I have never heard a difference that I would describe as "way better".

 

[Andrew Slater]

Have you tried using RTA to measure what you're listening to, using both mics, to see if there's any difference that way?

Yes I’ve done moving mic averages, they all look pretty much the same… even when running sweeps they all look really close to the same thing… of course very small differences but the general shape is right on top of each other

Expectation bias?

I have both mics amd have performed Dirac calibrations with both. I have never heard a difference that I would describe as "way better".

so it’s in a car very highly reflective environment… and very very susceptible to any minute changes… especially in the mid range. not sure if that makes a difference
definitely not expectation bias, all four of my preetss on my DDRC 22 each one has a different mic, default Target responce on all of them. (Not the auto Target found in 3.3 I’m still on 3.1)

It’s really the nitty-gritty… I get absolutely repeatables results if I stick with the same mic… change the mic change the sound stage, a lot. Each individual mic gives repeatable results every time

In overall smoothness, umik2 from cross spectrum just gives me the feeling noting is too loud or quiet, everything is smooth, properly placed in stage using test discs….
I would love to know more about mics and measurements

And yes for me “way way better” I’m definitely picky and want absolute precision. The umik2 I can achieve this with no fiddling with stuff after Dirac. To me that says a lot

 

[thothsong]

Yes I’ve done moving mic averages, they all look pretty much the same… even when running sweeps they all look really close to the same thing… of course very small differences but the general shape is right on top of each other

I didn't mean pink noise or sweeps. I meant use the UMIK-2 Dirac configuration, play real content, and use RTA to measure a section, once using UMIK-1 and once using UMIK-2, in the same measurement position.

 

[Andrew Slater]

I didn't mean pink noise or sweeps. I meant use the UMIK-2 Dirac configuration, play real content, and use RTA to measure a section, once using UMIK-1 and once using UMIK-2, in the same measurement position.

Hummm interesting…..

I’ve validated Dirac, RTA is wrong way to do it tho. An array of gated measurements was the only way Dirac and my measurements were the same. RTA seemed to add a lot more data. Although non Dirac manual tuning moving mic is the only way.

Okay so post Dirac rta with what noise source? PN or pink? Or are you taking about something else as “real content”

Like what am I looking for?

As far as PN averages, all mics look generally the same…. But it’s always changing and I can’t move mic exactly the same…

Stationary? With what noise? Like what’s real content? Music? That might be too difficult to see.

I can give that a shot if it reveals something.

 

[AustinJerry]

My apologies, I didn't know you were using Dirac in a vehicle. I have no experience using Dirac in this manner.