Question about CSL UMIK 1/2 vs Earthworks M23R

[sebna]

Hi All,

I have a question about downsides of using CSL UMIK-1 vs UMIK-2 vs Earthworks M23R for Home Theatre setup calibration using DIRAC / DART or just manual EQ with REW.

I believe that CSL UMIK-1 tracks pretty much the same as Earth M23R, but I also read that due to fact that UMIK-1 has no internal clock that there is some drift happening as it uses PC's USB clock and that UMIK-2 has it's internal clock and it is free from that problem.

But which aspects of REW readings and DIRAC calibration would it affect if any? I read that speaker distance would be affected by UMIK-1? I read that distortion is also not too accurate possibly for both UMIK-1/2 vs M23R?

Would love to understand the differences better.

Another question - would this be all I need, interface wise / preamp for Earthworks M23R for use with Dirac ART and REW?

Scarlett 2i2

The original studio-quality 2-in, 2-out interface, Focusrite Scarlett 2i2, has been remastered for the artist.

focusrite.com

EDIT: I guess I see the first potential problem with this interface:

• Frequency Response 20-20kHz ± 0.06dB (so that would be a bit of waste for such good mic and also taking into accounts that the idea is to measure subs going to single digits?)

Thanks

 

[AwesomeSauce2015]

In my experience, focusrite mic preamps / ADCs are pretty good. There are no issues there when it comes to measuring speakers. I remember doing a loopback measurement on mine and it goes below 20hz no problem. It does roll off, but any speaker roll off will be massively different anyways.
The biggest issue is that with the Earthworks + Focusrite combo, you need some way to get an SPL calibration. This can be as simple as having a calibrated meter and putting the mic right next to it to calibrate REW, but it is still an additional step and (if you don't have a modern phone with an accurate mic) more hardware.

Personally, if I were getting something for use with REW and Dirac, I would get a supported mic system like the UMIK-1 or -2. They are well known, seem to be reasonably accurate, aren't crazy expensive, and (If I understand correctly) come with an SPL calibration. Basically, they are what everyone else is using to do it, so it should work, and if it doesn't, someone can probably help!

So to answer your original question: Having a slight clock drift probably won't show up in measurements. Any mic to pc latency is low enough that I wouldn't be concerned about it messing up the "distance"* measurements, but if you want then go ahead and get a UMIK-2.
(* - These "distance" measurements are usually trying to time-align your speakers, but in DIRAC ART's case, it could very well need accurate actual distance measurements [IDK if that is the case, but given what they are claiming to do, they might]. In that case, I would invest in a super-sophisticated piece of technology: a tape measure.)

No home theater calibration system can "fix" speaker distortion, so there is no sense in being able to measure it accurately unless you are doing DIY speaker design or reviews. I mean yes, an Earthworks mic will have better distortion performance than a UMIK, but that won't really impact you when you go to calibrate your system.

Now, if you are trying to measure stuff down into single-digits:
Is your room, quite literally, a sealed concrete walled-bunker filled with subs? If so, then you might have a chance at getting single-digit performance. I guess exotic stuff like a rotary subwoofer might be capable of doing it in a conventional room, but I am not convinced.
If you have an exotic subwoofer, or a concrete bunker, then yes, you probably would need the earthworks mic to accurately measure down that low. But, I am not very experienced in measuring that low. My usual bass extension targets are 20 or 30hz. So long as I hit those targets, anything below that is just icing on the cake.

 

[bert stoltenborg]

Hi All,

I have a question about downsides of using CSL UMIK-1 vs UMIK-2 vs Earthworks M23R for Home Theatre setup calibration using DIRAC / DART or just manual EQ with REW.

I believe that CSL UMIK-1 tracks pretty much the same as Earth M23R, but I also read that due to fact that UMIK-1 has no internal clock that there is some drift happening as it uses PC's USB clock and that UMIK-2 has it's internal clock and it is free from that problem.

But which aspects of REW readings and DIRAC calibration would it affect if any? I read that speaker distance would be affected by UMIK-1? I read that distortion is also not too accurate possibly for both UMIK-1/2 vs M23R?

Would love to understand the differences better.

Another question - would this be all I need, interface wise / preamp for Earthworks M23R for use with Dirac ART and REW?

Scarlett 2i2

The original studio-quality 2-in, 2-out interface, Focusrite Scarlett 2i2, has been remastered for the artist.

focusrite.com

EDIT: I guess I see the first potential problem with this interface:

• Frequency Response 20-20kHz ± 0.06dB (so that would be a bit of waste for such good mic and also taking into accounts that the idea is to measure subs going to single digits?)

Thanks

If you want to measure easy stuff like room acoustics those UMIKs are good enough. If you want to measure speakers you'll get errors because there are two different clocks used. UMIK 2 can sample to 192 kHz but the mic only is efficient to 20 kHz. So if you want to measure serious stuff I'd buy a serious soundcard and a serious mic.

 

[sebna]

If you want to measure easy stuff like room acoustics those UMIKs are good enough. If you want to measure speakers you'll get errors because there are two different clocks used. UMIK 2 can sample to 192 kHz but the mic only is efficient to 20 kHz. So if you want to measure serious stuff I'd buy a serious soundcard and a serious mic.

But is what DART does considered serious stuff from perspective of measurements perspective and calculations based on them?

Or did you mean it I was building and setting up DIY speakers I would more serious stuff than UMIKs?

 

[sebna]

In my experience, focusrite mic preamps / ADCs are pretty good. There are no issues there when it comes to measuring speakers. I remember doing a loopback measurement on mine and it goes below 20hz no problem. It does roll off, but any speaker roll off will be massively different anyways.
The biggest issue is that with the Earthworks + Focusrite combo, you need some way to get an SPL calibration. This can be as simple as having a calibrated meter and putting the mic right next to it to calibrate REW, but it is still an additional step and (if you don't have a modern phone with an accurate mic) more hardware.

Personally, if I were getting something for use with REW and Dirac, I would get a supported mic system like the UMIK-1 or -2. They are well known, seem to be reasonably accurate, aren't crazy expensive, and (If I understand correctly) come with an SPL calibration. Basically, they are what everyone else is using to do it, so it should work, and if it doesn't, someone can probably help!

So to answer your original question: Having a slight clock drift probably won't show up in measurements. Any mic to pc latency is low enough that I wouldn't be concerned about it messing up the "distance"* measurements, but if you want then go ahead and get a UMIK-2.
(* - These "distance" measurements are usually trying to time-align your speakers, but in DIRAC ART's case, it could very well need accurate actual distance measurements [IDK if that is the case, but given what they are claiming to do, they might]. In that case, I would invest in a super-sophisticated piece of technology: a tape measure.)

No home theater calibration system can "fix" speaker distortion, so there is no sense in being able to measure it accurately unless you are doing DIY speaker design or reviews. I mean yes, an Earthworks mic will have better distortion performance than a UMIK, but that won't really impact you when you go to calibrate your system.

Now, if you are trying to measure stuff down into single-digits:
Is your room, quite literally, a sealed concrete walled-bunker filled with subs? If so, then you might have a chance at getting single-digit performance. I guess exotic stuff like a rotary subwoofer might be capable of doing it in a conventional room, but I am not convinced.
If you have an exotic subwoofer, or a concrete bunker, then yes, you probably would need the earthworks mic to accurately measure down that low. But, I am not very experienced in measuring that low. My usual bass extension targets are 20 or 30hz. So long as I hit those targets, anything below that is just icing on the cake.

Hi,

Thank you for taking time to answer my questions and for explaining it for me.

I already own CSL Umik-1 so I could use it to calibrate SPL for XLR mic + Interface.

I don't live in a bunker but I will have 4x 18" subs in the room so low teens should be no problem I would imagine. Maybe single digits was a stretch.

 

[sebna]

I just realized that some years ago I bought a 2nd hand Audient ID4 (mk1) so I already have a decent interface (I think it goes close to flat to 10hz).

I could pair it with one of the following mics:

• CSL Dayton EMM-6 (I think it is calibrated to 5hz) at which would come to about 160 USD delivered to EU
• Isemcon EMX-7150 (which is 381 USD delivered) (calibrated to 10hz)
• Earthworks M23R (which is 777 USD delivered) (calibrated to 6hz?)

Would Audient ID4 MK1 be good enough for the purpose of both DART and Subs config with manual PEQ / REW / MSO / DART / DLBC with one of these mics?

Would these interfaces and any of the 3 mics be free from the time drift problem?

Does anybody have experience with something like Isemcon EMX-7150? If so how does it compare to Earthworks M23r? Would it provide more repeatable and dependable measurements under 20hz (vs CSL UMIK-1, CSL EMM-6)? Is it a true middle ground option or is it on the level of M23R or maybe more of CSL EMM-6? Would it be as durable and lasting and not drifting over time as M23R (I read that M23R readings do not really change over 10yr of usage).

As I already have the decent enough, I would hope, interface for few extra quid I could get a mic for it and at least get rid of time drift problem of UMIK-1.

It is just that shipping dealer's Earthworks back and forth will cost me 1/3rd of the CSL EMM-6 mic, and I might want to repeat the measurements sooner or later (which would be another 1/3 of the price), maybe new DIRAC software version might require to remeasure. But then I am not a fan of going with cheapest options either, especially for a tool which might be still handy and used 10yrs down the line. That is why I wonder if it makes sense to spend more on EMX 7150 or even M23r and be done with it.

 

[AwesomeSauce2015]

But is what DART does considered serious stuff from perspective of measurements perspective and calculations based on them?

Or did you mean it I was building and setting up DIY speakers I would more serious stuff than UMIKs?

No. I wouldn't consider it to be a system requiring absolute measurement perfection.
Yes, when you are measuring speakers, often you will have a loopback measurement out of your DAC, and back into a spare mic input on your ADC. This allows the software to cancel out the interface's frequency and time response, from your measurement, as well as providing a consistent timing reference which can be important for gated measurements.
One other use for these super consistent mics is for compliance purposes. Some concert venues or jurisdictions have SPL limits, and you can be fined for exceeding them. And other places may have requirements for SPL monitoring. This is when you want a calibrated reference mic. You can then use it to calibrate your cheap mics on the go, and use the cheap mics to tune a PA, or for your compliance needs. The key here is that that calibrated reference mic has a traceable (reputable) calibration, and is specified to hold said calibration reliably over a period of time. So basically, if you get in trouble, you have the documents to prove that you are right.

For home theater room correction, you don't really need that. You aren't measuring at 120+db, and you don't need to be able to prove to a court that your mic is good.

I just realized that some years ago I bought a 2nd hand Audient ID4 (mk1) so I already have a decent interface (I think it goes close to flat to 10hz).

I could pair it with one of the following mics:

• CSL Dayton EMM-6 (I think it is calibrated to 5hz) at which would come to about 160 USD delivered to EU
• Isemcon EMX-7150 (which is 381 USD delivered) (calibrated to 10hz)
• Earthworks M23R (which is 777 USD delivered) (calibrated to 6hz?)

Would Audient ID4 MK1 be good enough for the purpose of both DART and Subs config with manual PEQ / REW / MSO / DART / DLBC with one of these mics?

Would these interfaces and any of the 3 mics be free from the time drift problem?

Does anybody have experience with something like Isemcon EMX-7150? If so how does it compare to Earthworks M23r? Would it provide more repeatable and dependable measurements under 20hz (vs CSL UMIK-1, CSL EMM-6)? Is it a true middle ground option or is it on the level of M23R or maybe more of CSL EMM-6? Would it be as durable and lasting and not drifting over time as M23R (I read that M23R readings do not really change over 10yr of usage).

As I already have the decent enough, I would hope, interface for few extra quid I could get a mic for it and at least get rid of time drift problem of UMIK-1.

It is just that shipping dealer's Earthworks back and forth will cost me 1/3rd of the CSL EMM-6 mic, and I might want to repeat the measurements sooner or later (which would be another 1/3 of the price), maybe new DIRAC software version might require to remeasure. But then I am not a fan of going with cheapest options either, especially for a tool which might be still handy and used 10yrs down the line. That is why I wonder if it makes sense to spend more on EMX 7150 or even M23r and be done with it.

Any XLR mic with an interface that doesn't have any weird driver issues could possibly be free from this supposed "time drift" problem. But, given your use case, any "time drift" in a USB mic is not going to be a factor. And even then, the interface also works over USB, so it will have latency. The measurement software should correct for this latency by itself, for both a USB mic and an XLR mic + interface.

The CSL calibrated EMM-6 will be a solid mic for probably a few months and maybe up to a year. After that, the mic itself will probably drift out of calibration. You would then probably want to have it calibrated again, in order to have reliable accuracy. But then again, you already have a UMIK-1, so I wouldn't waste my money on the EMM-6 as it is probably an inferior quality microphone. -- Not to mention that after shipping it back to you, the EMM-6 probably won't be in calibration anymore... (It just isn't that good of a mic)

I can't really comment on the other two mics, but what I will say is that I wouldn't trust the EMM-6 to hold its calibration or to be accurate down to 5hz. It just isn't a high quality microphone. The Earthworks mic will probably hold its calibration very well, as Earthworks makes high quality mics.

If you are really concerned about measurement accuracy, then your best bet is to get something that can be calibrated as a complete system like the UMIK-1 or UMIK-2, and have it calibrated on a semi-regular basis by a reputable organization. Either that, or get a high quality analog mic and interface, and still get it calibrated regularly.

Now what most people probably do is buy something like the UMIK 1 or 2, maybe get it 3rd party calibrated once, and then just use it. For Dirac measurements, a UMIK-1 will suffice, unless Dirac says otherwise. You could ensure accuracy by simply getting the UMIK calibrated, and if you want a UMIK 2 then go ahead. The simple truth of the matter is that either of the two UMIKs are good enough for what you are doing. You don't need the super consistent and accurate performance of those other mics, in your use case.

 

[bert stoltenborg]

But is what DART does considered serious stuff from perspective of measurements perspective and calculations based on them?

Or did you mean it I was building and setting up DIY speakers I would more serious stuff than UMIKs?

I always hear that UMIK-mics are sufficient for room correction purposes.
UMIK's only have an AD-convertor AFAIK, so when you want to monitor phase like in speaker developement you could get into trouble.
An Earthworks, Bruel&Kjaer, ACO etc electret mic is Class I and typical has a drift of less than 1 dB in 250 years. These cheaper mics seem to drift sometimes several dB's in a couple of years. At least that is what you read all over the net. I use a B&K 4007 for 35 years and I know guys using even older 4007's without problems.
These M23 mics are not excessively expensive so maybe they are cheaper over time than an UMIK or ECM 8000 you have to calibrate all the time.

 

[Blumlein 88]

Might want to read this from John at REW.

Designing an active 2-way speaker crossover

https://kimmosaunisto.net/Software/VituixCAD/VituixCAD_Measurement_REW.pdf "Note! Single channel measurement systems such as USB microphones with latency variations by default are not recommended for speaker engineering due to timing and phase variations and normalizations. REW should not be...

www.audiosciencereview.com

If the M23 is purchased and careful calibration for SPL is wanted, I'd probably buy the Umik 1 just for that purpose and then measure everything once you have calibrated level of the M23 with the M23. But there are ways to get reasonably close and just use the M23.

 

[Blumlein 88]

Another worth considering is the Line Audio Omni. It also is somewhat lower in sensitivity.

Line Audio Design - Made in Sweden

Line Audio Design

www.lineaudio.se

Finally I might think you would be just as well off with a Dayton EMM6 and the Audient. Either from Dayton or get a Cross Spectrum labs model assuming they are available. Basically a similar quality to the Umik, but not USB based. Uses 48 volt phantom power. I see you mentioned that above.

Dayton Audio EMM-6 Electret Measurement Microphone

Dayton Audio EMM-6 Electret Measurement MicrophoneThe Dayton Audio EMM-6 is a precision electret condenser microphone designed for measurement and critical recording applications. The extremely flat frequency response and true omni-directional polar pattern make it perfect for use with room...

www.parts-express.com

Cross·Spectrum - Calibrated Dayton Audio EMM-6 Microphones for Sale

Purchase calibrated Dayton Audio EMM-6 measurement microphones, low-cost option for performing precise and accurate sound measurements

www.cross-spectrum.com

 

[sebna]

Another worth considering is the Line Audio Omni. It also is somewhat lower in sensitivity.

Line Audio Design - Made in Sweden

Line Audio Design

www.lineaudio.se

Finally I might think you would be just as well off with a Dayton EMM6 and the Audient. Either from Dayton or get a Cross Spectrum labs model assuming they are available. Basically a similar quality to the Umik, but not USB based. Uses 48 volt phantom power. I see you mentioned that above.

Dayton Audio EMM-6 Electret Measurement Microphone

Dayton Audio EMM-6 Electret Measurement MicrophoneThe Dayton Audio EMM-6 is a precision electret condenser microphone designed for measurement and critical recording applications. The extremely flat frequency response and true omni-directional polar pattern make it perfect for use with room...

www.parts-express.com

Cross·Spectrum - Calibrated Dayton Audio EMM-6 Microphones for Sale

Purchase calibrated Dayton Audio EMM-6 measurement microphones, low-cost option for performing precise and accurate sound measurements

www.cross-spectrum.com

I was actually just reading the first link and John's post.

But does not the same problems apply to Earthworks M23 as it is connected to interface that communicates with a PC with REW over USB?

 

[Blumlein 88]

I was actually just reading the first link and John's post.

But does not the same problems apply to Earthworks M23 as it is connected to interface that communicates with a PC with REW over USB?

No because you can loopback inside the interface which is using the same clock.

 

[sebna]

No because you can loopback inside the interface which is using the same clock.

Thank you. Also came to same conclusions via my own research in the meantime.

But as far as I understand none of the DIRAC flavours allow, like REW does, to select loopback as a source of time reference.

I was told on other forum, that I don't need that with DIRAC as DIRAC does its own time reference but after reading about how problematic it can be with USB mics I wonder how well DIRAC handles it and if it is not the part of the reason why Trinnov's Optimizer is considered superior, with its slef powered XLR, skipping USB interface all together.

So I thought I found a way how to address clock and time drift (with loopback) but this option is not an option for DIRAC (which is known to do less than stellar job with time alignment at MLP when verified with REW, when Trinnov and Optimzer is said to do a perfect job in this regard).

 

[bert stoltenborg]

Thank you. Also came to same conclusions via my own research in the meantime.

But as far as I understand none of the DIRAC flavours allow, like REW does, to select loopback as a source of time reference.

I was told on other forum, that I don't need that with DIRAC as DIRAC does its own time reference but after reading about how problematic it can be with USB mics I wonder how well DIRAC handles it and if it is not the part of the reason why Trinnov's Optimizer is considered superior, with its slef powered XLR, skipping USB interface all together.

So I thought I found a way how to address clock and time drift (with loopback) but this option is not an option for DIRAC (which is known to do less than stellar job with time alignment at MLP when verified with REW, when Trinnov and Optimzer is said to do a perfect job in this regard).

The timing errors are significant when you f.e. measure a woofer and tweeter to design a X-over.
When you measure a room you have countless reflections that add to the primary source so phase/time is irrelevant.

 

[sebna]

The timing errors are significant when you f.e. measure a woofer and tweeter to design a X-over.
When you measure a room you have countless reflections that add to the primary source so phase/time is irrelevant.

Do you know if loopback mechanism in case of XLR mic addresses problem of USB clock drift (as USB is also a interface for communication with PC by interface)?

 

[bert stoltenborg]

Do you know if loopback mechanism in case of XLR mic addresses problem of USB clock drift (as USB is also a interface for communication with PC by interface)?

Yes it does. If you have a normal USB soundcard it sends signals and retrieves them using the same clock. Something cheap like a Behringer UMC 202HD does the trick, a nice mic like that Earthwork is a pro solution.
If you want to do real pro measurements you still have to have your equipment calibrated by a certified lab, over here once every 18 month, but for private measurements this is a killer setup IMHO.
If you are interested in theoretic stuff about this you could refer to the REW or ARTA manuals.

 

[sebna]

I have Audient iD4 mk1 and I wonder if it is possible to set it up with analogue cable connecting some inputs and outputs to achieve analogue loopback to be used as time reference for REW and measuring mic?

If so could someone advise which cable I need and which inputs outputs should I populate with it?

I have tried to find this info online but so far was unlucky. Probably too basic knowledge to get enough coverage

So I am asking if something like the guy in this video does would work with ID4 Mk1 and if so what should I connect with what ?

Video clip is timestamped.

Thanks