I've considered attempting to change my measurement process for a while now, but I think it would be like starting over. I never did because I was trying to avoid having to run 5 different programs to get a legit measurement. I've gotten very good with OmniMic, so I imagine at least knowing how to read the response(s) and knowing how to measure (above ~400 Hz) will definitely aid me in trying something knew. Sucks though because the OmniMic is so simple, but obviously not perfect. Wish someone sold a one-stop-shop measurement setup like OM, but better
Member
Joined 2003
Clio pocket is the one stop shop, but it has a price tag. Really though the main benefit in Clio or Omnimic is absolute SPL , which honestly isn’t a necessity for DIY work. Clio has the advantage of an internal clock reference for capturing timing information, so it’s leaps better than Omnimic for design work. It’s not hard to plug a mic into a usb audio interface though and use REW or ARTA, and in fact gets you closer to a one stop shop because the same equipment and software can be used for impedance and T/S parameters with a simple resistor jig.
I own an Omnimic and find the software very annoying. You can’t change any basic settings like adjust the windowing or smoothing without remeasuring. Moving to any other software where the impulse response can be saved directly and reprocessed anytime without remeasuring is hugely beneficial.
I own an Omnimic and find the software very annoying. You can’t change any basic settings like adjust the windowing or smoothing without remeasuring. Moving to any other software where the impulse response can be saved directly and reprocessed anytime without remeasuring is hugely beneficial.
Last edited:
Agree. Saving the PIR file is a big, big advatage.
I used OmniMic for several years before I started using an Audix XLR mic + Behringer USB interface + ARTA. Omnimic is a good way to start learning about acoustical measurements. It is a good first step. Eventually a designer will want to move to a two-channel rig.
I also found the factory calibration on the OmniMic to be the most accurate of any mic I own. Currently, I use the OmniMic for harmonic distortion measurements, and for room measurements.
j.
CLIO is good gear for sure.
But i think a better path would be a inexpensive dual channel setup. Use it with REW ....& good to have ARTA too for true dual channel..
UMC202HD sound card $80; EMC8000 mic $50; REW $ donate: ARTA $90
So $220 for a measurement setup it will take many years to outgrow.
This puts you on a more "open-source" type path, if i may use a bit of an analogy.
When and if you decide you might need a better mic, you'll know what you need by then...
Soundcard will still be sufficient. Measurement software will still be way more capable than the person using it (if you're at all like me
But i think a better path would be a inexpensive dual channel setup. Use it with REW ....& good to have ARTA too for true dual channel..
UMC202HD sound card $80; EMC8000 mic $50; REW $ donate: ARTA $90
So $220 for a measurement setup it will take many years to outgrow.
This puts you on a more "open-source" type path, if i may use a bit of an analogy.
When and if you decide you might need a better mic, you'll know what you need by then...
Soundcard will still be sufficient. Measurement software will still be way more capable than the person using it (if you're at all like me
Member
Joined 2003
For distortion I find the noise floor in Omnimic is very high, a good 10dB higher than what I get measuring with a Steinberg UR22mkii and Line Audio OM1 mic, so with Omnimic the volume usually needs to be cranked just to get the harmonics poking up above the noise.
I've read a little on CLIO but still a little unsure of the major benefits over OmniMic.
My OmniMic was calibrated by CSL (I actually worked with Herb directly), so I'm certain it's accurate. The only shortfalls I really have with it is...
*Reliable measurements below ~400 Hz
*Possibly a one stop SIM so I don't have to use separate programs (I use WinPCD), but this is minor
*No separate amp or CD for measuring would be handy but also not a deal breaker
*All in one for T/S parameters and FR would be handy but not a deal breaker.
So... What exactly does CLIO do better than the OM + DATS combination for crossover design?
My OmniMic was calibrated by CSL (I actually worked with Herb directly), so I'm certain it's accurate. The only shortfalls I really have with it is...
*Reliable measurements below ~400 Hz
*Possibly a one stop SIM so I don't have to use separate programs (I use WinPCD), but this is minor
*No separate amp or CD for measuring would be handy but also not a deal breaker
*All in one for T/S parameters and FR would be handy but not a deal breaker.
So... What exactly does CLIO do better than the OM + DATS combination for crossover design?
I consider one of CLIO's primary benefits that there's no screwing around. I'm not worried about specific drivers, specific sound cards, dealing with other hardware and whether it gets along with whatever other software, the zillion ways Windows can hose you, etc.
It's installed, it's done, it's right, and I don't have to worry about it. That's significant to me.
In addition to the above, it uses its own hardware for both signal generation and data acquisition, which gives you the proper phase/timing information that's lacking with OmniMic. It's "a one-stop-shop measurement setup like OM, but better." I thought that was the original request in post 21.
Is it everything to everyone? Obviously not. Each user has to decide what's important to them and what their time/patience is worth.
The Full version of Clio is really quite expensive, the pocket version is cheaper but not the same. I considered looking at the pocket version but have steered clear this post was the final clincher
https://www.diyaudio.com/community/...aker-build-abec-modelling.357792/post-6836124
You could use the omnimic to create your own calibration file for a non calibrated analogue mic
https://johnr.hifizine.com/2012/12/diy-mic-calibration/
Omnimic measures and represents the data in different ways, it uses or can use a blended measurement for low frequencies where the gate starts to run out of resolution. The problem with this is that it is no longer a quasi anechoic measurement but something else.
You need a long gate to have high resolution at 300 to 400Hz and that is the point where nearfield and baffle diffraction simulation starts to become unreliable. That is an area that is not well represented by a single on axis measurement, the power and off axis responses becomes quite important at those frequencies particularly how they are going to interact with the room.
You might not want to change your practices but these are some of reasons why you should.
I actually completely agree with your last statement, that was why I started this thread. I can live with things the way I'm doing them now, but forever I'll wonder if or how I could do better if I never upgrade my measuring system and technique. I definitely design/build enough speakers to justify the expense and learning curve. I'm just researching to find the best option. And as speaker builders, we all know that the most expensive doesn't necessarily equal the best.
Will CLIO solve the issue of sub-400 Hz measurements? Or will those still be quasi-simulated responses based on multiple measurements?
Reading reviews for CLIO I can see it being a frustrating start.
No I don't think so, Clio is still a mic and a soundcard in a package. 300 to 400Hz is tricky to do quasi anechoically if you are aiming for total accuracy. Below 200Hz most speakers are omnidirectional enough that a nearfield and basic baffle diffraction compensation gives a highly accurate result. Ports are another matter but there are calculations to use to scale the measurements and get a decent result. 300 to 400 is where the nearfield measurement starts to become unreliable depending on the size of the driver. There is also more directivity than you might think in many designs so single axis measurements are not going to give a full picture of what is happening there around the speaker. The best solution to getting good measurements in that range is to make the reflection free time in the measurement as long as possible, going outdoors, into a bigger space or using a ground plane.
This App Note from Audio Precision is good reading
https://www.admess.de/tl_files/admess-2013/pdf/Audio Precision/AppNote - Loudspeaker EA Measurements-1.pdf
Edit: My recommendation for getting a better measurement setup would be to buy a reasonably simple Audio Interface from Focusrite, Behringer, Steinberg, MOTU or any of the other well known brands. If you use them with an ASIO or WDM driver in REW or ARTA configuration should be reasonably straight forward and pair that with a calibrated analogue mic or calibrate one yourself. These can be cheap or expensive the differences are really quite small between them when calibrated. If you can afford the full CLIO FW setup then that is a good choice, if not what I said before will get you most of the way there for less money but more hassle, how much is hard to say.
Last edited:
The omnimic blended measurement is actually a pretty good guidepost to blending the gated far field to the near field. In other words, after blending NF and gated FF, the final result should be similar to the omnimic blended response.
I also find the omnimic software a convenient way to view FRD files.
Mattsk8 - As you start down the path of making better measurements, keep in mind that taking accurate, repeatable acoustic measurements is, in my opinion, one of the most challenging aspects of DIY speaker building. So don't be frustrated if you don't get it right on your first, or fifth, or tenth attempt.
Member
Joined 2003
Omnimic has "blended" response, ARTA has dual windowing, REW has frequency dependent windowing, SoundEasy has variable windowing...all similar methods trying to do the same thing. I'm sure these features have their use cases but I've not found much value in any of it for gathering driver data for a design.
Omnimic software is adequate for viewing frequency response data, but not better than VituixCAD IMO to just load the same data into the calculator tool or into a driver or overlay.
In any case, the mic and software used generally isn't the limiting factor in getting accurate reflection free response in-room, but rather the measurement method and response processing. The VituixCAD guides provide enough detail for a simple process that provides reliable results that only requires two software packages for complete loudspeaker design, one software to measure, and one to do the rest. The software you choose won't fix problems with measurement methods or response processing methods, which is why I posed from the start to just pick your favourite software and follow the same process outlined in the VituixCAD guides.
Omnimic software is adequate for viewing frequency response data, but not better than VituixCAD IMO to just load the same data into the calculator tool or into a driver or overlay.
In any case, the mic and software used generally isn't the limiting factor in getting accurate reflection free response in-room, but rather the measurement method and response processing. The VituixCAD guides provide enough detail for a simple process that provides reliable results that only requires two software packages for complete loudspeaker design, one software to measure, and one to do the rest. The software you choose won't fix problems with measurement methods or response processing methods, which is why I posed from the start to just pick your favourite software and follow the same process outlined in the VituixCAD guides.
My experience too.
The App Note fluid linked in#32, is one of the best I've seen, of an easy to read comprehensive review of measurement techniques.
I sincerely believe the physical techniques matter far more than whatever software or soundcard...or even mic!, that we use.
It seems to me, all the various software/hardware ends up cranking out the same Fourier math in the end....
So what is easiest to get repeatable results with, and makes the most $ sense, gets my vote.
Not mentioned in this thread but it is elsewhere...while you can get REW free and free version of ARTA, the paid version of ARTA (about $90) allows an automated process to take multiple measurements. If you build a DIY turntable then you tell ARTA what angles you will measure and add a delay in between and hit start. I tell ARTA I want to take measurements 0 to 180 degrees in 10 degree increments and give me 7 seconds of pause between each for me to run up the turntable and rotate it. I can take 0 to 180 degrees horizontal and vertical for tweeter and midrange in 30 minutes with 4 clicks of my mouse. Then I can full advantage of VituixCADs capabilities. I got the Steinberg MkII preamp used off eBay, Dayton EMM-6 used off a forum's classifieds, and paid for the ARTA license, all for a little under $250.
You guys are fantastic, I sincerely appreciate all the responses. I think before I drop coin on a whole bunch of new equipment, I should figure out how to get the quasi-simulated sub ~400 Hz response using my OmniMic. From there I can decide if / where / why I need to upgrade.
Although, being able to post response measurements with the little Clio trademark would be pretty sweet 😎
Although, being able to post response measurements with the little Clio trademark would be pretty sweet 😎
Whereas I agree with what has been said in this thread so far, it should be mentioned that measuring between 200 and 1000 Hz with some accuracy requires either a large distance between reflecting surfaces and the DUT, or averaged impulses from multi-spot measurements (space averaging). See both "Klippel" threads. I know the OP asked for LF measurements, but MF i.m.o. should be added to the discussion.