Rational Acoustics



John Paganelli
August 14th, 2012, 11:30 AM
Does anybody have any thoughts on submixing multiple measurement mics (set up at various locations throughout a test area) and feeding the mix into a single measurement channel input? My intent for this would be in the case of a distributed sound system in a large church where all of the speakers are the same and I want to get an overall average EQ of the system in the Sanctuary. I am currently still using SMARRT 6, so I am limited to only 2 channels. This would certainly be faster than moving (1) mic around and averaging individual measurement points. But, I am not sure if I would be getting reasonable data. I am also not sure how SMAART would determing the microphone delay - thus getting back to my concern about valid data.

Thanks

adam
August 14th, 2012, 01:05 PM
Does anybody have any thoughts on submixing multiple measurement mics (set up at various locations throughout a test area) and feeding the mix into a single measurement channel input? My intent for this would be in the case of a distributed sound system in a large church where all of the speakers are the same and I want to get an overall average EQ of the system in the Sanctuary. I am currently still using SMARRT 6, so I am limited to only 2 channels. This would certainly be faster than moving (1) mic around and averaging individual measurement points. But, I am not sure if I would be getting reasonable data. I am also not sure how SMAART would determing the microphone delay - thus getting back to my concern about valid data.

Thanks

Hey John,

Undertaking such a measurement will not yield useful data. You've only two options available to achieve your desired measurement. One is averaging multiple traces together with v6 and the other is to use v7. I realize you are aware of those options, I'm just stating them for those playing along at home.

-A

John Paganelli
August 14th, 2012, 03:14 PM
Hey John,

Undertaking such a measurement will not yield useful data. You've only two options available to achieve your desired measurement. One is averaging multiple traces together with v6 and the other is to use v7. I realize you are aware of those options, I'm just stating them for those playing along at home.

-A

Does v7 actually average all of the individual mic measurements or just simultaneoulsy capture individual measurements (that you still need to independently average)?

adam
August 14th, 2012, 03:49 PM
Does v7 actually average all of the individual mic measurements or just simultaneoulsy capture individual measurements (that you still need to independently average)?

Both. v7 allows for multiple simultaneous measurements as well as multiple real time live averages of measurements. You can of course still manually average captured traces, as with v6.

adam
August 14th, 2012, 03:56 PM
Both. v7 allows for multiple simultaneous measurements as well as multiple real time live averages of measurements. You can of course still manually average captured traces, as with v6.

To help with visualization, here is a screen shot showing spectral measurements for four mics, using two preamps. I also have four live averages running that are averaging the live mic measurements. All averages are in power, though they could be in dB.

Arthur Skudra
August 14th, 2012, 06:08 PM
Does anybody have any thoughts on submixing multiple measurement mics (set up at various locations throughout a test area) and feeding the mix into a single measurement channel input? My intent for this would be in the case of a distributed sound system in a large church where all of the speakers are the same and I want to get an overall average EQ of the system in the Sanctuary. I am currently still using SMARRT 6, so I am limited to only 2 channels. This would certainly be faster than moving (1) mic around and averaging individual measurement points. But, I am not sure if I would be getting reasonable data. I am also not sure how SMAART would determing the microphone delay - thus getting back to my concern about valid data.

ThanksI guess it boils down to what you think "reasonable" data is. If you were to do things right, then a digital mixing console would be needed for all those mics, each input having a separate delay to time all the other mics back to the furthest mic from the system. Things get a bit complicated when you have delay fills involved what do you reference your time to, and how do you account for different time delays from one part of the system to another. For the expense and trouble involved, you can upgrade to version 7 and buy yourself a multi-channel interface from Presonus or Roland and do multi mic measurements and averaging, along with a whole lot more! Take the plunge, you'll be glad you did!

martindale
August 15th, 2012, 10:42 PM
This is an interesting thread/ question. It's not necessarily true this multiple mic placement measuring "will not yield useful data."
I'm thinking of the way the Dolby guys measure and tune commercial film theaters. Some of my work includes tuning film mix rooms, and I've worked with Dolby learning their measuring methodology. I don't agree with everything they do, nor use it in every situation, but you can in fact get very useful information by looking at mics in multiple locations, and averaging the response you see.
How you obtain this averaging is the question. The Dolby guys use an interface which includes a multiplexor, scanning each mic (up to six), one after the other, over a specific time period.
This really only works for an RTA display, insofar as using SMAART. The output of the multiplexor is sent as the single data stream from the interface into SMAART.
There is no need to worry about setting time delays for each mic position--you are looking at a simple RTA readout.
The RTA display will, of course, keep changing, as the multiplexor looks at each successive mic. Your display decay time is critical in getting something useful from it. You have to learn how to "read" this changing display, and make sense of it, and if this is part of your gig, make EQ adjustments accordingly. If you're considering how a theater responds from different seats, this is quite interesting.

Like I said, I don't use this methodology myself for all work, and in fact only will do it when needing to conform to the requirements that Dolby makes when tuning a commercial theater. But I thought I would mention it, because it is a valid measurement technique, and can be used quite effectively.
BTW, I don't know of any commercially available mic multiplexor that would do exactly this; the box Dolby uses is a proprietary build.
Of course, you can also use SMAART v. 7 in the way described elsewhere in this thread.

Arthur Skudra
August 16th, 2012, 03:32 PM
This is an interesting thread/ question. It's not necessarily true this multiple mic placement measuring "will not yield useful data."
I guess it boils down to context. Keep in mind that cinema systems you're dealing with playback and sound reproduction mostly. The reason why Dolby/THX are successful in doing things the way they do is consistency of venue and equipment from sound stage to theatre, and their strict adherence to using certified equipment and similar layout and acoustics. It's a proven formula, so using a multiplexor, bunch of mics arranged in a certain way and RTA in one venue will allow one to achieve the same results in another. Do a search on www.film-tech.com to find some interesting manuals on what is involved.

On the other hand, in a typical house of worship, you doing live sound reinforcement, each is unique in their acoustical and sound system challenges, I wouldn't want to try to do any system alignment work without the benefit of a dual channel FFT. Though during a service, I wouldn't hesitate to use single channel spectrograph along with a dual channel transfer function to see what's going on. Again context is key.


BTW, I don't know of any commercially available mic multiplexor that would do exactly this; the box Dolby uses is a proprietary build.
Of course, you can also use SMAART v. 7 in the way described elsewhere in this thread.
Sencore used to make one, they seem to be getting out of the portable audio test equipment business and focusing on the video side of things, you can pick up one of their rebadged Terrasonde Audio Toolbox handheld units along with a mic multiplexor real cheap on eBay right now. Not that many left AFAIK. Goldline USA makes a mic multiplexor as well.

martindale
August 16th, 2012, 06:01 PM
I agree Arthur--as you said it's the context. As I said, I would not use this technique in other analysis and tuning scenarios. But using the 'plexor and learning how to interpret the display as the mics move around the room--and then applying eq to meet an averaged X curve-- is good experience for your ears and head. You take everything you learn and bring it to the next job.

Harry Brill Jr.
January 28th, 2013, 11:35 PM
This has been done with some success. Simply plug all of your mics into a digital console that has delay on each input. Once you have the gains set, mute all but one mic and find the delay time and insert it into that channel instead of Smaart's reference delay. Repeat this for all input channels, then turn all the mutes off and you will have your live spatial average. It is easier to buy Smaart 7 and a Roland Octapre, than to use Smaart 6 and a digital console with delay on each input. You get the advantage of seeing all the individual traces and the average at the same time.