Understanding More About B Format Measurements and CATT ReflPhinder

[Trdat]

Apologise about the title as what I actually want to know is if you can get directional understanding using CATT ReflPhinder with an Omni mic? (Where the reflection comes from)

And do we know about its spectral content from the software? Or what frequency that reflection is? And if that comes from a normal sweep or do I have to use a sinewave at a specific frequency to know that the reflection is from that frequency?

On the CATT website it says "reflection direction analysis based on synthesized classical 1st order polar microphones"

But I also read in some B format articles that you need a ambisonic mic to be able to get directional analysis.

 

[No. 5]

Apologise about the title as what I actually want to know is if you can get directional understanding using CATT ReflPhinder with an Omni mic? (Where the reflection comes from)

Like a Dayton OmniMic, or other omnidirectional microphone? Unfortunately not, as those types of microphones are purely pressure sensors and are unable "see" any directional data by themselves.

And do we know about its spectral content from the software? Or what frequency that reflection is? And if that comes from a normal sweep or do I have to use a sinewave at a specific frequency to know that the reflection is from that frequency?

I have not used that software, so I can't comment on what it can do, but you'll find that any reflection will have significant bandwidth so a sweep would certainly be preferred.

But I also read in some B format articles that you need a ambisonic mic to be able to get directional analysis.

They work well for it, but I doubt that an ambisonic mic is the only way to do directional analysis, you could probably do something useful with a binaural microphone as well.

 

[Blumlein 88]

The $399 MiniDSP ambisonic mic is probably the least expensive ambisonic microphone you can get.

ambiMIK-1

Introducing ambiMIK-1, an affordable First Order Ambisonic (FOA) microphone with embedded preamp, Analog to Digital converters (ADC) and USB processor. This compact solution enables high sample rates (up to 32bit/192kHz) in a portable package. With applications ranging from VR/AR & music...

www.minidsp.com

Don't know anything about that software.

 

[Trdat]

The $399 MiniDSP ambisonic mic is probably the least expensive ambisonic microphone you can get.

ambiMIK-1

Introducing ambiMIK-1, an affordable First Order Ambisonic (FOA) microphone with embedded preamp, Analog to Digital converters (ADC) and USB processor. This compact solution enables high sample rates (up to 32bit/192kHz) in a portable package. With applications ranging from VR/AR & music...

www.minidsp.com

Don't know anything about that software.

There is the Zoom H3 V3 is about $250

The only other question is do I need a 4 channel recorder to go with it? Well, with ambiosonic mic in general...?

The Zoom V3 I haven't totally understood I think it has its own recorder or not.

 

[Trdat]

Like a Dayton OmniMic, or other omnidirectional microphone? Unfortunately not, as those types of microphones are purely pressure sensors and are unable "see" any directional data by themselves.

Great, thanks for that. That is what I thought but the software write up mentioned 1st order polar mics so I didn't really understand what that meant and if it meant omni mics. I have a standard omni mic.

I have not used that software, so I can't comment on what it can do, but you'll find that any reflection will have significant bandwidth so a sweep would certainly be preferred.

I just wanted to understand if a tone had to be generated or a normal sweep would provide the directional analysis. It seems a sweep is fine.

They work well for it, but I doubt that an ambisonic mic is the only way to do directional analysis, you could probably do something useful with a binaural microphone as well.

Well, whatever is cheaper really. Do I need a 4 channel recorder with these types of mics? I haven't fully undertood the full implementation, I came across the need of a 4 channel recorder, I am presumng instead of a soundcard that an omni mic connects to with an ambiosonic you feed it into a 4 channel recorder, is this correct?

 

[No. 5]

Great, thanks for that. That is what I thought but the software write up mentioned 1st order polar mics so I didn't really understand what that meant and if it meant omni mics. I have a standard omni mic.

You bet! Yeah, it's an esoteric detail of a niche technology. There are, in fact, multiple "orders" of Ambisonic microphones with more capsules than the B-Format.

Well, whatever is cheaper really. Do I need a 4 channel recorder with these types of mics? I haven't fully undertood the full implementation, I came across the need of a 4 channel recorder, I am presumng instead of a soundcard that an omni mic connects to with an ambiosonic you feed it into a 4 channel recorder, is this correct?

Since a B-Format microphone is actually four mic capsules (three figure eights and an omni), yes, you do need four recording channels, one for each of the capsules. I've seen a couple instructional articles on DIYing this kind of mic, I haven't priced it out, but that might be an affordable option if you have a sound card with four input channels.

 

[Trdat]

You bet! Yeah, it's an esoteric detail of a niche technology. There are, in fact, multiple "orders" of Ambisonic microphones with more capsules than the B-Format.

Since a B-Format microphone is actually four mic capsules (three figure eights and an omni), yes, you do need four recording channels, one for each of the capsules. I've seen a couple instructional articles on DIYing this kind of mic, I haven't priced it out, but that might be an affordable option if you have a sound card with four input channels.

Okay thanks, really appreciate it.

Last questions, the minidsp ambiMik-1 its a USB mic that connects directly to the comp can that be used without a 4 channel recorder directly with REW? Or any other appropriate program?

And any other programs like ReflPhinder that can help identify where the reflection comes from and its spectral content? Maybe ARTA?

 

[No. 5]

Okay thanks, really appreciate it.

Certainly!

Last questions, the minidsp ambiMik-1 its a USB mic that connects directly to the comp can that be used without a 4 channel recorder directly with REW? Or any other appropriate program?

Yes to an AmbiMIK-1 being able to be plugged directly into a computer and used with appropriate software without a four channel recorder being needed, hard no on it being able to be used with REW. Although REW can use at least two of the mic channels from the AmbiMIK-1, it has no ability to process spatial audio and is incapable of giving you insight into arrival directions.

And any other programs like ReflPhinder that can help identify where the reflection comes from and its spectral content? Maybe ARTA?

ARTA has the same problem as REW. Last time I looked into this, I was unable to find much of any software that could do this, much less any that was affordable. If I find anything, I'll pass it on since I've been interested in doing the same thing for some time.

 

[Trdat]

Certainly!

Yes to an AmbiMIK-1 being able to be plugged directly into a computer and used with appropriate software without a four channel recorder being needed, hard no on it being able to be used with REW. Although REW can use at least two of the mic channels from the AmbiMIK-1, it has no ability to process spatial audio and is incapable of giving you insight into arrival directions.

So I think I have understood correctly, with the minidsp ambisonic USB mic and with a software that can manage the 4 recorded channels and of course provide 3d IR directional cues than it work without a 4 channel soundcard but it seems there isn't any programs appropriate at the moment, at least affordable that is.

It seems we both are looking for the same thing, an affordable way to do B format measuring. Please let me know when you reach your research goals. What about the CATT software that I first mentioned would that work directly with the minidsp ambimic? I know there is other options but prices are around 1k to 2k including software and mics that is a little steep just to work out where the reflection comes from. So yeh, our search will continue.

Lastly, what about the Zoom H3 V3 it seems that it records directly to its own program but then how do I convert that information to a REW type measurement. Or does it connect to REW or another software. Again presuming this doesn't need a 4 channel soundcard.

 

[No. 5]

So I think I have understood correctly, with the minidsp ambisonic USB mic and with a software that can manage the 4 recorded channels and of course provide 3d IR directional cues than it work without a 4 channel soundcard...

You got it!

but it seems there isn't any programs appropriate at the moment, at least affordable that is.

It does seem that way. Although this thread inspired me to do another quick search into possible options, and I found this. But it requires Matlab and A-Format Ambisonic data. I also found a fairly cheep capsule mount to DIY a Ambisonic microphone. It's not much, but it's something!

It seems we both are looking for the same thing, an affordable way to do B format measuring. Please let me know when you reach your research goals. What about the CATT software that I first mentioned would that work directly with the minidsp ambimic? I know there is other options but prices are around 1k to 2k including software and mics that is a little steep just to work out where the reflection comes from. So yeh, our search will continue.

It dose indeed seem that way, yes. I'll let you know if anything comes of it. I would imagine that the ambiMIK-1 would work with ReflPhinder, but I did come across an article today where someone used a Zoom H3-VR for doing exactly what we are talking about. And a licenses for ReflPhinder is 180 Euro's, so it's not actually too bad... although that's more than I would be comfortable spending on a project like this.

Lastly, what about the Zoom H3 V3 it seems that it records directly to its own program but then how do I convert that information to a REW type measurement. Or does it connect to REW or another software. Again presuming this doesn't need a 4 channel soundcard.

I don't have a Zoom H3 and I haven't read any of the documentation, so I can't give you anything really useful, I notice the Zoom has a USB output, so it might be able to be connected to the computer that way? The article I linked seems to indicate that you can. Otherwise recording test signals with the Zoom and them importing them into the software of your choice might be possible.

 

[Trdat]

It dose indeed seem that way, yes. I'll let you know if anything comes of it. I would imagine that the ambiMIK-1 would work with ReflPhinder, but I did come across an article today where someone used a Zoom H3-VR for doing exactly what we are talking about. And a licenses for ReflPhinder is 180 Euro's, so it's not actually too bad... although that's more than I would be comfortable spending on a project like this.

I don't have a Zoom H3 and I haven't read any of the documentation, so I can't give you anything really useful, I notice the Zoom has a USB output, so it might be able to be connected to the computer that way? The article I linked seems to indicate that you can. Otherwise recording test signals with the Zoom and them importing them into the software of your choice might be possible.

Yes, I read that article as well in SynAudCon website by Hadi Sumoro it seems that you can get away with just the Zoom H3 V3 and REflPhinder but not sure if it connects directly to the program or if you are loading the info into program after you have taken the measurement. On top of this is the data from the measurement ready to be transferred or it needs some conversion before we can upload it to ReflPhinder? (This wasn't mentioned in the article)

You can connect the Zoom with USB but not sure if that is directly to the third party software its more like just to transfer data. If this is the case then we are limited with the Zooms application if other third party software comes out.

 

[No. 5]

In my limited experience, analysis software often like using "known" signals (i.e. generated by themselves) rather than something prerecorded, but as I have nether ReflPhinder or an H3, I'd just be guessing.

The H3's documentation says that "you can connect it to a computer to use it as a 360° USB mic" and "4ch Ambisonics: This sends tracks 1–4, which are converted from input signals 1–4, as separate signals to the computer" so it looks to me like it can connect directly with whatever software is on the computer.

 

[Trdat]

In my limited experience, analysis software often like using "known" signals (i.e. generated by themselves) rather than something prerecorded, but as I have nether ReflPhinder or an H3, I'd just be guessing.

The H3's documentation says that "you can connect it to a computer to use it as a 360° USB mic" and "4ch Ambisonics: This sends tracks 1–4, which are converted from input signals 1–4, as separate signals to the computer" so it looks to me like it can connect directly with whatever software is on the computer.

I wonder if in the article the H3 is directly connected to the ReflPhinder with the software picking up the IR response automatically. But then how is the sweep generated?

My understanding is that you will need to tranfer the file across to a computer from the Zoom convert the A format recording to a B format on the free ambisonic software that comes with the Zoom. Then some process of trimming is involved because the sweep is generated from another source hence it needs to be trimmed to the actual lenght of the IR and then tranferred into ReflPhinder. A bit of a process but for $250 and a $180 I think its the best we are going to get but then again I could be wrong and ReflPhinder might automtically pick up the IR.

 

[No. 5]

Looking through ReflPhinder's documentation, it appears that that importing a recorded measurement sweep means importing the "dry" sweep and the recording from the Ambisonic microphone (so that the software has something to compare), or it can produce it's own signal. Also interesting, it seems that a "first order" Ambisonic microphone is limited to 10 degrees of angular resolution for direction finding. I still think it's worth trying out.

 

[Trdat]

Looking through ReflPhinder's documentation, it appears that that importing a recorded measurement sweep means importing the "dry" sweep and the recording from the Ambisonic microphone (so that the software has something to compare), or it can produce it's own signal. Also interesting, it seems that a "first order" Ambisonic microphone is limited to 10 degrees of angular resolution for direction finding. I still think it's worth trying out.

I spoke to an acoustician and he mentioned that ZOOM H3 does it own conversion internally but the better option is to convert it in the free mabosnic mix software that is provided and you will get a better conversion.

But when you say importing the dry sweep you mean the format A without it converted to format B?

And, the limit of 10 degrees will give us how accurate? So from one of the microphones Refphinder will only give you the location of the reflection maximum 10 degrees from that the centre of the microphone?

 

[No. 5]

I spoke to an acoustician and he mentioned that ZOOM H3 does it own conversion internally but the better option is to convert it in the free mabosnic mix software that is provided and you will get a better conversion.

Good to know.

But when you say importing the dry sweep you mean the format A without it converted to format B?

In order to determine any system's transfer function, the original ("dry") signal must be compared to the signal that came out of the system. So in this case, if you use a recording of a sweep to measure the room and capture the sweep playback with an Ambisonic microphone, the software will need the original recording of the sweep so that it can compare it to what was captured with the Ambisonic microphone.

And, the limit of 10 degrees will give us how accurate? So from one of the microphones Refphinder will only give you the location of the reflection maximum 10 degrees from that the centre of the microphone?

That's what it sounds like from the documentation. It's probably not "hard" steps of 10 degrees though, probably more like how 1/3 octave smoothing gently smooths a higher resolution spectral measurement.

 

[Trdat]

Good to know.

In order to determine any system's transfer function, the original ("dry") signal must be compared to the signal that came out of the system. So in this case, if you use a recording of a sweep to measure the room and capture the sweep playback with an Ambisonic microphone, the software will need the original recording of the sweep so that it can compare it to what was captured with the Ambisonic microphone.

That's what it sounds like from the documentation. It's probably not "hard" steps of 10 degrees though, probably more like how 1/3 octave smoothing gently smooths a higher resolution spectral measurement.

Can I ask what you want to use it for?

I ask as because you mentioned $250 plus the $180 for reflection Refphinder is a little more than what you would like to spend and I also trying to justify the cost.

 

[No. 5]

Certainly: we can localize sound and perceive a three dimensional sound field, but an omnidirectional microphone cannot. I have been coming to understand that this ability is a relevant and often overlooked part of acoustic analysis and our preferences for for a "sound". So this is something that I'd like to investigate further so that I can perhaps understand it better. Additionally, it would be helpful in analyzing a room to be able to know the where, when, and the level, of reflections.

I made myself a "starter" binaural microphone system a few dollars, and I'm hoping that will be a starting place to look into the spatiality of a sound field, but I can’t justify spending over $400 on my curiosity. I'm a fan of the DIY rout, so following build instructions or designing and building myself is the way I prefer to do some things, and it can be cheaper, additionally, I'm not a programmer, so open source options for the software would be my preference. So if there’s a low cost open source/DIY alternative to ReflPhinder and an H3, I’d probably jump on it.

 

[Trdat]

Thanks, it seems we are both interested in understanding more about the same concepts and both struggling to justify the costs involved. I am keen but I will sit on it for a little longer till I go with a the Ref and Zoom purchase.

The acoustician I spoke to mentioned that he uses IRspatial-Ambix(his program), to find direction of reflections. You can download it here, http://www.angelofarina.it/Public/B-format/Software/

If you can work out how this program actually finds reflections please let me know but I am baffled.

 

[No. 5]

Thanks, it seems we are both interested in understanding more about the same concepts and both struggling to justify the costs involved. I am keen but I will sit on it for a little longer till I go with a the Ref and Zoom purchase.

Imagine being one of those people who can spend hundreds of dollars on a whim!

The acoustician I spoke to mentioned that he uses IRspatial-Ambix(his program), to find direction of reflections. You can download it here, http://www.angelofarina.it/Public/B-format/Software/

If you can work out how this program actually finds reflections please let me know but I am baffled.

Interesting, thank you for sharing!

I installed it, but since I don't have any Ambisonic recordings to try it our with, I don't know how it would show reflections. I didn't see a Z axis... but that may be visible if it has a file loaded.