Insta360 Mic Pro - New wireless mic with 32-bit stereo internal recording

[palm0014]

[.....]
One take away-
Because the 3 MEMS in the Insta360 "puck" are positioned in a triangle layout, orientation of the puck (on-edge, rather than on-face) will matter with regard to the axis of the resulting mono cardioid, as well as the resulting stereo axis.  If all 3 MEMS were spaced along a single line instead of in a triangle, edge-on verses face-on orientation wouldn't matter based on sensor geometry, although the housing shape could have some influence.

I've spent a happy hour or two doing intensive testing (at home but as scientifically as I could) with the Mic Pro recording in stereo, mounted vertically vs mounted horizontally.  I first tested it on white noise from a Tannoy hifi speaker very close (obviously not actually white, but still a basis for comparison) in both alignments, and then using it to record stereo music from the hifi, on a stand, horizontal then vertical.  Lastly I tried it also in front of the hifi, mounted vertically on my chest, and then horizontally under my cap peak.

Overall, as Gutbucket suggested, it's best horizontal, but the differences between that and vertical are not a matter of good vs disaster.  Mounted vertically the stereo image is not seriously degraded, and while I think the response is flatter and slightly crisper when horizontal, I had to compare a lot to reach a conclusion. 

So - if you can mount it horizontally, that's best, but if it has to be used vertically, (front capsule up), then it's not a disaster.  And that's how I recorded that Cathedral pipe organ recital the other day, chest mounted.  On that occasion I did have a plan for subtle horizontal mounting - not polite to wear a cap in church! - I stuck a small metal plate onto the top of my umbrella handle and was going to attach the Mic Pro to that via its magnet, then sit with it kind of between my knees, effectively like a covert mic stand.  But the day turned out sunny, and the umbrella would have caused puzzlement to the Cathedral staff... so on my chest it went.

Thank you for these tests!  I was trying to figure out how to wear a nice hat to the National Arts Center in the fall, but now I don't have to overthink it and simply attach the mic to something more sensible.

[Ozpeter]

Something that might be of academic interest here, as it only affects mono recordings made with the small wind muff fitted, not stereo recordings with the large wind muff - but if you do use the Mic Pro in the wind in mono, just use the omni polar pattern as that only uses the omni mic at the front, under the small wind muff.  Cardioid pattern (etc) uses the side mics to construct the polar pattern simulation, and thus it picks up wind noise from those side mics.

But for stereo, in the wind, it's essential to use the optional larger wind muff.

Just sayin'....

[Gutbucket]

Vertical orientation got me thinking back on an old idea I had but never tried.. When open taping using a tall stand in situations where a wide mic-bar appropriate for spaced omnis isn't doable, I've considered wide spacing a pair of omnis vertically rather than horizontally, positioned above and below a regular horizontally oriendted coincident pair rather than being left/right of that pair.  No useful positional imaging that way from the omnis, but the directional pair would handle that, and should still achieve appropriate spaced omni phase decorrelation to a low enough frequency, improving the sound of the bass and providing a nice open ambient "air".  I should try that sometime.

I carry an umbrella on my walk to lunch here in summer, which serves as parasol more often than as an umbrella.  Best way I've found to manage the Florida sun and heat. Far better than a big straw hat, I'd be drenched in sweat without it. I do the same traveling and hiking, when it serves a third purpose as walking stick.  Carrying it does occasionally draw odd looks on sunny days (good call on avoiding puzzlement of the cathedral staff), but also sometimes knowing nods.   

When I was running an ambisonic TetraMic I considered building it into the head of a walking stick somehow to use exactly like you mention, but never took it farther than the concept.  Such a thing would be much easier with the Insta360.

[Ozpeter]

AI reminded me how to reset it - hold down the power button for ten seconds.

Now it behaves consistently in the following way -

LEDs on - pressing the record button starts / stops recording reliably (and it doesn't seem particularly fussy about button press time, within reason).

LEDs off - pressing the record button once makes the blue light come on for a few seconds, but recording does not start.  Press the red button a second time while the blue light is still visible, and the red light comes on for a few seconds and recording starts.  To stop recording, the record button has to be pressed once to get the lights back on, and immediately pressed again to stop recording.

palm0014, I wonder whether yours does actually have the same problem, which you interpreted as a problem with pressing the button too long?  I am wondering because if you pressed the button and nothing happened, maybe you pressed it again shortly after, thus satisfying the apparent requirement to press it twice to get recording started.

I do think that it's a bug....

You did help me figure out an issue I was having with stopping the recording!  I didn't know that I had to press it (or any button for that matter) once to get the lights back on, and then press again to stop the recording.  You solved a mystery of mine, thank you!  Last week while ending my recording, I was getting fed up and would just power down the device.

Now I'm starting to think that this might not be a bug persay, but the way it works.

If LEDs off - pushing any button will bring the device to life and then you can push any buttons to your heart's desire and they will work?

I'm glad you reported it as it would be good to get a response from Insta360.

Insta360 have now confirmed that this undocumented 'feature' isn't a fault, it's the way it is designed to work.  I can see that it has some logic to it (it could help to avoid accidental button presses, as you have to press twice), but the fact that users can only discover how to record with the LEDs off by trial and error is regrettable.  I've flagged it on the Reddit forum and will make a YT video tomorrow to draw at least some attention to the matter.

[Ozpeter]

Vertical orientation got me thinking back on an old idea I had but never tried.. When open taping using a tall stand in situations where a wide mic-bar appropriate for spaced omnis isn't doable, I've considered wide spacing a pair of omnis vertically rather than horizontally, positioned above and below a regular horizontally oriendted coincident pair rather than being left/right of that pair.  No useful positional imaging that way from the omnis, but the directional pair would handle that, and should still achieve appropriate spaced omni phase decorrelation to a low enough frequency, improving the sound of the bass and providing a nice open ambient "air".  I should try that sometime.

I carry an umbrella on my walk to lunch here in summer, which serves as parasol more often than as an umbrella.  Best way I've found to manage the Florida sun and heat. Far better than a big straw hat, I'd be drenched in sweat without it. I do the same traveling and hiking, when it serves a third purpose as walking stick.  Carrying it does occasionally draw odd looks on sunny days (good call on avoiding puzzlement of the cathedral staff), but also sometimes knowing nods. 

When I was running an ambisonic TetraMic I considered building it into the head of a walking stick somehow to use exactly like you mention, but never took it farther than the concept.  Such a thing would be much easier with the Insta360.

Wide stereo rigs can indeed be a bit over the top visually - that's why I always liked the MKH M/S rig I used for classical concerts, as it simply pointed front and back with minimum width.  One piano recital client was very surprised when she received her copy of her performance, as she thought I hadn't turned up, due to the low profile of the recording (and me).

The Mic Pro is indeed so adaptable to various low profile mounting methods.  I don't use its clip at all, just the magnet - helped by my discovery that I happened to have a  surprising number of magnet attracting mounting accessories in the cupboard.  Please don't laugh, but now I wear a magnetic disc on a thread round my neck under my top layer of clothes,  so that I can instantly whip the Mic Pro from my trouser pocket and attach it to my shirt if a sudden recording opportunity arises.  It's quite transformed my attitude to audio recording - no bag full of stuff required, no advance planning, it's much more of an everyday thing now.  I do hope you try one soon, Mr Gutbucket - or even two - I would very much like to hear your expert opinion on it in due course.  

A day or two back I tested my hearing - nothing left about 7kHz, sadly.  I really shouldn't be praising gear I can't really hear, let alone applying EQ to the result... but that's the way it goes.  Sorry, what did you say?

[mccordo]

Anyone who tries using multiple recorders simultaneously please let me know, thanks!

I've got the DFC stage lip table for a show at the local jazz club this Saturday with a local DJ playing with live musicians similar to DJ Logic & Friends. Normally I'd use my DPA 4061 > Deity PR-2, but I bought the Insta360 stereo pair with mixer on a Prime Day deal combined with a gift card. I'll run the stereo pair and post the results here day or so later. This will be my first show with them, as I brought the set along for a 5 show Phish run earlier this month but didn't run them.

[Gutbucket]

^Nice.  No worries if not, but sweet if you can run both rigs, as the comparison against 4061 in that kind of jazz club front-table stage-lip situation should be revealing.

[vwmule]

Posted this earlier but anyone want to take a crack at EQ this recording?

https://drive.google.com/file/d/1FEkOFeVuKNczqfZYROmUhmu-M4a9bFuh/view?usp=sharing

[mccordo]

^Nice.  No worries if not, but sweet if you can run both rigs, as the comparison against 4061 in that kind of jazz club front-table stage-lip situation should be revealing.

That's a great idea and can be easily done! I'll have the results on Sunday morning. 

[Ozpeter]

Posted this earlier but anyone want to take a crack at EQ this recording?

https://drive.google.com/file/d/1FEkOFeVuKNczqfZYROmUhmu-M4a9bFuh/view?usp=sharing

https://drive.google.com/file/d/1d2tIQamP-uLkHUQGhVB4kiUApoKy4YBr/view?usp=sharing

That's just the first bit.  EQ I used, and the result, are included below.  In the frequency response analysis, the green lines are the original, the red lines are after EQ.  I also slightly widened the stereo image and shifted it a bit to the left to make it more centred.

I guess the trouble with recording rock band events is that the sound is totally dependent on the on stage amps, the acoustic of the location, how far back you are, etc etc.  The item having the least effect is probably the recording gear!  Anyway, with a bit of processing this sounds fine to me.

[Ozpeter]

Anyone who tries using multiple recorders simultaneously please let me know, thanks!

I've got the DFC stage lip table for a show at the local jazz club this Saturday with a local DJ playing with live musicians similar to DJ Logic & Friends. Normally I'd use my DPA 4061 > Deity PR-2, but I bought the Insta360 stereo pair with mixer on a Prime Day deal combined with a gift card. I'll run the stereo pair and post the results here day or so later. This will be my first show with them, as I brought the set along for a 5 show Phish run earlier this month but didn't run them.

So how do you use the Mic Pro as a stereo pair?  Recording internally in mono on each one, then syncing to stereo after?  Or transmitting to the receiver and recording with an additional recording device from that (which would lose the 32 bit float advantage of course)?  A comparison with the sound of a pair and the sound of one recording internally to stereo would be interesting but I guess you have to do one or the other.

[Ozpeter]

I've now made a video showing the (not very difficult) method required to record when the Mic Pro LEDs are off. As Insta360 haven't documented this themselves I have saved them the bother - in the hope that others won't trip over this little catch and lose recordings.   Not monetised.

[Ozpeter]

And if you want to see the amazing amount of circuitry inside the Mic Pro, an owner whose device got run over by one or more cars has taken the remains to bits in a video - fascinating!

[mccordo]

Anyone who tries using multiple recorders simultaneously please let me know, thanks!

I've got the DFC stage lip table for a show at the local jazz club this Saturday with a local DJ playing with live musicians similar to DJ Logic & Friends. Normally I'd use my DPA 4061 > Deity PR-2, but I bought the Insta360 stereo pair with mixer on a Prime Day deal combined with a gift card. I'll run the stereo pair and post the results here day or so later. This will be my first show with them, as I brought the set along for a 5 show Phish run earlier this month but didn't run them.

So how do you use the Mic Pro as a stereo pair?  Recording internally in mono on each one, then syncing to stereo after?  Or transmitting to the receiver and recording with an additional recording device from that (which would lose the 32 bit float advantage of course)?  A comparison with the sound of a pair and the sound of one recording internally to stereo would be interesting but I guess you have to do one or the other.

There's a small charging case that holds both Insta360s with their clips and the receiver. The receiver is about the size of a small Anker battery like this one and plugs into the USB-C port of the phone just the same. I don't have it with me right now to verify, but I believe that the receiver syncs the pair, but files will record to each device separately.

For this show I'll have the same table for both the early and late sets, so I can compare recording a single and the stereo pair from set to set. Those can then be compared against my usual DPA 4061 > PR-2 rig that I'll also run. 

[Gutbucket]

^ That's it.

Higher quality and more robust to use the files locally recorded on the separate transmitters and combine them afterward than recording the wireless transmission to the receiver.

As you are aware, we can align two separate recordings in post and stretch/shrink the length of one to match the other if necessary, as tapers have done for decades using separate recorders.  If drift is minimal, alignment alone may be is all that's needed, which is less work.  But longer recordings will tend to drift more by the end and require stretching of one source to match the other.  Generally two identical recorders of the same model have a better chance of the clocks remaining in close sync, depending on the precision of the clock chip used.
^
We can run two of these that way of course, but ideally want to avoid having to do that extra work.

Most of these small recorders are primarily intended for lavalier recording work in support of video, so are designed to support the use of multiple units in close enough sync to match video frame rates.  To do that they all support TimeCode in some way or another, which also help us sync audio. 

That usually can be done in a rudimentary way on lavalier systems such as these without extra gear by syncing the two clocks via their control app prior to recording, after which they run freely and will start to drift after being sync'd. 

Less simply the clocks can maintain more active corrective sync with each other while recording, which in this case probably requires using the receiver.  In some cases is requires a separate timecode generator, but the Mic Pro receiver may be doing that.

Getting into the weeds-
Actual precision required to sync video is much looser than the degree of synchronization required for perfect audio - 30 frames per second compared to 48,000 samples per second (a difference of 3 orders of magnitude).  That's the primary difference in resolution between TimeCode (for video) and WordClock (for audio).  The question becomes, do we really need that degree of perfect sample-rate sync?  Depends on what we are doing with it.  Lip sync to video only requires relatively loose sync.  Sync between two separate non-correlated audio sources (say AUD mics + SBD) needs to remain somewhat tighter than that.  Sync between the two channels of a stereo pair needs to be much tighter because the two channels are phase-correlated.  In a stereo pair, sync between a pair of spaced mics can probably be a bit looser than between a coincident pair, because with a spaced pair some of the content is decorrelated with a randomized phase between channels) while in a coincident pair the phase-correlation is tight, with no phase difference between the two channels (or rather if there is a phase difference it remains constant).