Hi.

Here's a test that I done for a question that Ray asked on another post. I've created a new post for this test so as to not distract the other post, plus I think other users may find this test interesting.

The phase test portion of this video is absolutely conclusive. The Pocket does record an external mono microphone exactly the same to both the left and right channels of its recording.

Here's the blurb about the test that I did for my YouTube description.

DJI Osmo Pocket audio adaptor external mono microphone left and right channel analysis

This is a test primarily for members of the DJI forum but I've made it public on my YouTube channel as it may be useful for other Pocket owners.

This test is to see if the Pocket with its external 3.5mm microphone adaptor is recording a mono microphone dialogue source equally to the left and right channels, which is what's to be expected when a mono microphone is connected via a TRS 3 pole cable & jack and the first two poles, Tip & Ring are normalised.

This test does prove that both channels are equal which is seen by the metering in both my NLE, Edius and my DAW, Cubase. The point is further proven by a phase cancellation test where I reverse the phase 180 degrees of one channel, centre both channels and cancellation is achieved.

Please call back to my YouTube channel and have a look at my Osmo Pocket playlist as I will be adding further videos about the Pocket and various technical tests. I will also be doing some small/nano microphone builds specifically for the Osmo Pocket to take full advantage of its size.


Cheers,
Dave.

Well done again Dave.
I am totally convinced by this test that the two channels are equal in signal and amplitude.

I have subscribed and pressed the bell so I will not miss any of your future testing.

Bob

BobWinNV Posted at 3-23 18:15
Well done again Dave.
I am totally convinced by this test that the two channels are equal in signal and amplitude.

Hi Bob.

You shouldn't have clicked on that bell icon thing, my videos are dead boring

I'm glad you liked the test video, I just thought it was only fair to give some solid evidence of what the Pocket can do.

I've changed my tune a bit with regard the Pocket after tempering my expectations and am way more interested in what it can do and so far it's proving to be the best system I've ever had for a pro video/dialogue acquisition tool that fits in my pocket. I usually refer to such a setup as a nano recording system and it truly is.

While I know that there are issues, in fact, I've worked out some other stuff but it's just negative crap that would only help fuel those who'd rather bash the system than work out what it's good for. I won't be posting those results but will post stuff that I think will help people.

I'm really starting to enjoy this Pocket setup with the Clippy mic, so much so I'm thinking about becoming a vlogger

Cheers,
Dave.

Thanks Dave.
I think I'm going to have a look at my audio amps. Check to see if there is any cross talk or a breakdown in the physical audio chain somewhere.  I've some calibrated audio tones somewhere I can work from as it seems the channel imbalance is at my end. You certainly know your Cubase shortcuts, the speed you got through most of those
I'm still not sure why one channel would import only to a mono track on my system while the other went to a stereo track. It could account for the 6dBm difference on the meter reading of that side.
I will run through Cubase again and check the bus routings. I've had it set up quite simply. No talkback channel or throughput studio monitoring, just the main in and output busses.
Weirdly though as again at my end, the audio through Cubase sounds cleaner to me than when playing the video.
Thanks again for your efforts, Dave. Much appreciated.
Now all you have to do is prove we have two independently recorded channels inside the Pocket and not some weird hybrid output

I've had a quick check before retiring for the night Dave.
I do seem to be getting a very small amount of audio bleedthrough from somewhere minuscule as it is.
I don't have a lot of storage webspace left to upload a video but took a screenshot to show the inverted phase test came out similar to your own. I just get the odd transient bleed through which is annoying and must be related to my audio problem.
Thanks again for pointing me in the right direction and can confirm your findings.

Ray.

Ray-CubeAce Posted at 3-23 18:47
Thanks Dave.
I think I'm going to have a look at my audio amps. Check to see if there is any cross talk or a breakdown in the physical audio chain somewhere.  I've some calibrated audio tones somewhere I can work from as it seems the channel imbalance is at my end. You certainly know your Cubase shortcuts, the speed you got through most of those
I'm still not sure why one channel would import only to a mono track on my system while the other went to a stereo track. It could account for the 6dBm difference on the meter reading of that side.

Hi Ray.

Don't be fooled by that video, I only know two shortcuts that's why I was pulling up the wrong things Back in the old Atari MIDI only days I was really fast with it but I was a lot younger then and had more grey matter. It would appear that I've now swapped grey matter for grey hair

From what I've tested there are defo two independent/discreet tracks being recorded. Even if there are phase issues between them, this can only happen with two discreet channels. The problem is the input and not the output. But to be honest, I only got my adaptor this morning and I'm so happy with the results with the Clippy mic that I'm not really concerned if I don't get anything else working with it.

Or let me explain it this way, in case what I've just said sounds flippant.

For less than £400 this setup is capable of excellent UHD picture quality that's stabilised and with the mic adaptor and the Clippy lav gives great dialogue quality.

Yes, there are hardware problems, yes, there's software issues, and yes, a bunch of stuff needs ironing out that shouldn't be there. But during the last few weeks I've been using the Samsung Galaxy S10 and the GoPro Hero 7. Both of these devices are excellent examples and pinnacles in their respective fields and each also has their own issues. On balance, with individual issues taken into account and nothing being perfect, I'm just having too much fun with the Pocket and recording some stuff that's simply awesome.

Or another way to look at it. I used to have to pay £400 for just one 8MB RAM module for my old Akai samplers.

Cheers,
Dave.

Not surprising that a mono mic would be recorded mono.
I agree with Daves comments above - there are 2 hardware channels into the OP, but there is voodoo within.

I do not think the hardware in the audio path is mono.
It seems to be subtly stereo. Listen with headphones, close eyes.

Here is my independent test attempting to replicate Dave's mono mic left right channel test.
I am using the Omni Mono microphone Edutige ETM-001 and running my tests with my Osmo Pocket set to audio Auto and noise reduction off.  The audio  recorded by my Osmo Pocket was not processed in anyway.  The audio was brought into the Audacity Software directly from the Osmo Pocket .MOV file.

Dave, you did great, thank you so very much for the endeavor, the tests and videos are quite time consuming....
I totally agree with you, OP is a small tool that offers us freedom and joy when using it, I am looking forward to your coming video, OP with a miniature mic in a very simple setup :-)
And I love that line "negative crap that would only help fuel those who'd rather bash the system than work out what it's good for"

Have a nice weekend my friend.

Excellent work

lummox Posted at 3-23 22:04
I do not think the hardware in the audio path is mono.
It seems to be subtly stereo. Listen with headphones, close eyes.

Hi lummox,

Yes, I hear a little difference. But it is not clear, you always hear the sound on both channels. Maybe the other microphone should be completely covered for the test.

David_Harry Posted at 3-23 19:21
Hi Ray.

Don't be fooled by that video, I only know two shortcuts that's why I was pulling up the wrong things  Back in the old Atari MIDI only days I was really fast with it but I was a lot younger then and had more grey matter. It would appear that I've now swapped grey matter for grey hair

Hi Dave.
I think you are being modest. I had a keyboard with the shortcuts printed on them sometime back before it broke and still had trouble
It was never my intention to be one of the people knocking the Osmo Pocket but wanted simply to find out why certain things were happening to the audio channels. In the field, with the inbuilt mics, I noticed bird song, rapid water movement, and some speech high-frequency oddities while other high-frequency sounds seemed to be reproduced fairly accurately. In short, it bugged me and want to find a way to negate these small hiccups if possible, hence my early adoption of the audio adaptor in the hope that an external source would improve things. It's not a deal breaker and never has been but the less post processing I have to do at any one time the better.
I've just been really grateful anyone else can also be open-minded and do tests to greater accuracy than myself that is equally interested in the benefit of the product.

I have come to the same conclusion as yourself, whether it is some discrete component in the input Osmo Pockets channels that is causing a small problem with audio gain causing some form of random distortion into the recording or some sort of coding error in the software I'm not sure. It could just be an oddity no-one at the time could foresee with the chosen components. Maybe we will never get to the bottom of it, who knows? But what you have proven and what I suspected from my line level experiments is that given enough level on the input channels we can at least get identical waveforms on both channels and as more of the bit depth gets used, that the audio quality improves quite a bit despite the initial 192kbs audio rate with the AAC coding.
You have further proved that providing the correct mic capsules can be obtained that the Osmo Pocket can give a decent account of itself without the need for a powered microphone. That the background noise level when silent is of an acceptable nature.
As far as I'm concerned, all good news and I have not at any time thought you have been flippant.
The Osmo Pocket is good value for money and fills a niche that needed filling.
So, and sorry to put you on the spot once again but your experiences have outweighed my own.
Do you think adding a second capsule will give more audio gain per channel than just using one across both or do you foresee it causing another problem?
I have a two-day event to record the third week of April that I would like to be confident in getting the best output I can. That is when my experimenting and accumulation of best practice I can manage from the Pocket has to be sufficient. Up to that time I have the opportunity to play and experiment.

Ray.

Udo13 Posted at 3-24 00:04
Hi lummox,

Yes, I hear a little difference. But it is not clear, you always hear the sound on both channels. Maybe the other microphone should be completely covered for the test.

I tried that Udo, results are the same. On playback there is amplitude even on the covered up mic. And when talking into one mic while covering up the other channel, the uncovered mic cuts in and out. It’s strange.

BobWinNV Posted at 3-23 23:24
Here is my independent test attempting to replicate Dave's mono mic left right channel test.
I am using the Omni Mono microphone Edutige ETM-001 and running my tests with my Osmo Pocket set to audio Auto and noise reduction off.  The audio  recorded by my Osmo Pocket was not processed in anyway.  The audio was brought into the Audacity Software directly from the Osmo Pocket .MOV file.

Hi Bob.

I see what you were saying about the Edutige

BTW. That little bit of audio that's left over after the phase cancellation is likely to be the difference between the two tracks after the AAC encoding. As there's probably slightly different encoding artefacts on each channel and those differences won't cancel. It's a bit similar to how RFI & EMI etc. is rejected in a balanced mic system.

Cheers,
Dave.

David_Harry Posted at 3-24 20:04
Hi Bob.

I see what you were saying about the Edutige

Well we have a sample of two now to prove you're conjecture.   

I also think that the Edutige ETM-001 is a good back up to the Clippy EM-172.  Not quite as good but easier for folks on this side of the pond to get in their hands for under $30 US.

I have enough parts to make four microphones on order from micbooster.com and the Clippy Mono that I ordered yesterday as well.  

Bob

BobWinNV Posted at 3-24 20:21
Well we have a sample of two now to prove you're conjecture.   

I also think that the Edutige ETM-001 is a good back up to the Clippy EM-172.  Not quite as good but easier for folks on this side of the pond to get in their hands for under $30 US.

Hi Bob.

The Edutige in your example does sound very good and has a very healthy level and matches very well with the Pocket. I'd imagine that the main difference between it and the Primo, is that the Primo may have a lower inherent noise floor, maybe a test in their when you get your Clippy mic.

Cheers,
Dave.

David_Harry Posted at 3-24 20:29
Hi Bob.

The Edutige in your example does sound very good and has a very healthy level and matches very well with the Pocket. I'd imagine that the main difference between it and the Primo, is that the Primo may have a lower inherent noise floor, maybe a test in their when you get your Clippy mic.

Yes Dave, there will be a test but I am already convinced that the EM-172 does have a lower noise floor by some good bit.
Bob

Very good one!

Thanks for sharing!