After using the audio adapter a few times now I have two problems with my unit that I'm not sure are common problems or not.

The first is :

Using the adaptor adds another five and a half seconds to the start-up time while the microphone adapter initialises.
Which is odd. OK, the Osmo Pocket takes roughly four seconds to go through its initial wake up while the screen comes on and the gimbal does its own thing. OK, I sort of understand that. But adding the audio adapter means an additional five and a half seconds is needed for the adapter to 'talk to' the Osmo Pocket and record sound.

That means I have to wait twice as long before I get an audio input from the Osmo Pocket compared to using its own internal microphones.

This is very odd as the adapter allows a brief recorded audio burst and then cuts out for five seconds, and then comes back in again which I can show in the waveform below.



Why should this happen? Surely if it is picking up sound at the start of the recording, why then should it drop out for five and a half seconds? I do not believe this to be a fault but some sort of checking system. Yes, I can start the Osmo Pocket up earlier and anticipate when I need to record but it is anoying when the Osmo Pocket itself can be ready in just under half that amount of time.  I should add that this problem is repeatable time after time.

This is not something I have ever come across using any other product that allows the use of an external microphone.


Which brings me to my second and more problematic problem.

There appears to be a DC offset of the signal when recorded. I first noticed this after carefully looking at the signal waveform in my editor, which if you look at the image above, you can see is slightly in the negative voltage area when the signal is silent. To check it was really a DC offset problem I checked it with several other clips that used the audio adapter.

The image below is from one of those clips.




As you can see each channel has slightly different amounts of DC offset. This does not happen when recording directly from the Osmo Pocket.
This is problematic. It means the beginning and ending of clips can produce audible pops and more importantly when amplified would mean any connected loudspeaker is not coming to rest at zero volts when there is no audible signal. Possibly heating voice-coils or other components in the audio chain. Another byproduct of having DC offset is that meters cannot accurately read the output levels of the signal.

This problem can be rectified in post-production but is very time consuming and should not be needed. If I don't remove the DC offset and butt join clips I get audible pops where they meet.

Again. The output from the Osmo Pocket does not have this attribute. It is not the mics. I have tested four microphones that work well with this adapter giving reasonable output levels. More likely there are no blocking capacitors on the analog inputs of the ADC that would have removed this problem.  


If you have an audio adapter and can check for these things yourself please get back to me with your results.

I suspect results will vary. The actual amount of dropout time I have is around five seconds and twenty-two frames when recording at 60fps.

There seem to be all sorts of problems with this device - apart from low levels, the stereo image is poor or corrupted,  DC offset would be just another reason not to buy it.  I get the feeling that DJI may release a new version in due course which might even allow use with the wireless accessory, or maybe combine the two.  Certainly I wouldn't buy one now.  Dual system audio seems to be the way to go if possible.

fansfe82067d Posted at 8-9 14:45
There seem to be all sorts of problems with this device - apart from low levels, the stereo image is poor or corrupted,  DC offset would be just another reason not to buy it.  I get the feeling that DJI may release a new version in due course which might even allow use with the wireless accessory, or maybe combine the two.  Certainly I wouldn't buy one now.  Dual system audio seems to be the way to go if possible.

The Adapter does give two channels but it has never been stereo but rather a mixed mono output.
No idea why. The audio quality itself is reasonable as is the audio output level if you can find a microphone that matches the audio adapter's input. It's a mixed bag of results for sure though.

Hi Ray.

Those are very interesting findings that you have. I’m not sure that I’ve noticed some of them on my setup, which maybe suggests that some of the issues could be specific to your adaptor. I have seen videos that show other users having specific individual problems with their adaptors.

The start up time is something I notice, I’d say it’s probably some kind of initialising delay. I could see how this can be a problem for someone shooting opportunistic footage, for instance, some of those events that you’ve shot. As you wouldn’t want the Pocket wasting its battery, you’d have it powered off as much as possible between takes and then you’d want it back in record mode as quick as possible when there’s some action to film.

I’ll de doing some stuff over the weekend, so will check for the other issues you’ve shown.

I have to be honest though, for the best part, I’m totally happy with my Pocket and mic interface. There’s certain things that I’ve just come to accept and avoid, lack of two channel recording being the main one. There’s no denying that the Pocket’s audio ability is very limited but I’d never complain about the results I get with the internal mics, stereo issues aside, and I’m totally happy with the mono dialogue recordings I get with the adaptor. I’ve not had any problems either with any mics that I’ve connected to my adaptor. Now, I’m very careful to say ‘MY’ adaptor, as I’m fully convinced that there are definite manufacturing and QC issues with these adaptors, which is very evident from your use and that of other users.

I’d go so far as to say that I’m lucky with my setup. But there really shouldn’t be any element of luck with these things, every user should expect the same technical performance as the next user when using identical equipment. Because if this, I can totally understand why some users get very mad because of these problems, I would as well.

I sincerely hope that the Action doesn’t suffer these audio issues when it’s audio adaptor is released. I use my Pocket more as a replacement for a tradition camera, so can accept certain restrictions with that. But I fully expect the Action to be capable of vlogging and interviews by way of a proper audio system.

Cheers,
Dave.

David_Harry Posted at 8-15 03:45
Hi Ray.

Those are very interesting findings that you have. I’m not sure that I’ve noticed some of them on my setup, which maybe suggests that some of the issues could be specific to your adaptor. I have seen videos that show other users having specific individual problems with their adaptors.

Hi Dave.
Thank you for looking and understanding.
I agree that because of varied reports from various people over time, anyone can only refer to their own adapter when reporting possible issues which is why I'm interested in individuals feedback about their own adapters. I am also wondering if it is just my adapter. Hope not, I've had it too long now to complain about as in retro inspection all of my clips using this have shown signs of it that I didn't pick up on until I changed my editing methods.
I'm pretty sure the dropout is being caused by the audio adapter or the Osmo Pocket 'checking' that it is a verified match of 'authorised' components and why I'm hoping the next generation of Pocket is slightly larger to fit my hand better and have some of this tech built-in. Even if the cost is more. I accept there is a bootup time due to the nature of the Pocket but dismayed that an external adapter should lengthen that time.
The second problem means that most times if I butt-join my clips together I have roughly a one in three chance of getting a pop or click at the edit point due to the slight DC offset. Sometimes, stopping or starting a completed video at odd points within a video will also cause this to happen.
I'm not overly fussed about whether it's mono or stereo. It is what it is but the ability is there in the electronics.
I do believe as you do that any mic should have a high impedance and possibly a good dynamic range to work well. Frequency response is down to preference and usage. To think one mic may cover all eventualities is possibly not a good idea. I do love those EM 172 capsules for general use but they may be a bit high frequency heavy for voice if used for close up work. Really big dynamic range with a very wide frequency response for everyday event work.
Thanks for the heads up. I may at a later date get one of their lavalier mics which also look promising.

Ray.