External Mic Problems

[matthewlai]

I encountered this problem on my FDR-X3000, but from online posts it seems like this affects all Sony action cams with external mic input.

Basically the problem is that the firmware applies very high gain through AGC, and that generates a periodic popping noise because the amplifier saturates.

This problem has been reported by many people, and the cause is now well understood (links removed because forum software thinks this post is spam otherwise, but just search for "X3000 external mic", and you can find the dozens of report out there).

I have contacted Sony and they said they will not fix it because it supposed doesn't affect the 3 (THREE!) Sony microphones listed on the "compatible mics" page. They are all tiny (and very expensive) lav mics.

I happen to be an electronics engineer, so if all else fails, I'll design a hardware active attenuator to workaround this silly bug that they can easily fix on the firmware side (I know because I used to write camera firmware for another company). Very disappointed in Sony.

Before I do that, has anyone else found simpler workarounds?

The mic I'm using is Rode Videomic Go, but many people have reported problems with many other mics.

Thanks

Matthew

[IamNic]

Hello @matthewlai,

I have heard about this a few times on the Sony Action Cam subreddit.

Yet I am unaware of which trype of microphones shows this behavior.

Maybe we can try to compile a list of microphones, which overmodulate.

- Nic

[matthewlai]

Thanks Nic,

Yeah I have done a lot of reading on that subreddit as well. It's probably easier to compile a list of working microphones than non-working ones, since they seem to be much rarer. There are posts from people who have tried half a dozen microphones to find one that works.

My guess is it's the sensitivity - the microphone I have (Videomic Go) has -35dB sensitivity, and it doesn't work. Sony's ECM-CS3 has -38dB, and is on the compatibility list (though I don't know if they actually tested it, and I don't want to spend more money to buy their microphone to do their job for them). All the working mics I've seen so far are lavalier ("lapel") mics, and using an external attenuator to reduce the gain of any microphone also works.

There is most definitely a firmware bug here. AGC should not work like this. Someone seriously screwed up on the control algorithm, and it can probably be fixed by changing a few constants in the code. It's too bad that they are just not interested in fixing it.

According to one of the reviews on the product page (the one titled "Excellent video quality BUT TERRIBLE AUDIO..."), it doesn't work with at least one lav mic either.