Several errors from 8.0.1 version on ATEM Production Studio 4K

[nrkno]

Several errors from 8.0.1 version on ATEM Production Studio 4K

[Julusian]

As for protocol documentation, this project of mine is the most complete implementation I am aware of, but that wont really help in this case due to the jibberish being sent because of the lack of a dve

[Julusian]

Is it just the Production Studio 4K that is not working in 8.0.1, with the 2ME 4K working fine?

This sounds a lot like the issues in #15 and #24. The problem is that the atem sends some commands filled with jibberish, and we don't have a great strategy to filter them out right now.

The easy 'fix' is to add an error handler to the connection, and to ignore these 'errors'. It is on my todo list to figure out a better way of handling these expected command errors, but I havent gotten around to it yet.

[Kohbrax]

I really dont want to test an upgrade of our 2ME devices right now. We had it updated once (in early august, where the lib had not support for v8) and the downgrade worked. Before the update to 8 i had it on this library without any errors.

The other ATEMS are important for out production - thats why i cannot test it.

But thanks for the Link to the c# Project.

Can you give me an example how to catch these errors for a single connection? I havent seen any example for that.

Thanks for your fast response! :)

[Julusian]

I really dont want to test an upgrade of our 2ME devices right now

Thats reasonable. I am running 8.0 on my original 2ME without issues currently. And those errors normally only occur when the device doesnt have a DVE, but it is possible there could be edge cases with certain DVE/fly key values if you are ever setting those.

Something as simple as this will log the error, but won't interrupt the connection:

myAtem.on('error', function(err) {
  console.log(err);
});

[Kohbrax]

Something as simple as this will log the error, but won't interrupt the connection:

i was looking for this for ages! Thanks!

Sometimes i have the feeling, that the parsing of the tcp packets seems to have the wrong length. So its getting worse with the errors. But that is just a feeling - might not be right.

We had a problem, that we could not connect to the atem with the normal atem control software, after the atem-connection lib has connected. Just with this one small atem, and just in this weird case.

Is it maybe a valid fix to disable capabilities of parsing DVE Commands, when the ATEM does not have any?

Funfact: The 1 M/E Production Studio 4k is working fine..

[Julusian]

Sometimes i have the feeling, that the parsing of the tcp packets seems to have the wrong length. So its getting worse with the errors.

That should only happen when the protocol changes, but we have some tests in place to try and avoid errors from things like that

We had a problem, that we could not connect to the atem with the normal atem control software, after the atem-connection lib has connected. Just with this one small atem, and just in this weird case.

I have had that sometimes before. A restart of the atem usually solves it, but I have no idea why it happens, or if it is due to the library or something else.

Is it maybe a valid fix to disable capabilities of parsing DVE Commands, when the ATEM does not have any?

That is something we are discussing. We have an idea on how to do that, so will hopefully be able to give it a try soon.

Funfact: The 1 M/E Production Studio 4k is working fine..

That one has a DVE, so should be fine :)

[Julusian]

A prerelease of v2.0.0 has just been pushed to npm under the nightly tag yarn add atem-connection@nightly that should resolve this, along with other problems.
The parsing has been made a lot less picky on values to avoid errors like this, and any out of range ids are now debug warnings rather than errors.

We are happy that v2.0.0 should be stable, but want to spend a bit more time giving it some realistic testing before doing a proper release of it.

[Kohbrax]

Nice! Thanks for the Info. I'll update my staging system to this version.