Hamish Willee

Try to fix that case in https://github.com/mavlink/mavlink/pull/2042

@Ryanf55 Thanks for posting. I support the goal - we should be precise where possible. Whether we can do it largely depends on whether - this will break assumptions that...

@sypaq-nexton Yes, this is a known issue. The simple fact is that no one should be using `COMMAND_LONG` for positional commands. Firstly because of the loss of precision, but also...

I've responded to your post. I think this idea has merit - certainly just supporting commands in both types does not solve this problem. Let's see how the discussion in...

Multiple bytes are serialized little-endian, so I assume the last 8 bits are the bits with the lowest address, sent last. The definition seems to indicate 32 bits: ``` 16:...

@JonasVautherin I think so - because otherwise target_number would be much bigger. But I'm not a coder, so lets get @julianoes involved. FYI, ArduPilot code around this is here: https://github.com/ArduPilot/ardupilot/blob/master/libraries/GCS_MAVLink/GCS_Common.cpp#L2587

Probably. The other way to check is to grep QGC. If it has different code for both cases you'd be right for sure.

Hmmm. I would have thought it was somewhere in firmware update. But I'm only here for my charm and good looks.

So on PX4, MAVLink gets the value of this field using this function. ``` uint32_t px4_board_version(void) { #if defined(__PX4_NUTTX) return CONFIG_CDCACM_PRODUCTID; #else return 1; #endif } ``` For FMUv5 we...

@csswizardry Would you like to create a PR in line with https://github.com/mdn/content/issues/36376#issuecomment-2418543598 above?