@digitalentity do you have any flight logs?

Not really. I don't have the solid data to back the "delay" claim, but I have seen the "glitch prevention" code causing issues (especially on the MSP barometer).

In the past, I've thought about removing this filter... But the tests told me that the baro gets quite noisy! I don't know if this step is correct, since the INS is quite limited... I've just tested your modifications and the baro is really noisy. I can't say if in flight there will be any difference... Let's test this! My Hardware: MatekF405SE with baro DPS310.

Noise itself is not a problem and the median filter doesn't help reduce it much. Median filter is there because some baros were spotted in the past giving occasional extreme readings (plus/minus hundreds of kilometers).

Let's not merge this just yet, I'll replace median filter with something else that was tested on another code.

Let's not merge this just yet, I'll replace median filter with something else that was tested on another code.

Could you give a spoiler of what it is? 🧐 😈

Please disregard my last comment. The thing I was referring to is the LPF on the barometer before feeding it into the INS. We have that (1 Hz LPF on baro) inside the INS (shame on me for forgetting).

Yes we do... Now it might be useful... Glad you didn't let me remove it in the past 😆 🙃

@DzikuVx let's move this to 5.1?

@digitalentity ok, the release branch is there, please rebase

On other hand, this is a breaking change as it makes sense to put it in the next major, not minor

Ok, let's keep it scheduled for 6 then. This will also fix #8224