Jan Post

> The formula to calculate the frequency to input into a second order filter to achieve a specified -3dB cutoff frequency f0 is:  @LSchwerdt Thanks! This was driving...

@LSchwerdt wow, I didn't know it is okay to remain in the s-domain for this. I can see in the simulations that this frequency correction also holds true in the...

@ledvinap I marked the PR so that we can discuss it. Your comments make me think we should rather not merge this PR.

Just a small 15 seconds flight with debug mode `GYRO_SAMPLE` (not GYRO_SCALED) for me to look at the spectrograms. But treat it as if RPM filtering is disabled. Put enough...

Regarding `level_recovery_time`: This doesn't take into effect if the error is small enough or not after the set time. I guess it would be better to automatically quit recovery mode...

Very cool, I can directly get a use out of it now that I'm using ELRS.

@TonyBlit A PT1 will be better at rejecting high freq noise than a simple average. If a PT1 still doesn't cut it, you can try a biquad lowpass with a...

They are all occupied afaik. Maybe it makes sense to rearrange and make some room in the near future. We have another feature that would benefit from new OSD characters.

In this PR, in each case it was appropriate to use `lrintf()` inside of `DEBUG_SET` as far as I can see. The only thing I'm not sure about is `DEBUG_FEEDFORWARD_LIMIT`

I guess we could rebase the PR, include suggested changes, test that softserial still works on F7, F4, ... and then pull it in if there are no further suggestions....