[sv2fpv]

Vehicle oscillating in all directions in GPS modes.
With M8N GPS and 15.09 release with external compass no oscillating
https://forum.librepilot.org/index.php?topic=1110.0

With 16.09 and M8N big problem, oscillating and drift to North
back to 15.09 same problem.

16.09 with NEO M6 is fine no drifting , a little oscillating standard frequency.
Can i test the firmware?  Naza GPS.  Or does not.

[TheOtherCliff]

That firmware only helps DJI/Naza GPS issue.

One possible cause for your issue is bad mag calibration.  Mag calibration is continuous, even between pressing the button 6 times.  It gathers data all the time, not just when the progress bar is moving.

Use "aux mag only" setting (Attitude -> Magnetometer).  Mag must be green most of the time, even while flying.  Never ever red at all.

During mag calibration:
- setting the quad on the ground at any time during mag calibration is bad
- setting the quad on anything metal (like your car) during mag calibration is bad
- you really need to use RF telemetry to do calibration, not USB cable
- generally hold it chest high the whole time while someone else runs the GCS and tells you how to point the quad

[sv2fpv]

So I really do it.
This is a big story.
 "Basic (Complementary)" it flies brilliantly, no oscilating no drift.
I tried with different distances of the compass from the boat 10-15-20 cm, same problem
Many times and in different parts i did calibration(16.09), but always the same problem.
The best was with the NEO M6. The compass is always green at GCS only full throttle becomes orange.
PosHold very good position but little oscillating in all directions, GPS modes all fly very well

Now I expect them to try

http://www.ebay.com/itm/APM2-6-HMC5983-High-Precision-Compass-External-Magnetometer-Temperature-CA/142065025612?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m2749.l2649

http://www.ebay.com/itm/HMC5883L-GY-273-GY-271-Triple-Axis-Compass-Magnetomet-Sensor-3V-5V-For-Arduino/162491330905?ssPageName=STRK%3AMEBIDX%3AIT&var=461499489738&_trksid=p2060353.m2749.l2649

Have a nice day

[Brian]

Isn't it more accurate to say that Acks are used with some forms of telemetry, but never with "control only"?

Also, for long range unidirectional "telemetry only" you may want to try setting the model as Coordinator and making it's transmit power be the higher one.  This may or may not work well as the high priority coordinator task can fight with with the high priority stabilization task for time critical CPU access.  Is this task priority issue tracked to be worked on?

Unless something has changed, Acks are not used in any configuration of OPLink communications.

I believe the option for a one-way telemetry (data) connections has been removed, at least from the GUI.  Are you suggesting that you would like to have a telemetry link that is longer than your control link?  Do we / do we really want to support a totally autonomous mode where you have no control link to the aircraft?

[TheOtherCliff]

Some UAVOs are acknowledged?  They are retransmitted if not acknowledged.  Perhaps we are talking about two different levels, you the low level ack?

for long range unidirectional "telemetry only"

You can set the model to be coordinator, run it at high power, run the GCS at low power.  You get telemetry received, but GCS isn't successful when it sends.  But as I say, this seems to now have a bug in that the high priority RF task needs real time like the stab code does.