The project main page and the calibration dialog both state that the device must
be calibrated on device and the calibration stored in slot 0.
There are two problems with this.
If an on device calibration can prevent the device from sending invalid data,
then the Saver can apply the same logic when handling incoming data. If this is
not possible due to the multitude of devices supported and their different behaviours
then the current mitigation strategy should be employed with a clear warning.
Second, this requirement encourages users to perform on device calibration without
understanding the consequences.
Many users are unaware that the device will do extrapolation, for data between
calibration points, and assume that an on device calibration is always beneficial
without understanding how that degrades measurement accuracy.
This is evident on the NanoVNA groups.io forum, where users frequently ask
whether on device calibration is required.
Most replies show that users do not understand the extrapolation mechanism or its
impact, and often recommend that the user always do an on device calibration.
That practically defeats the purpose of a proper calibration in NanoVNASaver.
If this behaviour is intentionally left as is, Saver should clearly warn the user
that the device is not operating correctly and that they need to do an on device
calibration first, as a workaround for the issue. An explanation of the measurement
trade-offs involved should be included.
The project main page and the calibration dialog both state that the device must
be calibrated on device and the calibration stored in slot 0.
There are two problems with this.
If an on device calibration can prevent the device from sending invalid data,
then the Saver can apply the same logic when handling incoming data. If this is
not possible due to the multitude of devices supported and their different behaviours
then the current mitigation strategy should be employed with a clear warning.
Second, this requirement encourages users to perform on device calibration without
understanding the consequences.
Many users are unaware that the device will do extrapolation, for data between
calibration points, and assume that an on device calibration is always beneficial
without understanding how that degrades measurement accuracy.
This is evident on the NanoVNA groups.io forum, where users frequently ask
whether on device calibration is required.
Most replies show that users do not understand the extrapolation mechanism or its
impact, and often recommend that the user always do an on device calibration.
That practically defeats the purpose of a proper calibration in NanoVNASaver.
If this behaviour is intentionally left as is, Saver should clearly warn the user
that the device is not operating correctly and that they need to do an on device
calibration first, as a workaround for the issue. An explanation of the measurement
trade-offs involved should be included.