Evidently I didn't supply enough information on what I was trying to do for it to be clear. First off, yes, an airplane must have a compass, and yes, a GPS does not know what the heading is, it knows what the track was. However, if the aircraft heading is maintained, then the GPS track should be a reasonably accurate representation of the heading.
I've in the final throes of getting a homebuilt ready to fly. In addition to having a compass, the compass needs to be calibrated, errors corrected to the extent practical, and the final error recorded, normally on a little compass card. That's were I am now, The normal way of doing this is to use a compass rose, using north-south and east-west headings for calibration. The airport I'm on doesn't have one, and the nearest is about 100 miles.
Nor do we have n-s or e-w runways or taxiways. So I must look for alternative methods. And, yes, I know one can get a calibration of sorts using other headings. But for technical reasons, it won't be as good for calibration purposes.
My thought was that I would taxi in a fixed direction, and get the track made good data, repeating this until the gps said I had a north going track, align the compass, get a south going track, etc, When I tried this, the gps was unable, over the length of the area in which I could taxi, to develop a track made good. Thus, my question: for the iFly, how far (and perhaps how fast) must I do before I get a track made good?
The iFly does display the current track made good along the top of the screen as though it isa heading, which I agree, it is not really.