Fenlander; did you bounce the numbers on your two units against each other, or work straight off the map? Curious if the units thought they were in different places, or if they thought the same location was a different spot on the map.
Doesn't Garmin's display show a circle denoting the probability of error (an old GPSMap 76 I had did; don't know about their other models)?
As an aside, I had an early Lowrance NMEA2K antenna that would go nuts around radar sources. Any largish vessel running radar or shore based ship tracking would cause very strange errors (and in at least a few reported cases, killed the antenna, though that never happened to me.)