I just picked up a 1040 GPS because I had been using a different Venus638FLPx-L based serial GPS outputting raw NMEA strings, and the 1040's USB interface and API are really nice. I'm super happy with how easy the 1040 is to work with, but I'm seeing a fair amount of change in both position and speed when the 1040 is sitting still, and it does not settle down even after I let it run overnight.
To compare, I put the 1040's antenna and the antenna of the old GPS within a meter of each other, both stationary. The other GPS settles down within a couple of minutes so that the reported position does not change and speed is always zero, and the position reported is correct within a meter or two. The 1040's points are always at least two or three meters away from actual, they vary by more than 3 meters, and the variation does not span the actual location. Reported speed is rarely 0 and often exceeds 4km/h. I understand the specified 2.5m accuracy is best case, but the data I'm getting from the 1040 is much lower quality than what I'm getting from the other unit built around essentially an identical chip, so I'm wondering if I'm doing something wrong with the 1040. The only difference I can think of is that I have the other unit configured for 1Hz updates, whereas the 1040 says it defaults to 10Hz.