Here's a test where I wore the 405cx (red trace) on my wrist and the N95 (yellow trace) with sports-tracker running in my pocket. I didn't turn them both on exactly at the same time, but there does seem to be a difference in the distance and kCal values. In addition to the time/distance data the 405cx also uses heart-rate data for computing calories.
N95: 6.58 km, 1h 4 min, 576 kCal
405cx: 6.32 km, 1h 7min, 324 kCal
The N95 GPS data is much more wiggly, as seen in this picture of a long straight bit of the walk:
That probably also explains the extra distance on the N95, since the wiggles never go away:
Although the GPS-Filtering was set to "High" on the N95, it shows strange jumps along the shore-line where the satellite visibility to the South should be OK.
Here it looks like the N95 position (yellow) fix is first a lot to the east, and then significantly south of the bridge I actually crossed. Then for some reason the 405cx (red) and the N95 agree again. I think the 405cx samples position at 4 s intervals, so some cutting of corners is understandable.
This test wasn't entirely fair since the 405cx was on my wrist while the N95 was in my pocket. I need to re-do this test with the N95 and its antenna held at a similar height/visibility compared to the 405cx.
Garmin is doing a lot of things right with the 405cx: No extra GPS-pod like with the Polar products, a set of nice youtube-videos practically replace the manual, charging through the USB-port(why doesn't the N95 have this?), to name a few.