If you follow the advice given and simply delete a single track point in BaseCamp, the improvement in accuracy of the cumulative ascent is staggering. I tried this for the track from a recent bike/hike in Scotland:In the past, there has been a confusion about why the Trip Computer data differed from the downloaded track log. Garmin has made two changes to their devices to address this. One is they made the Trip Computer a little smarter. It no longer logs every position solution, but only those where it decides the device has changed locations. This improved the total distance travelled, particularly for those just meandering about. The second change was to include the Trip Computer summary data with the downloaded Trip Log data.
This second change means that you no longer see summary data based on the data in the track log, but that of the trip computer. So, the device and BaseCamp agree exactly; but in truth, this is not a good thing to do.
If you are seeing Trip Computer summary data in BaseCamp, you can force recalculation based on the logged track data by deleting a single point in the track or dividing the track and then rejoining the track fragments. The former removes one data point, the latter adds one data point. The added point is a duplicate of one of the original track points.
Original GPX loaded into Basecamp - ascent 2622m (WAY TOO HIGH!)
Same file with trackpoint 3 deleted to force a recalculation - ascent 1890m
This new figure is much closer to what I was expecting based on the map and guidebook.
Presumably Garmin's second change noted above also applies to FIT files too? I took a look at the FIT file for the walk:
I could not import the original FIT file for this hike directly into BaseCamp, so I had to convert it to a GPX first using GPSVisualizer. The initial converted GPX file had no elevation data, so I converted the FIT file again but adding elevation data from the GPSV's best digital elevation model. I then imported this new GPX file into BaseCamp. The 1880m total ascent using GPSV elevation data agreed very closely with the recalculated total having deleted a single track point and gives some validation to the number.
I'd be interested to see if you get the same results and if your recalculated data is close to what you expect. If the results are consistent, simply deleting a single track point could be a relatively painless way to get a more accurate estimate of total ascent.