Race 2 accuracy
-
@jjpaz said in Race 2 accuracy:
@isaac.net said in Race 2 accuracy:
Hi,
I’ve done a new comparison, this time of an old running activity. The activity on Sports-tracker shows a distance of 9.52km, and the GPS points of the route are almost perfect. I marked the same route with an online distance calculator (https://www.calcmaps.com/), and the result was 9.67km, practically exactly the 1.5% difference mentioned in the thread.Here are the screenshots:
If, in your SportTracker activity, you open the “graphs” menu, you can show activity information over the map. If you move to the right (end of the activity) maybe you could see a final distance close to the 9,67km (the “Raw” measured distance).
Yes, that’s right, in that case the distance it shows is 9.65 km. It’s clear that Race 2 correctly captures the GPS points but then processes them incorrectly. In my opinion, this should be a priority bug fix.
-
@isaac.net said in Race 2 accuracy:
@jjpaz said in Race 2 accuracy:
@isaac.net said in Race 2 accuracy:
Hi,
I’ve done a new comparison, this time of an old running activity. The activity on Sports-tracker shows a distance of 9.52km, and the GPS points of the route are almost perfect. I marked the same route with an online distance calculator (https://www.calcmaps.com/), and the result was 9.67km, practically exactly the 1.5% difference mentioned in the thread.Here are the screenshots:
If, in your SportTracker activity, you open the “graphs” menu, you can show activity information over the map. If you move to the right (end of the activity) maybe you could see a final distance close to the 9,67km (the “Raw” measured distance).
Yes, that’s right, in that case the distance it shows is 9.65 km. It’s clear that Race 2 correctly captures the GPS points but then processes them incorrectly. In my opinion, this should be a priority bug fix.
I don’t know if it process de GPS data incorrectly, for sure it does it in a different way in comparison to other models.
-
@jjpaz what I posted are raw data from JSON files in quantified self. One of the watch recordings is no longer in the Sunnto database. I believe 3% is the acceptable GPS error for handheld devices.
Vector/T6c/Vertical 2 Ti
-
@Brad_Olwin I think that is valid for GPS but not for GNSS dual band devices (and even more when It seems something strange is happening with the Race 2 but not the others). Someone correct me, if I’m wrong.
-
@Brad_Olwin In my opinion, an error of more than 1.2km in a marathon (3% of 42km) is unacceptable under any circumstances, especially if all other devices measure close to the actual target (even the cheaper ones). Especially considering that the position values recorded by the watch are very precise
-
I think while the actual position data of a point may have some error, the distance measurement by GNSS is quite precise these days. Also some filtering occurs to the data measured every second.
What we see is, that under some circumstances Rce 2 measures short, while this is not always the case (see John Korir’s Marathon distance in the other thread).
As this is not a special setup only in Race 2, there is not much to do about it. If it does not happen everywhere and every time, this is just an arbitrary deviation with the range of 0.5%.
Not sure what else to add to this thread now. -
@Egika Are you sure it only happens in certain circumstances? Since I became aware of the problem and started checking distances, all my routes have the wrong distance. Today, a run of 6.01km according to Race 2, after correcting the distance with Strava, was 6.10km. Again, that 1%-1.5% less distance…
The difference can also be seen in the graph of sports tracker, 6.01km in activity info vs 6.11km in the graph -
@Egika Yes, unfortunately it looks like we’ve come to the end of the road and Race 2 users are just going to have to live with it
-
@isaac.net maybe your individual device is a little different. We have seen a tendency but we have also seen perfect matches.
-
@Egika I’ve worked in software development for a long time, and with the correct track data, the solution is possible, since it’s not a precision problem but a distance calculation issue (software), as demonstrated on the Sports Tracker website. In the same activity, you can see the distance calculated by Race 2 and by the app on the graph. It’s a 100% software problem and therefore solvable. If the problem is due to processing, could it be that the routes with less error are straighter, while those with more error have more turns? I’ll try a completely straight route in the next few days to see what happens…
-
@isaac.net sorry, you are on the wrong track.
While probably everything might theoretically be solvable by software, I doubt that Suunto is going to invest a lot into this one. But who knows…
