In my previous post on the Garmin GLO vs QStarz 818XT I looked at the two units from a physical perspective without having done a real world test. In this article I’m going to look at and analyse the data produced by RaceChrono using both devices riding on Silverstone’s GP circuit.
For the test, I installed both GPS receivers on the bike. The Garmin Glo was strapped to the rear of the bike on the tail piece while the QStarz 818XT was taped between the top yoke and the instrument console on the head stop lock bracket (circled in yellow).
The test was conducted for one session with a detailed analysis of the fastest lap from that session. These are the raw numbers produced by RaceChrono for lap 4 of the test session:
|QStarz 818XT||Garmin Glo|
Taking the raw numbers from RaceChrono and plotting them against a schematic for the track is quite illuminating. The diagram below shows the two traces obtained from the GPS devices. The red trace is the QStarz 818XT while the blue trace is from the Garmin Glo.
The first thing to notice is how both traces cut the right handed corner between Vale and Club. Neither trace is correct as I’m pretty sure that I was close to the right hand side kerb but definitely not over it. This isn’t totally unexpected because the reported accuracy of these devices is down to around 0.5 – 2m depending on the quality of the satellite signal and fix.
The second thing to notice is that the Garmin Glo (blue plot) records my line through those two corners more accurately than the QStarz 818XT. If you look carefully, you’ll also notice that the red trace (from the QStarz 818XT) looks smoother than that from the Garmon Glo. You can see this better in the detailed view below.
The QStarz 818XT delivers a data point to RaceChrono every 1/5th of a second (hence 5Hz) without fail. The Garmin Glo is supposed to deliver a data point every 1/10th of a second (ie 10Hz) but fails to do so, often skipping one or more data points within each second – look for the gaps in the purple dots on the blue plot. Over the course of the 3 minute lap, the Glo delivered a data point every 0.133 of a second resulting in an average frequence of 7.5Hz – about 25% below the quoted 10Hz. As noted in my previous article, this could be a flaw in the Garmin or more likely the Bluetooth stack in the Android operating system used to run RaceChrono.
Comparing the two traces, it looks as though the QStarz 818XT (red) has come kind of filter or smoothing algorithm because of the smooth nature of its trace. The Garmin Glo’s trace (blue) shows some “jumping” between two consecutive data points; almost as if the bike jumped one bike width’s to the side. Rather than being a flaw, this indicates to me that the Glo doesn’t attempt to compensate, smooth and adjust fixes but just reports the fix as calculated.
Over the course of the lap, both devices report pretty similar traces, except for the trace recorded by the QStarz 818XT at Luffield. You can see this in the red trace that leaves the track completely for a while before finally rejoining the circuit. A more detailed view is show below.
It’s pretty obvious that the QStarz 818XT was unable to accurately locate itself possibly due to a temporary loss of view to the overhead satellites. It takes the device almost 8 seconds to re-establish its position back on track. This could have occurred because the QStarz 818XT was mounted in a position on the front of the bike where my helmet (and upper torso) could have been shielding the device from a decent view of the satellites. I checked the laps before and after and this problem didn’t recur so it’s likely to have been a one off that lap. What’s interesting is that a seemingly large error like this has relatively impact on the overall lap time and both devices report lap times within 1/10th of a second of each other for that particular lap.
|Lap||QStarz 818XT||Garmin Glo||Difference|
With the exception of the excursion off track described above, both devices show fairly similar traces around the circuit. In terms of position reporting, the Garmin Glo looks to be more accurate than the QStarz 818XT but that could be because the QStarz 818XT was located less favourably than the Glo. If you’re looking to plot your lines around the track, the Glo would be a better unit to use than that QStarz 818XT. If however, you’re just interested in your lap times, then both devices do a great job reporting very similar lap times as you can see above.
In my next article, I will analyse the speeds reported by the two devices at different parts of the circuit.