I went to a track today that gives you with timing beacons to put on your car and at the end of which they give you a print out of the laptimes. I fully expected there to be some discrepancies when compared with those records in Laptimer, but I was amazed at just how big they were. Its very strange but over approx 1.19 lap there were sometimes almost a second different, but there does not seem to be any consistency to gap as it fluctuates. I had a look at some of the data for the laps and noted that lap lengths varied quite a bit and despite running at 5hz there were only 300 or so GPS fixes (despite there being green lines for the GPS signal and the GPS brick itself having a clear line of sight to the sky (its a caterham so roof!)). Finally, the speed was fairly off as well, lagging behind (despite changing the settings regarding delay from the external GPS as you previously suggested) and also suddenly jumping higher and lower only vaguely matching what was actually going on on track.
I was running a camera as well and having looked at the laps in iMovie it seems that the track data is correct.
So what is going on? I really dont see how the lap times can be off, as surely its just measuring the time between instances of travelling through the POI. Surely that has to be pretty accurate.
Any advice would be gratefully received because as it stands right now I cant overlay any of the videos as the speed is wrong and so is the timing!
I am using an Ipod Touch running the latest iOS with a Dual GPS.
Thanks.
Timing and Speed problems
-
Themanalive
- Fewer than 10 Posts
- Posts: 8
- Joined: Fri Jun 01, 2012 6:27 am
Re: Timing and Speed problems
Hi,
Its probably best you send one of the laps that are off and let me know what you think is the right time. Please use the .hlptrl format as it allows full analysis. Overall distance varies a lot as it is summed up from the individual distances between two fixes each. This means that every inaccuracy will vary the overall result. As the overall distance is not used for timing, it will not influence timing accuracy. On speed: you posted that issue already in the past, thought you had it solved using the link I had sent to you?
- Harry
Its probably best you send one of the laps that are off and let me know what you think is the right time. Please use the .hlptrl format as it allows full analysis. Overall distance varies a lot as it is summed up from the individual distances between two fixes each. This means that every inaccuracy will vary the overall result. As the overall distance is not used for timing, it will not influence timing accuracy. On speed: you posted that issue already in the past, thought you had it solved using the link I had sent to you?
- Harry
-
Themanalive
- Fewer than 10 Posts
- Posts: 8
- Joined: Fri Jun 01, 2012 6:27 am
Re: Timing and Speed problems
Harry,
Thanks for replying. Below are two laps, one is a 1.40.88 according to the track timing but is a 1.41.82 in Laptimer. The lap is longer in time as there was a spin, but it would not effect the gap.
Having gone through the times with more detail, it more normal to be between 0.1 to 0.5 difference. Usually it is a faster laptime in Laptimer not always. The other attached lap was my fastest of the day according to Laptimer as a 1.18.17 but the track time was 1.18.432.
With the speeds, I did as suggested in the previous string but did not have the chance to properly test it. It seems that I havent solved the problem. Its frustrating as it one of the bet things to put on the overlay. I used to use Racechrono with a different GPS dongle and the data from that was very precise - I would have thought the detail from the Dual GPS given it is a couple of years newer would be better or at least the same. Would it be more accurate if I used an OBD sensor?
Thanks for replying. Below are two laps, one is a 1.40.88 according to the track timing but is a 1.41.82 in Laptimer. The lap is longer in time as there was a spin, but it would not effect the gap.
Having gone through the times with more detail, it more normal to be between 0.1 to 0.5 difference. Usually it is a faster laptime in Laptimer not always. The other attached lap was my fastest of the day according to Laptimer as a 1.18.17 but the track time was 1.18.432.
With the speeds, I did as suggested in the previous string but did not have the chance to properly test it. It seems that I havent solved the problem. Its frustrating as it one of the bet things to put on the overlay. I used to use Racechrono with a different GPS dongle and the data from that was very precise - I would have thought the detail from the Dual GPS given it is a couple of years newer would be better or at least the same. Would it be more accurate if I used an OBD sensor?
Re: Timing and Speed problems
Hi,
That's the plot from the first lap.
The data in the file submitted looks perfectly consistent, with the start and stop fixes exactly matching the reported lap time. Problem seems to be that direction reported while crossing the start / finish line is incorrect. Check the approximately 10 fixes just below the gray dot, draw a line through them and you see it gives a wrong direction. As direction needs to be used when doing the start / finish line interpolation, this error has an effect on timing too. It seems the GPS delivers incorrect data here - while reporting good accuracy. I'd suggest to forward this to Dual too. The second file you've sent shows the same anomaly by the way...
Two work arounds / further investigation you may try: 1) in case you have an iPhone4 or later, try a run using the internal GPS. It would certainly be interesting to see what the result is. 2) move the start / finish point a bit to the north - where accuracy seems to be better.
In case you get this effect in the same place regularly (the two laps show the same pattern), there is probably a reflection or something disturbing GPS here - but that's again an item for a GPS specialist.
OBD gives you better data for short time peak values while GPS is more accurate on longer constant movements. The good thing about OBD is it matches your car's dashboard - which means OBD will typically show too high values.
- Harry
That's the plot from the first lap.
The data in the file submitted looks perfectly consistent, with the start and stop fixes exactly matching the reported lap time. Problem seems to be that direction reported while crossing the start / finish line is incorrect. Check the approximately 10 fixes just below the gray dot, draw a line through them and you see it gives a wrong direction. As direction needs to be used when doing the start / finish line interpolation, this error has an effect on timing too. It seems the GPS delivers incorrect data here - while reporting good accuracy. I'd suggest to forward this to Dual too. The second file you've sent shows the same anomaly by the way...
Two work arounds / further investigation you may try: 1) in case you have an iPhone4 or later, try a run using the internal GPS. It would certainly be interesting to see what the result is. 2) move the start / finish point a bit to the north - where accuracy seems to be better.
In case you get this effect in the same place regularly (the two laps show the same pattern), there is probably a reflection or something disturbing GPS here - but that's again an item for a GPS specialist.
OBD gives you better data for short time peak values while GPS is more accurate on longer constant movements. The good thing about OBD is it matches your car's dashboard - which means OBD will typically show too high values.
- Harry
-
Themanalive
- Fewer than 10 Posts
- Posts: 8
- Joined: Fri Jun 01, 2012 6:27 am
Re: Timing and Speed problems
I will move the start finish point and give it a go next time I am at the track to see if it makes a difference. I should be up there in the next week or so and will give it a try. Hopefully that will solve the timing problem, I will make sure I post back on here when I get back. Thanks for the help.