Using LapTimer and AIM SOLO together

[swbca3691]

I have been used your LapTimer for about 8 track days over the last year and it has proven to be an excellent tool for finding the fastest lines using the lap "segments" built into LapTimer. I just tried an AIM SOLO laptimer for a day because its primarily for real-time use rather than data logging features.

The Question . . . the lap times recorded by the Harry's LapTimer on Galaxy GS5 and the AIM Solo are always within .2 seconds of each other (plus or minus) on a 2 mile 2 minute track. I read a lengthy thread on this forum about accuracy but I don't recall it talking about the actual time deviation around "perfect". Do you expect that a +/- 0.1 second range around "perfect" is the best GPS timing can do ? Obviously plus or minus .1s is really good for a training timer, so this question is only 'academic'. If it matters, the speed at the Start/Finish line on this track is 150 kph.

Thanks for your great product.

[Harry]

Please have a look into viewtopic.php?f=39&t=459 for some timing precision estimates. Timing accuracy for GPS devices is a function of position accuracy. Higher update rates help too, but especially when the start / finish line is passed fast, position accuracy is what counts. I do not know what chip set the AIM uses, but due to development costs, dedicated devices will often come with dated components. When using an external GPS like one of those listed on www.gps-laptimer.de/compatibility/android with LapTimer's sophisticated interpolation mechanisms, you can expect 0.05s accuracy for more than 95% of all recordings.

- Harry

[swbca3691]

Related accuracy question . . I spend time looking carefully at the track position of braking and accelerating as displayed in the G-Circle in critical turns when trying to evaluate fast results versus slower results. Is the system latency or dampening in acquiring and processing information from the android accelerometer a factor in the G-Circle being synced fairly well or really-well with the video, or do you know the latency and make adjustments when applying the overlay ?

[Harry]

Two considerations: 1) Data and video needs to be perfectly synchronized when looking into braking points. Search for "sensor delay" here on the forum to get some background on this topic. LapTimer provides a good default but can be optimized individually. 2) In case you record GPS with the internal GPS, acceleration is stored at a 1 Hz rate only. In video overlays, G values are interpolated and will generate an impression of higher precision than that actually available. This means peaks may be misaligned by 0.1 or 0.2 seconds. To eliminate this effect, you need to add a 5 or 10 Hz GPS option.

- Harry

[swbca3691]

Sorry, but when you say with internal GPS, acceleration is stored at a 1Hz rate, does that mean the G-Circle is based on a calculation of the rate of change in GPS coordinates ? I thought it was simply the phone's accelerometer. I know when I hold the phone on edge, it reads exactly 1G left or right. (This is my last question)

[Harry]

No, it is measured using the accelerometer and displayed at a by far higher rate in real time. However, storing is done jointly with GPS fixes, so this channel defines what you get from LapTimer's database later.

- Harry

[swbca3691]

Follow up . . concerning the synchronization of the Overlay with Video when studying throttle and braking points during turns in fast laps versus slower laps.

I just reviewed some of my videos from last year and this year, and the G-Circle was indicating acceleration >1 sec ahead of hearing the engine being powered up during turns. The same 1 second+ lead was consistent across all videos. This of course is assuming the audio track is synced with the video track.

I also observed the G-Circle and the SPEED indicator on the Overlay appear to be synced with each other.

When watching Youtube videos of other drivers, I have noticed their Laptimer Overlay was off by as much as 5 seconds. I have never had that happen.