Hi, thanks for explaining this case, indeed when car path when passing the lap counter is very different between two passes (your drawing is showing the worst case) then accuracy is worse than few 1/100s .
But this is not the normal use case of racer who is battling for a 1/100s .The error in this case is dependent of the speed, the faster the car goes the better is the accuracy, for example a car at passing the lap counter at 40km/h and a lap with 2 passes separated by 2 meters you can get an expected error of 6/100s
Also this error is not cumulative and will be canceled at next lap when going back on the common car path of the driver.
We tested on a track and compared to video timestamp and error was few 1/100 secs differences (usually less than 2-3/100s) you can check it on some of our youtube video
This can be improved with software but not yet implemented in our software but as I wrote above the case you describe is not the normal use case when you're doing your best laps and as you wrote and expect the next lap this error is automatically corrected
Originally Posted by
Totallyosho
Today there was a post on lapmonitor's facebook regarding differences of 1/100 sec which reminded me that when I tested it, for the local track, this effect was way bigger (tested by checking at the two extremes of the track's width when the car was detected).
Anyone knows if this is somehow handled by the software or hardware? I don't care, at all, it's a fantastic product, I'd just like to be able to now what is a significant difference and what is car placement between laps, and of course over multiple laps this will average out.