Autosport Labs

Hmm, interesting, thank you.

Can you downgrade to 1.2.7 and do a quick test there?

We are soon releasing the 2.0 firmware beta + App, I will check to make sure this problem doesn't exist here. Thanks for reporting it.

brentp wrote:Hmm, interesting, thank you.

Can you downgrade to 1.2.7 and do a quick test there?

We are soon releasing the 2.0 firmware beta + App, I will check to make sure this problem doesn't exist here. Thanks for reporting it.

Ok so running it at 1.2.7 produces the same problem at multiple frequencies (above 10Hz). I've noticed that the time increment is always either 0, 0.094, or 0.109 (this is the case with both 1.2.7 and 1.2..

I also lowered the frequencies down to 10Hz on the Accelerometers, and 5Hz on the GPS to see how it responded to multiple frequencies below 30Hz (since running everything at 10Hz has given the best results so far).
On both the 1.2.7 release and 1.2.8, running the clock and GPS at 5 Hz, accelerometers at 10Hz, I got fairly consistent time intervals (0.203, 1 in every 5 was 0.188).
Running the clock at 10Hz (all else run at 5 or 10Hz as listed above) resulted in time intervals of 0.094 or 0.109.

Any thoughts?

--
Evan

I was seeing the empty cell phenomenon recently too - http://www.autosportlabs.org/viewtopic.php?t=4115 though I'm not sure which firmware I was testing with. Different logrates as well - my GPS time was logging at 1Hz.

toga94m wrote:I was seeing the empty cell phenomenon recently too - http://www.autosportlabs.org/viewtopic.php?t=4115 though I'm not sure which firmware I was testing with. Different logrates as well - my GPS time was logging at 1Hz.

Have you discovered a solution to it yet?
I checked on one of the bench files I did (logging GPS time at 5Hz). In Race Analyzer, it is 6.300s, my logfile is 64 lines including header (which looks good so far), but the time stamp is 182707.203 - 182700.797 = 6.406 seconds. So, it's climbed 0.106 seconds already in a ~6 second log.

We changed the V2 firmware to have a different, more precise timing mechanism for the logging intervals. Let us check that out and see if we can reproduce the problem.

-Brent

Ok. Let me know what I can do to help, if anything. I'm looking forward to hearing what your results are.