Originally Posted by
Steamer
+1. First time I ever had it work was this past Saturday. Before, my 500 would always produce a corrupted file anytime I tried using it on a 300K or longer. I don't know why, other than to say it hasn't been a problem on shorter rides, and that I never let the battery run all the way down. It worked perfectly on this recent 300K, for some reason. The firmware has not been updated recently, either.
Again, it has
nothing to do with battery power and
nothing to do with available memory.
Newer firmware doesn't appear to fix this problem (I'm running the latest).
Originally Posted by
Steamer
FWIW, I use a gomadic charger (with the special cable that allows simultaneous use and charging).
It isn't worth anything (I'm trying to be clear not rude). Mentioning will only confuse people and encourage them to do vodoo to try to prevent the problem.
Originally Posted by
Steamer
Edited to add: I wonder if following a single big TCX or GPX file for a long course is the issue. Legend has it that following a course that is longer than so many waypoints is corrupt-file-inducing. I can't really verfiy that, however.
It's unrelated to that.
I had the problem using a split route. The waypoint limit is 200 (supposedly) and the total route (not the split ones) had about 100 waypoints.
Anyway, if the course loads, the device should be able to follow it. Navigation is (mostly) a separate process from recording data.
Keep in mind that this is a "black box", which means we can't be certain what the cause is.
I'm going to
speculate here but it looks like the device is keeping track of some data in internal memory that is
related to recording data. Some problem occurs with this "tracking" that is correlated with a larger data file (the longer the ride or the more data, cadence/speed/HR, that is recorded). When the problem occurs, it seems like the code overruns a buffer (or two) and causes a fault in the software. The problem doesn't appear to be related to a particular size the data file reaches. Given the kinds of software problems that I've seen reported, it would seem that the device is fairly memory constrained and some work is needed to get all of features to fit (but that's speculation too). This speculation isn't worth alot (it just is an attempt to explain why the behavior appears fuzzy/flaky).