On the new version, it seems pretty reliable that it breaks after 45-50 mins, and often the error is the invalid value report rather than a stuck core. The fact it happens after a fixed time seems odd.
I left it running overnight and actually it was still running this morning, so with a reboot after every error it is possible to run this version - though obviously with errors this frequent that's far from ideal!
It's interesting you see chip passed on thoses tests and I don't. But I've raised that problem on the SDK beta test thread so will leave discussion of it to there.
Cheers,
MikeStatistics: Posted by mikebell — Sun Mar 15, 2015 10:17 am
]]>