jks

Seeed had a batch of BBGs with bad Ethernet phy chips a while back (or bad magnetics in the RJ45 causing phy problems -- it was never clear). Although I thought it was more than a year ago. So if this was a Kiwi purchased from drop.com a year ago that's been sitting in the box ever since then it's quite possible it has a BBG with the known defect.

Well, you can force the AI to run at 500 MHz. But the Kiwi server may have problems at that clock rate. From a shell window (via ssh/PuTTY connection or admin console tab) use the command "cf5". Then "cf" to check that it actually switched. Remember that the BBB/BBG runs at 1 GHz (fixed).

Other commands are:

cf15 = 1.5 GHz
cf11 = 1.176 GHz
cf1  = 1 GHz

v1.377 contains a fallback that retries the git pull using a fixed ip address for github.com. Interestingly, about half way through checking the fix on Tony's Kiwi the git pull to github.com (using the domain name) started working. So that means this was probably a mirror server configuration issue by Github as opposed to an ISP issue. Anyway, the fix is in and will hopefully save someone some grief in the future.

I don't know. If there is no code running that requires GPU cycles then I can't imaging it burns any power on its own just sitting idle.

Hi John. Yes, probably. The algorithm chosen for the GPS update of the ADC XO frequency was checked by opening the door to the outside (on a winter's day) that is adjacent to where the Kiwi is developed and watching what happened. At the time the GPS time/position output was a lot noisier than it is now (thanks Christoph!) so that was the tradeoff. And of course the whole thing is not a PLL (and definitely not a GPSDO). It's more of a FLL.