I have this ... gut feeling shall we say ... that the test devices at work have something odd about them ... some sort of non-linear timing or harmonic about them or something. My problem is I don't have anything to compare them directly against i.e. the same hardware, more or less, but running Linux instead of BB10. So I figured, what the hell, I'll buy a cheap chinese Linux-on-a-USB-stick: a not-that-cheap Tronsmart T428 at $100 with a RK3188 chipset from the famous ultra-cheap Rockchip vendor. Why so expensive i.e. about twice the price of normal Linux-on-USB-sticks? Well this fellow has 2Gb of RAM and a quad-core Cortex A9 running at 1.8Ghz, and he's bleeding edge in ultra cheap and nasty chinese ARM SOCs. He actually ought to be relatively capable i.e. near the same ball park, albeit nothing like as good, to what I'm testing. Even for my own non-work hobby software projects, it could be very handy to have a native ARM test target around - though even with its four cores, I think I'll be cross-compiling to it as the RAM they fit these things is usually unbelievably slow.