... the fact that your (or any other) Bay Trail device is hopeless crippleware.
Bay Trail reboots into a completely different random universe.
each and every time the power button is depressed.
Conceivably, you might try
i915.modeset=1 in your grub options,
but don't get your hopes up.
Condolences are in order :
It is almost as if there is a hard-wired random number generator
attached to the power button.
It starts with faulty pin control, arising from the fact that some
of the power controls have default states that do not get reset
upon power on power off cycling.
And cannot be fixed. Ever.
Intel Bay Trail is a spectacular train wreck complete with super-size-me body bags,
due to some overlooked architecture issues which the Intel folks
seem to so nonchalantly overlook prior to market release.
" Intel has officially discontinued its Celeron N2820 Bay Trail chip
mere months after launch..."
Their 'release to market, oops, then fix the bugs modus operandi' is so much fun:
"Some of those flaws, it would appear, have been traced back to the
Windows has the same problem and it cannot be patched over - it just stutters a lot.
These crippleware kits always come with Win 8 installed,
I should have known better than to blithely assume because it had Win 8 we could
simply replace it and proceed at pace.
I paid good money for a (Kick me again if I ever buy another SoC cesspool from ACER)
Iconia W4-821 with 3740 Bay Trail and I have been on a rocket-launch learning curve.
Google "fedlet" or "bay trail kernel" to see what a bunch of us who deluded ourselves
into spending time and money on BAY TRAIL and you will see what is involved.
Kudos to the guys who got there first and shined a bright light on the issues:
Adam Williamson "Fedlet" (he is an ex-Mandriva rock star from way back).
Jerry Wells "Ubuntu on Asus Transformer Book T100"
...but the juiciest comments are in the kernel mailing list,
where some of the guys who are trying to patch the linux kernel
to accommodate faulty Intel hardware have some interesting perspectives:
Quoted - I kid you not ! - from
Re: Baytrail/T (ASUS T100 etc) regression from 3.13 onwards
Ugg, so reverting a real fix to make the T100TA work by accident...
this is ugly.
A module paramater to fix a bug? Ugh, that's almost worse than just
reverting the original patch, right?
Please don't do this, the distros will hate you even more than they
currently do :)
I could send you videos of how random this power reboot issue is
but it would be an utter waste of time and bandwidth.
Brick, uses power, and displays random color schemes.