[lbo-talk] Democracy Now website

ravi gadfly at exitleft.org
Fri Aug 25 12:57:01 PDT 2006


At around 25/8/06 3:11 pm, Andy F wrote:
> On 8/25/06, ravi <gadfly at exitleft.org> wrote:
>
>> I guess the BSoD does provide the fitness advantage of the much required
>> occasional reboot (to clear memory leaks and all other things that one
>> is too busy or lazy to program for) and would just crop up in another
>> form if the particular bug that triggered it is fixed!
>
> Doesn't or didn't Windows have some glitch where an effective timer
> started when it booted, and then roll over in 42 days or something,
> resulting in BSoD?
>

You are a sharp cookie, young man! Yes from my [meagre] Windows programming days, I recall a 32-bit uptime (since boot) counter in milliseconds that overflows/wraps-around (2**32 = 4.3 billion max value / 1000 = 4.3 million seconds / 3600 = 1193 hours / 24 = 49.7 days).

--ravi

P.S: I think newer versions of Windows provide a 64-bit counter that applications can switch to using, which I would assume Windows kernel and system modules already do. My rough estimate tells me that will wraparound in about 500 million years, by which time I hope humankind would have abandoned Windows ;-). But there is the flip side: as computing speed and fine granularity applications grow, time measurements in more fractional seconds (than milli) gain importance, bringing the above consideration back into relevance, but I exaggerate. Even at nanoseconds, if my calculation above is right, we have 500 years of uptime. (not to forget there are programmatic ways to deal with this stuff, and other non-uptime timers are already available).

P.P.S: Found it: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/sysinfo/base/gettickcount.asp

-- Support something better than yourself: ;-) PeTA: http://www.peta.org/ GreenPeace: http://www.greenpeace.org/ If you have nothing better to do: http://platosbeard.org/



More information about the lbo-talk mailing list