[Coco] OS9 Level I - Clock issue?

[William Astle]

On 2015-12-28 10:06, Gene Heskett wrote:
> Unfortunately for the hardware clock, we still run into the overflow and
> wrap back to zero in 2038 because the year byte is just that, a byte
> whose max value is of course 255 in decimal.

Actually, it would be 2155 in this case, wouldn't it? The year is really 
"years since 1900" (or at least I think that's how it's been interpreted 
these days). No doubt there are older programs that show "19115" or 
"19;5" or just "1915" instead of "2015".

The 2038 wrap-around is applicable to signed 32 bit unix time stamps 
which count seconds since roughly Jan 1, 1970 at 00:00:00 UTC, but that 
doesn't wrap any time in the likely relevant future on systems with a 64 
bit count.

OS9 (and NitrOS9) don't use the unix time stamp scheme as far as I 
recall so they won't be affected by the 2038 thing. If I recall 
correctly, OS9 actually stores all 6 components of the time/date 
separately in a 6 byte packet.