KH> You've probably already had a few replies on this, but the problem's
 KH> not with the 'year' byte - it's with the 'century' byte (offset 32H
Ahh yep I didn't notice the century byte :)
 KH> - also in BCD - see CMOS.LST). A lot of older ATs do not increment
 KH> the century byte with the passing of midnight, Jan. 31 1999 (mine
So, do they increment it at all?  Otherwise, it seems a bit redundant.
 KH> BIOSes in the late 486/early Pentium era - these require a
 KH> replacement BIOS to function correctly. AFAIK, most (if not all) of
Hmm.. You can't just increment the century?
 KH> one. A lot of the programs in use today don't even bother with
 KH> checking the century, and this is going to lead to a _lot_ of
Yes, but most software, you'd think, would use a full 4-digit year, so that 
shouldn't be a problem.  Also, most operating systems manage to figure it out 
that the year should be 2000.  At least, in my test MS-DOS 6.22 got it right. 
I haven't tried it under Win95.
Thanks for the reply!
Sam
--- FMail/386 1.20+
---------------