Re: Flaky timekeeping?

> Sounds like you've narrowed it down to a software issue.

I thought so too.  It now seems that either 1.4T has issues too, though
my evidence doesn't rule out the possibility that they're
hardware-based.  Specifically, when my build of the world finished, the
drift dropped from about .4% to about .1% - the job mix _does_ affect
timekeeping (though 4.0.1 still kept substantially worse time than
1.4T, so there's also a significant version difference).

> Now, if only you could merge the working SCSI support with the
> working time code!

I may try to backport the SCSI support at some point, though I also may
not bother since the main point (or at least the main driving desire)
of my messing with my NeXTs is a working printer, and it's possible the
NeXTprinters I have will turn out to be nonworking one way or another.

Right now, I'm off trying to track down a non-NeXT-related bug, though,
so the NeXT stuff will probably mostly lie fallow for a bit.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		[email protected]
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B