HP3000-L Archives

January 1999, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Gavin Scott <[log in to unmask]>
Reply To:
Gavin Scott <[log in to unmask]>
Date:
Sat, 23 Jan 1999 13:16:09 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (19 lines)
Larry writes:
> If you do this we might as well add the milliseconds.  Some time we will
> need them and while in that area of code go for both.

Except that milliseconds are already useless for generating any kind of
unique value.  Even microseconds are unsafe on today's systems which are
executing in the neighborhood of 1000 instructions every microsecond
(multi-issue processors running at 100s of MHz).  Today nanoseconds
(.000000001 second, or one American billionth of a second) are still
good, but this will change in the next few years.

A related problem is that once you get to enough resolution to ensure
unique/ordered time stamps, you can no longer represent these values in
convenient form (i.e. 32 bits), but have to go to 64-bit sized data
types, which are inconvenient to manipulate (but that's what MPE does
internally).

G.

ATOM RSS1 RSS2