home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.xenix.sco
- Path: sparky!uunet!decwrl!pa.dec.com!dynamix!david@uu3.psi.com
- From: david@dynamix.com (David L Jarvis)
- Subject: Re: Xenix considered harmful (was Re: SCO support - a success story)
- Organization: SOFTWARE / DYNAMIX
- Message-ID: <9209140723.AA28397@dynamix.com>
- Date: Mon, 14 Sep 92 7:23:30 EDT
- X-Received: by usenet.pa.dec.com; id AA00809; Mon, 14 Sep 92 10:11:24 -0700
- X-Received: by inet-gw-2.pa.dec.com; id AA24235; Mon, 14 Sep 92 05:30:40 -0700
- X-Received: from dynamix.UUCP by uu3.psi.com (5.65b/4.0.071791-PSI/PSINet)id AA08082; Mon, 14 Sep 92 08:26:36 -0400
- X-Received: by dynamix.com (smail2.5c)id AA28401; 14 Sep 92 07:23:30 EDT (Mon)
- X-To: comp.unix.xenix.sco.usenet (comp.unix.xenix.sco)
- X-In-Reply-To: <1992Sep07.084727.1644@kithrup.COM>; from "Sean Eric Fagan" at Sep 7, 92 8:47 am
- X-Mailer: ELM [version 2.3 PL11]
- Lines: 63
-
- > Excuse me?
- >
- > 1. You're not really likely to get "segmentation violation" (or "memory
- > fault") unless the processor has *some* sort of hardware memory protection,
- > which the '286 does. (The '286 has segment-level protection: read and
- > write/execute bits on each segment, and a whole nest of other things I don't
- > really want to get into without having my Intel manuals handy).
-
- Sorry, should have said "brain-dead" protection scheme :-) (note the :-)
- for the humor-impaired :-) )
- I'll admit my CPU architecture ignorance here ... all I know is
- that from the day I watched the first 286 Xenix systems go out right to
- today, I haven't seen many that are stable (if any) ... now that could be
- due to the 286, or the OS, or the applications - but the point remains ...
-
- > 2. If anything, the '186 was a bastardised '286: they both had the same
- > base instruction set and features, except that the '286 added a "protected
- > mode," four more bits of addressing (software and hardware), and, of course,
- > instructions to deal with all of that.
-
- Here I'll take you to task ... I worked for Tandy when the first 186 came
- to the computing world in the form of the Tandy 2000 ... it was *well*
- before *any* 286 box saw light of day ... I'd find it *very* difficult to
- believe the 286 was out first ... lets take this to private mail; tell me
- why you'd consider the 186 a bastardized 286
- (btw: my point there was that the 386 was, or seems to be ;-), a completely
- different architecture than the 8086, 80186, 80286)
-
- > For what it was intended, and the time and market, the '286 was mostly
- > adequate. It does not use a C/UNIX memory model, however, being closer to
- > something Pascal would have an eaiser time dealing with, but people have
- > coped. Intel did screw up in some areas on it, but not really the ones most
- > people claim.
-
- All I can base my opinions on is experience, which I have had plenty ... as
- for technical reasons why the 286 was so unstable, while I am interested in
- such things I'd have to say they are a moot issue at this stage ...
-
- > But, then, most people don't seem to think about Intel when they flame it;
- > they just seem to assume that, since everyone else *says* it, Intel *must*
- > be horrible, evil, and drain-bamaged.
-
- I would point out here that *much* of the technical information that was
- available to me while at Tandy came *directly* from Intel ... were I to
- flame Intel (which I have not, with my comments thus far), I would be more
- likely to flame them for their BLATENT disregard for consumers when they do
- things like disable the math-co in the 486 to create a market for the 487
- (which is actually a fully-functional 486, disabling completely the first
- 486 chip) ... or any of the other various atrocities they've committed
-
- >
- > Kinda like the U.S. political system, actually.
-
- It's not the system, it's the people running it, and the average joe-public
- who eats this crap-slinging up on the news every day ... much like problems
- in other countries I'd have to say ...
-
-
-
- #----------------------------------------------------------------------#
- # David L. Jarvis SOFTWARE / #
- # david@dynamix.com / DYNAMIX #
- #----------------------------------------------------------------------#
-