HP3000-L Archives

December 1998, Week 1

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:
Roy Brown <[log in to unmask]>
Reply To:
Roy Brown <[log in to unmask]>
Date:
Mon, 7 Dec 1998 14:03:57 +0000
Content-Type:
text/plain
Parts/Attachments:
text/plain (92 lines)
In article <[log in to unmask]>,
Denys Beauchemin <[log in to unmask]> writes
>X-no-Archive:yes
>I would like to take a few seconds and explain the reasoning behind my
>earlier message on this thread.  If you remember yesterday, I answered,
>rather tersely, "run far, run fast. . ." away from DOS applications on NT
>(or even 95/98.)
>
>When Roy compares buying DOS applications to buying MPE applications, he is
>very incorrect.

I may be wrong, Denys, but it's not because I don't know any of this
stuff below. Or does that just make it worse? :-)

>  A more proper comparison would be buying a (probably)
>PM-based application developed on MPE V/R, running in batch, and trying to
>use this on your brand new 997.
>
And it will run perfectly accurately, blazingly fast, and do everything
it was designed to do. 'Cos HP are better at MPE compatibility than M$
are as DOS compatibility.

However, it won't do what apps today can do. But what makes anyone so
sure that it needs to?

>Many people are under the mistaken impression to NT is built from or even
>on DOS.  This is totally incorrect.  NT is NT, and is derived from VMS, not
>DOS.  NT does handle many DOS programs, through a DOS emulator, but it
>fails to run them properly as often as not.  If your DOS program, in any
>way, tries to access the hardware, NT will reject it, hard.

Yeah, Dave Cutler, rings of privilege, etc. I reckon VMS could give MPE
a good run for its money in its day. Do we thank that one of these days
NT will do the same, or will Microsoft's QC ensure that never happens?
>
>If you DOS program works on NT, it will not be able to take advantage of
>long file names, or most of the features of NT, such has multi-threading
>and services.  The NT system will not be affected by the DOS program, as
>long as this program does not replace critical components of NT.  I have
>found that most of the time, the reason programs were not moved from DOS to
>NT is because the program wants to do something NT will not allow or the
>source code was lost.

Jim's salesman claims that his program runs on 95, 98 and NT. These are
claims whose veracity or otherwise is easily tested.

BTW, damned if I can see how to get MPE to do long filenames.
I can't even get Netware/iX to do them for our logical server....
>
>The situation is less critical on Windows 95/98 as these OS's were built
>from DOS and allow DOS to have full access to the machine.  Actually this
>was one of the goals of Windows 95, allow DOS programs to run.  On Windows
>95/98 DOS applications run single threaded and can still take over the
>whole machine in a runaway situation.  DOS is not a pre-emptive
>multitasking environment.

I have assisted the developers of a powder-flow measuring system
(dedicated kit plus an interface card and software on a PC) whose DOS
code ran fine on raw DOS or WIN 3.1, but doubled its speed in a DOS box
under WIN95. Which rather invalidated the calibration. Worked fine in
DOS mode on a WIN95 machine. Wrong choice of timers, reasonably easily
fixed, but shows that even on 95, DOS compatibility isn't all it's
cracked up to be.
>
>Overall, applications built for NT are more reliable than DOS versions, and
>they have the added benefit that they can run nicely on Windows 95/98,
>unless they depend on NT-specific things.  If the NT application also has a
>GUI, so much the better for the user.
>
Yes of course. But let's not forget the original question. Should Jim
let this unit buy a DOS app? One the salesman has demonstrated working
(though we don't quite know where)?

And sure, other things being equal, you'd pick an NT app over a 95/98
app over a DOS app. I guess. But *are* they equal?
I guess it comes down to: how much specific *app* functionality would
you sacrifice to gain general *OS* functionality?


But my main point was that Jim used the term FUD, and he was aware of
why he was using it. Now there are a lot of people out there who hear
'HP3000 app' and turn off through FUD, not through an objective
appraisal. Which, fair enough, may go against the HP3000 app.

But if us HP3000 types can't give an app a fair appraisal because it
uses an off-the-norm OS, what right do we have to expect it from others?
--
Roy Brown               Phone : (01684) 291710     Fax : (01684) 291712
Affirm Ltd              Email : [log in to unmask]
The Great Barn, Mill St 'Have nothing on your systems that you do not
TEWKESBURY GL20 5SB (UK) know to be useful, or believe to be beautiful.'

ATOM RSS1 RSS2