HP3000-L Archives

April 1998, Week 3

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:
Denys Beauchemin <[log in to unmask]>
Reply To:
Date:
Fri, 17 Apr 1998 13:04:51 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (88 lines)
X-no-Archive:yes
I have been out of town for the last few days and so I missed responding in
a timely fashion to this thread.  I only got the chance to view just a few
messages until this morning.

First off, I have heard of problems with Drive Copy.  There are fixes at
their web site.  Before I attempt to use such radical programs, I always
check at the web site for newer versions.  This has saved my hide more than
once.  I would have suggested the XCOPY and SYS method, but Costas preceded
me by a few days.  That is an excellent method and if you are only going to
copy a drive just one time, that is a good way to go.

Second, there are other utilities to do the job, GHOST from Ghost software
is a good one also.  However, Ghost of DriveCopy are good solutions for
people who need to crank out copies of disc very quickly.  They are much
faster than XCOPY and they are usually foolproof.  I would have liked to
have seen the miscopied disk drive and try to figure out what the problem
was.  My guess though, is that the latest version at the web site would
have solved the problem.

Third, the XCOPY solution will not completely work if you have dual boot
systems.  You will need to recreate the MBR with the ERD and setup
diskettes from NT.

Fourth, in answer to Bruce's question.  Bruce's suggestion to use a SCSI
enclosure and try out the disk drive before hand is an excellent one.  I
have a couple of SCSI enclosure and I have done this in the past.  It is an
excellent way to go.  You all should know I am a fan of SCSI-based systems.
 I participate in other lists dealing with NT and, up until recently,
Windows 95.  I admonish people to set up their PCs for NT solely on SCSI.
 There are so many advantages with SCSI VS IDE, it makes the decision a
no-brainer, especially with respect to servers and high-end workstations.
 I cannot make quite the same compelling case for regular desktops, and
unfortunately, no laptop has SCSI drives, though I do have a SCSI PC-CARD.
 On my NT server, the Adaptec 2940 card's BIOS does not allow me to boot
from any SCSI ID.  It will only boot from SCSI ID 0.  The documentation
does say that it will boot from any drive on the SCSI bus, while under BIOS
control, but reading the full documentation, and my experiences, show that
whilst any drive will boot, it must be assigned SCSI ID 0.  Other adapters
or other brands my work differently.

Fifth, I read Gavin's post this morning and let me write this about that.
 I always fix the drive letter of my CD-ROM to something other that the
next available letter after my last partition.  In Windows NT you can do
this using the disk administrator.  In Windows 95, you can do that with the
device manager, and properties of the CD-ROM.  One caveat for Windows 95.
 (I went around for a few days with a hoser on Win95-L about this.  He
insisted there was no downside to having real-mode drivers loaded for his
CD-ROM. Apart from the drive letter issue and the extra load in memory, he
was right :->.)  You must have protected mode drivers for your CD-ROM.  If
you do, you will be able to easily nail the CD-ROM to any sensible letter
you choose.  I chose F:.  If you insist on real-mode drivers, you will have
to diddle config.sys and autoexec.bat (2 files which we do not have on our
Windows 95 desktops or laptops.)

WARNING**WARNING**WARNING (too late for Gavin), never insert a drive letter
in the middle of a range of existing drive letters, this will cause untold
chaos.  Also, it has been pointed out at other places, that adding an IDE
drive on a system with existing SCSI drives, will cause the reassignment of
drive letters.  A fine state of affair indeed.

Sixth, Gavin goes on about the registry, and I agree with him.  After years
of dealing with the registry, I find it causes more trouble than it cures.
 The implementation is sloppy, the documentation, spotty and the chances
for corruption are way too high.  At least on Windows 95.  On Windows NT, I
find it is much more stable, but heed the warnings you encounter when
dealing with the registry.  I have attended classes on the registry, and I
have various books, some obscure on the subject.  I deal with it, because
out backup product deals with it.  It is still a pain.

I hate having products which have to dump all sorts of crap in the registry
and must be reinstalled in order to work.  I hate it even more when you
have to mess with the registry when you delete the application.  MS does
have a good registry cleanup utility, called REGCLEAN.  Get the 4.0a
version or better.

Bruce is entirely correct on this.  If the registry would be fully
documented by M$ it would simplify life tremendously.

Kind regards,

Denys. . .

Denys Beauchemin
HICOMP America, Inc.
(800) 323-8863  (281) 288-7438         Fax: (281) 355-6879
denys at hicomp.com                             www.hicomp.com

ATOM RSS1 RSS2