Info-IBMPC Digest V95 #101
Info-IBMPC Digest
Today's Editor:
Gregory Hicks - Santa Clara, CA <GHI...@arl.mil>
Today's Topics:
buffer size with smartdrv
Re: Buying a PC. (Which one to buy??) (3 msgs)
Re: cd rom drive needed (2 msgs)
changes for UUPC/extended 1.12o
Re: Clock trippler
Floppy Disk Drive Troubles (5 msgs)
Re: Help in decision (2 msgs)
Help: Question about IDE of Borland C++ 4.0
Re: new motherboard purchase
NFS for Windows (2 msgs)
Send replies or notes for publication to: <INFO-...@arl.mil>
For readers in the .MIL domain: Send requests of an administrative
nature (addition to, deletion from the distribution list, et al) to:
<INFO-IBMP...@arl.mil>. For all others, please send
requests to join, or signoff from, the list to an IBMPC-L BITNET
LISTSERV.
Currently these LISTSERV's are at hosts vmd.cso.uiuc.edu,
hearn.nic.surfnet.nl, taunivm.tau.ac.il and uga.cc.uga.edu. There are
others...
----------------------------------------------------------------------
Date: Wed, 10 May 1995 16:53:45 -0400
From: "J. Dashiell" <dash...@nadc.nadc.navy.mil>
Subject: buffer size with smartdrv
> There's a buffer size parameter that can be set in smartdrv.exe
>however like alot of microsoft stuff there's no guidelines for
>optimum installation on this parameter. So my questions are: .... some
>lines deleted ....
Hi Jude
I know this isn't exactly what you want to hear, but here it goes
anyway. Well, I played around once (or tried to) with smart's buffer
size and gave up because setting values for this thing other than the
default one would mess with the cache size settings. I mean the values
specified to disk caches would be interpreted accordingly to the
specified buffer size and I could see no logical relationship (wasn't
able to set a buffer size that would lead to a correct interpretation
of cache sizes by smartdrv). This was a very superficial attempt of
mine though. If you find it differently, I'd like to be told if you'd
be so kind and helpfull.
Any ideas or reactions? Is this another bug?
jude <dash...@nadc.nadc.navy.mil>
------------------------------
Date: Wed, 10 May 1995 10:01:12 -0500
From: "Micki S. Uhrig" <muh...@ozarks.sgcl.lib.mo.us>
Subject: Re: Buying a PC. (Which one to buy??)
Hi. My last pc was a PB (386sx, 16 mhz). Technical phone support was
good but upgrading was a major pain. My new pc I purchased from Royal.
The system has worked great but phone support is very poor.
Micki
muh...@ozarks.sgcl.lib.mo.us
muh...@cap1.capaccess.org
muh...@rain.gen.mo.us
------------------------------
Date: Wed, 10 May 1995 14:42:55 -0400
From: Michael <m...@netcom.com>
Subject: Re: Buying a PC. (Which one to buy??)
On Wed, 10 May 1995, Doug Becker wrote:
> My recommendation is NO!!!!! I know several people who have waited
>for several days to get the service on a dead PC from Packard Bell. I
>knw that there are other people who have had no problem with them but I
>try to stay away from them with out a doubt.
I have had friends with the same experience. One of the problems with
a PC like Packard Bell is they generally come "preconfigured" (ie, with
a sound card, video card, CD ROM drive, etc.), so you don't have much
choice/say in what your getting. The biggest "problem" seems to be
that the people I know who have purchased such "packages" (not on
Packard Bell) find that often the sound card, video card, etc. is not
always compatibe with a number of programs.
When I bought my last computer a couple of years ago, I went to
computer stores that configured PCs. I was able to choice the brand
sound card, video card, CD ROM, speakers, etc. I wanted. In most
cases, I found that such dealers would meet (or come very close to) the
prices of the preconfigured setups. (It doesn't hurt to take a
competitor's advertisement with you when you go shopping.)
Also, service is very important. Generally, the smaller, privately
owned computer retailers are more knowledgeable about what they sell
and are more responsive to questions.
Good luck!
Michael
(in Washington, DC)
------------------------------
Date: Wed, 10 May 1995 10:34:29 CDT
From: ERIC DAVIS <epd...@buzzman.b8.ingr.com>
Subject: Re: Buying a PC. (Which one to buy??)
look at acer computers. they seem to back their computers well and they
have been making pcs for people like toshiba. give them a look see.
------------------------------
Date: Tue, 9 May 1995 22:59:32 -0500
From: Paul Schmid <psc...@camelot.bradley.edu>
Subject: Re: cd rom drive needed
>I am a little confused about the interfaces. Creative labs says the
>interface on the sound card is proprietary made by Panasonic, but my
>Gateway 2000 tech rep says any CD rom (IDE interface) should work fine.
>They have a double speed Sony for $119. with cables.
Rick,
The E-IDE/ATAPI drives will not work on a sound card interface
unless the sound card specifically has the "ide" interface built in.
For instance, a Panasonic E-IDE/ATAPI drive *will not* work on the
Panasonic interface on a Sound Blaster 16.
If you want that type of drive, get an E-IDE controller card or
use the interface card supplied by the manufacturer - the card will
give you kind of a pseudo-E-IDE :)
Good luck,
Paul
------------------------------
Date: Fri, 12 May 1995 15:53:21 -0500
From: Rick Snyder <ri...@aac.msstate.edu>
Subject: Re: cd rom drive needed
Paul, and others:
Could you explain what E-IDE is, and is this different from IDE
interface?
Since your note, I think I understant that a CD ROM drive can connect
in any of 3 different ways:
1) connect to the sound card, if the interface is appropriate.
In my case, I have a value edition Sound Blaster 16, and the interface
on the card is proprietary for Panasonic, so only one of their CD-ROM
drives would work.
2) connect to the IDE cable, the same one that goes to the hard
drive(s). In this case, it doesn't matter what the interface is on the
sound card, because it won't be used anyway.
3) connect to a new card which comes with the CR-ROM drive.
The disadvantage to this is that it takes up a slot. The advantage is
that any interface will work (IDE, SCSI, other) because the drive comes
with its own.
What do any of you think of the above. Is this correct?
Also, what brands do you suggest? Is a double speed adequate for most
purposes? Is $119 for a Sony double speed IDE a good price?
Thank you.
Rick
------------------------------
Date: Sun, 07 May 1995 15:20:41 -0400
From: Soft...@kew.com
Subject: changes for UUPC/extended 1.12o
Here's the list of changes for the pending release UUPC/extended 1.12o.
Not listed below is that I'm actively working on the news documents.
UUPC/extended 1.12o will ship with at least newly updated drafts of the
documents, the 1.12b are getting far to creaky ...
Enhancements
The new Boolean option suppressLoginInfo is added to allow disabling
the display of the system version and login time during remote system
login.
The previously unimplemented Boolean option undelete now works under
OS/2. When noundelete is set or defaulted, files are deleted by ALL
UUPC/extended programs using a faster file delete call which bypasses
the OS/2 Undelete Cache.
Bug fixes
As usual, Kai Uwe Rommel contributed many of the news related fixes to
this release.
When no news server was defined in the UUPC.RC file and no SYS file
existed, an invalid SYS file was generated. The automatic generation
of the SYS file was corrected to use the mail server as the default
news server as well.
The unused Boolean option compressbatcxh is dropped.
When no parameter are passed to a program from UUXQT or other programs,
a bug in the IBM OS/2 C/Set++ compiler will generate invalid parameters
to the program. A workaround has been applied to various programs to
always pass at least a debug level to various programs such as RNEWS,
INEWS, abd SENDBATS.
The code added to 1.12n to determine the starting sequence number would
sometimes generate a negative sequence number for jobs. Corrected to
always limit the sequence number between one and one million.
Various programs tended to have stack overflows under the MS C DOS 8.0
compiler, especially when various common routines called memory
intensive C library functions such as sprintf. Various common routines
were revised to use more efficient formatting, and in addition various
DOS modules were changed from .COM files to .EXE files to allow
additional memory usage.
Previously, if an entire subdomain (denoted with an asterisk prefix)
was listed in the HOSTPATH file to be routed to a gateway program, the
subdomain entry was passed to gateway program by RMAIL. RMAIL is
corrected to pass the actual host name being gatewayed to the program.
If RMAIL processed a forward file with no aliases, it would exit with a
non-zero return code but no error message. It now bounces the mail
properly with a message reporting no addresses could be delivered to,
and then exits with return code zero.
When writing entries into the V-Mail server queue, RMAIL searched the
directory for a free file name, which because of a race condition could
result in a file name collision. RMAIL is modified to use normal
sequencing instead for more reliable and faster operation.
If an entry in the system aliases file was immediately preceded by a
comment line, the comment was processed as part of the alias. System
alias processing is corrected to completely igore the comment.
Under Windows NT, changes to the internal configuration table would
cause REGSETUP to fail because of references to NULL pointers.
If UUPOLL was running with auto-UUXQT (-U) mode where UUCICO normally
runs UUXQT in background, UUXQT was never run for thes local host.
This in turn caused news command which would locally queued to never
run. UUPOLL is modified to automatically run UUXQT for the local host
when in auto-UUXQT mode.
When a call grade was specified in the time field of the SYSTEMS file
and UUCICO was invoked to call any system with work ("-s any"), UUCICO
might try to call a system it has no work to send to.
The NIST clock setting code failed to work properly with the MS C DOS
8.0 compiler, and generally smelled bad. Restructured the code to
break apart system dependent and independent parts, corrected system
dependent code, also revised messages to make the action taken clear.
In some cases, various programs either overreported or undereported
problems with deleting files. Modified affected programs to exactly
report failed deletions of files.
If a group was removed from the active file by a rmgroup command, all
groups after that group were deleted as well. Corrected active file
processing routines to not truncate active file list.
A space in a message id would cause severe problems with the history
file database. Modified history file processing to properly handle
such ill-formatted fields.
When RNEWS is processing news from a pipe, it failed to properly
process the data. RNEWS is modified to (more) gracefully handle such
input.
Maximum hop processing news caused infinite loops in NEWSRUN.
Corrected this error and moved the check maximum hops to inside the SYS
file processor where it belonged.
Other changes
Corrections of various compiler warnings continued.
--
Drew Derbyshire UUPC/extended e-mail: soft...@kew.com
Telephone: 617-641-3452
------------------------------
Date: Sun, 7 May 1995 20:52:02 GMT
From: "David W. Gillett" <David_W....@livewire.com>
Subject: Re: Clock trippler
> Anyway, why I'm wondering about the upgrade to 100 is I heard that with
> IBOX (or any graphic interent windows program) the speed of your
> computer, not just modem speed, determines how fast it is because your
> computer becomes the server. Something like that anyway. I have noticed
> that in IBOX everything comes in 256 byte blocks and it likes to pause
> every once in a while. The block size itself seems very small to me
> after Zmodem and the speed of FTP when I'm doing it in text mode. Even
> with the double download (from the FTP site to Skynet and from Skynet to
> me) it's usually a lot quicker than getting a big file by FTP while in
> graphic Internet. How would going from a 486DX-33 to 486DX-100 affect
> the speed, or would it? Would it increase the block size, get rid of or
> shorten the pauses, or just process everything faster? I assume it would
> be a very noticable difference to, right? Even from a 486DX2-50 to a
> 486DX-100?
The performance of FTP of any sort is going to be dominated by line
speed. [Depending on details of the protocol, tuning block size can
bring actual bandwidth used *closer* to the bandwidth officially
available, but even a poorly-tuned protocol will run about twice as
fast at 28.8 as at 14.4. A second factor is going to be the speed of
writing to disk. Raw CPU speed will rate a distant third, except that
for instance the average XT can't keep up with 14.4 -- but the
difference between a 486-33 and 486-100 isn't likely to be noticeable
for FTP.
CPU speed *will* make a difference in graphics performance. (Just a
couple of months back, a column by Michael Abrash presented a
line-drawing routine for Pentiums which produces pixels faster than any
common PC video adaptor can display them; this is quite a feat.) This
has nothing to do with which box is called "server", but simply with
who's doing the work.
[I'm not specifically familiar with IBOX, but I suspect that the
256-byte block size and slower FTP in "graphic Internet" are very
likely to be symptoms of a design which allows you to continue to use
the Internet interactively *while* the FTP transfer is going on,
something that ZMODEM doesn't provide for....]
------------------------------
Date: Wed, 10 May 1995 12:54:26 +0000
From: Frederic Desire Elai Bisson <F.D.E.Biss...@lmu.ac.uk>
Subject: Floppy Disk Drive Troubles
Hello,
I've got two disk drives (A:=3 1/2 HD, B=5 1/4 HD) that worked very
welle until yesterday. Yesterday has seen my 3 1/2 HD drive died. The
tests performed by the AMI BIOS are all okay, the heads can be moved
normally, but when I tried to do something like reading or writing
(including DOS, BIOS and Hardware access) the drive doesn't read my
disks that are readable on any other computer. The problem is that the
drive seem not to find any 'adress mark' on the disks that appears
unformatted !
It's only my 3 1/2 HD drive that doesn't work, the 5 1/4 drive works
very well as usual !!!
Does anyone know where this trouble comes from ?
Thanks
BISSON Frederic
------------------------------
Date: Wed, 10 May 1995 09:29:12 EDT
From: "Sean F. Leinen" <sfle...@sean.jersey.ingr.com>
Subject: Re: Floppy Disk Drive Troubles
That anomaly you see *usually* is an alignment problem with the floppy
drive. This can happen when a drive is dropped or something hits it
hard (floppy drives don't like the sudden stop that occurs during a
freefall :-). I'm not saying that's what happened with your floppy
drive, however. It could very well be that something failed in that
drive, and the head alignment no longer matches that of any other
floppy drive out there.
One question/tip I'd like to pass on to you; can this drive read it's
own-formatted floppies? In other words, if you format a floppy in that
drive, can it *read* that floppy it just formatted? If so, can you
read that just-formatted floppy on another machine? If not, this is an
alignment problem, and with today's miniaturized floppy drive
technology, it's usually better and less costly to just replace the
drive (it would cost you more time and expense to align that floppy
drive than it would to replace it).
Hope this helpful,
--
Sean F. Leinen NY/NJ Senior Systems Specialist (201) 460-7421
Mailstop: NJRUT WWW: http://www.intergraph.com. ARPA: sfle...@ingr.com.
Intergraph Computer Systems, One Meadowlands Plaza, East Rutherford NJ
07073
------------------------------
Date: Wed, 10 May 1995 15:34:42 +0200
From: Intact - Petrofina <fi...@wise.nl>
Subject: Re: Floppy Disk Drive Troubles
The fact that the drives are still stepping during the
Power-On-Self-Test is not an indication that all's well. Most self
tests (as far as I know) only test if the drives respond to the step
commands, and then try to step 80 tracks forward and 80 tracks
backwards. The "track-0 sense" line is used to detect if an 80-track
drive is actually installed (in the case of a 5.25" drive). The actual
reading/writing operation is not tested; if it would have been tested,
you'd always see an error message if you start your PC without a floppy
in the drive(s).
From the fact that both your drives stopped working, I'd say the
read/write circuitry on the floppy controller is broken, not the floppy
drives. (The index-circuitry is only used while formatting, so that's
not it) I suggest trying another floppy controller or trying one or
both your drives on a different machine which has working drives.
Jac Goudsmit
CD-i/CD-ROM Software Engineer
Codim Interactive Media CV
Eindhoven, The Netherlands
j...@wise.nl
------------------------------
Date: Wed, 10 May 1995 16:36:18 +0200
From: Jyri Ruut <jy...@ttkt.tartu.ee>
Subject: Re: Floppy Disk Drive Troubles
The problem may be in a cable connecting your I/O card and the drive.
Maybe it helps when to remove the computers cover and make sure that
connections haven't loosened. Sometimes, some of the wires in the cable
may have brokened.
Ju"ri Ruut e-mail: jy...@ttkt.tartu.ee
Tartu Tervisekaitsetalitus ph.: (372) 7 434272
Tartu Public Health Service fax: (372) 7 430685
private: (372) 7 477930
------------------------------
Date: Fri, 12 May 1995 19:24:41 +1000
From: Ted Forsyth <tfor...@ozonline.com.au>
Subject: Re: Floppy Disk Drive Troubles
Hi all,
If MS-Dos reports a seek error, then if your lucky it's a cable
problem. If it's a general error, check your bios settings and
cabling anyway. You never know what you may find...
See ya...
Ted...;-)..
------------------------------
Date: Sun, 7 May 1995 21:05:03 GMT
From: "David W. Gillett" <David_W....@livewire.com>
Subject: Re: Help in decision
It sounds to me like your plan is something like:
1. Buy and install zip drive and a couple of disks.
2. Copy entire hard drive to zip disks.
3. Remove zip drive.
4. Sell/abandon old computer.
5. Bring zip drive and disks to U.S.
6. Purchase new computer.
7. Install zip drive, and copy stuff from disks.
Have you considered?
1. Remove existing hard drive.
2. Sell/abandon old computer.
3. Bring hard drive to U.S.
4. Purchase new computer.
5. Install old drive as second/slave drive, at
least temporarily.
Note that I'm assuming that your 200 MB drive, and the drive in your
new machine, are probably compatible IDE drives. If you're thinking of
changing controller type then this option becomes less practical, but
it seems to me that you might be able to save both money and effort.
[Your old computer is likely to fetch less, but what's the street
value of a used 200 MB drive? If you can find a *new* one here, it's
probably under $100....]
------------------------------
Date: Sun, 7 May 1995 17:09:32 -0400
From: John Garvey <JGa...@shore.net>
Subject: Re: Help in decision
It seems to me that it would be easier and cheaper to just borrow an
external tape drive and back up the entire drive on tape and then just
reinstall on the new machine when you get it. Maybe even better still,
why don't you buy the tape drive since frequent backups should be made
anyway. You would also be able to sell your old computer intact rather
than with a missing hard drive.
------------------------------
From: "SD...@DELPHI.COM" <SD...@news-feed.delphi.com>
Subject: Help: Question about IDE of Borland C++ 4.0
Date: 10 May 1995 19:32:48 -0400
Hi, I have a question about IDE of Borland C++ 4.0. When I use RUN
command in the Debug menu, it is supposed to run my application without
re-compiling if the program is not changed. I don't know why,
everytime I use RUN command, my application program is re-compiled and
re-linked, which drives me crazy since my program is long and takes
minimuts to finish these process. Does anyone know how to fixed it? I
really appreciate your help. Please email your suggestions to:
Thanks again!
Weiying
------------------------------
Date: Mon, 8 May 1995 11:25:27 EST
From: David Schreiber <DSCH...@ansc1.cag.uconn.edu>
Subject: Re: new motherboard purchase
> I am about to purchase a new motherboard. I have in mind an AMD 486
>DX4 100 MHz. Looking through the latest issue of Computer Shopper, I
>see vast price differences on this board. I was going to buy from TC
>Computers for $399 but then saw an ad from Galaxy Computers and Ledgend
>Computers that have the same board for ~$299-$320. What would make
>TC's so much more?
> Also - is it worth the extra $ to go with PCI over VL-Bus?
1. Before you assume that all motherboards are equal...is the onboard
RAM cache the same? you need at least 250 KB for a fast machine, and
500 is good. Do all the other features line up ie. upgradability etc.
Are you interested in the Pentium upgrade at some future date? The
last question leads into the answer to Question 2.
2. If you sell your computer when you take the step to a new CPU bus
(upgrade monitor, vid card, HD controller, etc), then you should get a
mobo like the cards you have ie. VLB. If you are upgrading the cards
also, then get PCI, and when you upgrade to pentium, you can swap the
CPU and keep the cards. Look into the Nexgen P5 mobo as an upgrade
option.
In short, determine your budget, then go shopping for the best set of
options you can find.
good luck...
Dave Schreiber dsch...@ansc1.cag.uconn.edu
Animal Science Dept
University of Connecticut
------------------------------
Date: Thu, 11 May 1995 11:01:18 +0200
From: Intact - Petrofina <fi...@wise.nl>
Subject: NFS for Windows
Does anyone know if there's any decent non-commercial NFS client and/or
server software for Windows for Workgroups?
I tried the Tropic NFS server which is pretty OK but it's a normal
program (not a VxD driver) so it has the same priority as all the other
Win programs and so if anyone is doing a lot of disk access, all
applications slowly grind to a halt.
I also tried the XFS32 client but it's very kludgy (only way to mount
drives is through an extra menu in the File Manager, mounting drives
via the usual "network" buttons in the common file dialogs is not
possible) and it uses so much resources that I lose the connection to
two network drives mounted via IPX/NetBios/NetBeui.
The purpose of all this is that I want to be able to use the Sun server
for general data storage from our WFW network.
I know there's PC-NFS (by Sun) but it's very expensive (we'd need about
10 copies of about $200 each) and we have an old version which is hard
to install, uses a lot of disk space and is too complicated for our
mere mortal users.
Thanks in advance...
Jac Goudsmit
CD-i/CD-ROM Software Engineer
Codim Interactive Media CV
Eindhoven, The Netherlands
j...@wise.nl
------------------------------
Date: Thu, 11 May 1995 06:39:33 -0500
From: Gloria S Martin <gl...@tenet.edu>
Subject: Re: NFS for Windows
There is software available that may fit your needs. It's called
"First Class". It's owned by midloth!coach_p...@news.onramp.net.
Good luck.
------------------------------
*********************************
End of Info-IBMPC Digest V95 #101