Searching \ for '[OT] HD Question' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=question
Search entire site for: 'HD Question'.

Exact match. Not showing close matches.
PICList Thread
'[OT] HD Question'
1998\07\06@144005 by Andy Kunz

flavicon
face
Anybody have a utility to completely wipe an IDE hard drive?  My boss has a
totally trashed partition table and cannot delete the partitions with FDISK.

Thanks.

Andy

==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================

1998\07\06@145625 by Sylvain Bilanger

picon face
>From:         Andy Kunz <spam_OUTmtdesignTakeThisOuTspamFAST.NET>
>Subject:      [OT] HD Question
>To:           .....PICLISTKILLspamspam@spam@MITVMA.MIT.EDU
>
>Anybody have a utility to completely wipe an IDE hard drive?  My boss
has a
>totally trashed partition table and cannot delete the partitions with
FDISK.
>
>Thanks.
>
>Andy

Hi, I think the only thing left if fdisk does'nt work is to low level
format you're HD. You can maybe try Partition Magic, but i dont it will
work. Low level format would be the solution.

Sylvain


______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com

1998\07\06@150546 by Mike Massen

flavicon
face
At 01:02 PM 6/7/98 -0400, you wrote:
>Anybody have a utility to completely wipe an IDE hard drive?  My boss has a
>totally trashed partition table and cannot delete the partitions with FDISK.

Try:-

FDISK/MBR

This is an undocumented feature of FDISK...


Rgds ~`:o)

Mike
Perth, Western Australia
http://www.wantree.com.au/~erazmus

Some say there is no magic but, all things begin with thought then it becomes
academic, then some poor slob works out a practical way to implement all that
theory, this is called Engineering - for most people another form of magic.

1998\07\06@155835 by Mark Willis

flavicon
face
What brand of Hard Drive?

 I ask because there are tools at some companies' web sites that
require their specific hardware be installed on the machine to use the
software.

 Actually it can be present, and need not be THE disk you are trying to
erase/"zero fill".  For example, if you have a Quantum HD on either IDE
channel, as either master OR slave, you can use
http://support.quantum.com/menus/soft_menu.htm and download OnTrack Disk
Manager (9.03 version), then (and this is important!) DO NOT INSTALL THE
PACKAGE! - just run the Utilities, Zero Fill section, on ANY of the
other disks on your machine.

 And you can move the drive to a machine that HAS an applicable HD, as
well.  (I had a Quantum HD that failed, they showed me this one as part
of their, ah, "stringent" RMA process.  Drive failed during zero fill,
as I suspected it would!  Got the drive replaced.)

 I'll start gathering those URLs together, if there's demand, but it's
not too hard to arrange this <G>  I'm pretty sure (but haven't checked)
that the Seagate, Western Digital, etc. versions of DM would have this
utility in there (I never use DM, myself, otherwise;  I just don't TRUST
anything where a 1-bit error can mangle a HD & my data.  My choice.)

 There are other tools (I use QAPlus/FE's verification routines to zap
& test my HD's when I get a new drive) but they're not free ($200ish!) &
I suspect you just want to get this DONE with, not spend money <G>

 Mark Willis, mwillisspamKILLspamnwlink.com

Andy Kunz wrote:
{Quote hidden}

1998\07\06@155841 by Ricardo Seixas
picon face
At 13:02 06/07/98 -0400, you wrote:
>Anybody have a utility to completely wipe an IDE hard drive?  My boss has a
>totally trashed partition table and cannot delete the partitions with FDISK.
>
Andy,

       Some time ago I had the same problem, I've tried a lot of things, but
the only that solved the problem was Linux fdisk.
       Basically you need 2 disks, a bootdisk and an imagedisk.
       You can get both on ftp.cdrom.com/...  <I don't remember where>.
       The disks are from Slackware distribution.
       If you need more help on this e-mail me privately.



Ricardo Seixas
.....rseixasKILLspamspam.....pobox.com


>Thanks.
>
>Andy
>
>==================================================================
>Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
>==================================================================
>
>

1998\07\06@161312 by Timothy D. Gray

flavicon
face
Low level format the drive. with the bios.. any good bios has a low level
format program built into the menu.

On Mon, 6 Jul 1998, Andy Kunz wrote:

{Quote hidden}

1998\07\06@163338 by Bob Blick

face
flavicon
face
Hi Andy,

I would not suggest a low-level format, and if DOS fdisk will not work,
what I have done is use a Linux fdisk.

If you have not installed Linux before, this will not be a fun thing to do
:-(

It can be done from a single floppy if you use a RedHat Linux disk. But
unless you've done it before I would not suggest it, because the install
disk is geared towards installing linux on your disk.

I have had some luck with DOS fdisk when there was a bad partition table,
to set the drive parameters(heads, cylinders, etc) to some random numbers,
and try fdisk again, and once was able to delete the partition table(be
sure to correctly set parameters before creating a partition).

Cheers,
bob

1998\07\06@164955 by Matt Calder

flavicon
face
Bob,
       I seem to recall having an fdisk.exe (different from DOS
fdisk.exe) on my Linux distribution. The reason I say this is because I
thought I partitioned my hard disk prior to installing Linux proper. So
assuming my memory is right, any Linux distribution includes a DOS
executable version of Linux fdisk, which in turn means one ought to be
able to find one on the web. Of course, I could be wrong.

       Matt


On Mon, 6 Jul 1998, Bob Blick wrote:

{Quote hidden}

/*****************************************/
/* Matt Calder, Dept. of Statistics, CSU */
/* http://www.stat.colostate.edu/~calder */
/*****************************************/

1998\07\06@175558 by J.Adams

picon face
Andy,

Try to get a copy of Disk Doctor. Same thing happened to me and I was
able to save 95% of the data....twice.

>Anybody have a utility to completely wipe an IDE hard drive?  My boss has a
>totally trashed partition table and cannot delete the partitions with FDISK.
>
>Thanks.
>
>Andy

John Adams ---------------------------- EraseMEelectronicsspam_OUTspamTakeThisOuTpobox.com
http://pobox.com/~electronics ----------- Check out the H.W.S
Internet Guide to Electronics book as well. Links on my site.

1998\07\06@183716 by Bill Cornutt

flavicon
face
-----Original Message-----
From: Sylvain Bilanger <eldamirspamspam_OUTHOTMAIL.COM>
To: @spam@PICLISTKILLspamspamMITVMA.MIT.EDU <KILLspamPICLISTKILLspamspamMITVMA.MIT.EDU>
Date: Monday, July 06, 1998 11:57 AM
Subject: Re: [OT] HD Question


{Quote hidden}

I ran a low level format for a MFM hard drive on a IDE drive.
The low level format worked.
The IDE drive never worked again.
Something about track information being on the disk?

But that was many years ago in a far off land.
And besides, I hear the computer is dead.

Bill C.    TakeThisOuTbillEraseMEspamspam_OUTcornutt.com

P.S.   Has anyone heard any good cow tipping stories.
I'm getting boored.

1998\07\06@200223 by rberdin

flavicon
face
Andy Kunz wrote:

> Anybody have a utility to completely wipe an IDE hard drive?  My boss has a
> totally trashed partition table and cannot delete the partitions with FDISK.

Most CMOS bios setup has a low level format on their menu.Norton diskedit also c
an manually repair a
sector on the hardisk.
But I wonder why FDISK fails, it should do the job.
Maybe the track zero of your boss' hardisk has a physical damage :(.

Regards,
Reggie

1998\07\06@204418 by Travis Veldkamp

flavicon
face
At 01:02 PM 7/6/98 -0400, you wrote:
>Anybody have a utility to completely wipe an IDE hard drive?  My boss has a
>totally trashed partition table and cannot delete the partitions with FDISK.
>

Try:
ftp://ftp.hddtech.ibm.com/hddtech/utility/ZAP.ZIP


--                              Travis Veldkamp
__ __|            _)            RemoveMEtveldkamspamTakeThisOuTbinary.net
  |  _| _` |\ \ / |(_-<
 _|_| \__,_| \_/ _|___/        Nebraska: Not much to look at,
                               but we sure have a lot of it.

1998\07\06@205257 by Mike Keitz

picon face
On Tue, 7 Jul 1998 07:59:32 +0800 Regulus Berdin <rberdinEraseMEspam.....skyinet.net>
writes:
>Andy Kunz wrote:
>
>> Anybody have a utility to completely wipe an IDE hard drive?  My
>boss has a
>> totally trashed partition table and cannot delete the partitions
>with FDISK.
>
>Most CMOS bios setup has a low level format on their menu.

Nearly all modern drives can't be "low-level formatted" without putting
them in a test jig at the factory.  They have servo pulses written in or
between the sectors which can't be recreated without additional hardware
to sense the position of the heads.  The old drives with stepper motors
didn't need servo pulses to put the heads in the right position for each
track.

Norton
>diskedit also can manually repair a
>sector on the hardisk.
>But I wonder why FDISK fails, it should do the job.

I guess FDISK is too smart (and too dumb) for it's own good, attempting
to deal with a corrupt table instead of just replacing it.  Using a
"write all zeros" utility as someone mentioned should make the drive look
like a brand new one, so FDISK will create the table.


_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]

1998\07\06@224729 by Timothy D. Gray

flavicon
face
Why not a low level format? from all my research you can introduce a
proper interleave and refresh the basic track information, (Yes magnetic
fields to deterioriate so a low level format will refresh those important
tracks.) In fact, low level format's have restored back to useful more
hard drives than any commercial program ever has. so, why not?

On Mon, 6 Jul 1998, Bob Blick wrote:

{Quote hidden}

1998\07\06@225942 by Timothy D. Gray

flavicon
face
Wow! Why did mine 4.5gig purchased 3 days ago low level format correctly?
Segate must make a better drive.
{Quote hidden}

1998\07\07@002300 by Herbert Graf

picon face
>Wow! Why did mine 4.5gig purchased 3 days ago low level format correctly?
>Segate must make a better drive.

   Maybe because it is not a true low level format, most IDE drives cannot
be low level formated out of the factory, and there is no need to as well
since the positioning system is closed system. TTYL

1998\07\07@090225 by Ron Fial

flavicon
face
OK, you guys, cut it out.  The absolute precision required to position the heads
on each cylinder is available only at the factory with a special machine.  The
head servo positioner is mechanically connected to it, and it positions the head
s and writes the information for each Cylinder/track for each side of each platt
er.  This information stays there forever.  The machine that does this is VERY r
igid, typically costs $250,000.  When you write data to the disk, the hard disk
controller always skips over this track I/D and position information. The FDISK
program only makes entries in a table to specify the start cylinder/head positio
ns for each virtual volume on the disk.  Keeping within those boundries, the for
mat that FORMAT performs is an higher level format that is operating system spec
ific, and  causes an MSDOS directory and File Allocation Table (FAT) to be writt
en to specific sectors in a partition.

The cylinder spacing on modern drives is so close that each surface must have se
rvo position information for each individual head.  (Earler generation drives so
metimes had a servo track on one surface of a platter, with a dedicated servo pi
ckup head, which provided information to position the entire set of heads.) Usin
g the servo track, it was possible to low-level format the other surfaces and wr
ite the sector mark headers, which are a specific bit pattern followed by the se
ctor number.  Other (earlier) drives used stepping motor positioners to determin
e cylinder position, and on those all surfaces could be low level formatted (no
servo necessary), however the density (spacing between cylinders) was poor.

Ergo, if you ever get an error that causes the lowest level format to be overwri
tten then your IDE drive is toast.  This can happen by accident or design.  Witn
ess the fiasco with the earlier Western Digital 1.6 Gig hard disks -- they had a
controller bug which screwed up the servo information, and WD replaced the driv
es if you complained to them, but never 'announced' that the drive was a time bo
mb.  (Our company lost two of them the first year, and it was painful).
 Regards,
   Ron Fial
-==========================================
At 11:11 PM 7/6/98 -0400, you wrote:
>>Wow! Why did mine 4.5gig purchased 3 days ago low level format correctly?
>>Segate must make a better drive.
>
>    Maybe because it is not a true low level format, most IDE drives cannot
>be low level formated out of the factory, and there is no need to as well
>since the positioning system is closed system. TTYL
>

1998\07\07@090230 by

flavicon
face
       My Western Digital Caviar drive has been low level formated many
times (by a previous owner) and it has been the most reliable drive I have
ever owned.  This was using the BIOS facility present in most modern PC's.

       Mike Rigby-Jones

> >Wow! Why did mine 4.5gig purchased 3 days ago low level format correctly?
> >Segate must make a better drive.
>
>     Maybe because it is not a true low level format, most IDE drives
> cannot
> be low level formated out of the factory, and there is no need to as well
> since the positioning system is closed system. TTYL
>

1998\07\07@090251 by Morgan Olsson

picon face
...
>since the positioning system is closed system. TTYL

Sorry, what is TTYL ?
/Morgan
/  Morgan Olsson, MORGANS REGLERTEKNIK, SE-277 35 KIVIK, Sweden \
\  EraseMEmrtspaminame.com, ph: +46 (0)414 70741; fax +46 (0)414 70331    /

1998\07\07@090258 by Lee Jones

flavicon
face
> Anybody have a utility to completely wipe an IDE hard drive?
> My boss has a totally trashed partition table and cannot delete
> the partitions with FDISK.

As several people have suggested, get the low level format
utility from the drive manufacturer.  This will either cure
it or show that you have a failed drive.

However, if you're trying to preserve some of the data on
the disk.  I'd recommend the Linux fdisk program.  I think
it's the best (most detail, most powerfull) partition table
editor around.  If you have a Linux boot disk and Linux root
disk image (2 floppy set), you can get Linux up enough to
use it's fdisk without touching your hard disk.

Another handy tool is disksave from the NT Resource Kit.  It
allows you to copy the master boot record (MBR) from the hard
disk to a file (i.e. usually on floppy) or the reverse.  It
will also do the same with the boot sector.  Simple, handy
little DOS program.
                                               Lee Jones

-------------------------------------------------------------------
Jones Computer Communications             RemoveMEleeEraseMEspamEraseMEfrumble.claremont.edu
509 Black Hills Dr, Claremont, CA 91711         voice: 909-621-9008
-------------------------------------------------------------------

1998\07\07@090307 by Caisson

flavicon
face
----------
> Van: Timothy D. Gray <RemoveMEtimgrayspam_OUTspamKILLspamLAMBDANET.COM>
> Aan: RemoveMEPICLISTTakeThisOuTspamspamMITVMA.MIT.EDU
> Onderwerp: Re: [OT] HD Question
> Datum: dinsdag 7 juli 1998 4:57
>
> Wow! Why did mine 4.5gig purchased 3 days ago low level format correctly?
> Segate must make a better drive.

Tell us, what command did you use to 'low level format' your drive ?
Maybe that will shed some light on the matter ...

Greetz,
 Rudy Wieser

1998\07\07@090315 by Caisson

flavicon
face
part 0 801 bytes content-type:application/octet-stream; name="Killmbr.com" (decoded 7bit)

Try this one.  Its the standard 'killer' for the MBR.  Source supplied also
:-)
Feed the KILLMBR.DAT to DEBUG (DEBUG < KILLMBR.DAT) to obtain the .COM
file.  Can be typed fairy quick in DEBUG also.

Greetz,
 Rudy Wieser

Content-Type: application/octet-stream; name="Killmbr.com"
Content-Description: Killmbr.com (MS-DOS-toepassing)
Content-Disposition: attachment; filename="Killmbr.com"

Attachment converted: wonderland:Killmbr.com (????/----) (0000EA41)
Content-Type: application/octet-stream; name="Killmbr.dat"
Content-Transfer-Encoding: 7bit
Content-Description: Killmbr.dat (DAT -bestand)
Content-Disposition: attachment; filename="Killmbr.dat"

Attachment converted: wonderland:Killmbr.dat (????/----) (0000EA42)

1998\07\07@090323 by paulb

flavicon
face
Herbert Graf wrote, himself quoting Timothy Gray:

>> Wow! Why did mine 4.5gig purchased 3 days ago low level format
>> correctly?
>> Segate must make a better drive.

>  Maybe because it is not a true low level format, most IDE drives
> cannot be low level formated out of the factory, and there is no need
> to as well since the positioning system is closed system.

 Obviously Seagate *does* make a better drive, as it's sufficiently
"idiot-proof" to answer a "low-level format" command with "Yes, it's
done" as indeed it has been; for once and for all.  The drive is smart
enough to know it never needs low-level formatting.

 Actually IDE drives, especially those with cache, are far smarter than
most users know.  Very occasionally you *may* notice the computer "hang"
temporarily whilst the drive makes a series of "gear change" noises.
What it is *actually* doing is varying its running speed up and down,
making a "last ditch" attempt to recover faulty data.

 AFAIK, its gymnastics include re-reading to obtain data from a crook
sector, re-writing the data if found, test-writing the sector and if
necessary, reading the *complete* track into cache before automatically
low-level formatting that entire track and writing all the data back.
(THat's why it needs teh cache to do this!)  It quite possibly moves
sectors to miss out "hard" errors.  In effect, a "smart" IDE drive has
SPINRITE built-in.

 A further very good reason for the drive to ignore a "format track"
command is that most IDE drives *lie* about their structure, so an
external format command *cannot* be executed because it's always
*wrong*!

 All of which is a bit long-winded.  Andy never really wanted to do a
low-level format in the first place.  Norton Disk Editor will allow you
to manually rebuild the partition table if you knew what it looked like
before.  If you didn't, you can probably patch the first partition to
be readable so the disk can be salvaged, and with enough work, find the
other partitions too.

 There should be - I'm sure there are, as they are so valuable -
programs to do this auto-magically.  FDISK /MBR will recover the
remainder of the boot sector unless a "boot manager" was used (probably
a good reason *not* to use boot managers) and if this combination
doesn't work, a new disk is called for.

 Last time (quite recently) I thought for a while that my boot sector
had been trashed, it wasn't.  The machine crashed with a funny
"ping" noise and thereafter hung just after the "Verifying DPMI pool
data" message, leading me to conclude it was failing to load the boot
sector.  Further investigation, using an old DOS drive which started
loading and *then* hung, revealed that it was in fact the external cache
memory that had died - the machine worked up to the point that the
internal cache was filled, but no more!  Check it out.
--
 Cheers,
       Paul B.

1998\07\07@091242 by Dr. Imre Bartfai

flavicon
face
On Mon, 6 Jul 1998, Herbert Graf wrote:

> >Wow! Why did mine 4.5gig purchased 3 days ago low level format correctly?
> >Segate must make a better drive.
>
>     Maybe because it is not a true low level format, most IDE drives cannot
> be low level formated out of the factory, and there is no need to as well
> since the positioning system is closed system. TTYL
>
>
Furthermore, they do a "fake" low level formatting, i. e. one can believe
it is done. But it isn't.

Imre

1998\07\07@091245 by Dr. Imre Bartfai

flavicon
face
Hi,
try with the Linux Fdisk.
On the other hand, the Norton Utilities can help you. I use the very
old 4.5 and it does all the stuff I want without annoying questions.

Imre


On Mon, 6 Jul 1998, Andy Kunz wrote:

{Quote hidden}

1998\07\07@121110 by Andy Kunz

flavicon
face
A "low level format" from the BIOS did the trick.  Thanks all.

Andy


==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================

1998\07\07@123635 by Timothy D. Gray

flavicon
face
> Tell us, what command did you use to 'low level format' your drive ?
> Maybe that will shed some light on the matter ...
>
Under my bios - choose low level format, chose interleave 1:1 press ok for
format.

1998\07\07@123723 by Timothy D. Gray

flavicon
face
>   Obviously Seagate *does* make a better drive, as it's sufficiently
> "idiot-proof" to answer a "low-level format" command with "Yes, it's
> done" as indeed it has been; for once and for all.  The drive is smart
> enough to know it never needs low-level formatting.

and they have a routene that makes the hd light blink alot, make the drive
generate track clicking noises and take approximately 15 minutes during
this low level format while counting through the heads and sectors.
Those smart guys!

1998\07\07@132905 by Sean Breheny

face picon face
On Tue, 7 Jul 1998, Timothy D. Gray wrote:

> >   Obviously Seagate *does* make a better drive, as it's sufficiently
> > "idiot-proof" to answer a "low-level format" command with "Yes, it's
> > done" as indeed it has been; for once and for all.  The drive is smart
> > enough to know it never needs low-level formatting.
>
> and they have a routene that makes the hd light blink alot, make the drive
> generate track clicking noises and take approximately 15 minutes during
> this low level format while counting through the heads and sectors.
> Those smart guys!
>

My guess is that it writes all zeros to the data area instead of really
doing a low-level format, since writing zeros will accomplish the closest
thing to a low-level format that the IDE can take.

Sean

1998\07\07@141106 by William Chops Westfield

face picon face
Well, *I* don't have any problems believing that there is something
lower-level than dos utilities can manage while still being above the
"$250K special fixture for writing servo tracks" level of "formatting."

BillW

1998\07\07@143632 by Peter L. Peres

picon face
On Tue, 7 Jul 1998, Sean Breheny wrote:

> On Tue, 7 Jul 1998, Timothy D. Gray wrote:
>
> > >   Obviously Seagate *does* make a better drive, as it's sufficiently
> > > "idiot-proof" to answer a "low-level format" command with "Yes, it's
> > > done" as indeed it has been; for once and for all.  The drive is smart
> > > enough to know it never needs low-level formatting.
> >
> > and they have a routene that makes the hd light blink alot, make the drive
> > generate track clicking noises and take approximately 15 minutes during
> > this low level format while counting through the heads and sectors.
> > Those smart guys!
> >
>
> My guess is that it writes all zeros to the data area instead of really
> doing a low-level format, since writing zeros will accomplish the closest
> thing to a low-level format that the IDE can take.

I think that the drive bothers and annotates/moves bad cylinders and maybe
rewrites the index and sector marks and so on in the process and also
scrubs the tracks by doing several writes and detecting flaky bits (better
than a user-level disk utility does). This can restore the sustained
transfer rate to specs, which is vital if mastering CDROMs from it, or
using it for video recording (you don't want hiccups in the data in this
case).

Re: pre-written cylinder servo tracks, I think that the technique is not
used anymore, but I may be wrong. I have heard about a method that allows
the drive to do the whole thing alone (no help from jigs). It parks in
track 0 (mechanical stop), and then starts writing tracks. After writing
the 1st track, it moves the heads inwards until it can no longer read
data, then writes another track, checks the 1st one, then moves over the
2nd track and further inwards until can't read data, writes the 3rd track,
back over the 2nd to check etc. The head geometry dictates the distance
between cylinders, and the drive gets to 'make' a format that best matches
the mechanics. Large cylinder jumps are handled by 'learning' the voice
coil A/D converter setting for max. stroke and scaling (this is the
shuffling sound at boot probably).

This is certainly done at the factory whicle connected to a special jig
(electronic) but it MIGHT be done @ home in some cases. It would also
explain why there are some drives that look identical and have very small
capacity variations between them. I think that it works like with chips:
You get sorted classes after this preformatting, good ones go for higher
capacity marking/controller presetting, worse ones are labelled for lower
capacity... e.g. 2.1 and 2.3 GB drives have too little difference between
them to justify a different head/controller/platter setup.

Which implies, that the highest capacity drive from a certain range is
also the one with the best quality subystems in it ;)

Peter

1998\07\07@143642 by William Chops Westfield

face picon face
Ok, let's get serious.  It's been an awful long time since I've looked at
the details of disk formatting, but my understanding goes like this:

1) There are bit patterns in the media (usually magnetic, but sometimes
  optical even for magnetic media) that delineate the sectors.
2) in addition to that, there is additional data in each "hard" sector
  that is software-related - sector id and other info.  This data accounts
  for the difference between the "unfomatted" and "formatted" capacity of
  a disk (ie 1.44M on a "2M" stiffy.)
3) There's the actual data in the sectors, including low-level filesystem
  info like partition tables and such.  All the "logical disk" info actually
  happens in the SW "data" portions of the disk (ie "partitions" are not a
  HW entity, or even SW independent.)
4) There's the file system.

This explains things like the hard-sectored floppies (remember them!) that
still required "formatting", and all the "large system" disk packs that had
whole separate servo surfaces (written at the factory) but still had lengthy
"format" operations that would cleanse your disks.

Each level wipes out all the data from ALL the higher numbers, of course.

So, DOS/Windows "format" deals with 4 (except on floppies), "fdisk" deals
with 3, bios "low level format" deals with 2, and the fancy fixtures are
only necessary for 1.

BillW

1998\07\07@152151 by Sean Breheny

face picon face
Hi Morgan,

I think that it is Talk To You Later

Sean


On Tue, 7 Jul 1998, Morgan Olsson wrote:

> ...
> >since the positioning system is closed system. TTYL
>
> Sorry, what is TTYL ?
> /Morgan
> /  Morgan Olsson, MORGANS REGLERTEKNIK, SE-277 35 KIVIK, Sweden \
> \  EraseMEmrtspamspamspamBeGoneiname.com, ph: +46 (0)414 70741; fax +46 (0)414 70331    /
>

1998\07\08@032153 by paulb

flavicon
face
William Chops Westfield wrote:

> So, DOS/Windows "format" deals with 4 (except on floppies), "fdisk"
> deals with 3, bios "low level format" deals with 2, and the fancy
> fixtures are only necessary for 1.

 Mmm, no, not quite.  Your 1) is not the bit patterns in the media, but
the track patterns complete with servo control data.

 A true low-level format does 2), but is neither available on IDE
drives nor possible since the drive *itself* manages the low-level
format *as required* ("on-the-fly") transparently to the BIOS.  It has
to as the heads, sectors and tracks aren't really the ones it tells the
BIOS it is using (called "geometry tranlsation" or similar).  You see
you *cannot* low-level format only *part* of a track.

 A BIOS routine may issue a "format track" command but the drive will
sensibly ignore it since the drive looks after that itself.  If however
the BIOS proceeds to "verify" that track, the track *will* indeed be
read, and reported as perfect (since if it was not, the drive itself
would have already repaired it).

 FDISK relates only to hard drives, and writes the partition table,
which also contains the MBR.  The MBR uses the partition table to map a
"virtual drive" into the partition, and proceeds to load the boot sector
exactly as the MBR was itself loaded.  FDISK only (used to) check that
it can find enough real sectors to build the partition.

 The DOS( Win) FORMAT command performs a variable degree of checking
for bad sectors - later versions will accept an already existing file
system if it passes a few basic checks, otherwise they go looking for
bad sectors.  But on an IDE "smart" drive, they rarely find any!

 It certainly does perform a low level format *if required* on a
floppy.

(Reminder:)
> 1) There are bit patterns in the media
> 2) data ... is software-related - sector id and other info.  This data
> accounts for the difference between the "unfomatted" and "formatted"
> capacity of a disk (ie 1.44M on a "2M" stiffy.)
> 3) low-level filesystem info like partition tables and such.
> 4) file system.

--
 Cheers,
       Paul B.

1998\07\09@140833 by scott.2

flavicon
face
Ooo, ooo.  I know an answer to one....

Andy:

Use DOS debug:

----------------------------------------------------------------
     Debug        Enter Debug       Comments
     Prompts      Commands
  ----------------------------------------------------------------
     -            A 100             Assemble from CS:0100.

     nnnn:0100    INT 13            Call interrupt 13.

     nnnn:0102    press the         (nnnn in the segment address).
                  ENTER key

     -            RAX               Replace AX register.

     AX 0000

     :            0301              Write on sector.

     -            RBX               Replace BX register.

     BX 0000

     :            0200              Start from ES:200.

     -            F 200 L 200 0     We want to write zeros.

     -            RCX               Replace CX register.

     CX 0000

     :            0001              Cylinder, 0, sector 1.

     -            RDX               Replace DX register.

     DX 0000

     :            0080              First physical hard disk, head
                                    0. (Substitute 0081 for this
                                    entry if you are clearing the
                                    table on the second physical
                                    hard disk, 0082 if you are
                                    clearing the third physical
                                    hard disk, and so forth).

     -            P                 Proceed (Debug will display
                                    several lines of information).

     -            Q                 Quit Debug.


Carefull, Carefull.  Use at your own risk.. This came from microsoft and
I've used it to delete non-dos partitions and DOS.

Scott

1998\07\13@013800 by Jorge Ferreira

flavicon
face
       Hi

       If you want to erase it, you can use low-level format. A utilitty for th
is
can be found on most modern BIOS (you access it via the SETUP procedure).
       If you weant to try toi restore it, maybe you can use a partition tabble
editor. In such situations I use the instalation disks (2x 1.44FDD) of the
Linux OS. Its free and have a very good partition editor, in the sense that
you can even define the partition id flag to anything you want and not be
limited to a list of more or less known OSs or worse, be limited to create
or erase partitions of the OS that you are using like with MS-DOS FDISK.

       Note the its not necessary to perform a Linux setup, you just need to bo
ot
with the setup disk and use the partition editor to correct your partition
tabble and them abort the instalation.



At 13:02 98.07.06 -0400, you wrote:
{Quote hidden}

===============================================================
cumprimentos / best regards
     Jorge Ferreira          //RemoveMEjorgegfKILLspamspammail.telepac.pt
------ Make sure brain is in gear before engaging mouth -------
===============================================================

1998\07\13@052455 by Eric Smith

flavicon
face
Sean Breheny <shb7STOPspamspamspam_OUTCORNELL.EDU> wrote:
> My guess is that it writes all zeros to the data area instead of really
> doing a low-level format, since writing zeros will accomplish the closest
> thing to a low-level format that the IDE can take.

My guess is that they are actually doing a low-level format.  Everyone has
been claiming that IDE drives *can't* do this because of the embedded servo,
but that's BS.  Embedded servo is used on nearly all disk drives now, whether
they have IDE, SCSI, IPI, HIPPI, Fiberchannel, or some other interface, and
almost all of them can be low-level formatted.  The low-level format procedure
simply avoids writing into the servo areas between the sectors.  That's not
a particularly difficult or demanding thing to do.

Note that this is different from writing zeros to all sectors, because it
writes the ID fields as well.  Thus it can potentially recover from some
problems that simply writing zeros will not fix.  Although these days some
drives don't have ID fields.

It is true that *some* IDE drives don't support low level formatting.  But
I've had no problems with low-level formatting Maxtor and Seagate drives
using the utilities provided by the manufacturers.

Of course, if your servo information gets trashed, a low-level format won't
fix it.  I've only experienced that on one drive, and I'm not sure how it
happened.

[ObPIC:]
I've actually had occassion to interface an IDE drive to a PIC.  It's fairly
easy, but unfortunately needs way too many I/O port pins, mostly for the
16-bit data bus.

I'm using it for data logging.  It originally stored 4G of data, but I
recently upgraded it to 11.5G.  :-)

I've been considering switching to a parallel-interface ZIP drive.  It will
only store 100M of data, but it's easier to eject and swap the disk than to
power cycle the whole system to swap out the IDE drive.  And talking to the
parallel ZIP doesn't need as many I/O port pins.

Cheers,
Eric

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...