News Group Posts

floppy drive fifo

From: bill@wjv.com (Bill Vermillion)
Subject: Re: We can't read from floppy on Pentium II, SCO Unix rel. 5.0.0b and 5.0.2
References: <Pine.SCO.3.96.1010521031529.19524A-100000@pp5.progplus.com> 
Date: Mon, 21 May 2001 10:32:12 GMT

In article <Pine.SCO.3.96.1010521031529.19524A-100000@pp5.progplus.com>,
Robert Weiner  <robert@progplus.com> wrote:
>On Sun, 20 May 2001, Bela Lubkin wrote:
>> On 1999-12-26, Robert Weiner wrote:

>> That was later fixed.  If Robert was having I/O errors with his
>> 5.25" floppy on OpenServer 5.0.5, it was probably because the drive
>> hardware was defective.

>Or motherboard issue?  Keep reading... 
...

>Still not the case. The same not-functioning 5.25" drive worked
>fine on different machines with different motherboards. ...

>Over the years I've learned the following: 

>a) Tyan changed the documentation for the S1832DL. The board by
>default ships with the 2nd floppy disabled (and the infrared port
>enabled). The jumper settings in older documentation were declared
>WRONG, meaning that if you followed the instructions, you did NOT
>enable the second floppy. The general rule of thumb to get a 2nd
>drive working is to change JP4/JP8/JP9 appropriately and set the
>correct floppy type for the 2nd floppy in the bios. I never did
>get a 5.25" floppy working correctly with SCO505 and an older
>S1832DL, but I did get it working with a newer S1832DL and WinNT.
>I'm not sure if I ever tried the 5.25" floppy with SCO505 in a
>newer S1832DL board.

Thansk for that research and information.

>c) I remember that under WinNT4, the 5.25" drive tended to function
>(at least from SP5 thru SP6a). The big test was performing a
>format. Any time I could format then write then read I knew things
>were ok. I remember that under SCO505, I could read, sometimes
>write, but could NEVER get a format to complete successfully.

That is not a good test.  MS systems will format a disk and will
mark any bad sectors and make them unavailable.   You will see this
in the format summary.  Unix systems OTOH most often write raw data
to a disk and thus require disks with NO flaws at all.  I had one
early Unix system that did not even retry on a write attempt during
format. [8" devices].  On some surplus Xerox disks I bought I had
over 90% failure.  Another OS [ LDOS for those who remember it]
was able to get about 50% format.  On CP/M [from which MS derived
so much] I had a 95% success rate.

Part of it had to do with the sloppy way formats were done under
CP/M - using the worst case single density format instead of the
worst case double-density format. The SD worst case format is 'e5'.
You could resurect lost files on MS by simply editing the directory
entry and replaceing the 'e5' as the first letter of an 'erased'
file by changing it back to any legal ascii value, not just it's
original value.  I haven't looked at an MS product at that low a
level in a few years - but they were still using that format then -
a format which will pass almost anything that will fit in the door.

Do not confuse MS 'take two aspirin and call me in the morning'
approach with the Unix "well run some blood tests and and a
cardiogram' method.  The former will do you in at times.

I still routinely format ALL the floppies I use on a Unix system
and then just dd track one [which I've sucked from a known good MS
formatted disk] back on at the end of a format so that I have
a 100% known good formatted disk I can use under Unix or MS.

>d) As a separate issue, I started wondering about the 5.25" media I
>had. Most of it was somewhat aged (3+ yrs) and I had a feeling that
>the oxide simply gave up. While going through various old backups
>and archives stored on 5.25" floppies (5-10 yrs old), I noticed
>that many were no longer readable. I've always wondered if the
>earth's magnetic field and radiations from computer equipment could
>demagnetize old floppies... assuming they were stored in the same
>spot for most of those years.

They really won't demagnetize in proper conditions.  Magnetic
materials are quite stable if not changed by an external influence.
Get media hot and the slightest magnetic filed will change them.

[This is technically called the Curie point.  At that point the
ease of changing the particle orientation is so easy this was used 
to 'contact print' video tapes.  Special recorders were made which
angled the video heads in a mirror image of players.  The master
tape - with a higher curie point - was sandwiched with a tape of a
lower point and run through rollers and heated up - and you had an
instant high speed duplication.  Expensive and duplicator opted to
go with herds of standard VCRs for this] [/topic drift]

However the manufacture of the coatings and the material used
did affect longevity.  In the bad old days there were disks that
you could not read 6 months later.  Brown made a lot of bad media
as I recall [that's not a retail brand but an OEM]. Bad choice of
binders can make the oxide come off on the heads or just come off
the disks.

Heat is a killer.  One hour at 150 degrees will do as much damage
as radiation 500 times higher than a lethal dose. [I had that
reference in some 3M/Scotch literature a long time ago - and I
haven't seen that reference since].  So that means that if you have
a bad fan and leave a floppy in the drive it could get hot enough
to make it easily damaged.  I've pulled warm floppies from systems
in the past when I'd gone to look at a problem.  Good indicator of
dead/bad fan or dust bunnies galore.

In the early days you could also erase a disk by accidentally
setting a phone on top of a disk and have the phone ring. [this was
when phones had bells in them]. I've always heard that but never
proved it.  The HD disks had a much higher coercivity and were
more immune to outside influence.  The 3.5" format had media quite
similar to the 5.25" HD format.  The latter is remarkable rugged.

In the last year I took archived data from 10 year old 3.5 disks
[written on a Xenix system] and moved it to CD.  Out of 200+ disks
I had about 3 or 4 unreadable files.  Never had any disk completely
unreadable.  The 5.25" HD disks are also almost as rugged.

>hard disk or 4mm tape in preparation of the media failing...
>I think I was too late. During my tests of floppy drive vs
>motherboard, I [eventually] took extra care to use floppies which
>still accepted a format.

Test them ONLY on Unix systems if you want to be sure.

>I'm pretty sure I was hit with multiple problems including:
>erroneous motherboard documentation, media oxide issues
>and possibly something that SCO505 does differently than
>SCO500/DOS/WinNT4 which compounded the problem (and you just
>indicated the auto-configure changed so something was different).

Oxide problems are most likely in the 360/720K formats.

>I also have a feeling that there could have been a Pentium II vs
>Pentium III issue on the motherboard which did something with
>the 2nd floppy (no matter how unlikely that would seem). A near
>duplicate machine to mine, at a customer, had a working 5.25" drive
>on a dual P-II/450 system where I ran the same configuration except
>for dual P-III/500s. I built both machines as twins and the only
>difference was in the CPUs. This is still an open issue for me and
>one day I will get around to doing some more testing on it.

I'd be interested in hearing the results of that testing.  We're
the chips of the same family or were they of Celeron or Xeon
variants.

-- 
Bill Vermillion -   bv @ wjv . com

Comments?

Jump to Comments

Many of the products and books I review are things I purchased for my own use. Some were given to me specifically for the purpose of reviewing them. I resell or can earn commissions from the sale of some of these items. Links within these pages may be affiliate links that pay me for referring you to them. That's mostly insignificant amounts of money; whenever it is not I have made my relationship plain. I also may own stock in companies mentioned here. If you have any question, please do feel free to contact me.

Specific links that take you to pages that allow you to purchase the item I reviewed are very likely to pay me a commission. Many of the books I review were given to me by the publishers specifically for the purpose of writing a review. These gifts and referral fees do not affect my opinions; I often give bad reviews anyway.

We use Google third-party advertising companies to serve ads when you visit our website. These companies may use information (not including your name, address, email address, or telephone number) about your visits to this and other websites in order to provide advertisements about goods and services of interest to you. If you would like more information about this practice and to know your choices about not having this information used by these companies, click here.