8 Oct
2024
8 Oct
'24
3:10 p.m.
Hello Tom.
We need a lot more of this type of preservation.
I've been doing my part and built an imagedisk machine with an Adaptec 1522CF
controller which allows me to read single density.
I've started specializing in systems with 8" drives.
Imagedisk along with 22DISK to read and copy CP/M files, and anadisk to dive into weird
formats.
All running DOS 5 via SD card on a machine that use to run Windows 95.
In my experience, old floppies read amazingly well after 45 years.
I've recovered almost all of mine and many others.
I run my SA850 open so I can see the head surface.
They are easy to clean and not near as delicate as we've always heard.
I worked on these in the 70s, then 5.25" drives at Percom in the early 80s.
Also in my experience, drives that have not been goofed with retain good alignment so
don't rush off to twist those stepper motors.
Reading SSSD 8" 28x128 77 track disks is fairly easy.
It's the crazy m2mfm formats from intel and DEC RX02 that are troublesome - I have
systems for those.
RogerArrick.com/computers
-- Roger Arrick -- Tyler, Texas, USA -- Roger(a)Arrick.com --
________________________________
From: Tom Stepleton via cctalk <cctalk(a)classiccmp.org>
Sent: Tuesday, October 8, 2024 3:45 PM
To: General Discussion: On-Topic and Off-Topic Posts <cctalk(a)classiccmp.org>
Cc: Tom Stepleton <stepleton(a)gmail.com>
Subject: [cctalk] Valuable floppy archiving: seeking current best practice
Hi folks,
I have a floppy disk data archiving project to undertake, and although I'm
aware that this can sometimes lead to spirited discussion (and hope to
avoid that!), I'm interested in current good practices for pulling data off
of hard-to-replace disks.
In this situation, the disks are 8" floppies likely in ordinary IBM
26-sector, 77-track, 128-bytes/sector, double-sided FM format. I have a
flux reader and will have a pair of Shugart 851 drives for the job; these
likely haven't been used for a while, though. At this stage I'll consider
the job done if I manage to get good low-level recordings from the disks:
assuming the FM data decodes well and sector CRCs look good, further
analysis can come later.
The main risk I'm worried about is physical damage to the media. While
reportedly the disks don't show visible defects (nb: they belong to someone
else so I can't inspect them myself right now), I'm still anxious about any
chance I might find the binder that secures the magnetic material to the
cookie degraded to the point of allowing the oxide to come free. Most other
situations I think I can deal with, but I'd like to have a more concrete
plan if I start to find oxide building up on the heads.
I'd be interested to know what precautions people might take for common
data recovery problems. One option is cyclomethicone for cleaning and
lubrication if necessary, but other than "you could use this", I'd be
interested to know details of how people put it to use if they are worried
about media failure. Meanwhile, I'm disinclined to use isopropanol or other
more aggressive solvents given received wisdom, although I know opinions
differ here.
The disks are boot media and other materials relating to the RSRE Flex
operating system <https://en.wikipedia.org/wiki/Flex_machine> as developed
for PERQ workstations. I'm not aware of other copies of this OS being
available, though it would relieve some of the pressure to learn that these
weren't the only ones. The disks themselves are primarily ICL-branded
although a few indicate manufacture or resale by Maxell, DEC, Inmac, among
other brands.
Is there anything else that people would advise me to look out for?
I did find this thread
<https://classiccmp.org/pipermail/cctalk/2018-July/040673.html>, but
practices may have advanced in the past six years. I know the whiteboard
cleaner that folks seem to have liked is difficult to find these days,
particularly here in Britain. I've also had a hard time finding Photo Flo
or similar photographic wetting agents and have used deionised water with a
drop of dish soap instead.
Thanks for any advice,
--T