20 Oct
2007
20 Oct
'07
7:52 p.m.
Yep, went through that at the museum because some
people were advocating
putting media on the archive shelves - but it's not an idea I'm a fan off; the
stuff's just too prone to damage and decay.
Unless you recover the data, what you have is a physical artifact of a magnetic
storage medium. There is absolutely no way to say what, in fact, is even on it
until you read it. Bits aren't preserved if they exist on only one physical medium,
which you may not be able to recover in the future.
20 Oct
20 Oct
8:16 p.m.
On 20 Oct 2007 at 19:52, Al Kossow wrote:
Unless you recover the data, what you have is a
physical artifact of a magnetic
storage medium. There is absolutely no way to say what, in fact, is even on it
until you read it. Bits aren't preserved if they exist on only one physical medium,
which you may not be able to recover in the future.
Exactly. I'm starting to see a trend with some brands of floppies
more than 25 years old where the oxide is starting to separate from
the substrate, leading to fouled heads and "see through" tracks.
Mostly on Wabash brand floppies currently, although a few off-brands
such as "Precision" seem to also be showing this behavior.
Time to get 'em archived.
Cheers,
Chuck
21 Oct
21 Oct
2:44 a.m.
Chuck Guzis wrote:
On 20 Oct 2007 at 19:52, Al Kossow wrote:
I'm glad it's not just me. Almost every single Wabash disk I've come across
seems prone to this - if I'm lucky, there's one chance to read the data, and
the drive heads always need a clean afterwards. More often than not, the
surface comes away though.
Other than Wabash, things seem to hold together pretty well, although I think
I recall having troubles with a batch of Parrot disks a while back (although
in that case they were ones that hadn't been stored in very nice conditions).
Unless you recover the data, what you have is a
physical artifact of a magnetic
storage medium. There is absolutely no way to say what, in fact, is even on it
until you read it. Bits aren't preserved if they exist on only one physical medium,
which you may not be able to recover in the future.
Exactly. I'm starting to see a trend with some brands of floppies
more than 25 years old where the oxide is starting to separate from
the substrate, leading to fouled heads and "see through" tracks.
Mostly on Wabash brand floppies currently, although a few off-brands
such as "Precision" seem to also be showing this behavior. Time to get 'em archived.
Indeed. Although I'm almost more worried about ST412 hard drives, a problem
which seems to often get overlooked (the same can be said of other hard disk
technology, such as SMD). The contents are often far more interesting, but a
sensible procedure for archiving them doesn't really exist short of finding
the original hardware with which they belong and somehow transferring data to
a modern system.
cheers
Jules
9:28 a.m.
"Chuck Guzis" wrote:
Mostly on Wabash brand floppies currently, although a
few off-brands
heh. they used to do that when new if you used them in a room over 85F :-)
I once (back when s-100 was current) tried to run several CP/M machines
with 8" floppies in a un-airconditioned room over the summer. big
mistake. We ended up with a lot of floppies with clear bands.
With an A/C running at 75F the problem went away, but some brands did
better than others.
-brad
9:53 a.m.
On 21 Oct 2007 at 12:28, Brad Parker wrote:
heh. they used to do that when new if you used them
in a room over 85F :-)
I once (back when s-100 was current) tried to run several CP/M machines
with 8" floppies in a un-airconditioned room over the summer. big
mistake. We ended up with a lot of floppies with clear bands.
Our storage is a relatively constant 65F at 40-50 percent humidity.
I'm more than a bit concerned that this is just the tip of the
iceberg and other brands will follow suit. Consider the situation,
for example, with old DC-xxx carts of approximately the same age.
It's akin to the situation with very old documents--you might have
handled them a lot when they were newer, but as they get older and
start decomposing, you realize that you'll have to capture the
document's contents and forget about the medium on which they're
inscribed.
Cheers,
Chuck
2:31 a.m.
Al Kossow wrote:
Exactly. There's not much point saying "I have this" when you don't know
if
"this" is still viable. There seems little point in keeping media on shelving
either, except for ease of sorting and identifying things which might be "the
same" (which is somewhat misleading anyway - I'm forever finding media where
the contents bear no resemblance to what's actually written on the label)
Again with museum hat on, one of my worries is not the archival of this stuff,
but the subsequent need to recreate the content - typically there's little or
no redundancy at the bridge between old world and new (i.e. I think we only
have one SCSI drive which can be easily hooked up to a PC in order to read and
write certain classes of disk pack - if it fails then it has implications for
the running of any system requiring those packs).
I think it was probably Dave D. who came up with the idea of some form of
'registry' listing who could handle reading and writing certain types of media
- in light of the above, that seems like a rather good idea...
cheers
Jules
Yep, went
through that at the museum because some people were advocating
putting media on the archive shelves - but it's not an idea I'm a fan
off; the stuff's just too prone to damage and decay.
Unless you recover the data, what you have is a physical artifact of a
magnetic storage medium. There is absolutely no way to say what, in fact,
is even on it until you read it. Bits aren't preserved if they exist on
only one physical medium, which you may not be able to recover in the
future. 
3:21 a.m.
On Sat, 2007-10-20 at 19:52 -0700, Al Kossow wrote:
Just on that note, there's a problem I've been struggling with for a
while. I'm trying to capture disk images for an Ensoniq Mirage sampler,
which uses a slightly odd format - five sectors of 1024 bytes followed
by one sector of 512 bytes.
I can read the disks with piece of software intended for a similar piece
of equipment, I ought to be able to write them but haven't tested this,
but I can't format them.
Now at the moment this isn't a huge problem, because I've got working
formatter disks for the Mirage - you boot up the appropriate disk, stick
a blank in, and it will format it. However, to write to the disk you
need to have a format on it in the first place, so you can't write a
formatter disk without having a formatted disk... Chicken and egg.
Gordon
Yep, went
through that at the museum because some people were advocating
putting media on the archive shelves - but it's not an idea I'm a fan off; the
stuff's just too prone to damage and decay.
Unless you recover the data, what you have is a physical artifact of a magnetic
storage medium. There is absolutely no way to say what, in fact, is even on it
until you read it. Bits aren't preserved if they exist on only one physical medium,
which you may not be able to recover in the future. 
3:26 a.m.
On 10/21/07, Gordon JC Pearce <gordonjcp at gjcp.net> wrote:
Just on that note, there's a problem I've been
struggling with for a
while. I'm trying to capture disk images for an Ensoniq Mirage sampler,
which uses a slightly odd format - five sectors of 1024 bytes followed
by one sector of 512 bytes.
I can read the disks with piece of software intended for a similar piece
of equipment, I ought to be able to write them but haven't tested this,
but I can't format them.
Now at the moment this isn't a huge problem, because I've got working
formatter disks for the Mirage - you boot up the appropriate disk, stick
a blank in, and it will format it. However, to write to the disk you
need to have a format on it in the first place, so you can't write a
formatter disk without having a formatted disk... Chicken and egg.
I'll be curious to hear about any future progress. I, too, have a
Mirage, and a box of sample disks, but I don't know if I have any
utility disks.
-ethan
22 Oct
22 Oct
1:54 a.m.
On Sunday 21 October 2007 11:26:04 Ethan Dicks wrote:
I'll be curious to hear about any future progress. I, too, have a
Mirage, and a box of sample disks, but I don't know if I have any
utility disks.
If you grab the SQ80 utilities from www.buchty.net then you can adapt them to
read the Mirage's single-sided disks (it's just a case of commenting out the
bit that toggles the "side select" and making it not loop over side 0/side
1/) and get usable images. I haven't tried writing them back to formatted
disks yet.
There is an application that runs under DOS that will format SQ80 disks, which
work just fine in the Mirage - it will cheerfully ignore side 1, even though
many of them have DSDD drives - but it's a PITA to find a machine to install
DOS on just for one thing. It must be possible, therefore, to format the
disks on a PC controller. I know the Mirage uses a WD1772 controller.
I suspect that the formatter needs to write some "magic bits" to the directory
block before the Mirage will actually talk to the disk. Using the Geibler
utility I can format a disk but I need to write an image to it before I can
save anything.
Gordon
1:59 a.m.
On Sunday 21 October 2007 11:26:04 Ethan Dicks wrote:
I'll be curious to hear about any future progress. I, too, have a
Mirage, and a box of sample disks, but I don't know if I have any
utility disks.
If you grab the SQ80 utilities from www.buchty.net then you can adapt them to
read the Mirage's single-sided disks (it's just a case of commenting out the
bit that toggles the "side select" and making it not loop over side 0/side
1/) and get usable images. I haven't tried writing them back to formatted
disks yet.
There is an application that runs under DOS that will format SQ80 disks, which
work just fine in the Mirage - it will cheerfully ignore side 1, even though
many of them have DSDD drives - but it's a PITA to find a machine to install
DOS on just for one thing. It must be possible, therefore, to format the
disks on a PC controller. I know the Mirage uses a WD1772 controller.
I suspect that the formatter needs to write some "magic bits" to the directory
block before the Mirage will actually talk to the disk. Using the Geibler
utility I can format a disk but I need to write an image to it before I can
save anything.
Gordon
21 Oct
21 Oct
6:21 a.m.
Gordon JC Pearce wrote:
Now at the moment this isn't a huge problem,
because I've got working
formatter disks for the Mirage - you boot up the appropriate disk, stick
a blank in, and it will format it. However, to write to the disk you
need to have a format on it in the first place, so you can't write a
formatter disk without having a formatted disk... Chicken and egg.
But you could format it, if you would have a bit image of a formatted
disk. (preferably take the bit image of a freshly formatted disk, and
hope for no bad block)
9:14 a.m.
On 21 Oct 2007 at 11:21, Gordon JC Pearce wrote:
Now at the moment this isn't a huge problem,
because I've got working
formatter disks for the Mirage - you boot up the appropriate disk, stick
a blank in, and it will format it. However, to write to the disk you
need to have a format on it in the first place, so you can't write a
formatter disk without having a formatted disk... Chicken and egg.
It can be done with ordinary PC hardware--there are just a few tricks
that need to be employed. Think IBM XDF disks (which could be copied
on a standard PC using the right software).
Or use a WD17xx controller to do it.
Cheers,
Chuck
23 Oct
23 Oct
8 p.m.
New subject: Ericsson Desktop PC
I got another email about an older Ericsson Desktop PC. If anyone is
interested, please contact them directly.
Bob
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Hi, Bob:
I have a very rare (Ericsson manufactured/sold VERY few PCs) vintage PC
and I wondered if you would be interested in it. And I could really use
help determining its value, so ANY suggestions on that would be greatly
appreciated. Some specifics follow. If you need any more information,
just E-Mail me at the address below.
-Ericsson Desktop PC (Ericsson manufactured/sold VERY few PCs),
including the following (all are original items provided with the
original PC):
-Color Monitor, Keyboard, Dual Floppy Tower with stand
-All cables, connectors
-Complete set of SCHEMATICS
-Full set of diskettes (DOS 2.11B)
-Full set of original documentation, each set in pristine (who actually
reads manuals?) condition, in original binders and original cardboard
holders:
-Users Manual and Users Guide
-DOS Manual and Users Guide
-GW Basic Manual and Users Guide
-Original shipping containers/packing material (Yes, I actually have the
original cardboard "Ericsson" boxes, and they are in excellent condition!)
-Mint condition outside
-Fully functional/operational when last used (NO operational problems
experienced)
-Manufactured ~ 1984
-Beige/Putty color
In summary, I have every item that came with the original PC purchase,
with the possible exception of a few pieces of plastic wrap or foam used
for packing.
If you need any other information, please don't hesitate to contact me
at the address below. I'm in the Bay Area.
Thanks,
MCraggs
sunpun at earthlink.net <mailto:sunpun at earthlink.net>
22 Oct
22 Oct
12:12 p.m.
Al Kossow wrote:
Yep, went
through that at the museum because some people were
advocating
putting media on the archive shelves - but
it's not an idea I'm a
fan off; the
stuff's just too prone to damage and decay.
Unless you recover the data, what you have is a physical artifact of a
magnetic
storage medium. There is absolutely no way to say what, in fact, is
even on it
until you read it. Bits aren't preserved if they exist on only one
physical medium,
which you may not be able to recover in the future. from 2002 through 2005. My current system has used a
C:
drive with approximately 2 GB of files for about the past
5 years. GHOST compresses that to an image file of just
under 1 GB which allows me to fit 4 images per DVD or 3 DVDs
per year with one image saved each month.
About once a year, I take a whole day and read all of the old
files back to the hard drive and check for a valid file by
producing and comparing the current MD5 value with the original
MD5 value. Thus far this exercise turned up only one bad file
which could not be read back - at the extreme edge of the DVD -
which had a clear fingerprint at the edge of the DVD media.
I am considering making up a DVD with all of the December 31st
images since after 4 years, over 50 different end of month images
hardly seem very important any more.
In addition, I have all of the RT-11 binary distributions on
a CD and a second CD with the DECUS RT-11 CD from Tim Shoppa.
Every so often, I also read these media as well. The ISO files
from both CDs are on a DVD with other RT-11 archived
files.
Since the CD is over 20 years old and still supported on current
DVD burners, I suspect that the single layer DVD will also be
around for at least another 10 years. At that point, an upgrade
to the current media type may be needed. Since I have used the
following as backup media over the past 30 years, it will not be
unexpected:
RX02 - 1/2 MB (around 1980)
TK25 - 32 MB (around 1985)
TK70 - 256 MB (around 1990)
5 1/4" optical Sony SMO S-501 compatible - 290 MB on each side (around 1995)
CD - 600 MB (around 1995)
DVD - 4 GB (around 2002)
Sincerely yours,
Jerome Fine
--
If you attempted to send a reply and the original e-mail
address has been discontinued due a high volume of junk
e-mail, then the semi-permanent e-mail address can be
obtained by replacing the four characters preceding the
'at' with the four digits of the current year.
6270
days inactive
6273
days old
14 comments
11 participants
participants (11)
-
aek＠bitsavers.org -
bobalan＠sbcglobal.net -
brad＠heeltoe.com -
cclist＠sydex.com -
cisin＠xenosoft.com -
emu＠e-bbes.com -
ethan.dicks＠gmail.com -
gordonjcp＠gjcp.net -
gpearce＠curiousgroup.co.uk -
jhfinedp3k＠compsys.to -
julesrichardsonuk＠yahoo.co.uk