Date: Tue, 1 Nov 2016 06:39:04 -0400
From: Peter Cetinski <pete at pski.net>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Cc: "General Discussion: On-Topic Posts" <cctech at classiccmp.org>
Subject: Re: Imaging Old Disks Advice Needed
> Well, I made a number of these this weekend and I just left the end
open. The hub
> keeps the cookie in place. Thanks to everyone for the input.
>
> So, I was able to image half of the disks without issue. The others all
had a few
> bad tracks. On most of those I don't see any physical damage so I was
wondering
> if there were any other techniques to possibly recover those tracks?
Baking the
> cookie? Is there a good tool to merge tracks from multiple disks if I
find another
> copy of the software that has the missing tracks?
Pete:
I've had pretty good luck during a data restoration project by ensuring the
drive heads are clean and then also cleaning the cookie VERY GENTLY with
good quality rubbing alcohol (I'm using 91%, some say you should use better
stuff) on a cotton ball. My method has been to put the cookie on a clean,
dry, soft surface. The disks I'm working on are Apple ][, so flippy (if
the jacket is punched), otherwise I only need to concentrate on the
"bottom" side of the cookie. Essentially, the side without the hub ring.
I use a very fine-tipped needle top bottle (found it in the baking section
of a local big-name craft store) and apply a reasonable amount of rubbing
alcohol then use a fresh cotton swab and gently clean the surface.
Depending on how it goes, I'll clean it a second time.
Note, some of the disks look just fine, but still have some issue reading.
This cleaning process has been highly successful (maybe 85%?) for me even
with disks that looked clean.
Also, on a few prior attempts I've had inexplicable results in trying
different drives. I don't know if alignments, speeds, magnetic sensitivity
or other factors were at play but it's worth trying as another trick to
have in the collection. Sometimes it worked better, sometimes it was worse.
I am not 100% sure, but I thought the "baking" was for media that was
shedding or at some risk of having a physical issue with delamination, etc.
Also, I can't really help on the merging question. I'd think you could
just cut the resulting images together using a hex editor or similar.
Naturally, you'd have to know where the bad parts were in the file. Maybe
you could start with a 'fc /b filename.bin filename.bin' at a Windows
command prompt, note the offsets of differences between multiple reads and
visually compare those sections in a hex editor?
Good luck with the data restoration! See ya.
-Todd
> From: Philipp Hachtmann
> that one posting sounded a lot like that, sorry.
OK.
> Do you have a source where there are still 30k chips sitting and
> waiting?
It was ~30K a couple of months ago. I checked about a week ago, and it was
down to ~26K (IIRC).
Although, like I said, I doubt they have all 26K in stock themselves; based on
comments they made when we bought a large group, I think that's the total
number available to them across a number of suppliers, in a network which
shares inventory information.
Noel
> From: Don North <north at alum.mit.edu>
> .. the hardware bootstrap reads track 1 sectors 1, 3, 5, 7
Ah, thanks for that. Starting to look at the code, I had missed the
interleave.
So does DEC do anything with track 0, or is it always just empty?
Noel
