26 Oct
2006
26 Oct
'06
1:46 a.m.
Hey folks,
Does anyone have any guidelines about making archival images of
CD-ROM media? For years, I've been using dd on Solaris or Linux
to dump the raw bytes of CD-ROMs, and cdrecord if I want to make
a copy from the raw image, but I don't know if this is considered
"good enough" to preserve all the associated filesystem and partition
structures of the disc. It has been for me, so far, but I'd like to
hear other opinions on the matter. How do you archive CD-ROM images?
Does anyone use any special software or have any tips?
-Seth
26 Oct
26 Oct
9:03 a.m.
Seth J. Morabito wrote:
Hey folks,
Does anyone have any guidelines about making archival images of
CD-ROM media? For years, I've been using dd on Solaris or Linux
to dump the raw bytes of CD-ROMs, and cdrecord if I want to make
a copy from the raw image, but I don't know if this is considered
"good enough" to preserve all the associated filesystem and partition
structures of the disc. It has been for me, so far, but I'd like to
hear other opinions on the matter. How do you archive CD-ROM images?
Hmm, won't most tools create an ISO filesystem image from a CD? I would expect
that's good enough in nearly all cases, although possibly not for things like
games consoles where I believe the original media's screwed around with as an
anti-piracy measure.
Personally I find that I only ever use a few percent of the commercial
software on a CD, though - e.g. I'll just install the app but not all the
bundled junk that comes with it. Saving an archive image means you get
everything, warts and all. For a true archive you should save everything, but
if it's really just for personal use I'd be tempted to just save what you
think you'll need.
11:04 a.m.
On Thu, 2006-10-26 08:03:44 -0500, Jules Richardson <julesrichardsonuk at
yahoo.co.uk> wrote:
Seth J. Morabito wrote:
For a pure data CD, an ISO image is probably good enough. (Though you
may want to additionally get hold of all the TOC/ATIP information
that's not inside the ISO image.)
For anything else (CD+G, CDI, ...) you need to understand how a CD
works and at which layer this type of CD is operating.
Especially for "copy-protected" CDs, which may contain premastered
broken sectors, you probably also want to get hold of the C2 level
error bits.
Does anyone have any guidelines about making
archival images of
CD-ROM media? For years, I've been using dd on Solaris or Linux
to dump the raw bytes of CD-ROMs, and cdrecord if I want to make
a copy from the raw image, but I don't know if this is considered
"good enough" to preserve all the associated filesystem and partition
structures of the disc. It has been for me, so far, but I'd like to
hear other opinions on the matter. How do you archive CD-ROM images?
Hmm, won't most tools create an ISO filesystem image from a CD? I would
expect that's good enough in nearly all cases, although possibly not for
things like games consoles where I believe the original media's screwed
around with as an anti-piracy measure. Personally I find that I only ever use a few percent
of the commercial
software on a CD, though - e.g. I'll just install the app but not all the
bundled junk that comes with it. Saving an archive image means you get
everything, warts and all. For a true archive you should save everything,
but if it's really just for personal use I'd be tempted to just save what
you think you'll need.
I started to write some helper library to "easily" access SCSI devices
(to which virtually all CD drives belong, even those that are
connected to an IDE cable for transport.) That one can read images
that contain most of all available information, but that's far from
being production-ready...
MfG, JBG
--
Jan-Benedict Glaw jbglaw at lug-owl.de +49-172-7608481
Signature of: Gib Dein Bestes. Dann ?bertriff Dich selbst!
the second :
12:33 p.m.
Jan-Benedict Glaw wrote:
For a pure data CD, an ISO image is probably good
enough. (Though you
may want to additionally get hold of all the TOC/ATIP information
that's not inside the ISO image.)
OpenVMS CDs and Ultrix CDs are not ISO9660 just for starters so I
think you really do need a dump of the entire CD (via dd or any
of the CD/DVD-writing programs out there). Even if it looks like
an ISO9660 CD that's not enough. I've written CDs that have both
an ODS-2 filesystem and an ISO9660 filesystem (in my case the data
was shared between the two but there could easily have been data
that existed in only one or the other).
For anything else (CD+G, CDI, ...) you need to
understand how a CD
works and at which layer this type of CD is operating.
Especially for "copy-protected" CDs, which may contain premastered
broken sectors, you probably also want to get hold of the C2 level
error bits.
Ideally you would want a dump of the raw CD "bitstream" (I think it
has a proper name but I don't know it) but for most CDs that's far more
than is needed. If the CD device for the target system never made that
much info available you probably don't need it. Even having all that
info won't necessarily be enough to recreate a CD (the PS copy
protection
scheme you mention relies on deliberate error data and I think even
modern
writers don't give you enough control to reproduce such a disc).
Antonio
12:47 p.m.
On Thu, 2006-10-26 17:33:51 +0100, arcarlini at iee.org <arcarlini at iee.org>
wrote:
Unfortunately, you cannot access it at the lowest level with regular
CD drives. (Well, at least not with the documented MMC-5 commands...)
Jan-Benedict Glaw wrote:
Are they multi-session?
For a pure data CD, an ISO image is probably good
enough. (Though you
may want to additionally get hold of all the TOC/ATIP information
that's not inside the ISO image.)
OpenVMS CDs and Ultrix CDs are not ISO9660 just for starters so I
think you really do need a dump of the entire CD (via dd or any
of the CD/DVD-writing programs out there). Even if it looks like
an ISO9660 CD that's not enough. I've written CDs that have both
an ODS-2 filesystem and an ISO9660 filesystem (in my case the data
was shared between the two but there could easily have been data
that existed in only one or the other). For anything
else (CD+G, CDI, ...) you need to understand how a CD
works and at which layer this type of CD is operating.
Especially for "copy-protected" CDs, which may contain premastered
broken sectors, you probably also want to get hold of the C2 level
error bits.
Ideally you would want a dump of the raw CD "bitstream" (I think it
has a proper name but I don't know it) but for most CDs that's far more
than is needed. If the CD device for the target system never made that much info available you probably don't need it.
Even having all that
info won't necessarily be enough to recreate a CD (the PS copy
protection
scheme you mention relies on deliberate error data and I think even
modern
writers don't give you enough control to reproduce such a disc).
These are two distinct problems: Preserving all information and
recreating a new disk from it.
Even while we might not be able to reproduce an errorneous disk with
today's CD burners, we really might want to preserve the information.
Just in case the original CD gets lost. Or think about emulators.
They could be taught to emulate errors. I didn't understand all of the
CD formats (information about these is quite sparse and I don't have
some free money to buy the standards) and I know even less about the
burning process. But I remember that there are some (Windows based)
tools around to burn CDs with errors, so I guess it's possible to some
extent...
MfG, JBG
--
Jan-Benedict Glaw jbglaw at lug-owl.de +49-172-7608481
Signature of: Eine Freie Meinung in einem Freien Kopf
the second : f?r einen Freien Staat voll Freier B?rger.
6481
days inactive
6481
days old
4 comments
4 participants
participants (4)
-
arcarlini＠iee.org -
jbglaw＠lug-owl.de -
julesrichardsonuk＠yahoo.co.uk -
sethm＠loomcom.com