[Vox-Outreach] Testing of CD Images

[Peter Jay Salzman]

On Mon 12 May 03,  4:47 PM, Mike Simons <msimons@moria.simons-clan.com> opined:
> [perhaps we should move this to vox-tech]
> 
> On Mon, May 12, 2003 at 01:31:45PM -0700, Peter Jay Salzman wrote:
> > point of reference - the only time i've *ever* burned coasters is when i
> > played a game like quake3 while writing a cd.  other than that, i've had
> > a 0% failure rate.
> 
> > On Mon 12 May 03,  4:25 PM, Mike Simons <msimons@moria.simons-clan.com> opined:
> > >   I would prefer the disks be verified before the event...
> > 
> > you don't even have to verify it.  just:
> > 
> >    mount /cdrom; umount /cdrom
> > 
> > and getting a prompt back quickly is enough.
> 
> - Could you explain how getting a prompt back quickly means that the 
>   disk image is correctly burned?

sure, it's easy.

i found that the vast majority of burns are either all or nothing.  if
the burn failed, it's *very* rare the cd will mount.

is it 100% foolproof?  no.  nothing is.

however, i have extensive experience with my own and other peoples' cd
writers, and when i say "vast majority", i mean exactly that: "vast
majority".

in fact, i believe the only time i've seen anything to the contrary of
this, it was with cases of extreme overburning.

and if we're talking about 20 or even 50 burns, using something like
md5sum is just fscking obnoxious.

pete

-- 
GPG Instructions: http://www.dirac.org/linux/gpg
GPG Fingerprint: B9F1 6CF3 47C4 7CD8 D33E 70A9 A3B9 1945 67EA 951D