The following message is another opinion on the longevity of CD's, posted to
the photoconservation list, that is very informative and less alarmist than
the news report I posted previously to the list.
Iona McCraith
AAO Preservation Consultant

----- Original Message -----
From: <[log in to unmask]>
To: <[log in to unmask]>
Sent: Thursday, May 06, 2004 10:33 AM
Subject: Re: [photoconservation] Permanent retention of scans, data, etc.


> I think we need some clarification about CD-R lifetime and CD lifetime in
> general.  True that most of the hard data out there is manufacturer
> produced and some claims are for 100 to 200 years of life.  I think these
> are high.  However, claiming 2.5 years life for a quality product is not
> accurate.
>
> I have performed stability research and tested many hundreds of CDs over
> the past 6 years.  I have CD-Rs samples that analyze the same as they did
6
> years ago.  Read-only CDs have been around since the early 80s and most
are
> still playable.  For sure those poor quality ones with cheap metal layers
> and poor quality lacquer layers have suffered, but we know this was a
> weakness back then and can be a weakness nowadays with a poor quality
> product.
>
> I believe that if you stick with the proper combination of variables you
> can produce CD-Rs that will outlast the technology to play them on.  How
do
> you achieve this?
>
> Choose brand name products.  It is true that in most cases you get what
you
> pay for.  Sticking with a good quality manufacturer lessens the risk of
> getting a poor quality product.  Although some frown on this
> recommendation, it is a good way of increasing your odds of getting a
> reliable product.  I have definitely seen this trend in most of the discs
> that I have tested.
>
> Choose discs with the phthalocyanine dye and gold metal layer.  With a
gold
> metal layer, laser rot is not a problem and the phthalocyanine dye is much
> more stable (light and heat) than cyanine or azo dye.
>
> Choose discs with extra tough top layers in order to prevent penetration
of
> moisture and pollutants to the metal layer and to provide more durable
> physical protection than normal lacquer layers.
>
> Discs need to be recorded with a low error rate, otherwise they can fail
> rather quickly and/or will not play well in different players.  To achieve
> this, record at around 4x (or at least 4x to 10x).  My tests have shown
> that 4x seems to work best for many discs.  10x is not bad.  Other speeds
> like 1x or 48x just don't produce good discs.  You also need to ensure
that
> your disc and recorder are compatible.  If they are then you will record
> discs with low error rates.  Before embarking on a large project have some
> sample recorded discs tested to ensure the error rate is low.
>
> Sticking with the above recommendations should produce media that will
> outlast its technology and provide several decades of life.  Ignoring
these
> recommendations leads to the possibility of discs failing within 5 years
> and these are the discs that the media and others are all excited about
and
> giving CD-Rs a made reputation.
>
> Joe Iraci
> Senior Conservation Scientist
> Canadian Conservation Institute