Date: Thu, 21 Jun 2001 14:47:56 -0500
Subject: Re: arsclist cd roms - which are closest to archival quality?
From: Konrad Strauss <konrad59@xxxxxxxx>
Content-type: text/plain; charset="US-ASCII"
on 6/17/01 4:56 AM, M. Sam Cronk at scronk@xxxxxxxxxxx wrote:
> My question: which cd rom "standards" are closest to archival quality?
> PhthaloCyanine gold-reflective surface? Are those with blue-dye
> reflective surface now comparable? We need to make a rather large purchase
> and a good long term decision.
This is a complex question. There are basically 2 components to a CD-R. The
reflective surface, which can be either gold or silver, and the dye
formulation, either cyanine or phthalocyanine.
Silver is less expensive so most CD-Rs use a silver reflective surface.
However silver can tarnish when exposed to air reducing it's reflectivity,
possibly increasing the error rate. Gold will not tarnish, but is more
expensive so the so-called "archival" CDs use a gold reflective surface.
In simplistic terms cyanine dye, which can be green to blue in color, is UV
sensitive, so it can deteriorate when exposed to light, increasing error
rates. Phthalocyanine, which is clear to the naked eye, is not UV sensitive
so theoretically it is more robust. However there's more. Phthalocyanine
needs a tighter laser power tolerance and generally works better at higher
burning speeds, whereas cyanine works within a wider range of laser power
and works better for lower speed burning. But there's even more.
Manufacturers are constantly changing their dye formulations to both reduce
production cost and to optimize it for the higher speed burners. So even
though one particular brand works well for you, the next batch you buy may
not have the same dye formulation and may not work as well.
A third factor is how the top of the CD-R is sealed. The original CD-Rs
simply had a lacquer sealer, like commercial CDs. but this proved to be too
fragile. Most good discs now have a resin sealer, however it is still
possible to get lacquer-sealed CD-Rs - for silkscreen printing. Make sure
the discs you get are resin-sealed.
There's another part to this equation. Your CD burner. Every CD-R has it's
manufacturer and lot number written in the lead-in area. Every CD burner has
a table of these CD manufacturers and dye formulations burned into ROM.
Prior to actually burning the CD, the burner writes to, and then reads back
from a special area on the CD-R to determine the proper laser power, based
on the aforementioned table. So, if your burner is a year or two old, and
you haven't updated the firmware, and you're using contemporary CD-Rs, the
table in your burner may not have the necessary information for the burner
to properly calibrate its laser, so it'll use a generic setting, possibly
increasing the error rate.
Finally, you have to consider the speed at which you burn the CD-R. Nowadays
most CD-Rs are optimized for high-speed burning. 4x and higher. In fact many
new CD burners (computer based that is) do not even burn at 1x. SO if you're
burning your CDs in real time you have to be especially careful about your
choice of media.
What to do? As far as the choice of CD-R, Phthalocyanine with gold
reflective surface is probably the most archival. However if you're burning
at 1x you may find the error rate a bit high, in which case you might want
to consider cyanine dye. You must also use a new CD-burner with updated
firmware. Get a number of discs from different manufacturers from your
supplier and burn some program onto them and send them to be tested by a
facility which has a CD-R verifier - Most manufacturing plants and large
mastering facilities have these. When you get the results you'll clearly see
which discs work best. Ideally you should look for a block error rate (BLER)
of 10 or lower. You should also look at where the errors occur. Generally,
marginal discs have a much higher error rate in the first couple of minutes,
which reduce as the disc progresses. So you want to check the running tally
and make sure the BLER rate isn't significantly higher at the start of the
disc. Any E22 or E32 errors indicate error concealment or interpolation,
and those discs should be rejected.
I have found that the Mitsui Gold (phthalocyanine) and the Taiyo Yuden
(cyanine) discs are consistently the best. As far as burners go, Plextor is
considered to be the best computer-based burner. For stand-alone, I'm
partial to the Alesis Masterlink. It allows limited editing of track IDs,
burns at 4x and has the added advantage of being able to burn 24-bit
(assuming you recorded at 24-bit resolution) CD-ROMs in AIFF format -
readably by almost any computer/software platform.
Recording * Editing * Mastering * Production