mclaren's opinion on CDR (fwd)

This is the text of email to me from Dana Parker in response to mclaren's
post on CD Recordable which I forwarded to her.


mclaren-

I hardly know where to begin to correct the misinformation, myth, and utter
nonsense in your message about CD-R technology.

Dvorak on CD-R: Anyone who placed his or her trust in what John Dvorak says
about CD related technology would be CD-ROM-less today. In 1989, Dvorak
wrote a column about CD-ROM technology that was just as negative - and
just as wrong -as his March 1996 column about CD-R. There is a reason behind
this animosity on Dvoraks part - it took him six months to successfully
install his first CD-ROM drive, by his own admission. As a result, he
dismissed the entire technology as worthless, which says more about
Dvorak than it does about CD-ROM or any of its related technologies.

Lets take your claims one-by-one.

1. The problem with the writeable CD technology is that you have to make an
`image' to be written to the disk. If you want to back up 650 MB, you
want to have another 650 MB on your hard disk empty to store the image,
which is then transferred to the CD-R.

---False. You have the option, supported by most if not all CD-R software
packages,to record "on-the-fly" from a virtual image. A virtual image
takes from 1% to 5% of the space of the data to be recorded, depending on
the software. Creating a virtual image is virtually transparent to the user
- you drag and drop the files to be included, then click on record.

2. You get *one shot at writing to the disk* from beginning to end. You
can do multiple sessions, but that's a *real hassle too.*

---False. When Dvorak says something is a "real hassle", what hes really
saying is "this is something I dont understand and wont spend five
minutes figuring out, let alone attempt to explain to you". Again,
depending on the software, creating multiple sessions is a simple
drag-and-drop, point-and-click operation. Its not rocket science.

3. If there's a glitch, brownout, operating system hiccup, horribly
fragmented file, or any number of things, then the CD doesn't get burned,
and you end up with a *blown session.* These are common, and *the result
is a totally useless disk* that must be discarded.

---False. Dvorak is right that any of these factors, or more likely, a
combination of them, can ruin a recording session by causing a buffer
underrun. It is not, however, a common occurence, it does not necessarily
render a disc useless, and it is certainly not unavoidable. In over six
years of using CD-R, I have never cut a coaster or had a buffer underrun.
It is certainly not unavoidable. In over six years of using CD-R, I have
never cut a coaster or had a buffer underrun.

4. It takes a couple of hours to prepare your files to be burned into the
disk. If something changes, you have to start over. And if your hard disk
isn't fast enough, you can't even burn the disk at all.

---False. You can prepare files for recording on CD-R in a matter of
seconds. A large dataset might take longer. A dataset intended for
distribution might take longer yet, but only because of normal
precautions that would be part of preparing any dataset for distribution:
virus checking, defragmentation, directory structure, file placement.
Most CD-R manufacturers recommend a minimum hard drive speed of 19 ms.

5. "One vendor told me that the quad-speed CD-ROM writers *were kept off the
market* because the system requirements were outrageous, and tolerances for
error was nil.

---False. This must have been a vendor of 2x CD-Recorders. 4x
CD-Recorders have been on the market since 1993, and they function
perfectly well on a 386 with 8MB RAM.

---In short, everything Dvorak says is wrong.

6. The test reported on this forum in which a 15-minute audio file was
written to CD-R is, alas, meaningless. The acid test: how often does the
drive and disc combo produce a bad burn when a full 680 megs of audio
data are written non-stop to CD-R disk?

---Apparently not often enough to prevent numerous state, federal, and city
governments as well as legal firms from using CD-R as an archival medium for
irreplaceable data, or to prevent hospitals from using it to store permanent
medical records, or to prevent Jet Propulsion Laboratories from converting
irreplaceable data, or to prevent hospitals from using it to store permanent
medical records, or to prevent Jet Propulsion Laboratories from converting
their 9-track tape libraries of raw satellite data to CD-R. Apparently
not often enough to prevent sales of CD-Recordable drives from increasing
by a factor often in the past 12 months, or to prevent dozens of CD-R
service bureaus from basing successful businesses on creating quality
music and data CD-R discs by the thousands every month for their customers.

7. Reports suggest that the error rate right now is around 60%, depending
on the brand of CD-R disc. That is, 6 out of 10 discs produce a bad burn
and a blowndisc which must be discarded.

---Id be very interested in seeing those "reports". It would be an amazing
contrast to the independent testing being sponsored by organizations such as
SIGCAT (Special Interest Group for CD Application Technology), OSTA (Optical
Storage Technology Association) and JPL. The OSTA prelimianry test results
impressed even those of us who knew CD-R was a reliable storage medium - it
showed disc reliability to be greater than 96%, regardless of the disc brand.

8. Worse still, my gearhead friends inform me that the writeable CD
technology is so flaky that to get good results, people who burn their
own CDROMs must check the grapevine to see which brand name of writeable
CD-R disc has a bad error rate this month, and which manufacturer came
out with a good batch. One month TDK writeable CD-R discs produced
near-100% bad burns, the next month it was Mitsui, the next month Sony...
And it changes from one month to the next. You get the idea.

---I get the idea you or your gearhead friends are making this up. As
someone who has been burning her own CD-Rs, providing technical support
for other CD-R users, and writing about the technology for publication
since 1990, I can assure you this does not happen among "people who burn
their own CD-Rs". The media is relabeled and distributed by hundreds of
companies. There is no way of knowing whether a given disc was made in a
certain month, and often there is no way to tell who the original
manufacturer was. CD-R users couldn't exchange this information even if
they wanted or needed to. The problem these days isn't getting a certain
brand of media, it's getting any media at all. There is currently a
worldwide shortage of several months' duration, and every CD-R media
manufacturer is building new plants and adding production lines to keep
up with the increased demand, which far exceeded their wildest
expectations. This doesn't sound like evidence of a "flaky" technology
to me.

9. Worse still, many CD-R writeable drives will burn discs that other
brands of CD-ROM read-only drives cannot read.

---This would be more accurately expressed as "some CD-ROM drives are
incapable of reading some CD-R discs." This is not because there is
necessarily something wrong with the discs. It?s because some CD-ROM
drive manufacturers cut corners and narrow the required parameters within
which the drive will read to the point that some CD-R discs - as well as
many pressed CD-ROM discs - cannot be read, even though they are within
specifications. It is also not attributable to the drives that recorded
the discs. Recorders use a built-in Optimum Power Calibration before
every writing instance to ensure that what is recorded can be reliably
read by any CD-ROM drive or CD audio player that is capable of reading
within the full range of disc parameters.

10. Despite all the nonsense about the photoreactive dye lasting 10
years, 20 years, 500 years, or what-have-you, the brutal reality is that
CD-R discs don't last very long. A year, maximum. Probably less.

---I have a backup of my hard drive on a CD-R disc recorded in 1990. It?s
still good, as are all the other discs I have recorded since then. I have
not lost one disc to age. The fact is that CD-R discs will probably last
longer than pressed CDs, because they use gold, rather than aluminum, as
a reflective layer. Aluminum tends to corrode due to reactions with
impurities in the polycarbonate and in the aluminum itself. Gold does
not. However long the data on CD-R lasts, whether it is 10 years or 300,
the fact remains that CD-R is the most durable medium ever developed for
the storage of data, with the possible exception of papyrus sealed in
clay jars.

11. Fingerprints are permanent. They cannot be burnished off. Try to
do so, and you'll destroy the CD-R because you'll smudge the
photoreactive dye.

---Have you ever actually seen a CD-R disc?
Fingerprints can be wiped or washed off. I polish off fingerprints on my
pants leg. The dye - which is an organic dye polymer sensitive to a certain
wavelength of laser - is protected on one side by a 1.2mm thick layer of
clear polycarbonate (the same stuff used to make motorcycle helmets); on the
other by a thin layer of gold and a UV-cured lacquer, often with an
additional layer of protective coating and/or printable surface. Your
fingers will never come in contact with the dye layer.

12. The photoreactive dye is laid down in a single continuous spiral
track 1/10 the width of a human hair. As the disc endures thermal stress
in winter and summer, the single continuous spiral track distorts and
finally becomes unreadable.

---The dye polymer is applied via spin coating over the entire surface of the
disc. Track pitch is 1.6 micrometers, the same as for pressed CDs, and
the track itself is .6 micrometers wide. On an unrecorded disc, the
spiral track is in the form of a "wobbled pregroove" molded, as a
sinusoidal wave with a .3 micrometer excursion, into the polycarbonate
itself. The pregroove guides the recording laser. The spiral track on a
recorded disc is composed of optical marks produced by a chemical change
and/or degradation in the dye layer. "Thermal stress" is not a factor in
data loss as much as prolonged exposure to extremely bright light, which
can theoretically, over a period of years, affect the
comparative reflectivity of marked and unmarked areas of the disc.

13. If the spiral dye channel is even microscopically off-center, the CD-ROM
drive will accumulate so many errors as the drive head runs from the inside
toward the outside fo the CD-R disc that the disc will be unreadable.

---The pregroove is molded into the disc and does not change or become
distorted. If the molded pregroove is microscopically off-center beyond the
allowed parameters, the affected disc will never make it out of the
plant, nor will the thousands of other discs pressed from the same mold.

14. Thus, unfortunately, the writeable CD technology is not ready for
prime time. I researched the technology 6 months ago:it was dismal.
Packet-writing drives are now beginning to become available, but it
doesn't seem to have helped the abysmal error rate of the CD-R media, nor
has it helped lengthen the ridiculously short lifespan of dye-based CD-R
discs once burned. Those of you who are contemplating buying a CD-R
writeable drive to burn your own microtonal audio CDs would do well to
wait another 3-5 years. Perhaps the technology will have been debugged
by then.

---I have been working with and researching CD-R technology for most of
the past seven years, and I am forced to conclude that your research is
what could more properly be described as "dismal" and plagued by an
"abysmal error rate". If you depend on the baseless opinions of has-been
PC industry pundits, or the properly be described as "dismal" and plagued
by an "abysmal error rate". If you depend on the baseless opinions of
has-been PC industry pundits, or the advertisement-driven mainstream PC
industry magazines for your information, in fact, Id have to say the
appropriate term for your CD-R research is "non-existent" - and your
"findings" are glaring proof of that.

You and John Dvorak could both benefit from doing some real research on CD-R.
I'd hate to think that someone might actually decide not to try CD-R
technology because he or she mistakenly thought you actually knew what
you were talking about.

Regards,

Dana J. Parker
Contributing Editor/Standards Columnist
CD-ROM Professional magazine
Co-author, CD-ROM Professional's CD-Recordable Handbook





Received: from ns.ezh.nl [137.174.112.59] by vbv40.ezh.nl
with SMTP-OpenVMS via TCP/IP; Fri, 6 Sep 1996 02:53 +0200
Received: by ns.ezh.nl; (5.65v3.2/1.3/10May95) id AA23327; Fri, 6 Sep 1996 02:54:35 +0200
Received: from eartha.mills.edu by ns (smtpxd); id XA22930
Received: from by eartha.mills.edu via SMTP (940816.SGI.8.6.9/930416.SGI)
for id RAA01854; Thu, 5 Sep 1996 17:54:34 -0700
Date: Thu, 5 Sep 1996 17:54:34 -0700
Message-Id:
Errors-To: madole@ella.mills.edu
Reply-To: tuning@eartha.mills.edu
Originator: tuning@eartha.mills.edu
Sender: tuning@eartha.mills.edu