[MacDV] Re: Digital Video Quality

[Richard Brown]

In reading this string, I note people are worried about the archival 
nature of their burned CD-R/DVD-R media. One solution for getting 
better archival nature is to abandon the sexy super fast burn speeds of 
modern burners. Faster is NOT better, NOT as archival. You need not 
look too far to find a variety of white papers on this subject. Pit 
depth suffers at faster burn speeds (shallower) and promote early data 
loss. This may also include the chemical nature of the "certified" fast 
burn speed CD-R / DVD-R material. I have decade old 1x and 2x discs 
which have maintained 100% data integrity, and expect them to last 
quite a few more years. All these were noted "premium" or "archival" 
discs as purchased, and burned sloooooooow.  Music for example, as in 
AIF audio CD's, will get funky in many players should the pit depth be 
too shallow. The etching power does NOT increase with speed. It is just 
like exposure in a camera. One second is a LOT more exposure than 
1/1000 second given the same f/stop and lighting. Burning at 52x with 
the same laser energy as 1x cannot lead to paradise, only instant 
gratification with short-lived duration.

Given the drop in hard drive prices, I agree with others that to 
archive these days implies buying archive hard drives. The problem with 
this idea, of course, is that an archive drive, by definition, is one 
used for backing up important data only, and thereafter is shelved in a 
safe place to keep that data safe. This implies the drive will be 
booted many, many, many, times as files are accumulated. This defines a 
unique problem: most drives suffer the MOST wear and tear on bootup, 
and are thereafter stable with little wear and tear until the next shut 
down and reboot. Archive drives will, by definition, be amongst the 
hardest use drives in any workflow.

So, there are caveats, both ways, excepting: IF an archive hard drive 
fails, and IF the data maintains integrity through the failure (no head 
crash nor magnetic aberration) you will be able to have a forensic hard 
drive specialist extract all your data. The same is NOT true of 
CD-R/DVD-R media once compromised.

There are caveats both ways, and with the growing needs of MASSIVE 
storage due to the "digital lifestyle" influencing professional and 
personal concerns, it is a real problem seeking an answer which will 
stand up to geologic time. Currently, barring a breakthrough of 
archival, digital proportion, the solid bet remains using the best of 
the new breed of large hard drives, but keeping aware of the ongoing 
fluctuation within drive manufacturers in terms of build and material 
quality. It's best to check with people who go through a lot of hard 
drives - and ask them which manufacturer currently heads their "don't 
buy" list.

Seagate, for example, once legendary in the SCSI drive market (we only 
recently retired some very, very old Cheetahs and Barracudas), has gone 
through buyouts and such, resulting in a lousy IDE implementation. We 
had THREE D.O.A. drives, in the same batch, which cancelled all future 
use of Seagate products. Who cares who made the drives, but they 
sullied Seagate's name in perpetuity. And now Hitachi has taken over 
IBM's drive business, but this apparently included the same development 
and manufacturing facilities. They have announced a 7200 RPM / 400GB 
drive, which will prove interesting. Maxtor has had recent issues, 
moreso than Western Digital. Basically, it remains a mixed bag, without 
a defined winning scenario.

In short backup here, and there, and there, and maybe there, and you 
have a shot at keeping your data.

Richard Brown



On May 27, 2004, at 5:53 PM, Peter Tattersall wrote:

> There are a number of effects which cut down the life of CDRs and 
> DVD±Rs. Since the data is burned in by light, we should not be 
> surprised that (sun)light can burn them out. Since the discs are built 
> in layers held together by glue, we shouldn't be surprised when the 
> layers move in relation to each, or when the glues fail under stress. 
> As for no-name vs brand name, some of the no-names are the same as the 
> brand name media, just without the brand. Price and "brandedness" 
> don't seem to be a reliable indicator of quality these days. Ar! It 
> were different when I were a lad!
>
> On 27-May-04, at 8:01 PM, Norm Lamoureux wrote:
>
>>      Why would 20 - 30 % of DVD-R media only last
>> about 10 years or less? They claim to last many
>> decades. No-name media is just as good as brand name
>> media?