cinemakinoeye
Well-known member
Wise words, I'd add MIRROR THE CRITICAL STUFF, since these days drives are so cheap, it's folly not to take advantage of things like RAID-1 mirrored pairs to reduce the impact of drive failure.dvInsight said:[...] BE PARANOID and BACK UP OFTEN. Because stuff fails, falls and farts in real life all the time.
I've recently had a 250GB external FireWire 800 drive with 10 hours of video become corrupted and I did not freak out a bit, since it was part of a RAID-1 mirror pair, I simply replaced the drive with a fresh one and rebuilt the pair. Things break, that's the nature of electro-mechanical devices, that's why drive manufacturers quote MTBF (Mean Time Between Failures) numbers, they are telling you they know it will fail at some point... there's always a statistical probability of data loss due to catastrophic hardware disk failure or corrupted disk structures. If you live with that as a fact of life, you'll never stress when a disk becomes corrupted or hardware fails. I work with this rule: design your workflow in such a way that you can pull any single disk out of the system and the work can continue without missing a beat. With drive prices as low as they are these days, there's no need to have a delicate electro-mechanical device like a hard drive be the weakest link of the chain. This is why RAID-0 for capture/storage makes me nervous.