On Thursday 22 February 2007 14:21, John Andersen wrote: > ... > > Yet some here persist (you know who you are Kai, ;-) in echoing the > dogma of yesteryear as to why scsi is better. ...
I suppose "better" can be measured many ways. I can find 15,000 RPM SCSI drives but only 10,000 RPM SATA drives. I put one of each in my newest system. The 15,000 RPM SCSI is Ultra 320 and used as the system / root file system drive. The SATA drive uses the newer (SATA II, is it?) 300 Mbps interface. The SCSI drive is much smaller--only 36 GB--while the SATA drive holds 150 GB. The system is zippy, though. Let's hope the drive lifetimes are decent. One thing I've noticed is that as drives get faster (especially head positioning speed), mounting them properly seems to be more important. If the drive can physically vibrate within its mounting, it appears that ordinary head motion can lead to a head crash when the drive as a whole begins to move in response to the head be repositioned. My hunch is that sometimes head activity pattern creates a resonance with the drive as a whole and the magnitude of the drive's motions (in its ill-fitting mount) become sufficient to cause a head crash, or at least to increase the probability of one significantly. It's all a hunch though, but I had a couple drive blow-outs (full-blown head crashes--the data recovery people said there was a lot of brown dust inside when they opened it up!). Both of them happened on a tower I had assembled rather shoddily. Up 'til then, I'd never thought about the possible consequences of letting a drive simply rest, unsecured inside the box. Since then I've been careful to mount my hard drives very securely. > ... Randall Schulz -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
