Comments on: It’s October…

By: Anonymous

Anonymous — Thu, 04 Oct 2001 01:08:44 +0000

I’m no computer expert, but, logically speaking, if something is squeezed too hard, it eventually busts.

By: Anonymous

Anonymous — Tue, 02 Oct 2001 07:02:02 +0000

Tried spinrite on it yet?

………Curtis

By: Anonymous

Anonymous — Mon, 01 Oct 2001 23:14:21 +0000

It's a Seagate Medalist Pro, 7200 rpm, a couple of years old--the model that was notorious for heat. It's not a screamer at all by today's standards but it has all the heat associated with the modern screamers.

I've never had a problem with the cable or card I used before (the cable's pretty new--only used once before I believe--and the card, a venerable Adaptec 2940, has always been rock-solid for me).

I do find it interesting that it causes a lot less damage when the drive fails in Linux than in Windows...

By: Anonymous

Anonymous — Mon, 01 Oct 2001 19:39:05 +0000

You don’t mention what type of SCSI drive you got. Is it one of those fast spinning screamers? If so then heat is bound to be a problem especially if the drive is of an older version (later versions of the screamers are better at keeping heat in check).

By: Anonymous

Anonymous — Mon, 01 Oct 2001 12:41:20 +0000

Heat or power – probably; but let us not forget the controller as a possible source of error, and even the cable between the two, or connections – including the slot in which the controller sits. Remember Jerry Pournelle’s rule that 90% of SCSI problems are cables. You’re probably right – things with moving parts DO tend to either break down, or generate frictional heat. However, contacts can subtly corrode over time, and that increases resistance, and that increases heat, and that can make them more susceptible to other heat. I’ve also heard of high-speed PC internal cables generating their own interference with their own signals if they’re doubled back on themselves.