Path: utzoo!mnetor!uunet!seismo!sundc!pitstop!sun!imagen!fjd
From: fjd@imagen.UUCP (Fred Drinkwater)
Newsgroups: comp.sys.ibm.pc
Subject: Seagate ST251 Reliability
Message-ID: <1583@imagen.UUCP>
Date: 28 Dec 87 18:09:46 GMT
Distribution: na
Organization: Imagen Corp., Santa Clara CA
Lines: 51
Keywords: Hard Disk Reliability
Summary: Summary of responses to my query

A couple of weeks ago I posted a query about possible reliability
problems with Seagate ST251 hard disks.  My questions were
prompted by some disturbing information in MicroCornucopia
volumes 38 & 39, which suggested that a lot of these disks
were failing within three months.  I am happy to report (especially
since I had one on order at the time) that the responses have 
been uniformly favorable:

From Dominick Samperi ihnp4!cmcl2!manhat!samperi:
I've been using two ST251's in my AT compatible for about 4
months without any problems.

From David F. Carlson atari!ames!rochester!ur-valhalla!micropen!dave:
We standardized at this site aver a year ago that *all* our 40meg
drives would be ST251's.  With several machines running UNIX 24 hours
a day we have yet to have any downtime or any new bad blocks on any of
our ~6 ST251s.  In short, a very good and reliable disk for my money.

From Karl Denninger hplabs!sun!laidback!karl@ddsw1.ARPA:
Got a dozen or so in the field and no problems so far at all.
Ages run from 2 weeks to 6 months or so.

Other similar reports...

Only one complaint:

From Angus Wang atari!ames!cs.duke.edu!aw:
I have one of these in an AT clone and have had no problem with it
but it does have, in my opinion, a large number of bad tracks.  It
came that way so there's not much I can do.  I had mine in almost
continual use for the first 3 months working on a development project
and used up about 10M for data and programs.  This project involved
a lot of reading and writing to the disk so I feel that if it was 
going to fail, it would have done so by now.  ONe thing I dislike
about the 251 is that it has a long average seek time.

(Regarding the above: speaking as a former disk controller designer,
my experience with various drives between 36 and 120 MB was that
a typical drive would have from 10 to 20 bad sectors found by
the manufacturer and reported on the drive's rap sheet.  Those
manufacturers who handled bad sectors by flagging the entire
track as bad, we chose not to do business with.  The manufacturer's
test procedure typically found 2-3 times as many bad sectors as
did the low-level formatter (because the mfg uses analog tests).
I never had a sector go bad after receiving a drive, for which I
am eternally grateful, since most OS's just panic if that happens.)

Fred Drinkwater {ucbvax,decwrl}!imagen!fjd

Disclaimer: The opinions above are not those of my employer or
my cat, who has more sense than to expose himself to a libel suit.