kindred@ea.ecn.purdue.edu (Donald Kindred) (05/02/89)
Has anyone really tested the 2090A for data integrity? I have seen
a number of articles posted about performance, and a number of articles
about problems the old driver had, but almost nothing recently about
anyone really trying to methodically test the 2090A. When I say
"really tested" I mean something fairly substantial such as:
1. write a substantially large test pattern (more than 90% of the
disk's capacity) into a file (or files).
2. verify the test pattern.
3. repeat until a substantial amount of test data has been written.
(Perhaps a GB or two)
I tried this recently with my Quantum Pro 80S, and would be interested
in hearing if anyone else has tried something like this. I would also
be particularly interested in what types of testing CBM has performed,
and what, if any, bugs CBM is aware of. My tests were only for the
case of a normal interlaced workbench screen (2 bitplanes), AmigaDOS
1.3 in ROM on a B2000, and two 40 MB test files on FFS partitions.
Are there things that aren't documented in the 2090A manual or disk
that one should be aware of? Specifically, something that might
cause relatively sparse errors (once every 200 MB or so). An error
a typical user might not realize occurred.
I have not yet determined the source of my errors; I have not ruled
out the possibility that the errors are due to a blunder on my part.
Don't interpret this as a bug report; it's just an inquiry.
--Don
--
Don Kindred | Internet: kindred@ea.ecn.purdue.edu
ACN, Purdue University | Phone: (317)494-0120