[comp.sys.ibm.pc.hardware] Diffs between 5.25 DD & HD disks

jin@spdcc.COM (Jerry Natowitz) (04/16/91)

In article <1991Apr16.092251.23313@cc.tut.fi> public@cc.tut.fi (PD Software Group) writes:

>Yes - it works only on 3.5". The differences between 5.25" DD and HD
>disks are so big that when you format a 360kb disk to 1.2mb disk,
>you'll get at least 500kb of bad sectors. The rest of that disk is
>also very untrustworthy, since the media is really different. But
>converting 3.5" DD to HD is usually quite safe.

I must either be lucky or your information is wrong.  I agree that attempting
to use vanilla format on DD disks will result in tons of bad sectors;
but I spent a day recently using fdformat on three different batches of
5.25 floppies and found that at least some of them will format and hold
data at 1200 kb.

My "test plan" was to format with fdformat (as a side line I was experimenting
with the x and y sector slide parameters) and use Norton Disk Doctor II to
check for bad sectors (and take the time used as a performance measurement).
I then used selected disks for backup/restores from Fastback, and finally
copied a bootable HD floppy to my 3.5 drive where I had formatted a floppy
to 1200 kb and then back again to a DD floppy formatted to 1200 kb.

The TDK HD disks, as expected, were error-free.  The TDK DD disks
were also error-free.  The Building 19 disks (from a local salvage store,
$2.99 for a bag of 10) had one bad track in the lot, but 5 of them had
soft errors (I could hear a re-seek in NDD, but no errors were reported).
I decided to use the Building 19 disks at 720 kb.

I found the optimal sector slide to be y:3 (3 sectors between  tracks)
and x:1 (1 sector between heads).  Your milage may vary on those parameters.
I further found that sector slide is consistent between 3.5 and 5.25 floppies.

Would I store critical data on a 1200 kb DD?  No, I use 3.5 for that and my
drive dutifully refuses to format 3.5 DD floppies to 1440.  I do intend to
do some longer term tests on my 1200 DD floppies and will report on results.
-- 
     Jerry Natowitz
     Guest user on:
ARPA jin@ursa-major.spdcc.com
UUCP {ima,harvard,rayssd,linus,m2c}!spdcc!jin