27 Dec
2006
27 Dec
'06
11:58 a.m.
On 27 Dec 2006 at 11:11, dwight elvey wrote:
Ahaa. I'd thought that it might be different
coercivities. I've been using
1.44M disk in my Cat without issues. It has a disk test mode that writes
patterns and reads. It was able to run for several hours without failure.
I know that 1.2Meg disk don't work well in 360K drives but the drive they
used in the Cat must be more forgiving.
You can use 1.44Ms as 720Ks because they're far closer in coercivity
than 360Ks and 1.2Ms. However, it makes sense to cover over the HD
sensing hole in 1.44Ms before using them, as you never know when
you'll try to read them in a drive capable of 1.44M which senses the
hole.
Don't try to use a 1.44M drive in HD mode to record 720K--it won't
work reliably--I believe there's some high-pass filtering involved in
auto-sensing drives that will get in the way. To illustrate this,
try this experiment: Cover over the density-sensing hole on a 1.44M
and format it up as 720K in an auto-sensing 1.44MB drive on a PC--
write some files to it. Now, remove the density sense cover and try
to read it. It won't read reliably. The PC is unaware of the
density hole--it's essentially blind to what density the drive thinks
the diskette is.
I have seen early operating systems that use the ready/not ready
transition to discover diskette changes, but these have mostly been
on 8" drives. On 5.25" drives, I've written drivers that poll a
drive (with motor on or off) to sense the write-protect status change
as a diskette is inserted or removed. This was tied into a scan of
the open file table to flash a display message that essentially said
"Put that back!". Not quite up to a locking door latch or auto
eject drives, but it worked very well.
Cheers,
Chuck