Raymond Check discusses MS DOS Floppy Disk Cache

[Fred Cisin]

>>> https://devblogs.microsoft.com/oldnewthing/20190924-00/?p=102915

On Tue, 24 Sep 2019, Chuck Guzis via cctalk wrote:
> I don't quite get the bit about the NAND gate, but back in the days of
> 5.25" floppies there was a way around that one.
> One need only monitor the drive write-protect status every tick or so.
> If it toggles, invalidate the cache.
> I implemented it on a 8085 system, checking the write protect status
> every 250 msec.  The drive needn't be spinning.  If you insert or remove
> a 5.25" disk, the status will change.
> We even went a bit further and checked to see if the disk contained any
> files open for writing.  If so, we sounded an alert and froze the system
> until the subject disk was re-inserted.  This reduced the number of
> disk-swapping related errors essentially to zero.

And, it presumably worked just fine.
They tried TWO of their staff, and neither could change a disk in less 
that 2 seconds.

COULD somebody go from write protected 8" or non-write-protected 5.25" 
(did you sense door open?) to disk out of drive in under 250ms, and then 
lso go from door open, no disk to non-write protected disk in drive (door 
still open?) in under another 250ms?   I can't.  But if somebody 
could, then you just need to change that user's settings to about 100ms.