Re: Hardware musings

["Michael J. Mahon" <mjmahon@aol.com>]

Ed Eastman wrote:
> Michael J. Mahon wrote:
>
> > Data cannot be read reliably until the disk is near its correct speed,
> > and drives differ considerably in the time required to reach full speed,
> > so it would be difficult to use the acceleration period reliably across
> > different drives.
>
>
> However there are disk protection schemes out there that read data bytes 
> fromt eh disk without the high bit set.  They are timing critical and 
> are there to detect a signature.  This was some on some later games that 
> have differing numbers of 0 bits between the bytes.
>
> This varying number of bits between bytes can be hard to detect and 
> write accurately if you writing program assumes 1 or 2 bits uniformly.

However, the Disk ][ hardware is not guaranteed to repeatably read
more than two consecutive "0" bits, including trailing "0"s after
longer-than-32-cycle nibbles.

So a 40-cycle nibble with its low bit set is the longest nibble that
allows the following nibble to be reliably read--since the drive
read amplifier may insert spurious "1" bits after two "0"s.

"Bit insertion" protection schemes should all abide by the "not more
than two consecutive '0's" rule.  There is still plenty of room for
clever timing tricks within this limitation.

It is, of course, possible to detect *any* nibble length, subject to
the limitation on reliable reading, and, of course, it is possible to
write any nibble length as well--whether it can be read back reliably
or not.

-michael

NadaNet networking for Apple II computers!
Home page:  http://members.aol.com/MJMahon/

"The wastebasket is our most important design
tool--and it's seriously underused."