Subject: RE: Early 3.5" Floppy Drives
From: Fred Cisin <cisin at xenosoft.com>
Date: Thu, 15 Dec 2005 13:43:43 -0800 (PST)
To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at
classiccmp.org>
On Thu, 15 Dec 2005, Allison wrote:
Any PC controller that can do 720k 3.5"
format can do
8" as it's the same data rate. it's not what chips was
used it's how it was used.
It's rare enough that Allison (or Tony) make a mistake,
that it is a rare opportunity to be able to disagree with
any confidence.
Any PC controller that can do "standard" 360K, can do 720K 3.5".
Only difference is whether the OS and/or BIOS are happy with the
trivial differences.
The software interface is a seperate issue. Very few media smaller
than 8" used 26 sectors per track FM or MFM.
But unless we allow some "re-programming with
solder and dead bugs",
many of the 8 bit FDC boards are hardwired to MFM, yet have the data
transfer rate that the 8" would want for FM.
The original IBM FDC board for the 5150 could be modified for 8"
(Flagstaff Engineering did so), but it's a lot of extra wires.
The IDE8 FDC were not hard wired for MFM. Amen. The selection
of FM/MFM is a bit in the command byte. It's bit 6. Write data
Fm is 05h and write data MFM is 45h. Now what the logic connected
to pin26 does with the signal is possibly unknown but the resulting
output to the drive always follows the command byte.
On the other hand, most FDC boards that support 1.2M
can do 8" DD. The ones that also support FM can usually
handle 8" SD. 'course there are SOME that are designed so weird that
they still can't.
Boards for 5150 that support 8" include the popular Compaticard,
Flagstaff Engineering, Maynard, Vista, J-Disk, MMF, etc.
Yes there are a few that pin26 goes nowhere and the resulting FM
data rate is then twice what you would expect. For some cases
that is an advantage. ;)
Allison