25 Nov
2005
25 Nov
'05
7:29 p.m.
Subject: Re: Pinout for SED9421
From: "Chuck Guzis" <cclist at sydex.com>
Date: Fri, 25 Nov 2005 10:36:33 -0800
To: cctalk at classiccmp.org
On 11/25/2005 at 9:59 AM Allison wrote:
Tandys EI design got me a few trips to FtWorth to massage their heads
with a bat for what they did.
Yes, 1meg was way above the usual 190k-380k common for for DDDS drives.
The highest I got to with MFM was 800k (80tr 2sided).
The all time worst was the 1771 internal data sep.
Tandy initally did
that to save parts. Really bad.
I remember once asking our FDC design guy (he cut his teeth at Sperry ISS)
about the 1771 on-chip DS. He just laughed and gave a thumbs-down. He
built his board around the 1781, using GCR and a bit rate of about 380 KHz
on DS media. The drive electronics needed the low-pass filter tweaked a
bit to work, but he managed to get about a megabyte on a 5.25" DS floppy.
Not too shabby for the time (1979). Drives back then tended to have more problems than the
controllers did.
In particular, we gave up on Micropolis because of their stubborn
insistence on using a leadscrew positioner, long after almost everyone else
went to the faster taut-band. I still have a 96 tpi Micropolis drive that
employs buffered seek (goes not ready while seeking). It also mounts the
drive electronics board and the positioner on the movable part of the disk
clamping assembly. Only the drive motor and tach circuit is mounted on the
stationary drive frame.
The disk that Shugart used while innovative was really bad for repeatability
of position and slooowwww.
Allison