Parallel drive (was: USB 5.25" floppy drive - do it

From: Steven Canning <cannings_at_earthlink.net>
Date: Sun Jun 13 17:50:09 2004

Actually, truth be known, I think I would just "opt" for a catweasel and
call it good. There is one on Ebay now with two days left currently, at $26!
http://cgi.ebay.com/ws/eBayISAPI.dll?viewItem&category=4598&item=5101020898&
rd=1&ssPageName=WDVW

Best regards, Steven

----- Original Message -----
From: "Steven Canning" <cannings_at_earthlink.net>
To: "General Discussion: On-Topic Posts Only" <cctech_at_classiccmp.org>
Sent: Saturday, June 12, 2004 4:00 PM
Subject: Re: Parallel drive (was: USB 5.25" floppy drive - do it


 Instead of speculating, lets run some numbers with the following
presumptions;

 1.. The spindle motor is running and the disk is up-to-speed (300 RPM)
2.. The head(s) have been stepped to track "zero" so we know what track we
are on
3.. We will be getting data from head "zero"
4.. For the moment we will ignore the directory, so we don't know the
starting sector #
5.. Worst-case through-put for a 1.44 MD 3.5 " Disk (500 K"bits"/second
minimum)
6.. Tracks per disk "80" (per side)
7.. Sectors per track "18" (1440 total per side)
8.. 512 bytes per sector (737280 "raw" bytes per side with overhead)
9.. One track contains 9,216 bytes (73,728 bits) with overhead
10.. 300 RPM equals 5 Rev / Second
11.. Absolute "best-case" scenario 5 Rev times 73,728 bits = 368640 bits /
sec or 46 KB/sec
12.. A modern EPP (Enhanced Parallel Port) can easily support >1 MB/sec
The parallel port should easily be able to support the data from a Floppy
Disk Drive if a DSP or FDContoller converts the data to parallel. If you
dumped a tracks worth of data you would have to write software to put the
sectors in the proper sequence (most likely contiguous but not necessarily
in ORDER) because your file can be spread across sectors all over the disk
(even the backside).

 Best regards, Steven



 ----- Original Message -----
 From: "Dwight K. Elvey" <dwight.elvey_at_amd.com>
 To: <cctalk_at_classiccmp.org>
 Sent: Friday, June 11, 2004 5:35 PM
Subject: Re: Parallel drive (was: USB 5.25" floppy drive - do it


  Hi
   I doubt one can fetch data with most PC's from the parallel
  port fast enough to keep from being overrun, even on a byte
  wise basis. That is why I've suggest the DSP. May of these can
  run fast enough to do it on a BIT wise basis and
  require no external hardware, other than buffers.
  Dwight

 From: "Steven Canning" <cannings_at_earthlink.net>

 I've been looking into this for some time. The parallel port lacks the
 "through-put" to take the data on and off the floppy as serial data (as it
 comes off the drive "raw") but if you added some hardware (like a Western
 Digital FD controller) it will separate the data and convert it to
 "parallel" data which the parallel port can support. The inverse is also
 true (parallel data back to serial to fed the drive). The FDC can handle
the
 Single density issue. Processing power of the computer is not an issue
 unless you have a painfully slow machine. I wish I had more time to work on
 this project. Anyone have the Kilobaud article were someone connected a FDD
 to a Heathkit ET-3400 ?

 Best regards, Steven

 ----- Original Message -----
 From: "Fred Cisin" <cisin_at_xenosoft.com>
 To: "General Discussion: On-Topic and Off-Topic Posts"
 <cctalk_at_classiccmp.org>
 Sent: Thursday, June 10, 2004 2:55 PM
 Subject: Parallel drive (was: USB 5.25" floppy drive - do it


 On Thu, 10 Jun 2004, Jules Richardson wrote:
 Not sure if such as a PC parallel port is fast enough to cope with the
 data rate of a floppy drive and leave enough time for the CPU to do the
 processing though... but that'd be nice; little more than a cable and a
 bit of glue logic hooked up to a parallel port that could be quickly
 swapped between machines.

 MicroSolutions (DeKalb IL) in their "BackPack" line, made parallel port
 floppy drives. I have a 2.8M 3.5" from them, but they also made a lot of
 other models.
Received on Sun Jun 13 2004 - 17:50:09 BST

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:36:57 BST