18 Mar
2007
18 Mar
'07
9:36 p.m.
Today I stumbled across something that may explain why some
people I've corresponded with have had such a hard time
getting single-density to work with disk imageing.
For my own setup, I have a couple of trusty Aopen AP5T
mainboards which do single-density nicely, installed in
cases with an internal 3.5" HD (1.44M) drive as A:, and
an external cable allowing me to connect any of the
following as drive B:
- 5.25" DD 40 track (Teac 55-G)
- 5.25" DD 80 track (Teac 55-F)
- 5.25" HD 80 track (Teac 55-G)
- 8" DS (Qumetrack 242)
I'm also quite a stickler for using the proper drive when
I read/write images - so if it's 5.25" 40 track, the 55-G
gets connected etc. This has worked very well for me.
Today I was setting up another system to do 5.25" disks
only, and I wanted it to be self contained within one case.
To minimize the drives, I decided to modify a Panasonic
JU-475 with a switch on the front bezel to force it to 300
rpm to serve as both the DD/80 and an HD/80 drive.
To verify that the drive was good, I hooked it up and tried
to read/write some disks - to my surprise I could not do
single density (at 300kbos). Tried a couple other HD drives
with the same result. On a hunch, I modded it to 300rpm, and
sure enough, I can read/write single-desnity fine at
300rpm / 250kbps.
(Yeah, I know the SD rate is really only 1/2 - by 250/300kbps,
I am referring to the MFM settings for the AT controller data
rate select register).
To rule out some odd ImageDisk quirk, I tried several versions
of TeleDisk - with the drive at 360rpm and configured as a 1.2M
HD drive, none of them could read a single-density disk either.
So at least for my AP5Ts, it appears that the internal controller
can do single or double density at 250kbps, but only double-
density at the "at compatible" 360rpm rate of 300kbps. My guess
is that the data separator does not work at that rate (works
fine at 500kbps however).
I'm curious to know if this is a charactistic specific to the
machines I am using, or if it is common among PCs that do single-
density to not work at 300kbps. Has anyone here read and/or
created single density 5.25" disk on a 1.2M HD drive spinning
at 360rpm?
Regards,
Dave
PS: I've added a note to the ImageDisk help file and manual
about this (that you may need a DD drive to do single-density),
and I've also put up a brief page describing the modifications
I made to add the speed switch to the Panasonic drive on my
images page.
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
18 Mar
18 Mar
9:43 p.m.
an external cable allowing me to connect any of the
following as drive B:
- 5.25" DD 40 track (Teac 55-G)
- 5.25" DD 80 track (Teac 55-F)
- 5.25" HD 80 track (Teac 55-G)
- 8" DS (Qumetrack 242)
That first one should read "(Teac 55-B)"
Dave
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
9:46 p.m.
Trivial typo nit-picking for clarity:
On Sun, 18 Mar 2007, Dave Dunfield wrote:
- 5.25" DD 40 track (Teac 55-G)
55-B ?
I'm also quite a stickler for using the proper
drive when
I read/write images - so if it's 5.25" 40 track, the 55-G
gets connected etc. This has worked very well for me.
55-B ?
19 Mar
19 Mar
5:20 a.m.
Trivial typo nit-picking for clarity:
On Sun, 18 Mar 2007, Dave Dunfield wrote:
Yes, in both cases, the DD 40 track drive should read "Teac 55-B"
(It's interesting how one's fingers will repeat the same typo
multiple times in a single document)
Dave
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
- 5.25" DD 40 track (Teac 55-G)
55-B ?
I'm also quite a stickler for using the
proper drive when
I read/write images - so if it's 5.25" 40 track, the 55-G
gets connected etc. This has worked very well for me.
55-B ? 
18 Mar
18 Mar
10:24 p.m.
Dave, that's very strange. I just went and tried it on one of my
"regular" systems, a box with an Intel TC430HX mobo in it. The
controller on this board is a National PC87306B (serial/parallel,
clock, floppy, IrDA and keyboard) that does just fine with FM. I
repeated on another "regular" box with an Intel PD440FX board (uses
the NS PD87307) with the same results.
In both cases, the drive was a Teac FD55GFR running at 360 RPM.
'Matter of fact, I have to be very careful with the first machine--
it's got a Toshiba 1.44M 3.5" on it that will automatically change
spindle speed if a 1.3MB 3.5" diskette is inserted without giving any
external indication.
In both machines, I coiuld format, write and read 48 and 96 tpi SD
formats in the 55GFR.
(Those intel 400-series mobos are great because they use the NS Super
I/O chip--it's 8473-compatible, which means it'll also handle writing
128-byte MFM sectors, unlike its NEC-based relatives.)
I wonder what's up with your setup?
Cheers,
Chuck
19 Mar
19 Mar
6:35 a.m.
Dave, that's very strange. I just went and tried
it on one of my
"regular" systems, a box with an Intel TC430HX mobo in it. The
controller on this board is a National PC87306B (serial/parallel,
clock, floppy, IrDA and keyboard) that does just fine with FM. I
repeated on another "regular" box with an Intel PD440FX board (uses
the NS PD87307) with the same results.
Hi Chuck,
Thanks for the response. So you have not seen this before (single-
density working at 250kbps but not 300kbps)?
The board in question is an Aopen AP5T which has an SMC FDC37C669
controller.
I just tried the same drive on another system with an NS PC87306
controller, and it does work pefectly at 300kbps.
The question is - how common is this? - I'll have to dig out some
of my other systems that do SD and test them. Problem is - most of
them only support one physical FDC so they only have a 3.5 installed.
Dave
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
12:40 p.m.
On 19 Mar 2007 at 5:35, Dave Dunfield wrote:
Thanks for the response. So you have not seen this
before (single-
density working at 250kbps but not 300kbps)?
No, but it's not a practice I've followed much. Many, if not most,
of the FM-encoded 5.25" (exclusive of the 100 tpi format) formats use
48 tpi diskettes, so we've always recommended a 48 tpi drive to do
the work. And I don't think we have a single system here with your
SMC controller, although I think I still may have the datasheet
somewhere here.
It might have happened to us at some point with a customer, but we'd
have assumed that his was one of the many controllers that don't
handle FM at all, so our advice would have been the same--find
another mahine to try. The problem may have been masked entirely by
this.
When doing your testing, remember to separate the non-FM capable
controllers from the FM-capable-but-not-at 300K ones. Don Maslin and
I did start doing a fairly exhaustive survey of LSI single-function
diskette controllers to see what could handle FM and 128-byte
sectored MFM, but that was long ago, and we didn't include the very
new multi-I/O chips.
Maybe it's time for a new census, although I wonder--many systems
made since about 1995 won't support more than a single floppy anyway--
and I suspect that floppyless desktop PCs aren't that far off from
becoming the rule (many are still shipped with a 34-pin header on the
mobo just in case, but that won't stay around, I suspect).
Cheers,
Chuck
3:49 p.m.
From: "Chuck Guzis" <cclist at
sydex.com>
---snip---
Maybe it's time for a new census, although I wonder--many systems
made since about 1995 won't support more than a single floppy anyway--
and I suspect that floppyless desktop PCs aren't that far off from
becoming the rule (many are still shipped with a 34-pin header on the
mobo just in case, but that won't stay around, I suspect).
Cheers,
Chuck
Hi Chuck
Your correct, floppies are on the way out. Media conversion will become
more of an issue in the next 10 to 15 years. Right now, I have a machine
that is one of the last laptops to have both CDROM and floppy that I use
for reading email and downloads.
None of the new ones have floppies.
I take my 3.5 floppy to a machine that can handle 5.25 and FM.
I still find it easier to do serial transfers to get to 8 inch disk but
serial
ports are on their way out as well.
Windows has just lanched Vista. How much support for drivers will there
be for this OS, if past history is any indication. I get the feeling that at
some
point, the only interface left that seems to be common is paper and
keyboard.
Just not sure what to do in the future.
Dwight
_________________________________________________________________
Get a FREE Web site, company branded e-mail and more from Microsoft Office
Live! http://clk.atdmt.com/MRT/go/mcrssaub0050001411mrt/direct/01/
4:56 p.m.
When doing your testing, remember to separate the
non-FM capable
controllers from the FM-capable-but-not-at 300K ones. Don Maslin and
I did start doing a fairly exhaustive survey of LSI single-function
diskette controllers to see what could handle FM and 128-byte
sectored MFM, but that was long ago, and we didn't include the very
new multi-I/O chips.
Do you still have the data you collected. This could be useful.
'
Maybe it's time for a new census, although I
wonder--many systems
made since about 1995 won't support more than a single floppy anyway--
and I suspect that floppyless desktop PCs aren't that far off from
becoming the rule (many are still shipped with a 34-pin header on the
mobo just in case, but that won't stay around, I suspect).
I think it is, and I plan to write some tools to make it fairly easy for
people to test these capabilities.
Modern machines are rapidly getting to the point where they will not
be useful for diskette imaging, however knowing what old machines to
look for is equally valuable - P1 class and earlier machines can be
had for "Free" - but who wants to collect a big pile of boxes and
take the time to scrounge up memory, drives misc other bits, and
get it running only to test it and find it's yet another brain-dead FDC.
If you know what to be looking for it could save a a lot fo time...
Dave
--
dave06a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.classiccmp.org/dunfield/index.html
4:48 p.m.
On 19 Mar 2007 at 15:56, dave06a at dunfield.com wrote:
Do you still have the data you collected. This could
be useful.
I'd have to go back to checking old BBS email--yeah, that's how far
back it was. Dial-in BBS, no internet. Since scrounging old PC's
with ISA diskette controllers is getting into "real antque" status, I
don't know how useful the information will be. The note that I made
on the Intel mobos was only because I had 'em and tested them. I've
also got some roughly contemporary Amptron VIA chipset boards that
don't do FM at all.
Cheers,
Chuck
7:05 a.m.
On Sun, 18 Mar 2007, Dave Dunfield wrote:
To verify that the drive was good, I hooked it up and
tried
to read/write some disks - to my surprise I could not do
single density (at 300kbos). Tried a couple other HD drives
with the same result. On a hunch, I modded it to 300rpm, and
sure enough, I can read/write single-desnity fine at
300rpm / 250kbps.
That's a known problem with many FDCs, they can't handle FM at 150 kBit/s,
only at 125 kBit/s. So either jumper your drive for dual speed (so it will
spin at 300rpm for 125/250 kBit/s and at 360rpm for 250/500 kBit/s) or
jumper it for single speed at 300rpm (therefore making a 5,25" drive look
like a 3,5" drive to the BIOS).
Christian
11:13 a.m.
Remove dan.desjardins at viewpix.com
Dan Desjardins
Director
Redactive Archive Management
1407 Willow Trail
Middleton, WI 53562
Phone: +1.608.204.9848
Fax: +1.608.204.9848
Mobile: +1.608.445.NEWS (6397)
-----Original Message-----
From: cctech-bounces at classiccmp.org [mailto:cctech-bounces at classiccmp.org]
On Behalf Of Christian Corti
Sent: Monday, March 19, 2007 6:05 AM
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: 1.2M HD / 300kbps / single-density
On Sun, 18 Mar 2007, Dave Dunfield wrote:
To verify that the drive was good, I hooked it up and
tried
to read/write some disks - to my surprise I could not do
single density (at 300kbos). Tried a couple other HD drives
with the same result. On a hunch, I modded it to 300rpm, and
sure enough, I can read/write single-desnity fine at
300rpm / 250kbps.
That's a known problem with many FDCs, they can't handle FM at 150 kBit/s,
only at 125 kBit/s. So either jumper your drive for dual speed (so it will
spin at 300rpm for 125/250 kBit/s and at 360rpm for 250/500 kBit/s) or
jumper it for single speed at 300rpm (therefore making a 5,25" drive look
like a 3,5" drive to the BIOS).
Christian
6:23 p.m.
Remove dan.desjardins at viewpix.com
How about you use the unsubscribe address in the headers instead?
--
--------------------------------- personal: http://www.armory.com/~spectre/ ---
Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com
-- UNHOLY UNION: Heavy Metal and Electronics -> Twisted Trans Sister ----------
6330
days inactive
6330
days old
12 comments
7 participants
participants (7)
-
cc＠corti-net.de -
cclist＠sydex.com -
cisin＠xenosoft.com -
dan.desjardins＠redactive.com -
dave06a＠dunfield.com -
dkelvey＠hotmail.com -
spectre＠floodgap.com