Tandy 1000 5.25" External Floppy
Neil
If anyone is looking for a Tandy 1000 5.25" External Floppy it looks
like there's one on eBay:-
VINTAGE TANDY 128K Color Computer..NR!! Item # 3409606312
(I know - but take a look. That's not a Coco drive).
Neil
Stephen Bendzick
Could this be an external drive for a 1000 EX or HX? I think I remember
recently reading something, maybe on this newsgroup, about Tandy selling
external drives for only the 1000 EX and HX, no others in the 1000
series. The drive looks identical to the Teac DSDD drive (model
FD-55(?)) standard in the Tandy 1000 SX. The SX and TX and have two
exposed 5.25" drive bays, so I can't see why either would need an
external floppy drive, and I don't remember anything to indicate one
might have been offered. Most notably, I don't recall it in the list of
upgrade options at the beginning of the SX manual, not that in the TX
manual.
Stephen
Neil
"Stephen Bendzick" <step...@catholicexchange.com> wrote in message
news:3E8FCD6C...@catholicexchange.com...
Some models used an external 5.25" that hooked up to the printer
port, IIRC. I believe they were all models that only had 3.5" bays.
Neil
Mo
label left nor serial number. I've been puzzled over this one 'cause it has
a card edge plug with a ground wire and eyelet coming out of the plug. Any
ideas which system this is designed for? I've opened it up and it looks like
it gets it power through the data cable. Thanks,
Mo
"Neil" <wh...@nothere.com> wrote in message
news:phQja.15102$B54.1...@news1.telusplanet.net...
Neil
"Mo" <mo...@atariland.com> wrote in message
news:OWQja.10028$Q27.5...@twister.austin.rr.com...
> Hey, I've got an off-white 720K 3.5" external floppy drive. It's
got no
> label left nor serial number. I've been puzzled over this one
'cause it has
> a card edge plug with a ground wire and eyelet coming out of the
plug. Any
> ideas which system this is designed for? I've opened it up and it
looks like
> it gets it power through the data cable. Thanks,
Yup. That sounds like a T1000 external all right. I don't specialise
in the 1000, but I'm sure there are websites out there with more
info.
N
Rick
They also made an external 5.25" "generic" drive that could be used on
any computer with a parallel port. That drive used it's own power
supply. It could have been one of those as well, but I don't recall what
error message the driver might kick out if you tried to load the driver
with no device connected. I bought one of these on close out nearly a
decade ago. IIRC this model was meant for use with lap tops.
But some of the 1000 series computers did use a proprietary card edge
interface for an external drive as noted. If anyone has a computer
catalog circa 1991 it has a cross reference chart for all the internal &
external floppy options for all computers made up to that point. If the
particular computer had a proprietary interface for an external floppy,
that's the direction the catalog would point you to. Not to forget all
the proprietary variations for the internal drives as well... The
parallel port model was released a lot later in the product line than
the card edge drive.
Jan Vanden Bossche
<3E8FCD6C...@catholicexchange.com>...
>> If anyone is looking for a Tandy 1000 5.25" External Floppy it looks
>> like there's one on eBay:-
>> VINTAGE TANDY 128K Color Computer..NR!! Item # 3409606312
>> (I know - but take a look. That's not a Coco drive).
>
It is. I have some 5.25" and 3.5" of those luing around. But the flatcable
is also special.
>The drive looks identical to the Teac DSDD drive (model
>FD-55(?)) standard in the Tandy 1000 SX.
It is. Exactly the same one.
>Most notably, I don't recall it in the list of
>upgrade options at the beginning of the SX manual, not that in the TX
>manual.
You could equip the EX - wich had a standard 5.25" drive build in - with an
external 3.5" or 5.25". The first to have a connection towards the future
disk format, the latter to facilitate disk-copy. You could start from either
drive, pressing F3 at boot.
The HX was another beast, because you could have 2 3.5" drives in there.
These were NOT standard, because - like the later TL, SL/2 and TL/2 took
their voltage from the cable. The effect of adding such an external drive to
a 2-drive HX was that the 3rd drive became drive C: ! And that you couldn't
format it under DOS 2.11. Luckily, you could swap the boot order of the
drives.
Oh yes, there is NO support for 3.5" drives in MS-DOS 2.11, so how did Tandy
do it ? Patched it, I suppose. And you don't have to tell an EX or a HX that
there is another type of drive attached, he knows. How ? Scanning the drive
at bootup and the setting some flags in memory ?
Well, they did, but if you put a DOS 3.x-formatted 3.5" disk into a DOS 2.11
HX, He can't write to it. The other way round is no problem.
>Stephen
Jan
Starting a 1000 SX, EX or HX:
F1 Monochrome
F3 Switch drives
F4 Slow speed
Jan Vanden Bossche
>Hey, I've got an off-white 720K 3.5" external floppy drive. It's got no
>label left nor serial number. I've been puzzled over this one 'cause it has
>a card edge plug with a ground wire and eyelet coming out of the plug. Any
>ideas which system this is designed for? I've opened it up and it looks
like
>it gets it power through the data cable.
Clearly an external 3.5" drive for the Tandy 1000 EX or HX. It's a standard
720 KB drive though, you could put it in any computer. But the size is a but
odd: it's indeed as wide as any 3.5 drive, but it's higher.
>Thanks,
>Mo
Jan
Jan Vanden Bossche
>> > If anyone is looking for a Tandy 1000 5.25" External Floppy it
>> >looks like there's one on eBay:-
>> > VINTAGE TANDY 128K Color Computer..NR!! Item # 3409606312
>> > (I know - but take a look. That's not a Coco drive).
>
>Some models used an external 5.25" that hooked up to the printer
>port, IIRC.
Well, it WAS a card-edge connector, but it was NOT the printer port. It was
a custom-build Tandy floppy-disk port: less than 34 pins, and including 5 &
12 volts over the flatcable. Voltage and signals were split in the box by a
small board. No power supply, no extra power supply cable necessary.
>I believe they were all models that only had 3.5" bays.
Indeed, but that type of drive existed in size 5.25" AND 3.5", and could be
hooked up to either the 1000 EX - with one 5.25" drive internally - or 1000
HX - with one OR two drives internally.
>Neil
Jan
Stephen Bendzick
>
[snip]
>
> The HX was another beast, because you could have 2 3.5" drives in there.
> These were NOT standard, because - like the later TL, SL/2 and TL/2 took
> their voltage from the cable.
The Tandy 1000 TX also has this nonstandard arrangement of power through
the ribbon cable for its 3.5" floppy drive. Any floppy drive that can
get power through the data/control cable connector, and not a dedicated
4-pin power connector, is not a standard drive. The standard floppy
drive interface has no provision for power lines. I am guessing that
Tandy used some of the few "unused" pins for +5 and +12 volt power.
They also may have used some of the odd numbered pins, which are
normally all grounds. Notably, the floppy drive in the TX /has/ a
separate power connector, it just isn't used; it's probably just a
slightly modified standard drive.
When installing any drive that can receive power through the data cable
into a standard system (which may include the Tandy 1000 SX), it is
important to find out which wires on the drive cable carry power and
disconnect those wires (or make a special adapter). Otherwise, the
drive could short out the system power supply. Most power supplies have
protection circuits to shut them down in this event, but if yours
doesn't, this is not the best way to find out.
Stephen
Neil
> They also made an external 5.25" "generic" drive that could be used on
> any computer with a parallel port. That drive used it's own power
> supply. It could have been one of those as well, but I don't recall what
> error message the driver might kick out if you tried to load the driver
> with no device connected. I bought one of these on close out nearly a
> decade ago. IIRC this model was meant for use with lap tops.
>
> But some of the 1000 series computers did use a proprietary card edge
> interface for an external drive as noted. If anyone has a computer
> catalog circa 1991 it has a cross reference chart for all the internal &
> external floppy options for all computers made up to that point. If the
> particular computer had a proprietary interface for an external floppy,
> that's the direction the catalog would point you to. Not to forget all
> the proprietary variations for the internal drives as well... The
> parallel port model was released a lot later in the product line than
> the card edge drive.
3½” and 5¼” Floppy Drives
NEW LOW PRICE! 3½" 720K Internal Floppy Drive. Was
$139.95 in RSC-22A. 25-1075 109.95
3½” to 5¼” Adapter Bracket. For above and 25-4053.
25-1076 34.95
3½” EX/HX 720K External Floppy Drive.
25-1061 279.95
NEW LOW PRICE! 3½” 1.44MB Internal Floppy Drive.
Was $169.95 in RSC-22A. 25-4053 129.95
NEW LOW PRICE! 5¼” 360K Internal Floppy Drive. Was
$169.95 in RSC-22A. 25-1063 139.95
5¼” 360K Internal Floppy Drive.
25-4051 139.95
NEW LOW PRICE! 5¼” 1.2MB Internal Floppy Drive. Was
$169.95 in RSC-22A. 25-4050 159.95
NEW! 5¼” 1.2MB External Floppy Drive.
25-1087 299.95
Weltec 5¼” External Apple IIc Compatible Drive.
900-2221 199.95
|
|
Computer | |||||||
|
Drive Format |
1000
EX |
1000 |
1000
RL |
1000
TL/3 |
3000 |
3000
NL 4025 LX |
1000 |
1000
RLX HD |
|
3½” to 5¼” 360k |
N/A |
Internal |
N/A |
Internal |
Internal |
Internal |
N/A |
N/A |
|
5¼” 1.2MB |
N/A |
N/A |
External |
Internal |
Internal |
Internal |
External |
External |
|
3½" 720k |
Internal |
Internal |
Internal |
Internal |
Internal |
Internal |
Internal |
N/A |
|
3½" 1.44MB |
N/A |
N/A |
N/A |
Internal |
25-4053 |
Internal |
Internal |
N/A |
Larry Fosdick
This is from the '92 Catalog:-
snip...
Neil, does that catalog you have give any info on the 25-1076 adapter? I just picked
up a few 25-1075 floppy drives with the adapters attached. Looks like the adapter
splits out the power into a traditional 5.25" drive power connector, but it has TWO
34-pin card edge connectors. One is flipped 180 degrees from the other. I found the
floppy info on the Tandy web site, but they only mention the adapter without
description. I wonder if the edge connectors are different other than their
orientations? I'm going to check with an ohmeter before I try to hook it up to
anything, but if the catalog says anything, please let me know.
Thanks,
Larry
Neil
"Larry Fosdick" <larry_...@hotmail.com> wrote in message
news:js2na.228483$Zo.41860@sccrnsc03...
That's pretty much it - doesn't even say you can have two drives!
Neil
Leonard Erickson
<jan8...@yahoo.com> wrote:
>The HX was another beast, because you could have 2 3.5" drives in there.
>These were NOT standard, because - like the later TL, SL/2 and TL/2 took
>their voltage from the cable. The effect of adding such an external drive to
>a 2-drive HX was that the 3rd drive became drive C: ! And that you couldn't
>format it under DOS 2.11. Luckily, you could swap the boot order of the
>drives.
XT class systems could have *four* floppies. The hard drive was
assigned C: *or* the first free drive letter. This caused a few
problems with doing installs on an XT box with 3 drives, as the
software wanted to install to drive C:, which was a floppy.
It wasn't until the AT that the first HD was "fixed" at C:.
>Oh yes, there is NO support for 3.5" drives in MS-DOS 2.11, so how did Tandy
>do it ? Patched it, I suppose. And you don't have to tell an EX or a HX that
>there is another type of drive attached, he knows. How ? Scanning the drive
>at bootup and the setting some flags in memory ?
Drivparm is supposed to go back that far. And it's all you really need
for a 720k drive. But they probably stuck something in the BIOS.
--
Leonard Erickson (aka shadow)
sha...@krypton.rain.com <--preferred
leo...@qiclab.scn.rain.com <--last resort
Jan Vanden Bossche
><jan8...@yahoo.com> wrote:
>
>>The effect of adding such an external drive to
>>a 2-drive HX was that the 3rd drive became drive C: ! And that you
>>couldn't format it under DOS 2.11. Luckily, you could swap the boot
>>order of the drives.
>
>XT class systems could have *four* floppies.
Well, I suppose ANY computer can have four floppy drives, if you have a
controller that supports it. The HX is AFAIK the ONLY XT-class computer
where you can even consider adding a 3rd floppy drive. Most XT's had
diskcontrollers with 2 connectors and that was it. I've NEVER seen a XT with
2 diskcontrollers. But it is probably possible. And on wich IRQ that would
have been ?
>The hard drive was
>assigned C: *or* the first free drive letter.
So, If you have only one diskdrive in an XT, you HD is B: ?
Well, in a 3-drive HX you do have 4 drives: A: & B: (internal), C:
(external) and D: (ROM) !
>This caused a few
>problems with doing installs on an XT box with 3 drives, as the
>software wanted to install to drive C:, which was a floppy.
Well, if I run the standard FORMAT C: from DOS 2.11 on my HX, it says:
'cannot format fixed disk" or something.
>It wasn't until the AT that the first HD was "fixed" at C:.
Well, in the AT it was cemented into BIOS, wasn't it. But why were all XT
HD's also designed as C:, even if there wasn't a (physical) drive B: ?
>>Oh yes, there is NO support for 3.5" drives in MS-DOS 2.11, so how did
>>Tandy do it ? Patched it, I suppose. And you don't have to tell an EX or a
>>HX that there is another type of drive attached, he knows. How ? Scanning
>>the drive at bootup and the setting some flags in memory ?
>
>Drivparm is supposed to go back that far.
There is no DRIVPARM on the HX. You start the machine without ANY floppy in
it, and it can read an write without problems its own 3.5" disks.
>And it's all you really need for a 720k drive.
And if you look at the ROM drive, from where the machine actually boots,
there is no DRIVPARM there.
>But they probably stuck something in the BIOS.
Yes, I think they did. Wasn't there something like a "disk-description
table" in DOS (like the DCT 'drive code table' in LS-DOS) I think they
filled it up and whacked it in RAM at startup, depending on the setup of the
machine. In that setup, you do have to say that you have 1, 2 or 3 disk
drives. But curiously, when you add an external drive, wich could be either
3.5" or 5.25", YOU DON'T HAVE TO SAY WICH TYPE IT IS! The machine knows, but
how ? You just have to confirm that it's there.
>Leonard Erickson (aka shadow)
Jan
Frankie Zsitvay
Jan Vanden Bossche wrote:
> Well, I suppose ANY computer can have four floppy drives, if you have a
> controller that supports it. The HX is AFAIK the ONLY XT-class computer
> where you can even consider adding a 3rd floppy drive. Most XT's had
> diskcontrollers with 2 connectors and that was it. I've NEVER seen a XT with
> 2 diskcontrollers. But it is probably possible. And on wich IRQ that would
> have been ?
IBM's standard controller for the PC and XT had an internal
bus connector (for the two internal drives via the cable with a
twist) and an external bus connector using a DB37 which could
also support two drives, but with a straight through cable.
Thus, four drives on one controller.
You could put a second controller into an XT if it is one of
those high density 8 bit controllers which has its own BIOS ROM
extension, jumpered as a second controller. However, it was
probably more advantageous to put in a second hard disk
controller card so you could have another pair of hard disks if
you needed storage. In both cases, you could use IRQ 7 as long
as you didn't use the line printer spooling feature. IRQ 3 or 4
would work if you didn't need to use interrupts on the serial
ports.
> >The hard drive was
> >assigned C: *or* the first free drive letter.
>
> So, If you have only one diskdrive in an XT, you HD is B: ?
Nope - and it gets real confusing what MSDOS does depending on
what hardware you have. Part of this has to do with the system
vendor and how they specify things - my DEC Rainbow calls the
hard disk "W" (for Winchester, ironicly an IBM drive nickname)
which MSDOS maps as drive E:. Drives A: through D: are defined
as floppies, whether they are present or not. Accidentally
referring to C: when the drive isn't installed generates an
error. Keeps you on your toes.
MSDOS assigns the drive letters at boot time.
A: and B: are always the first two floppies, and B: could be
an alias for A: if you only have one drive. These are defined
by the ROM BIOS on the main logic board.
C: and D: are the primary partitions on the first two hard
drives.
E: and F: are the second pair of floppy drives, if present.
Then come drive letters for the primary partitions on the
second hard disk controller, if present. Then come drive
letters for the logical drives in the extended partition on the
first hard disk controller, then the logical drives in the
extended partition on the second hard disk controller, if
present, etc.
Last comes the drives defined by device drivers in config.sys
and autoexec.bat.
If Stacker is involved, I think your machine just explodes.
But you get the general pattern - first drives defined by the
BIOS (floppies) have dedicated letters, then drives defined by
BIOS extensions (ROM chip on adapter card) get the next letters,
then logical drives on those drives get letters, then drives
defined by loadable device drivers get letters.
> Well, in the AT it was cemented into BIOS, wasn't it. But why were all XT
> HD's also designed as C:, even if there wasn't a (physical) drive B: ?
MSDOS uses B: as an alias to A: if only one drive is
installed.
If you have a MSDOS system with a single floppy, try it. Type
"B:" at the command line prompt and it will instruct you to
remove the disk in drive A: (if one is in there) and place disk
B: into the drive.
> >>Oh yes, there is NO support for 3.5" drives in MS-DOS 2.11, so how did
> >>Tandy do it ? Patched it, I suppose. And you don't have to tell an EX or a
> >>HX that there is another type of drive attached, he knows. How ? Scanning
> >>the drive at bootup and the setting some flags in memory ?
> >
> >Drivparm is supposed to go back that far.
>
> There is no DRIVPARM on the HX. You start the machine without ANY floppy in
> it, and it can read an write without problems its own 3.5" disks.
DOS 2.11 could support any drive defined by the ROM BIOS.
That's what the ROM BIOS is for. MSDOS was used on more than
just IBM clone machines, there were S100 boxes with 8 inch DSDD
floppies (1.2 megs each) running the same MSDOS.SYS as was used
on your HX.
> Yes, I think they did. Wasn't there something like a "disk-description
> table" in DOS (like the DCT 'drive code table' in LS-DOS) I think they
> filled it up and whacked it in RAM at startup, depending on the setup of the
> machine. In that setup, you do have to say that you have 1, 2 or 3 disk
> drives. But curiously, when you add an external drive, wich could be either
> 3.5" or 5.25", YOU DON'T HAVE TO SAY WICH TYPE IT IS! The machine knows, but
> how ? You just have to confirm that it's there.
Disk parameter headers and disk parameter blocks. Ripped
straight from CP/M with some differences, mainly to accomodate
the FAT allocation method. Defined in the BIOS and used by
MSDOS to figure out where to put stuff, the kind of FAT to use,
etc.
If, by any chance, anyone can come up with the MS internal
docs that explain how to port MSDOS to different hardware
platforms, specifically how to write an IO.SYS module for any
given version of MSDOS.SYS, I'd really like to know about it.
5.25" and 3.5" low density drives have the same electronics,
so of course you don't have to tell the system what kind of
drive is there. It can figure out the track count at system
start. Home the disk head, step out 80 tracks, step back in 40,
if the head is at track 0, it's a 40 track drive. If not, then
it's 80 track. Not advised with older 8 inch drives as this can
physically jam the head carrier assembly.
-Frank
Leonard Erickson
<jan8...@yahoo.com> wrote:
>Leonard Erickson wrote
>><jan8...@yahoo.com> wrote:
>>
>>>The effect of adding such an external drive to
>>>a 2-drive HX was that the 3rd drive became drive C: ! And that you
>>>couldn't format it under DOS 2.11. Luckily, you could swap the boot
>>>order of the drives.
>>
>>XT class systems could have *four* floppies.
>
>Well, I suppose ANY computer can have four floppy drives, if you have a
>controller that supports it. The HX is AFAIK the ONLY XT-class computer
>where you can even consider adding a 3rd floppy drive. Most XT's had
>diskcontrollers with 2 connectors and that was it. I've NEVER seen a XT with
>2 diskcontrollers. But it is probably possible. And on wich IRQ that would
>have been ?
The floppy controller for the PC & XT had a second connector on the
back of the card, which could be used for a couple of external drives.
That's how we hooked up the external 3.5" drive that showed up as
drive C.
Also, it was dead easy to replace the original cable with one with 4
connectors, and no twist. Then you just set the drive select jumpers
to 0, 1, 2 & 3 for the 4 drives.
XT floppy controller support 4 drives *by default*. You just need the
right cable.
And many had that connector on the back (a 35-pin D connector, as I
recall) so you could have two internal, and two external. I've got
several of those controllers.
>>The hard drive was
>>assigned C: *or* the first free drive letter.
>
>So, If you have only one diskdrive in an XT, you HD is B: ?
Nope, because if you only have one floppy DOS (or maybe the BIOS?)
assigned B: to it as well. This was so you could do things like:
COPY A:*.* B:
and you'd get prompted to swap disks as needed. Ditto for commands
such as FORMAT, DISKCOPY, and many others. You could have a DOS disk
in the drive (as A:) and type FORMAT B:, and you'd get prompted to
change disks.
So the minimum number of drive letters *assigned* to floppies is 2.
>>This caused a few
>>problems with doing installs on an XT box with 3 drives, as the
>>software wanted to install to drive C:, which was a floppy.
>
>Well, if I run the standard FORMAT C: from DOS 2.11 on my HX, it says:
>'cannot format fixed disk" or something.
Doesn't the HX have a ROM drive that has DOS on it? That'd be C: if
it's enabled. On my TL/3 & TL/2, if I leave the ROM drive enabled and
have an HD, the ROM drive is C:, and the HD is D.
I forget what happens when I have three floppies (the TL/2 & TL/3 will
take two 3.5 and 1 5.25 floppies) and the ROM drive enabled *and* the
pair of HDs (one XT IDE using the controller on the motherboard, and
one AT IDE using the rare Tandy card that lets you use an AT IDE drive
in an XT class or T1000 class system)
Let's just say things get complicated.
>>It wasn't until the AT that the first HD was "fixed" at C:.
>
>Well, in the AT it was cemented into BIOS, wasn't it. But why were all XT
>HD's also designed as C:, even if there wasn't a (physical) drive B: ?
Because there was always a *logical* drive B:
>>>Oh yes, there is NO support for 3.5" drives in MS-DOS 2.11, so how did
>>>Tandy do it ? Patched it, I suppose. And you don't have to tell an EX or a
>>>HX that there is another type of drive attached, he knows. How ? Scanning
>>>the drive at bootup and the setting some flags in memory ?
>>
>>Drivparm is supposed to go back that far.
>
>There is no DRIVPARM on the HX. You start the machine without ANY floppy in
>it, and it can read an write without problems its own 3.5" disks.
I'm saying that DOS 2.x has the (undocumented) DRIVPARM command
*available*. And that it's all you need to support a 720k drive.
>>And it's all you really need for a 720k drive.
>
>And if you look at the ROM drive, from where the machine actually boots,
>there is no DRIVPARM there.
DRIVPARM is a built in config.sys directive in versions of DOS from
2.?? on up. Maybe DOS 1.x. I'd have to dig up the references I posted
here a month or so back.
>>But they probably stuck something in the BIOS.
>
>Yes, I think they did. Wasn't there something like a "disk-description
>table" in DOS (like the DCT 'drive code table' in LS-DOS) I think they
>filled it up and whacked it in RAM at startup, depending on the setup of the
>machine. In that setup, you do have to say that you have 1, 2 or 3 disk
>drives.
That table is what DRIVPARM modifies.
>But curiously, when you add an external drive, wich could be either
>3.5" or 5.25", YOU DON'T HAVE TO SAY WICH TYPE IT IS! The machine knows, but
>how ? You just have to confirm that it's there.
Bet you a cookie that one or more of the "unused" pins on the drive
cable is used for determining the drive type. That or one of the
"standard" signals is different between 5.25 & 3.5 drives (also
possible).
Leonard Erickson
wrote:
> MSDOS assigns the drive letters at boot time.
>
> A: and B: are always the first two floppies, and B: could be
>an alias for A: if you only have one drive. These are defined
>by the ROM BIOS on the main logic board.
>
> C: and D: are the primary partitions on the first two hard
>drives.
>
> E: and F: are the second pair of floppy drives, if present.
>
> Then come drive letters for the primary partitions on the
>second hard disk controller, if present. Then come drive
>letters for the logical drives in the extended partition on the
>first hard disk controller, then the logical drives in the
>extended partition on the second hard disk controller, if
>present, etc.
>
> Last comes the drives defined by device drivers in config.sys
>and autoexec.bat.
It gets even worse if you have certain accessories installed on a
modern system.
In Device Manager you can actually *assign* particular letters to CDs
and any other device that gets a drive letter but isn't an actual
floppy or HD partition.
I have drive J set aside for the Jaz drives on my systems. And I
assigned drive I to the Zip disk (I already had drive Z being used as
a network drive by Netware).
So it was natural to make the CD drive K:.
Than I added a USB HD, and a 6-in-1 memory card reader. The card
reader takes up *4* drive letters. And the USB drive takes up another
letter. So does my MP3 player when it is connected.
So, that's *nine* drive letters that aren't floppies or "normal" HD
partitions. And if I don't specify letters in device manager, most of
them get added in the order that their *drivers* are loaded. Which is
partly dependent on what Windows feels like. Ick.
> If Stacker is involved, I think your machine just explodes.
Don't forget mapped network drives...
> But you get the general pattern - first drives defined by the
>BIOS (floppies) have dedicated letters, then drives defined by
>BIOS extensions (ROM chip on adapter card) get the next letters,
>then logical drives on those drives get letters, then drives
>defined by loadable device drivers get letters.
Yep, and there are so *many* loadable drivers. But again, note that
even IDE/SCSI drives that *aren't* considered HDs or "normal" floppies
can be assigned drive letters as long as the HD & Floppy assignments
don't grab them first.
> If, by any chance, anyone can come up with the MS internal
>docs that explain how to port MSDOS to different hardware
>platforms, specifically how to write an IO.SYS module for any
>given version of MSDOS.SYS, I'd really like to know about it.
Source code for DR-DOS was available for a fee. I think it still is
from the new owners.
Jan Vanden Bossche
Leonard Erickson wrote in message ...
>>
>>Well, I suppose ANY computer can have four floppy drives, if you have a
>>controller that supports it.
>
>back of the card, which could be used for a couple of external drives.
>That's how we hooked up the external 3.5" drive that showed up as
>drive C.
Are you talking about the Tandy HX, or the IBM XT ?
>Also, it was dead easy to replace the original cable with one with 4
>connectors, and no twist. Then you just set the drive select jumpers
>to 0, 1, 2 & 3 for the 4 drives.
I didn't know that. Are those 4 select lines connected ? On all controllers
? Would they still be, nowadays on modern systems ?
>XT floppy controller support 4 drives *by default*. You just need the
>right cable.
Didn't know that. I thought that was a fundamental difference between the
model 4 and a PC: the ability to drive 4 diskdrives.
>And many had that connector on the back (a 35-pin D connector, as I
>recall) so you could have two internal, and two external. I've got
>several of those controllers.
Well, my IBM PC XT has such a controller. My IBM PC does not.
>>>The hard drive was
>>>assigned C: *or* the first free drive letter.
>>
>>So, If you have only one diskdrive in an XT, you HD is B: ?
What is the smiley for a sarcastic remark ? I thought I had at least the
reputation of being an old dog in this computer scene, with plenty of
computers in my collection. So I would know such a thing. And I do!
>So the minimum number of drive letters *assigned* to floppies is 2.
Where is that defined ? In DOS or BIOS ?
>>Well, if I run the standard FORMAT C: from DOS 2.11 on my HX, it says:
>>'cannot format fixed disk" or something.
>
>Doesn't the HX have a ROM drive that has DOS on it?
Yes.
>That'd be C: if it's enabled.
If there are 3 diskdrives, the ROM is D:. And you really can't FORMAT drive
C:.
>On my TL/3 & TL/2, if I leave the ROM drive enabled and
>have an HD, the ROM drive is C:, and the HD is D.
????
Could you please, please, please check that ? On my Tandy 1000 TL, the HD is
C:, the ROM is D:
>I forget what happens when I have three floppies .. and .. *and* ..
>Let's just say things get complicated.
[G]
>>But why were all XT
>>HD's also designed as C:, even if there wasn't a (physical) drive B: ?
>
>Because there was always a *logical* drive B:
OK, makes sense. On a TRS-80, a HD can be anywhere.
>>>Drivparm is supposed to go back that far.
>I'm saying that DOS 2.x has the (undocumented) DRIVPARM command
>*available*. And that it's all you need to support a 720k drive.
Hmmmm. In the invisible, simple and fixed BIOS that an XT happens to have.
>>>But they probably stuck something in the BIOS.
>>
>>Yes, I think they did. Wasn't there something like a "disk-description
>>table" in DOS (like the DCT 'drive code table' in LS-DOS) I think they
>>filled it up and whacked it in RAM at startup, depending on the setup of
the
>>machine. In that setup, you do have to say that you have 1, 2 or 3 disk
>>drives.
>
>That table is what DRIVPARM modifies.
OK. That is about the same thing that I derived, and what Frankie Zsitvay
said. So, that must be true.
(learning something every day. You're never too old to learn)
Actually, Tandy invented with the HX what IBM later re-did with the AT ? (or
was the AT before the HX ?)
>>But curiously, when you add an external drive, wich could be either
>>3.5" or 5.25", YOU DON'T HAVE TO SAY WICH TYPE IT IS! The machine knows,
but
>>how ? You just have to confirm that it's there.
>
>Bet you a cookie that one or more of the "unused" pins on the drive
>cable is used for determining the drive type. That or one of the
>"standard" signals is different between 5.25 & 3.5 drives (also
>possible).
Verdomme, you're probably right! Because the connection between the EX/HX
isn't standard, and uses a pretty complicated connector IN the external
drive. It might very well be like you say...
>Leonard Erickson (aka shadow)
Greetings from the TyRannoSaurus
Jan-80
Jan Vanden Bossche
>> controller that supports it.
>
>bus connector (for the two internal drives via the cable with a
>twist) and an external bus connector using a DB37 which could
>also support two drives, but with a straight through cable.
I just checked my (real) IBM PC: it doesn't have such a connector. Neither
does my IBM PC XT model 286 (5162). But my IBM PC XT does have such a db-37.
>Thus, four drives on one controller.
OK.
> You could
..
>> >The hard drive was
>> >assigned C: *or* the first free drive letter.
>>
>> So, If you have only one diskdrive in an XT, you HD is B: ?
Apparently, nobody saw the sarcasm in this question.
> Nope - and it gets real confusing [..]
>
> MSDOS assigns the drive letters at boot time.
>
> A: and B: are always the first two floppies, and B: could be
>an alias for A: if you only have one drive. These are defined
>by the ROM BIOS on the main logic board.
What ROM BIOS ? We're talking XT-class computers here. So would there also
be an invisible BIOS for XT-computers ? That could explain the behaviour of
the HX: a combination of programmed hardware-detection and setup program
might correct the internal BIOS behaviour. (see the possible disk
combinations below)
> C: and D: are the primary partitions on the first two hard
>drives.
>
> E: and F: are the second pair of floppy drives, if present.
Hmmmmm.
Ah well, since I have never seen an XT with more than 2 drives - save the
HX - I can't confirm nor deny that. So I suppose you're right.
> Then come drive letters [..]
>
> Last comes the drives defined by device drivers in config.sys
>and autoexec.bat.
Pffff.
> If Stacker is involved, I think your machine just explodes.
Did anybody EVER have such a complicated machine ?
> But you get the general pattern - first drives defined by the
>BIOS (floppies) have dedicated letters, then drives defined by
>BIOS extensions (ROM chip on adapter card) get the next letters,
>then logical drives on those drives get letters, then drives
>defined by loadable device drivers get letters.
Sounds correct.
>> Well, in the AT it was cemented into BIOS, wasn't it. But why were all XT
>> HD's also designed as C:, even if there wasn't a (physical) drive B: ?
>
> MSDOS uses B: as an alias to A: if only one drive is
>installed.
Yes, I know ...
> If you have a MSDOS system with a single floppy, try it.
I have plenty of those, thank you, I know. I have a collection of over 50
machines, most of them Tandy/TRS-80, and a lot of those are MS-DOS machines.
ANd I am an IT professional, with a lot of MS/PC-DOS experience.
>> >>Oh yes, there is NO support for 3.5" drives in MS-DOS 2.11, so how did
>> >>Tandy do it ?
>> There is no DRIVPARM on the HX. You start the machine without ANY floppy
in
>> it, and it can read an write without problems its own 3.5" disks.
>
> DOS 2.11 could support any drive defined by the ROM BIOS.
>That's what the ROM BIOS is for.
Well, Tandy MS-DOS 2.11 from the HX doesn't run on any other XT ...
>> But curiously, when you add an external drive, wich could be either
>> 3.5" or 5.25", YOU DON'T HAVE TO SAY WICH TYPE IT IS! The machine knows,
but
>> how ?
>
> Disk parameter headers and disk parameter blocks. Ripped
>straight from CP/M with some differences, mainly to accomodate
>the FAT allocation method. Defined in the BIOS and used by
>MSDOS to figure out where to put stuff, the kind of FAT to use,
Doesn't answer the above question. Or I didn't understand it.
> 5.25" and 3.5" low density drives have the same electronics,
>so of course you don't have to tell the system what kind of
>drive is there. It can figure out the track count at system
>start. Home the disk head, step out 80 tracks, step back in 40,
>if the head is at track 0, it's a 40 track drive. If not, then
>it's 80 track.
This does, I think. Still, a great bit of programming on the HX. You could
have the following possible disk-drive combinations on the HX.
1. internal diskdrive 3.5" A:
Romdrive C:
2. internal diskdrive 3.5" A:
external diskdrive 3.5" B:
Romdrive C:
3. internal diskdrive 3.5" A:
external diskdrive 5.25" B:
Romdrive C:
4. internal diskdrive 3.5" A:
internal diskdrive 3.5" B:
Romdrive C:
5. internal diskdrive 3.5" A:
internal diskdrive 3.5" B:
external diskdrive 3.5" C: [*]
Romdrive D:
6. internal diskdrive 3.5" A:
internal diskdrive 3.5" B:
external diskdrive 5.25" C: [*]
Romdrive D:
With these combinations, you can't format C:
>Not advised with older 8 inch drives as this can
>physically jam the head carrier assembly.
Ah yes, the credo "software can't damage hardware" is a myth propagated by
people that don't know a lot about computers.
> -Frank
Frankie Zsitvay
Jan Vanden Bossche wrote:
> >> >The hard drive was
> >> >assigned C: *or* the first free drive letter.
> >>
> >> So, If you have only one diskdrive in an XT, you HD is B: ?
>
> Apparently, nobody saw the sarcasm in this question.
Sarcastic smiley = ;)
Or something like that. The comma of the semicolon to denote
a wink.
But sarcasm is kinda difficult to express through straight
text, so my apologies for not seeing it.
> > MSDOS assigns the drive letters at boot time.
> >
> > A: and B: are always the first two floppies, and B: could be
> >an alias for A: if you only have one drive. These are defined
> >by the ROM BIOS on the main logic board.
>
> What ROM BIOS ? We're talking XT-class computers here. So would there also
> be an invisible BIOS for XT-computers ? That could explain the behaviour of
> the HX: a combination of programmed hardware-detection and setup program
> might correct the internal BIOS behaviour. (see the possible disk
> combinations below)
Ok, I think you are confusing the motherboard setup program
you see on modern machines with the BIOS. The two aren't the
same, and a little history lesson would probably help shed some
light on why things are today the way they are.
When you turn on your computer, the microprocessor needs
instructions to tell it what to do. These instructions are
kept on a chip known as a Read Only Memory, or a ROM. Hobbyist
and home machines like the TRS-80 Model 1 had the program for
the BASIC language in this chip, while business machines like
the TRS-80 model II had a short program that clears the screen,
runs a few simple hardware tests to verify that everything is
working, and then loaded the first sector of the disk into
memory and executes it.
IBM (with some help from Microsoft) chose to do both of these
things with the IBM PC. The IBM ROM consists of three parts -
the Power On Self Test (POST), the Basic Input Output System
(BIOS) and the BASIC interpreter.
POST is responsible for checking the hardware and reporting
any faults. This part does the lengthy memory test, checks to
see if the keyboard is unplugged and tells you to press the F1
key on it to continue, beeps in a morse-code like pattern if
things are so badly screwed up that it can't even display
anything, etc.
The BIOS (also called the ROM BIOS because it is in ROM) is a
collection of subroutines that directly control the hardware in
response to requested from other programs. The idea was that
other programs wouldn't have to know how to control the
hardware. Those programs call the BIOS, and the BIOS does the
work. The BIOS in the PC and XT was configured through a group
of switches on the main logic board. The first subroutine of
the BIOS was called by the end of the POST to set up the
operation of the computer by reading these switches. The data
that defines how the machine is set up is located starting at
address 0000:0400H, and this data is written by this first
subroutine.
Shortly after the PC was introduced, a revision of the BIOS
was created which provided a way for additional devices to be
integrated into the BIOS in the ROM. The first subroutine was
modified so that it would scan the memory region between
C000:0000H and E000:0000H looking for a two byte signature every
2048 bytes. If this signature was found, some tests would be
run to determine if it was a valid program ROM chip, and if it
was, then control would be passed to it.
This program ROM would have code that integrated the operation
of the board it was on with the BIOS on the main logic board, so
that programs wouldn't have to deal with two different BIOS
ROMs. A call to the BIOS ROM on the main logic board would be
intercepted by the ROM on the add-in board, and if this call was
for the add-in board, it would handle the request. If the call
wasn't for the add-in board, then the call was redirected back
to the BIOS ROM on the main logic board, and handled there.
The hard disk of the IBM XT was integrated into the operation of
the BIOS this way, and such ROM chips are known as BIOS
extensions.
The BASIC interpreter was provided because the original IBM PC
was a cassette tape based computer, with the disk drives as an
option. When the IBM XT was released, the cassette tape port
was dropped, but the BASIC interpreter remained as the disk
version of BASIC used the BASIC in ROM for the bulk of its code.
Fast forward to the IBM AT, and IBMs desire to get rid of
setting up the machine using hardware switches. IBM shipped a
program disk with each AT machine that was used to set up the
machine, with the machine configuration stored in the clock chip
on the main logic board, which had a battery to keep it running
when power is off. Some early clones of the IBM AT also did
this, but it didn't take long for them to realise that, because
they didn't have the Microsoft cassette BASIC interpreter in
ROM, they had a place where the setup program could be kept.
All they had to do was put a bigger ROM chip on the board.
Beginning with the IBM AT, a hard drive was standard
equipment, so the BIOS code to control it was included with the
BIOS ROM on the motherboard. This made the combination
hard/floppy controller board a little cheaper to produce.
However, you still find the BIOS extension system at work
integrating display adapters, network interface cards, and SCSI
cards.
Eventually IBM stopped putting the Microsoft cassette BASIC
interpreter into their machines and put the machine setup
program in there instead. It is this setup program that most
people seem to think of as the BIOS of the machine, when in
reality it is just the program that sets various parameters that
the BIOS uses to run the machine. On the PC and XT it was done
with hardware switches, on the AT and PS/2 machines it was done
with a program loaded from disk, and on most 286 and later
clones, the setup program was built in.
With the memory management capabilities of the 386 and later
chips, the setup program was no longer limited to the 128K space
allocated in the AT and earlier machines, so today these
programs can be quite involved.
> > C: and D: are the primary partitions on the first two hard
> >drives.
> >
> > E: and F: are the second pair of floppy drives, if present.
> > Then come drive letters [..]
> >
> > Last comes the drives defined by device drivers in config.sys
> >and autoexec.bat.
> >
> > If Stacker is involved, I think your machine just explodes.
>
> Did anybody EVER have such a complicated machine ?
I once had an XT clone with two hard disk controllers and four
ST-4051 drives. It was contained in three discarded IBM XT
cases, the top one holding the computer and floppies, the bottom
two cases holding two drives each. These were full height
drives. I also had Doublespace loaded when I got DOS 6.2. The
floppies were A: and B:, the compressed volumes were C:, D:, E:,
and F:, and the host drives for those compressed volumes were
H:, I:, J:, and K:. When running Uniform (a utility for reading
and writing to CP/M disks) I had a drive G: as well, which
referred to a CP/M disk in drive A:.
Before that I was running DOS 3.3, which was limited to 32
megabyte partitions. Each drive was partitioned into two
logical drives, so I had floppies A: and B:, and the first hard
drive was C: and G:, the second was D: and H:, the third was E:
and I:, and the fourth was F: and J:. Uniform added drive K:,
again referring to a CP/M disk in drive A:.
Chances are I was not the only one to do such a thing. Jerry
Pournelle had a Compupro box running DOS 2.xx on a hard drive
system that was something like 450 megabytes total. He used
every letter in the alphabet.
> > DOS 2.11 could support any drive defined by the ROM BIOS.
> >That's what the ROM BIOS is for.
>
> Well, Tandy MS-DOS 2.11 from the HX doesn't run on any other XT ...
To be expected - remember, there are two parts to MSDOS,
MSDOS.SYS and IO.SYS. MSDOS.SYS is the same for all machines
running that version of DOS, while IO.SYS was provided by the
manufacturer of the machine and generally was specific to that
machine. Of course, if it is in ROM, I don't know how you would
be able to try it on another machine, unless it can SYS a floppy
from ROM, and knowing Microsoft and Radio Shack, they probably
wouldn't have allowed that.
> 1. internal diskdrive 3.5" A:
> Romdrive C:
> 2. internal diskdrive 3.5" A:
> external diskdrive 3.5" B:
> Romdrive C:
> 3. internal diskdrive 3.5" A:
> external diskdrive 5.25" B:
> Romdrive C:
> 4. internal diskdrive 3.5" A:
> internal diskdrive 3.5" B:
> Romdrive C:
> 5. internal diskdrive 3.5" A:
> internal diskdrive 3.5" B:
> external diskdrive 3.5" C: [*]
> Romdrive D:
> 6. internal diskdrive 3.5" A:
> internal diskdrive 3.5" B:
> external diskdrive 5.25" C: [*]
> Romdrive D:
>
> With these combinations, you can't format C:
Probably a limitation of the format program supplied with that
version of DOS. If you want to experiment, try downloading a
public domain format program from Simtel and it probably will be
able to format the external floppy drive. Most such format
programs don't use the BIOS at all, preferring to talk straight
to the disk controller chip.
I always thought the ROMdrive was a pretty neat idea. That
way you can start up and use the disk capability of the machine
even if you didn't have a disk with the operating system on it.
Definately beats having cassette BASIC in ROM. As long as there
is a way to disable it so you can boot a later version of the
operating system, I can't see how it would get in the way.
Alas, with the advent of cheap hard drives, it was one of those
good ideas that didn't see widespread use.
-Frankie
Leonard Erickson
<jan8...@yahoo.com> wrote:
>Leonard Erickson wrote in message ...
>>>
>>>Well, I suppose ANY computer can have four floppy drives, if you have a
>>>controller that supports it.
>>
>>The floppy controller for the PC & XT had a second connector on the
>>back of the card, which could be used for a couple of external drives.
>>That's how we hooked up the external 3.5" drive that showed up as
>>drive C.
>
>
>Are you talking about the Tandy HX, or the IBM XT ?
PC & XT refers to the IBM PC & IBM XT. And the floppy controller cards
were usable in any clone they'd fit into.
>>Also, it was dead easy to replace the original cable with one with 4
>>connectors, and no twist. Then you just set the drive select jumpers
>>to 0, 1, 2 & 3 for the 4 drives.
>
>I didn't know that. Are those 4 select lines connected ? On all controllers
>? Would they still be, nowadays on modern systems ?
No idea. The *drives* used to support them. I'm not sure about some of
the newer drives where setting them for the right select lines could
be a problem.
>>XT floppy controller support 4 drives *by default*. You just need the
>>right cable.
>
>Didn't know that. I thought that was a fundamental difference between the
>model 4 and a PC: the ability to drive 4 diskdrives.
>
>>And many had that connector on the back (a 35-pin D connector, as I
>>recall) so you could have two internal, and two external. I've got
>>several of those controllers.
>
>Well, my IBM PC XT has such a controller. My IBM PC does not.
But you could put the controller card from the XT into the PC and it'd
work.
>>So the minimum number of drive letters *assigned* to floppies is 2.
>
>Where is that defined ? In DOS or BIOS ?
Not sure.
>>>Well, if I run the standard FORMAT C: from DOS 2.11 on my HX, it says:
>>>'cannot format fixed disk" or something.
>
>>Doesn't the HX have a ROM drive that has DOS on it?
>
>Yes.
>
>>That'd be C: if it's enabled.
>
>If there are 3 diskdrives, the ROM is D:. And you really can't FORMAT drive
>C:.
You can if you have a Format program that isn't broken. :-)
>>On my TL/3 & TL/2, if I leave the ROM drive enabled and
>>have an HD, the ROM drive is C:, and the HD is D.
>
>????
>Could you please, please, please check that ? On my Tandy 1000 TL, the HD is
>C:, the ROM is D:
I can't check the TL/3 at the moment, because I blew the keyboard
"fuse".
And the TL/2 is buried, but that's how I recall it.
It's possible to have the ROM enabled, but boot off the HD. And I
think that changes things. The ROM is Drive C if you are booting off
it (or so I recall)
>>>>Drivparm is supposed to go back that far.
>
>
>>I'm saying that DOS 2.x has the (undocumented) DRIVPARM command
>>*available*. And that it's all you need to support a 720k drive.
>
>
>Hmmmm. In the invisible, simple and fixed BIOS that an XT happens to have.
No, it's in *DOS*, not the BIOS.
The drive table isn't in the BIOS, though the default one may be
getting loaded into RAM *from* the BIOS ROM.