Has anyone got SCP 86-Dos Running on any Machine
Robert J. Stevens
TIA
Bob in Wisconsin
USE
treb...@execpc.com
datapackrat
Bill
"Robert J. Stevens" <ztre...@execpc.com> wrote in message
news:CrqdnSrb3Kwxv5LVnZ2dnUVZ_ovinZ2d@choiceonecommunications...
hha...@hartetec.com
get it running in SIMH/AltairZ80. We need to know what disk
controller the disk image expects, and then implement it in the
simulator. I already did CompuPro DISK1A (765-based) and several
others (Vector, N*, etc.) If it runs on a Cromemco controller (as I
think I heard somewhere) then the thing to do would be simulate the
179x and the rest of the Cromemco specifics, and get RDOS or Cromemco
CP/M working to verify the simulated hardware.
That's how I got MDOS and OASIS running. Since I didn't know anything
about those OS, I started by getting CP/M to run on the simulated
hardware, and then moved to the new OS. It's much easier to debug the
software in simulation than in real hardware...
If an 86-DOS disk image can be made (in any file format, but ImageDisk
(IMD) preferred) then rhis is all doable. I figured out the Vector
skew, etc, by looking at text files in the disk image, and writing a
small program to "fix" the skew (ie, get to 1:1) and then re-skew them
to the original. This was necessary because the images I had were
made with Dave Dunfield's CPT program running on a different version
of CP/M than the original disk was made with, and they had different
skews. So the disk image files were essentially double-skewed, which
made no sense, but was fixable after some exploration using a HEX
editor.
Also, to get it working in simulation, as long as we have the disk
controller ROM, or ROM Monitor with the bootstrap, then it is easy to
figure out where the system tracks are. Turning on some debugging
output in SIMH will show the disk seeking and sector reads. Once the
ROM bootstrap loads the secondary bootstrap off the disk, then that
can be disassembled to see what it does.
Once 86-DOS is working in simulation, you can build a "real" system
with the appropriate hardware, write a working disk image back to an
8" floppy using ImageDisk, and then it should come up on the real
hardware. If you don't have the correct real hardware, then 86-DOS
can be patched using the simulator, and run on the hardware you have.
I'd say the disk controller would have to match, but other stuff like
memory and console I/O should be easily patchable. One nice thing
about the simulator is that you can set it up to halt on unhandled I/O
accesses, so it is easy to figure out where the code is that does I/O
and modify it as necessary.
On Apr 25, 5:22 am, "datapackrat" <datapack...@example.com> wrote:
> Does anyone have a disk image for SCP 86DOS?
>
> Bill
>
> "Robert J. Stevens" <ztreb...@execpc.com> wrote in messagenews:CrqdnSrb3Kwxv5LVnZ2dnUVZ_ovinZ2d@choiceonecommunications...
>
>
>
> > Has anyone been able to get SCP's 86-Dos Running on any Machine at all
> > TIA
> > Bob in Wisconsin
> > USE
> > trebo...@execpc.com- Hide quoted text -
>
> - Show quoted text -
Al Kossow
> If a disk image for 86DOS is available, then it should be possible to
> get it running in SIMH/AltairZ80.
Have you added 8086 support to SIMH?
** Posted from http://www.teranews.com **
hha...@hartetec.com
I didn't, but Peter Schorn did. Using the CompuPro hardware
simulation (DISK1A, DISK2, Selector Channel, MDRIVE-H), it is possible
to boot CompuPro CP/M-86 on it (as well as, of course, CP/M 2.2.) The
System Support board simulation is not complete (ie, no RTC, etc) but
it is good enough to use for a console.
I got CP/M 2.2 working, then Peter added the 8086, and modelled it
after the CompuPro CPU8085/8088, and got CP/M-86 to work. The DISK1A
has bootstraps for 8085, 8088, and 68000 in ROM, and the same ROM
image is used in the DISK1A simulation.
-Howard
Barry Watzman
But I still have everything .... all hardware and software components,
stuff I originally bought from SCP in 1980 or so .... I believe that I
could get it running with about 10-20 hours of work, and if things went
"well" (nah, would never happen) maybe as little as 3-4 hours of work.
Barry Watzman
version from 0.33 to 2.0
Good? Who know, but probably (all of my other well stored 8" disks are
still readable).
I've had thoughts of selling them on E-Bay. How much do you think they
are worth?
Barry Watzman
Double Density controller. I believe that the CompuPro Disk 1 and the
Cromemco 16FDC were also supported.
Barry Watzman
similar, in many ways (e.g. has things like 8259's), to the CompuPro
System Support board, but it's not the same.
I used to run 86-DOS on a system whose CPU card was the CompuPro
8085/8088 card with the SCP CPU support board (I had a custom firmware
ROM that I no longer have to handle startup CPU and OS selection).
hha...@hartetec.com
> For 86-DOS, you would need to simulate the SCP "CPU SUPPORT BOARD"; it's
> similar, in many ways (e.g. has things like 8259's), to the CompuPro
> System Support board, but it's not the same.
>
> I used to run 86-DOS on a system whose CPU card was the CompuPro
> 8085/8088 card with the SCP CPU support board (I had a custom firmware
> ROM that I no longer have to handle startup CPU and OS selection).
>
Thnks to you Barry, the SCP "CPU Support Board" manual is scanned and
on-line. The CPU Support Board can have an EPROM on-board, so it
would be nice to read the EPROM from a real board. Which disk
controller do your disks support?
Bill
>Thnks to you Barry, the SCP "CPU Support Board" manual is scanned and
>on-line.
Where?
Is this a new scan, or the same one posted maybe 4-5 years ago?
Also - I use Firefox. When I put the mouse over links on your website,
about half of them point to your local hard drive, not your URL. Of
course, if one knows to grab the tail end and paste....but that's not
a common skill, I've found. So.....Voice Operated Joy Stick?
What you want to know about BIOSes?
Would you happen to have a complete text-file list of Phoenix
functions, even better some way to figure out which ones are
present in a given implementation? They don't always seem to
be in the same place/order.
Bill
hha...@hartetec.com
The manual is the one Barry scanned about four years ago:
Given that the manual is almost 30 years old, and probably hasn't been
revised lately, the scan is probably still good. I don't know much
about modern BIOSes, but the source for the IBM PC BIOS is in the IBM
Technical Reference Manual.
Robert J. Stevens
> about modern BIOSes, but the source for the IBM PC BIOS is in the IBM
> Technical Reference Manual.
>
I would like to get the Listing for the IBM PC BIOS
Also interested in where on the floppy the BIOS is stored on the early
Implementations of 86-DOS and SCP-DOS and MS-DOS2.0
TIA
Bob in Wisconsin
Bill
<ztre...@execpc.com> wrote:
re: IBM PC BIOS is in the IBM Technical Reference Manual.
>Does anyone have a scan of the Manual.
>I would like to get the Listing for the IBM PC BIOS
>Also interested in where on the floppy the BIOS is stored on the early
>Implementations of 86-DOS and SCP-DOS and MS-DOS2.0
That would be two DIFFERENT concepts of BIOS
The first one referenced, ie the IBM PC BIOS, is stored in a ROM
in much the same way most early CP/M machines handled the
job of presenting some way to interface to CP/M - putting the
basic hardware handling routines into a piece of hardware.
CP/M then used a CBIOS (Customized/Customizable BIOS) to
interface to the actual hardware, with this piece living on disk.
I know CP/M-86 used something similar, but can't say for sure
about the others. Maybe it's buried inside something else -
MSDOS.SYS is too small, maybe somewhere in COMMAND.COM?
Also, the degree of compatability is far higher among ms dos
computers than was the case with earlier CP/M ones.
Time to dig up a pointer to WIM's BIOS Pages.......
Here ya'll go... http://www.wimsbios.com/index.jsp
Bill
John Elliott
: That would be two DIFFERENT concepts of BIOS
: The first one referenced, ie the IBM PC BIOS, is stored in a ROM
: in much the same way most early CP/M machines handled the
: job of presenting some way to interface to CP/M - putting the
: basic hardware handling routines into a piece of hardware.
: CP/M then used a CBIOS (Customized/Customizable BIOS) to
: interface to the actual hardware, with this piece living on disk.
: I know CP/M-86 used something similar, but can't say for sure
: about the others.
MS-DOS (up to version 6) and PC-DOS use a similar structure - MSDOS.SYS /
IBMDOS.COM roughly corresponds to the BDOS, and IO.SYS / IBMBIO.COM to the
CBIOS. On a PC-compatible IO.SYS/IBMBIO.COM calls down to the ROM BIOS, but
it doesn't have to be that way. Implementations of MS-DOS exist (eg, for the
Apricot PC, or the Sanyo MBC-550) where the ROM is only used for booting,
and after that all the I/O is handled by IO.SYS, which bangs on the hardware
directly.
For ease of disambiguation, it may be as well to do like Amstrad and refer
to the software in the boot ROM as the ROS (ROM operating system) rather
than the BIOS.
--
John Elliott
Thinks: This is what a nice clean life leads to. Hmm, why did I ever lead one?
-- Bluebottle, in the Goon Show
Bill
<j...@seasip.demon.co.uk> wrote:
> For ease of disambiguation, it may be as well to do like Amstrad and refer
>to the software in the boot ROM as the ROS (ROM operating system) rather
>than the BIOS.
Just changing the subject a little further off kilter ...
wondering...has anyone here used the Linux ROM
bootloader approach? IE, there's a Linux variant that
puts almost nothing in the system ROM save for some
sort of bootloader (but I think it can be EVERY possible
bootable device) which then loads and intializes, among
other things, CPU and support chip registers normally
done from ROM. They claim boots in a couple seconds.
No more ROM programming (and virus/trojan flashes)
Everything is software ... files in/on your boot device.
The other thing we really need is non-alterable by
miscreants (virus writers) boot devices like CD
Even if something got into your 'soft' and writable
storage device(s), you could always start with a
clean, non-infected boot. Is it really that hard to do?
Several times now I've been hit, and simply stuff in
a Knopix CD, to repair enough damage to get my
ancient LitePC/Win98/hacked system running again.
Did I mention, it usually boots in under 15 seconds,
and shuts down in under three?
Bill
Barry Watzman
Indeed; I believe I was the one who scanned it.
The disks that I have are for the Tarbell double density board, but I
believe that the CompuPro Disk 1 and the Cromemco 16FDC were also
offered, and possibly [likely] others.
It's possible that there was a listing of the ROM monitor that was in
the CPU support card in some of the SCP documentation (in fact I'm
pretty sure that there was, because I modified it, which means that I
had source code). If so, I probably have it .... somewhere. [Actually,
I think that the source code may have been on the 86-DOS distribution
disks.]
Barry Watzman
that I gave it to Howard Harte to post about 4 years ago.
Barry Watzman
86-DOS did not run on a PC, it ran on S-100 systems. So any references
to "BIOS" would be to a disk-based BIOS, not to a ROM bios (on a PC or
any other machine, although there would likely be some type of "boot
code" in any machine).
Conversely, on a PC, there is a BIOS on the motherboard, and the
listings of that for the IBM-PC, PC-XT and PC-AT were published by IBM
and are in the manuals. I would assume (but don't know for a fact) that
you can find those online, but they won't help you with 86-DOS, in fact
they are completely irrelevant.
hha...@hartetec.com
So, to simuulate 86-DOS in SIMH, we are missing a few things:
1. Most importantly, an 86-DOS disk image
2. The SCP ROM (probably not critical)
3. Tarbell CP/M disk image to get the simulated disk controller
working) also not critical, but would be extremely helpful.
Any volunteers to make an 86-DOS disk image using Dave Dunfield's
ImageDisk tools?
It is amazing to me that 86-DOS has not been run in simulation given
its historical significance.
Bill
<Watzma...@neo.rr.com> wrote:
>A lot of different things are being confused here.
>
>86-DOS did not run on a PC, it ran on S-100 systems. So any references
>to "BIOS" would be to a disk-based BIOS, not to a ROM bios (on a PC or
>any other machine,
Not really accurate. First off, the bus means essentially nothing, as
software isn't gonna care where or what it is. The Seattle Computer
boards system used a ROM BIOS/Monitor that got the CPU awake
and running, along with several 'monitor' functions, really little
apps which just happen to be useful even if you don't (yet) have
disk drives attached or running.
> although there would likely be some type of "boot
>code" in any machine).
Unless it had a front panel and you wanted to key in a boot routine
Most pre-IBM PC computers light up the 'console' and ask what
drive you want to boot from, or even IF you want to boot, or to 'hit
enter' or things like that. Call this the 'boot' function - get the
boot device going. Usually, a floppy drive. Hard drives were
uncommon at that time.
But it was also common to have other functions in that ROM.
The most frequent was probably a memory test of some sort.
Anyway, the purpose of the hardware BIOS (stored in a ROM) was
also to more or less 'define' how operating systems were supposed
to talk to the actual hardware. Most had a jump table that looked
a lot like the table in CP/Ms BDOS. If you wanna do THIS, then
jump THERE (return address previously pushed onto the stack
by a CALL from within the operating system)
Bill
Barry Watzman
You also don't need Tarbell CP/M. The Tarbell DD disk controller was
just a Western Digital 179x FDC (the board could be operated in either
of 2 modes, programmed data transfer, or DMA using an onboard 8257, but
DMA and using the 8257 was optional).
What you will have to do is get a disk image, then replace the BIOS in
the disk image with one that does appropriate I/O (both console and
disk) for the simulator. Since source code for the BIOS was distributed
on the distribution disk, along with instructions for integrating a new
BIOS into the system, this should not be that difficult.
I am not volunteering to do this, but one way that the image could be
obtained:
1. Get suitable disk imaging software that run on a Zenith Z-100 under
Z-DOS (MS-DOS). Ideally, the Z-100 should have both 5" and 8" drives,
and a hard drive. Put the imaging software and MS-DOS on the hard
drive. [I suggest the Z-100 because it has a hard drive, 5.25" floppy
drive(s) and 8" floppy drive(s) and it runs a version of MS-DOS that is
media compatible with a standard PC].
2. Put a blank (formatted) MS-DOS 5.25" floppy into the Z-100's 5.25"
drive.
3. Put the SCP 86-DOS disk into the Z-100's 8" drive
4. Run the image software, image the SCP 8" diskette to an image file
on the 5.25" diskette (the SCP disks were SSSD, and the 5.25" disks are
standard PC format 360k 5.25" diskettes which are of course larger than
a SSSD 8" floppy disk).
5. Take the 5.25" disk, which is standard MS-DOS format, put it in a PC
with a 5.25" drive and the image now belongs to "the modern world" and
you can E-Mail it, post it online or whatever.
6. Again, I'm not volunteering (indeed I have some very selfish reasons
for actually not wanting this to happen), but I actually have everything
except the imaging software (and I may have that ... I do have Teledisk,
although I've never used it).
Barry Watzman
used the SCP Monitor ROM once it was booted. This is from memory, and I
could be wrong, but that is my recollection.
I think that the only thing that the monitor did was boot the disk drive
from the supported disk controller, and many of the controllers had
their own capability to do that independent of the monitor.
By the time 86-DOS came out, most "PCs" (S-100 systems) did NOT have
front panels. And in fact the only common front panels were on Altair
and IMSAI machines, and those could NOT be used in an 86-DOS system
because they were 16-bit only hardware, and limited the entire bus to
16-bit use, while 86-DOS required that the hardware support a 20-bit
address bus.
Robert J. Stevens
> Does anyone have a disk image for SCP 86DOS? Bill
>
I checked them with DUTIL and they look good even tho I can't read them
Under CP/M
If anyone who might want to make a Disk Image from them can send me a
GOOD SSSD 8" floppy I can make another copy.
Looks like its for a Cromemco system however.
datapackrat
news:48195380$0$5722$4c36...@roadrunner.com...
> Re: "the SCP "CPU Support Board" manual is scanned and on-line"
>
> Indeed; I believe I was the one who scanned it.
>
> The disks that I have are for the Tarbell double density board, but I
> believe that the CompuPro Disk 1 and the Cromemco 16FDC were also
> offered, and possibly [likely] others.
>
> It's possible that there was a listing of the ROM monitor that was in
> the CPU support card in some of the SCP documentation (in fact I'm
> pretty sure that there was, because I modified it, which means that I
> had source code). If so, I probably have it .... somewhere. [Actually,
> I think that the source code may have been on the 86-DOS distribution
> disks.]
I should check this newgroup more often...
Being an "old fogie", I started using SCP 86-Dos back at version 0.10
(serial #11). When they used 16-byte directory entries (you don't want
to know the problems people encountered when they went to32-byte
directory entries).
Yes, there was a ROM monitor on the support board. I still have the source,
and the SCP 86-Dos assembler to go with it.
The "Bios" was more like a driver, where you wrote the low-level stuff to
make the system calls (like INT 25H / INT 26H disk read/write, INT 21H
AH=1 get key, INT 21H AH=2 write character) actually talk to the hardware.
86-Dos generously allocated 1K (1024 bytes) for this. Good luck if you
needed
more...
I was running the CPU, support board, 4 16K RAM boards (very expensive...),
Cromemco 4FDC controller, parallel port keyboard and a (mumble) memory
mapped
video board.
In the years since then, I work for an anti-virus company, and have
disassembled and
analyzed several thousand computer viruses. I still have my tools, too...
So, if someone can post a disk image, I can probably get it running.
Bill
Robert J. Stevens
> "Barry Watzman" <Watzma...@neo.rr.com> wrote in message
> news:48195380$0$5722$4c36...@roadrunner.com...
>
> I should check this newgroup more often...
>
> Being an "old fogie", I started using SCP 86-Dos back at version 0.10
> I was running the CPU, support board, 4 16K RAM boards (very expensive...),
> Cromemco 4FDC controller, parallel port keyboard and a (mumble) memory
> mapped
> video board.
>
>
> So, if someone can post a disk image, I can probably get it running.
>
Would a copy of the 8" Floppy do any good.
I was able to get some good copies using DUTIL and other programs on my CompuPro 8-16 system.
I can't put the 8" to a 5.25 since I can't read the Directory.
Robert J. Stevens
1.00 dated 1981
The floppy is Skewed at four but if Some one has a program that will
read a 8" floppy as DOS and write to a CPM 8" floppy the files may be
extracted
Bob
datapackrat
> > Bill; Would a copy of the 8" Floppy do any good.
> > I was able to get some good copies using DUTIL and other programs on
> > my CompuPro 8-16 system.
> > I can't put the 8" to a 5.25 since I can't read the Directory.
> > Bob in Wisconsin
> > USE
> > treb...@execpc.com
> >
> I just re-Checked with DUF05 and the Floppy is for SCP 86-DOS Version
> 1.00 dated 1981
> The floppy is Skewed at four but if Some one has a program that will
> read a 8" floppy as DOS and write to a CPM 8" floppy the files may be
> extracted
I do not need or want a physical 8" floppy (I have no way to read it).
What I need is a dump of the floppy, with heads (assume 1), cyls, sectors,
and skew (if not 1) (you'll have to forgive me if I don't remember the media
byte mapping from the top of my head). From there, I can set up a "virtual
disk" to extract files, and should be able to get 86-Dos running from the
virtual
disk. With some luck, I should even be able to get 86-Dos running from
"modern" media, like a 3.5" floppy.
Bill