Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

A S100 CP/M system resurrected to life ....

29 views
Skip to first unread message

Michael

unread,
Mar 12, 2009, 8:01:59 AM3/12/09
to
Here is my contribution to the S-100 world.

http://www.michael-george-hart.com/computerscience/imsai8080-8088.html

Let me know if there is anything I can do to improve the site.

I hope the read is as enjoyable as my efforts in getting the system
functional.

Michael

John Crane

unread,
Mar 12, 2009, 6:54:37 PM3/12/09
to

"Michael" <michael.g...@gmail.com> wrote in message
news:d2d2cc84-eaac-49c6...@j8g2000yql.googlegroups.com...

Very, very nice! A top-notch system and I'm sure it represents a lot of
hard work.

-John


Bruce Morgen

unread,
Mar 12, 2009, 9:08:32 PM3/12/09
to
Michael <michael.g...@gmail.com> wrote:

A great read, "coding"
errors in that obscure and
inconsistent language,
"English," notwithstanding.

Good work, Michael -- most
enjoyable!

Bill H

unread,
Mar 13, 2009, 6:41:51 AM3/13/09
to

One of the better reads I have seen recently

Bill H

Michael

unread,
Mar 13, 2009, 1:24:15 PM3/13/09
to
On Mar 12, 9:08 pm, Bruce Morgen <edi...@juno.com> wrote:

My apology with regards to the English. I did as a few friends to
review the document before publishing. They like it, but I can only
guess that their abilities at written English may not be much better
than mines.

By the end of next week when the article is not fresh in my head I
should be able to do a better job of proofing it.

I have managed to find parts to build another system that will be
based on a 80286 CPU. It is my intention to port LINUX to that system,
thereby showing how anyone how to get LINUX running on any S-100/
IEEE-696 system.

Perhaps you will be a proof reader before publication of that article
to the web.

gri...@foni.net

unread,
Mar 14, 2009, 6:03:36 AM3/14/09
to
Quoth Michael <michael.g...@gmail.com>:

> I have managed to find parts to build another system that will be
> based on a 80286 CPU. It is my intention to port LINUX to that system,
> thereby showing how anyone how to get LINUX running on any S-100/
> IEEE-696 system.

You might want to try porting Minix first - it supports the 80286 while
Linux needs at least an 80386.

Just my $.02
Jens
--
You can't out-sarcasm reality.

Michael

unread,
Mar 14, 2009, 6:43:08 AM3/14/09
to
> You might want to try porting Minix first - it supports the 80286 while
> Linux needs at least an 80386.

I know that this is not the proper place to start such a discussion.
But here goes.

I think I still have the very first release of Minix on 5" Floppies if
anyone out there should need them; and I do mean need then and not to
sell on Ebay or the like. I have had them since the mid 1980's so I
don't know if the media is still good.

I like MINIX and consider it a good OS. But from a professional point
of view porting LINUX will be better for me getting contracts and
having fun at the same time. There seems to lots of embedded type
LINUX work being offered to those who know

My current understanding of LINUX works on any 16 bit architecture.
The following link details this fact.
http://en.wikipedia.org/wiki/Linux_kernel_portability_and_supported_architectures#List_of_supported_architectures

What would be real show the world type project would be the porting of
LINUX to a 8 bit architecture.
That a challenge that I personally will not be taking up anytime soon
unless I have nothing better to do.

Michasel

All...@localhost.net

unread,
Mar 14, 2009, 12:02:05 PM3/14/09
to

You might want to look up UZI unix as that was a implementation for
8bitters.

Generally putting *nix or Linux on 8bitters is a problem of minimal
resources such as only 64k of ram unless there is a banking scheme or
Z180/Z280/Ez80 that has a memory extension.

Then again there is uClinux but from my expereince with it on ARM and
Blackfin getting it into a tiny space is problematic.

Allison

Bruce Morgen

unread,
Mar 14, 2009, 3:09:39 PM3/14/09
to
All...@localhost.net wrote:

Michael intends to port it to
a 286, which should be quite
doable with a little work.

Bruce Morgen

unread,
Mar 14, 2009, 3:11:30 PM3/14/09
to
Michael <michael.g...@gmail.com> wrote:

If it's not too lengthy,
that'd be my pleasure!

Dave Griffith

unread,
Mar 15, 2009, 5:10:07 PM3/15/09
to
gri...@foni.net wrote:
> Quoth Michael <michael.g...@gmail.com>:
>> I have managed to find parts to build another system that will be
>> based on a 80286 CPU. It is my intention to port LINUX to that system,
>> thereby showing how anyone how to get LINUX running on any S-100/
>> IEEE-696 system.

> You might want to try porting Minix first - it supports the 80286 while
> Linux needs at least an 80386.

It might be more interesting to use a Motorola 68K CPU for that.
CompuPro made a very nice one that optionally included an MMU.


--
David Griffith
dgr...@cs.csbuak.edu <-- Switch the 'b' and 'u'

gri...@foni.net

unread,
Mar 15, 2009, 6:58:25 PM3/15/09
to
Quoth Dave Griffith <dgr...@cs.csbuak.edu>:
[80286 system, Minix or Linux?]

> It might be more interesting to use a Motorola 68K CPU for that.
> CompuPro made a very nice one that optionally included an MMU.

Even more reasons for building a 68K-based board for the ECB
system I'm resurrecting at the moment - the 6809-SBC by Andrew
Lynch is designed around the same basic idea: to use a Z80-based
system as a host for other CPU cards, basically like the TRS-80
Model 16. I've worked out a simple design for data transfer
between the CPUs; now I'm sketching ideas for the protocol.

Michael

unread,
Mar 16, 2009, 3:54:51 AM3/16/09
to
On Mar 15, 5:10 pm, dgri...@cs.csbuak.edu (Dave Griffith) wrote:
> grif...@foni.net wrote:
> > Quoth Michael <michael.george.h...@gmail.com>:

> >> I have managed to find parts to build another system that will be
> >> based on a 80286 CPU. It is my intention to port LINUX to that system,
> >> thereby showing how anyone how to get LINUX running on any S-100/
> >> IEEE-696 system.
> > You might want to try porting Minix first - it supports the 80286 while
> > Linux needs at least an 80386.
>
> It might be more interesting to use a Motorola 68K CPU for that.  
> CompuPro made a very nice one that optionally included an MMU.
>
> --
> David Griffith
> dgri...@cs.csbuak.edu  <-- Switch the 'b' and 'u'
I very much wanted to use a 68K board, but those are hard to come by.
So I use the best I can currently get. If I happen to find a 68K or
better yes 386 board while I am
ramping up for the project then I will switch over.

Anyone has a 68K or a 386 IEEE 696 card they want to give up for a
good price?

The people that I know who actually have them and not even using them
for anything
just like keeping those types of boards around.

Michael

unread,
Mar 16, 2009, 4:01:24 PM3/16/09
to
Richard P.
The answer is yes and I agree to all your terms.

Also could you double check your e-mail address for e-mails I sent to
you from
my yahoo and gmail accounts give "Mail Delivery Status Notification
(Failure)
I have tried send you e-mail several times today.

Anyway, with that I though you should know how rare your board is.
So here is a story for you and the group.

I am certian the board is is the Macrotech satellite board. It is not
a s-100 perminant
bus master. It should have a full megabyte of 32 bit wide dual ported
100 ns
4-way interleaved dynamic RAM, a 16Mhz 80386 and a socket for an
optional
80387 math coprocessor.

The reason I know of this board is I worked with Macrotech testing
and
eventually writing an article for the Spring 1989 issue of
"Supermicro
Journal of VME, MULTIBUS, and S-100 systems" titled "RUN MS-DOS AS AN
S-100
SATELLITE TASK"

In those days I had a LOMAS DATA Thunder 186 board that very much a
single
board PC that you could place into your S-100 system. It had
connectors for
VGA, floppy and keyboard along with if I recall correctly  512K static
RAM.
Able to boot CCDOS and MS-DOS. At the time board was state of the
art.
That board claim to fame, at least the reason why I bought it, was it
passed
the IBM PC compatibability test, which at that time was the ability to
run MS
Flight Simulator. I had this board in my very first IMSAI-8080 with
modified front panel.
Been looking for another one for many year now.

Getting back to your board The first set of MI386S worked flawlessly.
I initially had one
of those and base my article and software for the article on that
board. For some reason
late in the writing of the article Macrotech had me exchange that
board for another MI386S
with a different revision number. In retrospect It was probably
because they no longer knew
how to make the newer releases of the MI386S function reliably. It was
my understanding the
engineer who originally designed the board had left the company. At
the time I speculated he
took with him some nuances about what was need in all those PAL's on
the MI386S board
to make if function reliably. They need that original board to reverse
engineer the PAL's. I am
only guessing here.

The new board had a number glitches and after them sending me several
replacement PAL that they hoped would have fixed the boards problem I
think
they realized they had a serious reliability issue with the MI386S.

I completed the article and it was published in the spring 1989 issue
of
Supermicro.

Initially I could have purchased the board at at a significant
discount
because of my article.  I never bothered since at that time the
boards
I had never  functioned reliably after the swap out with the initial
MI386S
given to me for the article. I just assumed by the time the article
was published
they would have the board released to the public at large functional
and reliable.


After the article was publish I got a call phone call from who I think
was the
owner of LOMAS DATA PRODUCT. She told me that a number of clients were
having
difficulty with their MI386S and they made a point of citing my
article to the fact that
their MI386S board should also have worked with their LOMAS DATA
PRODUCT boards.

Apparently Macrotech were selling boards with the same issues I was
having when writing the
article. It seemed that no one at the company had stressed tested
board in the CCDOS environment.

At the time they were pushing idea that the board worked great in a
Concurent DOS environment.
I guess lot of there customer were disagreeing with them.

Anyway, I told the owner of LOMAS DATA of my experience with the
board, which
must of spread like wild fire and killed the product and I think
Macrotech
for that matter since I never saw a whole page add having to do with
the
Macrotech 386 Satellite board again and for that matter any Macrotech
product
after 1989.

This happened many years ago. I am only telling the story as best as I
can
remember it. But has always bothered me that I may have killed a
company even
if they were selling a product that simple did not work reliably.

So if you actually have a MI386S board that actually works you do have
one of
the rarest of IEEE-696 boards around. But if you don't have the
documentation
that maybe rarer yet. :-)

Anyway send me an e-mail with your updated e-mail address.

Later
Michael.

Dave Griffith

unread,
Mar 18, 2009, 12:27:51 AM3/18/09
to
Michael <michael.g...@gmail.com> wrote:
> I very much wanted to use a 68K board, but those are hard to come by.
> So I use the best I can currently get. If I happen to find a 68K or
> better yes 386 board while I am
> ramping up for the project then I will switch over.

> Anyone has a 68K or a 386 IEEE 696 card they want to give up for a
> good price?

> The people that I know who actually have them and not even using them
> for anything just like keeping those types of boards around.

I have one for my own use. Once in a while I do come across them in the
stash of stuff I got from Ridgecrest.

--
David Griffith
dgr...@cs.csbuak.edu <-- Switch the 'b' and 'u'

0 new messages