Olympia Olytext 20 OS question.

From: Dale F Applegate <daleapple_at_jps.net>
Date: Tue Dec 21 15:53:20 1999

-----Original Message-----
From: Dwight Elvey <elvey_at_hal.com>
To: Discussion re-collecting of classic computers
<classiccmp_at_u.washington.edu>
Date: Tuesday, December 21, 1999 11:13 AM
Subject: RE: Olympia Olytext 20 OS question.


>"Ernest" <ernestls_at_home.com> wrote:
>> Hi Don. I may be able to get a copy of CP/M-80 but it won't be for my
>> specific system. That's the problem. I don't have the OS for it. That's
why
>> I was curious about how difficult it would be to set up a generic copy of
>> CP/M to run on it. I've never assembled (?) CP/M for a computer, but I've
>> been reading a how to on doing it, and interested in giving it a shot.
I'm
>> not even sure of how to load/run the system monitor to ID the various
>> address' but the thing is, as with anything, there's always a first time,
>> and you have to start somewhere.
>>
>
>Hi Ernest
> Most of CP/M is generic but the part called BDOS has to
>be specifically written to work on your computer. This
>part contains both the disk, printer, punch and serial I/O.
>The most important are the disk and serial. Usually the
>serial is easy because there were only a limited number
>of serial chips used. You don't have to make the serial
>interrupt driven. It is easier to make it a simple polled
>system at first. I've left mine as a polled system because
>I see no need to setup serial when I'm the only user.
> The disk I/O is a little more difficult. You'll need
>to know how the disk interface works. Most of the newer
>disk interfaces use one of the Western Digital series
>of chips. The reading and writing of sectors is relatively
>easy. You'll still need to have some understanding of
>disk formats and how the sectors are addressed.
> If you can find examples of code for the interfaces you
>have, 95% of the battle is won. You'll need a way of getting
>the code into the memory and then to the disk of your machine
>or someone else with a running machine can create the
>disk. The IMSAIs have the advantage that they have front
>panel switches. The machine may have a ROM based monitor
>that can be handy. If not, you'll need another machine
>to help boot strap your machine. The other machine doesn't
>need to be the same type, it only needs to be able to
>write the same disk format.
> I brought an old IMSAI back to life by doing a BDOS for
>it. It had an AY-3-1015 serial chip that was typical
>of the older machines. These are generally easy to work
>with since there is no software initialization ( other
>than clearing power up trash ) that needs to be done.
>The 8251 and 8250 type parts require initialization
>but they are also more versatile. The disk interface, on
>my machine, was an early TTL interface that had DMA to the
>machines memory. This took me quite a while to figure out
>and it wasn't until I found a schematic that I was able to
>figure it out ( it also was broken ). Luckily for you, most
>of the newer interfaces use standard chips and are easier to
>deal with. In many cases, you can make progress without
>schematics and only the chip manufactures spec sheet.
> It is also possible that many of the functions you need
>are already in ROMs on your machine.
>Dwight
>
>
Received on Tue Dec 21 1999 - 15:53:20 GMT