Bit of CP/M trivia needed

Dwight K. Elvey dwight.elvey at
Tue Aug 30 16:29:16 CDT 2005

>From: "Allison" <ajp166 at>
>>Subject: Re: Bit of CP/M trivia needed
>>   From: "Dwight K. Elvey" <dwight.elvey at>
>>   Date: Tue, 30 Aug 2005 10:14:15 -0700 (PDT)
>>     To: cctalk at
>>>    I don't think so. CPM was first written by Gary Kidall for the Intel
>>>MDS-800 (go read the CPM docs) and they NEVER had DMA capability. They used
>>>a 2k bit bootstrap loaded in a 1702 EPROM. Here is a picture of a MDS-800
>>>with THE actual 1702 Boot EPROM
>> Yes, what I meant to say was for the S-100. I was told that the first
>>S-100 implementation used this type of board. I knew Gary used a
>No.  Most (if not all) of the first S100 system had front pannels 
>(ALTAIR AND IMSAI Were the first!).  The few that didn't had some form 
>of rom monitor to simulate a fronpannel via TTY such as Poly-88, 
>Processor tech SOL and others.

 What does a front panel have to do with what I stated?
 I am familiar with the MDS800 having worked for
Intel and written both assembly and PLM80 for it.
I was also responsible for testing of the 1036 card
( the PLL used in the M2FM board set ). So, I do understand
the disk I/O of the MDS800. I made no claim as to
how a MDS800 works( please read previous mails ).
As I corrected, I was not intending to make a claim about
the first ever CP/M ( again please read previous mail in
context ) but felt that from the context of the conversation
that we were talking about S-100 machines. I tried to clarify
this misconception that I was talking about the MDS800 in
a previous mail ( again please read ), I was not talking about
the MDS800.
 All I'm saying is that I was told that this particular
DMA based interface was used for the first implementation
on the S-100 of CP/M. This was told me by another fellow
many years ago. I don't recall the fellows name but he
seemed at the time to be honest enough since he had nothing
to sell or gain by such a comment. He claimed to have some
involvement in that process. If you don't like that you
can just shoot me and be done with it but, please, don't
put words in my mouth.

>I have the history of being Altair owner SN00200 an early one with a 
>full set of warts and germs.  I was in the unenviable position by the end of
>January 1975 of having an assembled and debugged machine with 4k and NO
>software.  Needless to say I was very heads up for software of any kind. 
>>><>. The 1702 is mapped
>>>into and out of the memory space by the BOOT switch on the MDS front panel.
>>>Allison's description is dead on. 
>Should be with the hours spent using MDS800 and having one they are familiar.
>>>Dave Mabry is the expert of the Intel MDS
>>>and he can provide more details about how they're booted and the BIOS table
>>>is loaded. FWIW Gary had a contract to write PL/M for Intel and wrote CPM
>>>as a file handling system for use in developing PL/M. He offered Intel CPM
>>>but they didn't want it since they were working on their own OS.(BIG
>>>Mistake!) Intel's OS was released as ISIS and was only used in the Intel
>>>MDS systems (although I've heard they did have a PC version for a PC based
>>>MDS). ISIS is very crude compared to CPM.
>>>   MDS 800  <>
>>>   You might also want to take a look at 'CP/M and a brief history lesson'
>>>here;  <>. I used to have the exact
>>>quotes from Gary but unfortunately the link to them is now dead.
>Good synopsis.

More information about the cctech mailing list