On 2019-10-23, Jim Wilkins wrote:
"Clare Snyder" wrote in message
news
On Wed, 23 Oct 2019 08:48:33 -0700, Bob La Londe
wrote:
On 10/23/2019 6:33 AM, Jim Wilkins wrote:
"Terry Coombs" wrote in message
...
On 10/23/2019 6:34 AM, Jim Wilkins wrote:
"Steve W." wrote in message
...
...
My first was a 1984 RatShack CoCo2 with ADOS, DSDD drives, and OS7
as
well as the "C" compiler
Or you sure that wasn't OS-9? The one by Microware which made
the MC-6809 into a multi-user multi-tasking system (if you has additional
serial ports.)
I used it (along with DOS-69 on a SWTP 6809 system, after using
the MC-6800 on both an Altair 680b and a SWTP 6800. The latter made it
easier to wire-wrap custom interfaces -- and mechanically/electrically
compatible with the SWTP 6809).
I preferred OS-9, once I had it, but started with SSB'sj DOS-68
and DOS-69 for the 6809 system.
- I added an e-prom burner and a composite
video output to replace the RF output. Still have it (and the 1984
MC10 portable version).
Next was a Sanyo 550 which I upgraded to IBM compatible video - but
it
could not handle enough RAM for Lotus so I built my first PC XT
clone
What did you think of the CoCo?
They got a lot out of very little hardware. But the bit-banger
serial interface was terrible -- toss in some MC-6850 serial port chips
and it got a lot more usable for multi-user operation.
I was very impressed with the 6809's powerful instruction set, after
struggling to turn a homebrew wire-wrapped 8080 machine into something
resembling the IBM PC. The 8080 lacks relative jumps and is more
suited to embedded control than general purpose computing with
loadable programs. It provided good training in computer hardware
design, though.
the MC-6800 was pretty good already, but Motorola did the right
thing with the MC-6809, tossing away backward compatibility -- though
they made an assembler which could convert MC-6800 code to MC-6809 code,
as long as you stayed clear of things playing with analyzing the stack,
as the monitor (EP)ROM did. They (SSB) had a really nice later monitor
EPROM -- while the SWTP 6800 used Motorola's MIKBUG as the monitor ROM,
and tricky addressing for the early SSB floppy controller card EPROMs to
allow it to be interleaved with the I/O address space.
The company was right at the leading edge of high speed memory chip
testing so I learned a lot about transmission lines and impedance
matching that helped greatly with digital radio design later at Mitre.
The memory testers were so fast that there would be three test vectors
(address & write data) travelling out within the coax between the
machine cabinet and the test head at the wafer prober, and three
results coming back. We manually trimmed the cable lengths to match
their propagation delays within a few picoseconds, 16ths of an inch at
the speed of light.
https://en.wikipedia.org/wiki/Wafer_testing
The CoCo had an elegantly simple video controller that I borrowed, in
monochrome form, after giving up on Don Lancaster's Cheap Video.
That controller was another custom Motorola chip, IIRC.
Although I didn't use the 6809, studying it helped a lot when I had to
design a 16-bit A/D converter board to go in a 68000-based NuBus
Macintosh and the Apple Certified Programmer assigned to write its
driver quit.
I did a lot of assembly language work with the 6800 and 6809.
Very little with the 68000, where most of my examples were Sun
workstations and servers, with good C compilers. The first system was a
COSMOS CMS-16/UNX with v7 unix and a terrible C compiler. Later system
were the AT&T Unix-PC/7300/3B1 using the MC-68010 CPU, and the later
Suns (before SPARC came into the game) were mostly the MC-68020, and one
or two examples of the MC-68030, which I never had.
I could program UVPROMs on the Automatic Test Equipment we were
building at work, I wrote a routine to do it quickly for practice, but
the engineers gave me slow, pre-production samples of 2816 flash
memory that's pin-compatible with the 2716 UV PROM, and the 6116 CMOS
static RAM I was using.
http://cva.stanford.edu/classes/cs99...ts/at28c16.pdf
I built a wire-wrapped computer at work using a MC-68B00 and a
superset of the SSB monitor ROM. I did the assembly using a Tektronix
MDL -- but at first I programmed the EPROMs with a suitcase-mounted
manual prom burner (I forget the maker at the moment) until I
wire-wrapped a burner for the 2716 to live in the system and expanded
the monitor to include burn instructions. I could assemble the code in
the MDL, and through a probe which plugged in where the CPU normally
lived, I could load the program into a block of memory and then burn the
EPROM on that with the MDL acting as the CPU, then switch back to the
normal CPU and continue to use it without the help of the MDL.
Enjoy,
DoN.
--
Remove oil spill source from e-mail
Email: | (KV4PH) Voice (all times): (703) 938-4564
(too) near Washington D.C. |
http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---