it's been a few years since I released version 1.2, which completed the
simulation of the processor and rudimentary I/O of the Burroughs 220
computer. It correctly ran the machine diagnostics and had sufficient
I/O capability to run an assembler and numerous sample programs.

What it could not do is run the Burroughs Algebraic Compiler for the 220
(BALGOL), nor could it simulate the interactive graphics I/O devices
that extended the B220 installed at Caltech--a machine that I came to
love as my second "personal computer" (after the IBM 1620). Both early
machines took up a room-sized space, but I was able to use them
directly, sitting at their consoles, just as I would use my Apple ][+
seventeen years later!

I was pretty happy with the functionality of B220SIM v1.2, but I had
always hoped to be able to run BALGOL, an amazingly complete and fast
load-and-go compiler for Burroughs' variant of Algol 58 that was first
shipped to customers in 1961. It was a beautiful creation of several
early giants of computer programing, a team led by Joel Erdwinn,
including Donald Knuth (who wrote a compiler for the same language for
the Burroughs 205).

BALGOL required a fairly complete implementation of the B220 mag-tape
subsystem, including the ability to search the tape for specific blocks
(pretty advanced for its time--more like a "linear disk" than the IBM
tape model).

One reason the version 1.2 tape implementation was so basic is that 64KB
was just about filled with just a simple implementation, and had no
buffering. So I started version 2 with the assumption of a 128KB Apple
//e or IIc and proceeded to learn about writing bank-switched code.

The v2.1 simulator itself is in AUX memory, and main memory is used for
the human interface, graphics buffer, I/O buffers, and the mapping of
B220 I/O devices to ProDOS disk files. In the process, I simulated the
DEC 340 CRT display that Caltech attached and the light pen and
specialized interactive keyboards that were stationed with the CRT display.

If any of this sounds interesting, take a look at B220SIM v2.1 azt my
website, where you can read more about the simulator and the B220,
download the simulator and its source (and some sample programs and
utilities), and take a look at the actual assembly code.
--

-michael

NadaNet 3.1 for Apple II parallel computing!
Home page: http://michaeljmahon.com

"The wastebasket is our most important design
tool--and it's seriously underused."

Great work, Michael. Thanks for sharing,
av

On 8/22/21 2:17 AM, Michael J. Mahon wrote:
> What it could not do is run the Burroughs Algebraic Compiler for the 220
> (BALGOL), nor could it simulate the interactive graphics I/O devices
> that extended the B220 installed at Caltech--a machine that I came to
> love as my second "personal computer" (after the IBM 1620).  Both early
> machines took up a room-sized space, but I was able to use them
> directly, sitting at their consoles, just as I would use my Apple ][+
> seventeen years later!

Michael -

What an amazing update, and your heartfelt fondness for this machine
(and the one it runs on!) definitely shines through your implementation.
Congratulations!

- David

David Schmidt <schmidtd@my-deja.com> wrote:
> On 8/22/21 2:17 AM, Michael J. Mahon wrote:
>> What it could not do is run the Burroughs Algebraic Compiler for the 220
>> (BALGOL), nor could it simulate the interactive graphics I/O devices
>> that extended the B220 installed at Caltech--a machine that I came to
>> love as my second "personal computer" (after the IBM 1620).  Both early
>> machines took up a room-sized space, but I was able to use them
>> directly, sitting at their consoles, just as I would use my Apple ][+
>> seventeen years later!
>
> Michael -
>
> What an amazing update, and your heartfelt fondness for this machine
> (and the one it runs on!) definitely shines through your implementation.
> Congratulations!
>
> - David
>

Thanks, David and Antoine—it is a labor of love, to be sure!

I’ll never forget turning off the lights in the 8,000 sq. ft. computer room
(which also housed the 7094 and 7040) after everyone had left for the
night, and watching the dance of the hundreds of neon bulbs on the 220!

In fact, I wrote my first music performance program around 1964 for the
220. It played through radiated noise picked up by a transistor radio I
placed near the console. The first piece I wrote for it was “Stormy
Weather”! (Unfortunately I seem to have lost the tape recording I made.)

When the CRT and keyboards were installed in early 1965, they were placed
out of sight of the 220 console, so lots of visual feedback became
unavailable. So I brought the transistor radio in and put it near the
CRT, and it provided very useful audio feedback about the system. Every
keypress and loop had a distinctive sound—particularly an infinite loop! A
sorting loop sounded like noise—evidence of the entropy of the data.

Fun times!

When personal computers began to appear in the late 1970s, I knew it was
only a matter of time, and the Apple ][+ pushed me right over the edge!
;-)

--
-michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com

I hope you will find the tape, Michael, I would love to listen to it!
Antoine

Antoine Vignau <ntn.vignau@gmail.com> wrote:
> I hope you will find the tape, Michael, I would love to listen to it!
> Antoine
>

I’ll keep looking—it may yet turn up!

The “voice” of the notes is reedy, with some background noise.
Interestingly, the loops for the notes were easy compared to the loop for
timed tests. I used a single instruction loop that ran at a high enough
frequency to be a passable, but hissy, rest.

I also implemented counted repeats of note sequences, and I increased the
durations of the last few notes to create a rallentando at the end!

--
-michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com