7 Mar
2005
7 Mar
'05
3:13 p.m.
Hey all!
I am beginning to write a simulator for a "microcomputer", it will
feature a
curses (terminal) front panel. it will be a 12 bit machine. You will be
able to
issue commands to the simulator without stopping the program (for
mounting
virtual paper tapes, printing to files etc)
Shown on the front panel will be a switch register, an Accumulator and
the
program counter (all 12 bit) a "running" light and a Overflow light.
Insert = sw -> pc home = sim command Pgup = examine
([pc] -> sw
delete = (deposit) sw ->[pc] end = halt pgdn = run
Obviously a 12 bit computer suggests a PDP-8, but the function keys are
going
to be the "bit flip" front panel keys and they are grouped in three
sets of
four. I am not against a PDP-8 but it's instruction set is better
expressed
in octal right?
Were there any other 12 bit computers? preferably with Hex friendly
instruction sets?
Operational choices - should examine increment PC?, should deposit?
I am open to other meanings of the operation keys... they are arranged
as above on
my thinkpad.
The program will be written in C.
7 Mar
7 Mar
3:21 p.m.
Ron Hudson wrote:
Hey all!
I am beginning to write a simulator for a "microcomputer", it will
feature a
curses (terminal) front panel. it will be a 12 bit machine. You will
be able to
issue commands to the simulator without stopping the program (for
mounting
virtual paper tapes, printing to files etc)
Shown on the front panel will be a switch register, an Accumulator and
the
program counter (all 12 bit) a "running" light and a Overflow light.
Insert = sw -> pc home = sim command Pgup = examine
([pc] -> sw
delete = (deposit) sw ->[pc] end = halt pgdn = run
Obviously a 12 bit computer suggests a PDP-8, but the function keys
are going
to be the "bit flip" front panel keys and they are grouped in three
sets of
four. I am not against a PDP-8 but it's instruction set is better
expressed
in octal right?
Were there any other 12 bit computers? preferably with Hex friendly
instruction sets?
I don't know about hex, but the other 12 bitters were the CDC 160
http://www.cs.uiowa.edu/~jones/cdc160/ and the LINC/PDP 12 machines
that I can think of for now.
Operational choices - should examine increment PC?,
should deposit?
I am open to other meanings of the operation keys... they are arranged
as above on
my thinkpad.
The program will be written in C.
Forget C ... program in HEX :)
Ben alias woodelf
3:42 p.m.
On Mar 7, 2005, at 12:21 PM, woodelf wrote:
I don't know about hex, but the other 12 bitters
were the CDC 160
http://www.cs.uiowa.edu/~jones/cdc160/ and the LINC/PDP 12 machines
that I can think of for now.
I looked up the CDC and it looks very hex-ish. I might just do that.
the LINC/PDP12 is just a PDP-8++ right?
4:17 p.m.
Ron Hudson wrote thus:
On Mar 7, 2005, at 12:21 PM, woodelf wrote:
The LINC is actually a different architecture, and was born slightly
before the PDP-5 (the PDP-8's "dad"); so it's not a PDP8++. DEC
later designed the LINC-8, which was sort of a LINC and a PDP-8 sharing
the same apartment, and later the PDP-12 which could intermingle the
two different instruction sets.
But if you're looking for a hex-based 12-bit architecture to play with,
the LINC architecture probably isn't what you're looking for either. :)
-O.-
I don't know about hex, but the other 12
bitters were the CDC 160
http://www.cs.uiowa.edu/~jones/cdc160/ and the LINC/PDP 12 machines
that I can think of for now.
I looked up the CDC and it looks very hex-ish. I might just do that.
the LINC/PDP12 is just a PDP-8++ right?
7:16 p.m.
Ron Hudson wrote:
On Mar 7, 2005, at 12:21 PM, woodelf wrote:
No the LINC is a nice computer for lab experiments.
The PDP 12 is a computer that has a dual instruction set PDP 8 or LINC.
Ben alias woodelf
I don't know about hex, but the other 12
bitters were the CDC 160
http://www.cs.uiowa.edu/~jones/cdc160/ and the LINC/PDP 12 machines
that I can think of for now.
I looked up the CDC and it looks very hex-ish. I might just do that.
the LINC/PDP12 is just a PDP-8++ right?
8:37 p.m.
I have found some pictures...
Has anyone ever used one of these? (CDC 160?)
I don't see a bank of switches... can you fat finger in code?
It looks like there is a block input instruction, perhaps instead of
fat fingering
code you produce a binary tape and run it in to boot the machine?
It looks like the three big displays match the A, S, and PC registers
and are in octal...
right?
thanks.
ron.
9:50 p.m.
On Mon, 7 Mar 2005, Ron Hudson wrote:
I have found some pictures...
Has anyone ever used one of these? (CDC 160?)
I don't see a bank of switches... can you fat finger in code?
It looks like there is a block input instruction, perhaps instead of
fat fingering
code you produce a binary tape and run it in to boot the machine?
It looks like the three big displays match the A, S, and PC registers
and are in octal...
right?
Ron,
These are all questions that you'll find the answers for in the CDC-160A
operations manual. And you'll most definitely *have* to read it if you
plan to simulate it. You do plan on reading the manual, don't you?
--
Sellam Ismail Vintage Computer Festival
------------------------------------------------------------------------------
International Man of Intrigue and Danger http://www.vintage.org
[ Old computing resources for business || Buy/Sell/Trade Vintage Computers ]
[ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]
3:53 p.m.
On Mar 7, 2005, at 12:21 PM, woodelf wrote:
Hmmm. Looking at some of the other classes of instructions, two groups
divide the instruction word into two six bit fields.
Oh well, there are already a few PDP emulators... :^) unless anyone
can speak up for another 12 bit architecture perhaps it will be the
cdc-160.
4:19 p.m.
Ron Hudson wrote:
Hey all!
I am beginning to write a simulator for a "microcomputer", it will
feature a
curses (terminal) front panel. it will be a 12 bit machine. You will be
able to
issue commands to the simulator without stopping the program (for mounting
virtual paper tapes, printing to files etc)
Have a look at http://www.gjcp.net/~gordonjcp/pdp8.tar.bz2 - it's a
pretty crude PDP8 emulator that I wrote. ODT runs.
Gordon.
8:32 p.m.
On Mon, 7 Mar 2005, Ron Hudson wrote:
I am beginning to write a simulator for a
"microcomputer", it will feature a
curses (terminal) front panel. it will be a 12 bit machine. You will be able
to
issue commands to the simulator without stopping the program (for mounting
virtual paper tapes, printing to files etc)
Cool! If you haven't, check out Donald Knuth's fictional MIX
computer. It's a mishmash of a bunch of "contemporary" (1970s)
architectures. The cool thing about is there is SOFTWARE written
for it, even though it never physically existed.
There was no physical representation, if you were to make one,
it would add to the coolness of it.
9:48 p.m.
New subject: ot? Simulate something....
In message <20050307172917.L635 at localhost>, Tom Jennings writes:
On Mon, 7 Mar 2005, Ron Hudson wrote:
While it doesn't have the same classic factor as MIX, there's
something to be said for his newer MMIX design. It's a pretty
nice design, though being 64-bit it would be a rather bigger
project. Simulators (with video) and assemblers exist. GCC
has even been targeted to it.
Check out:
http://www-cs-faculty.stanford.edu/~knuth/mmix.html
Brian L. Stuart
I am beginning to write a simulator for a
"microcomputer", it will feature a
curses (terminal) front panel. it will be a 12 bit machine. You will be able to
issue commands to the simulator without stopping the program (for mounting
virtual paper tapes, printing to files etc)
Cool! If you haven't, check out Donald Knuth's fictional MIX
computer. It's a mishmash of a bunch of "contemporary" (1970s)
architectures. The cool thing about is there is SOFTWARE written
for it, even though it never physically existed.
There was no physical representation, if you were to make one,
it would add to the coolness of it.
10:24 p.m.
On Mar 7, 2005, at 5:32 PM, Tom Jennings wrote:
On Mon, 7 Mar 2005, Ron Hudson wrote:
Cool is it 12 bit, best expressed as 3 hex digits?
I am beginning to write a simulator for a
"microcomputer", it will
feature a
curses (terminal) front panel. it will be a 12 bit machine. You will
be able to
issue commands to the simulator without stopping the program (for
mounting
virtual paper tapes, printing to files etc)
Cool! If you haven't, check out Donald Knuth's fictional MIX
computer. It's a mishmash of a bunch of "contemporary" (1970s)
architectures. The cool thing about is there is SOFTWARE written
for it, even though it never physically existed.
There was no physical representation, if you were to make one,
it would add to the coolness of it.
11:30 p.m.
Ron Hudson wrote:
On Mar 7, 2005, at 5:32 PM, Tom Jennings wrote:
No ... I think it was decimal or base 64 depending on the computer
instruction
set used. I like decmal machines but thay don't make decimal addressed
memory.
His lateset version is risc machine, a machine I don't like because
they have too
low a instruction density. Now does it really matter if you use hex or
octal or what
ever?
Ben alias woodelf
On Mon, 7 Mar 2005, Ron Hudson wrote:
Cool is it 12 bit, best expressed as 3 hex digits?
I am beginning to write a simulator for a
"microcomputer", it will
feature a
curses (terminal) front panel. it will be a 12 bit machine. You will
be able to
issue commands to the simulator without stopping the program (for
mounting
virtual paper tapes, printing to files etc)
Cool! If you haven't, check out Donald Knuth's fictional MIX
computer. It's a mishmash of a bunch of "contemporary" (1970s)
architectures. The cool thing about is there is SOFTWARE written
for it, even though it never physically existed.
There was no physical representation, if you were to make one,
it would add to the coolness of it.
8 Mar
8 Mar
9:21 a.m.
On Mar 7, 2005, at 8:30 PM, woodelf wrote:
Ron Hudson wrote:
only that the function keys, which will be used to invert bits in the
switch register
are grouped in fours. only reason. could to a PDP-8 otherwise.
On Mar 7, 2005, at 5:32 PM, Tom Jennings wrote:
No ... I think it was decimal or base 64 depending on the computer
instruction
set used. I like decmal machines but thay don't make decimal addressed
memory.
His lateset version is risc machine, a machine I don't like because
they have too
low a instruction density. Now does it really matter if you use hex or
octal or what
ever? On Mon, 7 Mar 2005, Ron Hudson wrote:
Cool is it 12 bit, best expressed as 3 hex digits?
I am beginning to write a simulator for a
"microcomputer", it will
feature a
curses (terminal) front panel. it will be a 12 bit machine. You
will be able to
issue commands to the simulator without stopping the program (for
mounting
virtual paper tapes, printing to files etc)
Cool! If you haven't, check out Donald Knuth's fictional MIX
computer. It's a mishmash of a bunch of "contemporary" (1970s)
architectures. The cool thing about is there is SOFTWARE written
for it, even though it never physically existed.
There was no physical representation, if you were to make one,
it would add to the coolness of it. Ben alias woodelf
3:48 p.m.
New subject: decimal-addressed memory (was Re: ot? Simulate something....)
Woodelf wrote:
I like decmal machines but thay don't make decimal
addressed
memory.
Sure they do. I've got plenty of extra, so I'd be happy to sell you as
much as you want. Do you want the BCD, excess-3, or 4-2-2-1 addressed
parts? What access time and capacity do you want? I think the ones I
have the most of are the BCD ones with 10,000 words of two digits each.
If you need to store parity or word marks as well, you can use them as
one-digit-per-word instead. They're perfect for building your own
IBM 1620 clone.
Eric
10 Mar
10 Mar
7:52 p.m.
>>>> "woodelf" == woodelf
<bfranchuk at jetnet.ab.ca> writes:
woodelf> No ... I think it was decimal or base 64 depending on the
woodelf> computer instruction set used. I like decmal machines but
woodelf> thay don't make decimal addressed memory.
Sure they do, or did. An IBM 1620 had 5 digit addresses, with your
choice of 20k, 40k, or 60k (where k == 1000 of course) digits of core
memory.
paul
1:56 p.m.
Paul Koning wrote:
> Sure they do, or did. An IBM 1620 had 5 digit addresses, with your
> choice of 20k, 40k, or 60k (where k == 1000 of course) digits of core
> memory.
On Thu, 10 Mar 2005, woodelf wrote:
I was thinking 'solid state' memory ... Sigh!
If you really need decimally-addressed memory, and the decimal
address has a binary-weighted presentation, simply apply the
decimal address to the "binary" address lines. It will work
perfectly.
There will be unused memory cells, but who cares?
0000 0
0001 1
0010 2
0011 3
0100 4
0101 5
0110 6
0111 7
1000 8
1001 9
1010 unused
1011 unused
1100 unused
1101 unused
1110 unused
1111 unused
10 Mar
10 Mar
7:45 p.m.
>>>> "Tom" == Tom Jennings
<tomj at wps.com> writes:
Tom> On Mon, 7 Mar 2005, Ron Hudson wrote:
> I am beginning to write a simulator for a
"microcomputer", it will
> feature a curses (terminal) front panel. it will be a 12 bit
> machine. You will be able to issue commands to the simulator
> without stopping the program (for mounting virtual paper tapes,
> printing to files etc)
Tom> Cool! If you haven't, check out Donald Knuth's fictional MIX
Tom> computer. It's a mishmash of a bunch of "contemporary" (1970s)
Tom> architectures. The cool thing about is there is SOFTWARE written
Tom> for it, even though it never physically existed.
It's even a supported target in the GCC compiler.
paul
7069
days inactive
7073
days old
18 comments
9 participants
participants (9)
-
bfranchuk＠jetnet.ab.ca -
blstuart＠bellsouth.net -
eric＠brouhaha.com -
gordon＠gjcp.net -
ohh＠drizzle.com -
pkoning＠equallogic.com -
ron.hudson＠sbcglobal.net -
tomj＠wps.com -
vcf＠siconic.com