I was going to write a simulator for Bell Labs "Cardiac" (cardboard
slide the slides and move bits be the cpu your self) computer, but all I
can find is the Instruction Set, none of how the instructions are
implemented.
So, I decieded to write my own...
Here is the instruction set, I thought it best to consult with you all
before beginning to code. What
do you think?
Opcodes
0 1 2 3 4 5 6 7
Z N o1 o2 o3
=================================================================
Z N 0 0 1 i r r get copy (indirect) register rr to acx
Z N 0 1 0 i r r put copy (indirect) acx to register rr
Z N 0 1 1 o o o Mathop
0 0 0 nop
0 0 1 add acx + acy ==> acx
0 1 0 sub acx - acy ==> acx
0 1 1 and acx & acy ==> acx
1 0 0 exor acx xor acy ==> acx
1 0 1 or acx | acy ==> acx
1 1 0 cmp acx cmp acy set Z and N
1 1 1 cpz acx cmp 0 set Z and N
Z N 1 0 0 p p p inp copy port ppp ==> acx
Z N 1 0 1 p p p out copy acx ==> port ppp
Z N 1 1 0 i r r swp acx ==> (ind) Rn == > acx
Z N 1 1 1 0 o o acx op group
0 0 0 not r1 <== ! r1
0 0 1 shl Shift left
0 1 0 shr Shift right
0 1 1 inc Increment r1 <== r1 + 1
Z N 1 1 1 1 o o Jump Group
1 0 0 jmp ind [pc] pc <== [pc]
1 0 1 jsr ind [pc] R2 <== pc , pc <== [pc]
1 1 0 rtn pc <== R2
1 1 1 Halt,reset [pc] <== 0
Zero and Negative flags
On execution of the cmp and cpz instructions the Z and N flags are
set, following instructions can test these flags and choose to execute
or not.
Z N Meaning
0 0 Execute if acx was positive on test (acx > 0 / acx > acy)
0 1 Execute if acx was negative on test (acx < 0 / acx < acy)
1 0 Execute if acx was zeron on test (acx = 0 / acx = acy)
1 1 Always execute
Registers
00 r0 acx 1st operand
01 r1 acy 2nd operand
10 r2 ptr Pointer register
11 r3 pc Program Counter
here is an example of an assembled program....
Assembly of a simple program
org $10 ; program starts at 10 10
get ind pc ; load value1 acy 10 11001111 00010000
adr value1 11 00011001 00010001
swp acy 12 11110110 00010010
get ind pc ; load value2 acx 13 11001111 00010011
adr value2 14 00011010 00010100
add ; sum values 15 11011001 00010101
put ind pc ; store sum 16 11010111 00010110
adr sum 17 00011011 00010111
hlt ; end of program 18 11111111 00011000
value1 byt 19 00011001
value2 byt 1A 00011010
sum byt 1B 00011011
end ; end of code
I know, it only has 256 words of 8 bit memory, but isn't this as much as
some of the early machines?
I hit a sort on Jay West's name and only came up with his single,
important post. I take this to mean that all other related messages
are more rants or nearly so. There are 90 new messages in less
than a day. This will likely make me hit the (A cursory look at
Richards posts lately show OnTopicism--thanks) hit the delete key
fast and loose.
Why a thread like "GW Bush Cancelled VCF 5.0 VCF"
didn't halt immediately is a mystery.
If noone said it already, here's me: Moderation takes time. If there
is no-one to do it this MAY lead to list shutdown --Bad!
Self moderation saves time --everyone else's.
Thanks to those that do it.
John A.
weird characters, cursor control, bells and whistles
I believe the "\n", "\t" and other conventions are linked to the printf
function in C. For example if you want to a print a non scrolling list of
lines processed you could send out lines to printf a 5 character number and
then use "\b\b\b\b\b" to reposition the cursor at the start of the number so
that the next print command overwrites it. The effect is handy with
terminals kind of a pain on a LA34 where the printing is overwritten on the
paper
Mike
mmcfadden(a)cmh.edu
On Tue, 21 Aug 2001, Adrian Vickers wrote:
> >This leads to one mind-boggling possibility. If one etches custom boards
> >with eight Z80s each, and pumps all eight each clock, and one takes
> >several thousand of these boards, wouldn't that be a rather interesting
> >MPP supercomputer? Would it be more or less money than one made with,
> >say, IBM POWER4s, or Alphas, or StrongARMs?
>
> Heh! I like it... The chips would be cheaper in bulk as well, and with a
> bit of heatsinking it should be possible to overclock them somewhat.
How hot do Z80s get under normal operation? Would you be able to
overclock them using heatsinks without fans? That would simplify things a
bit.
> Mind you, even with discounts you're looking at twenty grand just to sort
> out your processor farm; and you still need the i/o logic...
Yes, but how fast a computer would you get for $20,000? Would it be
faster than other machines at that price? I don't think the I/O would be
a big deal. You could use off-the-shelf components to do it.
> Still, 8000 Z80s running at 4MHz could theoretically yield the equivalent
> of 1 Z80 running at 32GHz.
That's assuming that the task you're running can be parallelized to that
degree. And plus, if you were to run each of these hypothetical 8-Z80
boards as a single CPU, you would have a 64-bit system.
Peace... Sridhar
> Would Manic Miner be playable at that speed? ;)
> Cheers!
> Ade.
> --
> B-Racing: B where it's at :-)
> http://www.b-racing.co.uk
>
Richard Erlacher was inquiring about ancient LaserJet IID's.
Here is some info stored in my overloaded brain, maybe slightly garbled.
In duplex mode the HP LaserJet IID printed the paper on one side then
ejected the paper out the rear and then reinserted the paper to print the
other side. It was kind of a triangular contraption that held the paper on
the way out and then allowed it to be pulled back in.
>From about 1988-1995 we didn't use MSWindows, only DOS. Ours were connected
via parallel interfaces and downloading fonts took memory and time, the font
cartridges offered the chance to have "standard" fonts already installed.
The other problem was that occasionally somebody would send/download non HP
fonts to the printer and then garbage would print for 20-50 pages. It was
also difficult to track whose fonts were in the printer at any one time. We
always reset between jobs.
We also didn't install lots of printer memory.
We later got a HP 2000 printer but we had nothing but problems,
There were barcode fonts, graphic fonts, foreign languages, and line drawing
fonts.
I personally owned a IID and was given another but I donated them to the
local not-for-profit computer surplus.
Mike
mmcfadden(a)cmh.edu
Hello all,
I have a few S-100 motherboards that I'd like to make use of, but I have no
decent linear power supply for them. I have little experience designing
power supplies, so I hesitate to try to build a linear supply. Also, the
cost of the transformer alone would most likely be prohibitive. Then
there's the heat issues, big caps that go boom when my design fails, etc.
I would like to ask if any of you know of an existing, available, commercial
switching power supply that could be used to power an S-100 computer?
Here's the wrinkle: S-100 requires around +8V and around +/- 16V. The
"around" is because traditionally S-100 computers were supplied by
unregulated linear supplies of these voltages, and the S-100 cards
themselves carried the voltage regulators necessary to regulate the supply
voltages to +5V and +/- 12V. Also, the current must be high on the +8V side
(say 20+ amps, especially for a 20-slot motherboard), and about 5+ amps on
+/- 16V.
I realize I could get a +5V and +/- 12V supply, and just remove the
regulation circuitry on the cards, but given the amount of them that I have,
that gets very labor-intensive, and I'd rather not have to hack out parts of
these cards...
Thanks for any pointers...
Rich B.
On Wed, 22 Aug 2001, Kent Borg wrote:
> On Tue, Aug 21, 2001 at 06:10:37PM -0400, Master of all that Sucks wrote:
> > I think the distinction between supercomputers and mainframes is even
> > deeper than that. Supercomputers are designed to do things very very
> > fast. Mainframes are designed to do many many things at once. The two
> > goals frequently aren't convergent.
>
> Very much so. Maybe a supercomputer is to a dragster as a mainframe
> is to an 18-wheeler. Both might have roughly the same magnatude
> engine with the dragster doing somewhat more power (I am making this
> up--don't fault me on the specifics), but the intended uses are very
> different.
Your analogy is quite good. If we continue this analogy, we can say that
your proverbial dragster has a lot more horsepower than the
tractor-trailer, and the semi has a lot more torque than the dragster.
You could force a supercomputer to act like a mainframe, or vice versa,
but it wouldn't work very well.
> > Mainframes are very good at serving thousands, if not hundreds of
> > thousands of *simultaneous* transactions.
>
> Yes, though earlier mainframes were doing batch jobs with cards and
> later mag tape. But always mainframes are big iron for business and
> that makes them very different from the shortlived big thoroughbred
> iron of supercomputers.
They were still sitting there chugging away doing many simultaneous (not
very quickly) with data obtained from mag tape or punched card. There
were, of course exceptions to this paradigm, but I still think it's a good
way of looking at it.
> -kb, the micro-oriented Kent who used to be puzzled over what made
> mainframes different, and though he still doesn't know that much about
> mainframe architecture, he at least understands their function is
> quite different.
>
Peace... Sridhar
On Aug 21, 12:16, James Rice wrote:
> On another list, there has been a long thread about SO's and room for
> the hobby. Does anyone else keep their collection set up and
> functioning? Does your SO understand?
Mine isn't all up and running, but in the office I have:
SGI Indy R5000 (always on, it does mail, NFS, backups, etc)
Acorn Archimedes A440 (always on)
Sparcstation 1+ (rarely on!)
ancient PC (occasionally on, for proprietary data CDs and Word docs)
BBC B with floppies, hard drive, various extras
Exidy Sorcerer (for Galaxians etc)
SGI Indigo
and an ISDN router, managed hub, HP1600CM colour inkjet which are also
always on, and a laser printer which is turned on when needed.
Next door, my wife has an SGI Indigo Elan which she uses for mail, surfing,
etc.
There are few oddments on shelves, such as a Microwriter AgendA, a Z88, a
Psion Organiser, a 2901 microcode tutor, and some others.
There are two more Indys in the workshop, one of which is usually on, and
another BBC and another PC (used for Linux and 22Disk). There's also
another managed hub and another router (Ethernet and FDDI), a GatorBox, a
couple of terminal servers, ...
Most of the working part of the collection is in the workshop:
PDP-11/23 in office rack with 2 x RL02, running 7th Edition Unix or RT11
microPDP-11/83 running BSD 2.11
microVAX-II running VMS somthing-or-other (at least it was, last time I
looked)
6' DEC rack containing an 11/34, two 11/73's, front-loading 1/2" magtape
unit, large Fujitsu winchester, RX02, and a couple of other 8" floppies.
These run RSX-11 and RT-11.
There's a pair of shelf bays which have the following set up and runnable:
Apple Mac IIvx
Sage II
Dragon 32
2 more Indigos, one XZ, one basic LG graphics (Song and Dance Machine,
if you know what that is)
Atari Mega ST
VT/78
microfiche reader and 'fiche boxes
Acorn R260
Amiga 500
Commodore PET 2001-8K (except it's only 7K since one chip went)
Commodore 8050 dual drive for the PET
KIM-1
home-grown Z8-based SBC
Amstrad PCW 8512
Sinclair Spectrum
Sinclair QL
Acorn Atom
Nascom (not strictly runnable, it needs some work)
Apple ][ Europlus
Apple //e
Apple Mac Plus and hard drive
NeXTstation
Vaxstation 3100
Around the workshop, there are a few terminals, an Epson FX100, a laptop,
and various oddments, along with my test gear, spares, etc. The rest of
the collection languishes in the attic, including more Sinclair stuff,
Acorn Electron, C128, Vic-20, Sharp MZ80K, Acorn Archimedes A310, another
Acorn R260, ...
I'm not sure about "understanding", but Liz is reasonably tolerant,
providing the collection remains in designated areas. Which it does, most
of the time (there's another 11/23 reputedly with TSX-11 installed, and a
VT100, and a large Eurocard cage in the garage at the moment). She'd
rather have a modern PC than an Indigo on her desk, but there's no way I'm
putting a full-time Windows machine on my network :-)
--
Pete Peter Turnbull
Network Manager
University of York
The lady below would like to receive offers on the systems listed. Please
reply directly to her:
Reply-to: Norma Shulman <shulman22(a)rcn.com>
---------- Forwarded message ----------
Date: Sat, 4 Aug 2001 08:38:40 -0400
From: Norma Shulman <shulman22(a)rcn.com>
Subject: More information needed
I have 4 old systems:
(1) Atari 800XL system including a disk drive, printer, games, and software.
(2) Mac 512 with a Tecmar 5 MB external hard disk.
(3) Mac 128K (upgraded with the 512K board from a machine we upgraded
to a Plus) original machine (I have the sales receipt) with numeric
keypad, external floppy drive, ImageWriter I, padded carrying case,
much commercial software, and a complete public domain software
library from the Boston Computer Society on 400K floppies.
(4) dual-floppy Mac SE with a 40MB hitchhiker hard disk (connects
directly to the SCSI port on the back - no cable), Jasmine 20 MB
external drive, padded carrying case, ImageWriter II, Apple personal
modem.
Thank you,
Norma Shulman
shulman22(a)rcn.com
--
Sellam Ismail Vintage Computer Festival
------------------------------------------------------------------------------
International Man of Intrigue and Danger http://www.vintage.org
At 03:37 PM 8/18/01 -0400, you wrote:
>I have one of the Codar boards... it is part of the 11/73 which
>used to run the heating/cooling systems at the DEC mill complex.
The TOY-11 has a copyright date of 1983 on the board and the ic's have
date codes for 1986. Isn't old good!
>If you can check the battery and replace it if needed, put it
>in a qbus machine and see how many registers there are after 176600
>(open 176600 and then simply hit <LF> until it complains).
I used the MAP function in my 11/23+ boot rom to find the registers, there are
two used by the TOY-11. 176670 and 176672.
>Keep track of the contents and see if they correlate to the
>current date/time... keep in mind the registers may have the
>info in BCD...
>
>In fact... I just found some datasheets for the chip on the web and
>can see that the info is in BCD... and it is in successive registers
>on the chip... the question is whether the qbus board implements
>a direct one-for-one relation to the chip registers or if there is
>a control register used as an index (write the register number to
>read) and a buffer register which is read to read that chip register
>and written to set that chip register.
It appears that the TOY-11 implementation is very simple. Address
176670 is write only and sets up the address code on the clock
chip. 176672 is read/write and I assume that means if you read you
get the current counter setting and if you write you set the counter.
At least that is what I have been able to do by writing codes to '670
and then reading '672.
>Oh, another thing... it could be that the registers are in successive
>*BYTES*, not words.
It looks like the contents of 176672 contain a byte of data, 2 BCD digits.
>Keep me informed... sounds interesting... I might be able to help
>with a program which reads the info and sets the RT system date/time
>on bootup...
I'm going to look at the software for the Codar board, maybe that can be
adapted to work with the TOY-11.
