Alan Perry <alanp at snowmoose.com> wrote:
>
> This also included the Burroughs/Unisys A-series systems that were
> made into the early 90s.
>
> The A-series systems used a 52-bit, tagged word. 1 bit of parity, 3
> bits of tag and 48 bits of data.
>
> I don't recall how the addressing worked and Al has my papers on the
> architecture, so I can't look it up ...
The A-series were direct successors to the B5000/6000/7000 series and
used the same architecture. Unisys still sells the ClearPath MCP
systems, now using Intel processors and running the MCP operating system
(under emulation I think).
The architecture has so-called display registers, each of which pointed
to a stack frame for a lexicographical level in the code, i.e. the
procedure call level. Data is addressed as an address couple consisting
of a display register number and an offset, stored in a so-called
Indirect Reference Word (IRW). Data outside of the stack can be
addressed by means of descriptors. Data in another program's stack is
addressed by means of Stuffed Indirect Reference Words (SIRW), which
include a stack number and an offset. The operating system keeps track
of individual program stacks by means of a tree of pointers to job
stacks. Also, to keep track of the procedure calling and return linkage,
Mark Stack Control Words (MSCW) are created whenever a procedure is
called. The Display Registers point to the MSCWs.
All this is designed to support block-structured high level languages, in
fact all the operating system software is (was) written in various ALGOL
dialects.
There is quite a lot of information about this on the net. A book which
describes it is Computer Systems Organization by Elliott Organick (ACM
Monograph Series, 1973. LCN: 72-88334).
-Jonas
So about a month ago I purchased some Wang branded MSDOS software off of eBay. Much to my delight, when it arrived, it was still factory sealed. I showed my wife the date of manufacture on the label and she suggested I wait to open it. I reluctantly agreed to wait. So it's been about a month and today is the day. Please enjoy some pictures of the moment.
http://gallery.me.com/irisworld/100109
Rob
Rob Borsuk
email: rborsuk at colourfull.com
Colourfull Creations
Web: http://www.colourfull.com
I sent a question to the seller, asking for clarification about the
serial number. This is the (insane) reply that I got:
------------------
evan_k, thank you for asking about this fine microwaved masterpiece of
fine art. This was a perfectly working complete unaltered Mac 512k
before it was microwaved by the AMIR 9000 One Ton Microwaving Robot, now
a first of it's kind. It is now an incredible work of art considered to
be priceless and even better than before it was microwaved. The artist
also known as your professional microwave operator waited 25 years to
microwave this computer. Some things only just get better with age
especially when it comes to the world of art. You can be assured that a
fine connoisseur of art will acquire this work of art. The artists has
sold artworks to collectors which include Dale Djerassi [son of the of
Carl Djerassi who invented the birth control pill, Georgia Frontier just
to name a couple. He has also been featured internationally in numerous
prominent publications that have included HGTV, OPRAH, TLC, People
magazine, Mal Sharp's Weird Rooms Book, CNN, Good morning show, 100
Hojdare Sweden, Urdu Magazine, Los Angeles Times and so on. He also was
featured at the most prominent modern art museum in Baltimore called the
AVAM.
You can be sure this art is authentic, original & sign and dated by the
artist & guaranteed microwaved.
>>> In hindsight, could the DEC-20 or RISC version of it
>>> had marketing and other bad luck changed ,and
>>> today replaced the PC in many applications?
>>
>> Six bit based architectures all died with the rise of the IBM S/360
>> family. DEC, CDC, Univac.
>>
>> It also did not help that many of these old line mainframe
>> architectures were getting hard to extend. For the DEC 36 bitters -
>> what does one do after the 18 bit address pointer, with two neatly
>> stuffed into a 36 bit word? A kludge. What does one do after the 36
>> bit address pointer? A worse kludge.
>
> I kind of suspected that was the case. Did anyone make a large 48 bit
> machine?
>
If you mean physically large, then yes. However on the 1301, addresses range only from 0 to 3999 (decimal).
Probably not what you meant really.
FYI, for the Amiga developers around here....
-------- Original Message --------
Subject: [Mandrakeot] Hey Gene - Amiga stuff
Date: Mon, 21 Dec 2009 21:54:21 -0500
From: Ric Moore
Reply-To: mandrakeot at mdw1982.com
To: mandrakeot at mdw1982.comhttp://feeds.arstechnica.com/~r/arstechnica/open-source/~3/KX0p_xQ7x7g/know…
Thought you'd get a kick outa this one. Ric
--
--- Dave Woyciesjes
--- ICQ# 905818
--- AIM - woyciesjes
--- CompTIA A+ Certified IT Tech - http://certification.comptia.org/
--- HDI Certified Support Center Analyst - http://www.ThinkHDI.com/
"From there to here,
From here to there,
Funny things
are everywhere."
--- Dr. Seuss
Hi,
Where are you located?
Can you use DDS (DAT) media?
I'm one of the "too busy" people, but perhaps I can get some time in between now and the
New Year.
On Dec 17, 2009, at 9:45 PM, g-wright at att.net wrote:
> I'm back to try this again. I need a boot tape for a HP 3000
> Micro GX machine. Does anyone have a working machine
> or a tape. I can use almost any media. I had 2 list members
> say they would help but they seem to be too busy and its
> been 6 months. I would really like to get this going.
>
> Thanks, Jerry
> g-wright at att.net
>
under
http://bitsavers.org/pdf/intel/MDS4
thanks to Bob R. for the loan of the originals
still looking for a couple of other multibus hardware manuals
isbc286/10, 386, 486S, 214, 215, 221 disk, 550,552 ethernet
nj
On 12/14/09, Alexandre Souza - Listas <pu1bzz.listas at gmail.com> wrote:
> Tony Duell wrote:
>> I suspect such players are too expenice _new_ for me even to consider
>> buying, and contain suffieicently complex custom ICs that it's a bad idea
>> to buy one for which manufacuters spares are no longer available.
>
> Want something hackable? Look for an AP660 from Apex. No service manual
> but full listing of the ROM :oD
I have one of those. Unlike most cheap-o consumer devices, there _is_
detailed internals data available - detailed enough to retool the
firmware.
It's not full circuit diagrams and datasheets, but from the software
side of things, much, much better than most devices.
It also has a bog-standard IDE DVD transport, unlike less expensive
devices. That's the model I was contemplating mounting the IDE-CF rig
on for the Click! experiment (since I already have some
3.5"-drive-bay-PCMCIA adapter frames on hand).
My main AP660 is on the fritz right now, but it's probably PSU issues.
Fortunately, I have a parts unit I can use to verify that everything
else is good with. I've had one since it was new and been very happy
with it.
-ethan
>
> From: Ben <bfranchuk at jetnet.ab.ca>
>
>> Because the Burroughs B5500 and B6700 computers used a word
>> containing 48 bits of data, but also extra bits that described the
>> type of the data which played an important role in programming the
>> machine, I would consider that series sufficiently out-of-the-
>> ordinary to attempt to discuss here;
This also included the Burroughs/Unisys A-series systems that were made
into the early 90s.
The A-series systems used a 52-bit, tagged word. 1 bit of parity, 3
bits of tag and 48 bits of data.
I don't recall how the addressing worked and Al has my papers on the
architecture, so I can't look it up ...
alan
