>=20
>=20
>=20
> "The atmosphere of academic freedom at MIT allowed some nontraditional
>=20
> research to take place: Graduate students began playing Space War =96
>=20
> the first computer game =96 on the giant TX-2 computer."
>=20
>=20
>=20
> I designed a color graphics system for the PDP-11, memory mapped with
>=20
> barbaric AMS 1K differential DRAMs. I wrote the first app for it in
>=20
> assembly, Conway's Game of Life.
>=20
>=20
>=20
> http://en.wikipedia.org/wiki/Conway%27s_Game_of_Life
That's cred.
Reply by John Larkin●July 2, 20132013-07-02
On Tue, 2 Jul 2013 10:59:31 -0700 (PDT), Frank
<fatherhaskell@yahoo.com> wrote:
http://www.guidebookgallery.org/articles/thefatherofcomputergraphics
"The atmosphere of academic freedom at MIT allowed some nontraditional
research to take place: Graduate students began playing Space War �
the first computer game � on the giant TX-2 computer."
I designed a color graphics system for the PDP-11, memory mapped with
barbaric AMS 1K differential DRAMs. I wrote the first app for it in
assembly, Conway's Game of Life.
http://en.wikipedia.org/wiki/Conway%27s_Game_of_Life
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Reply by Frank●July 2, 20132013-07-02
On Wednesday, June 19, 2013 5:34:51 PM UTC-4, John Larkin wrote:
> John G schrieb:
>
>> All those big IBM mainframes of the 60s, 70s and 80s had a maze of >
>> yellow wire wraps on the back panels and some other colours, blue mostley.
>
> Hello,
>
> not only IBM and not only mainframes. Wirewrapping was used for a lot of
> computers in that time. PCB backplanes were used for computers produced
> in very large numbers.
>
> Bye
I worked with large Collins racks made for NASA. Maze of white Teflon
wires. I'm not thinking they were wire wrap.
I was pretty good at screwing up wirewrap wiring changes in equipment. Once
tied 5 MHz to ground. Everywhere, 5 MHz. TTL.
Greg
Reply by Uwe Hercksen●June 24, 20132013-06-24
John G schrieb:
> All those big IBM mainframes of the 60s, 70s and 80s had a maze of
> yellow wire wraps on the back panels and some other colours, blue mostley.
Hello,
not only IBM and not only mainframes. Wirewrapping was used for a lot of
computers in that time. PCB backplanes were used for computers
produced in very large numbers.
Bye
Reply by John Larkin●June 23, 20132013-06-23
On Sun, 23 Jun 2013 15:20:19 +0300, upsidedown@downunder.com wrote:
It would be good for them. It's a beautiful instruction set.
>Not much assembly programming was than on PDP-11's in 1980's, so the
>youngest PDP-11 assembly programmers would have been about 20 years
>old in 1990. At 2050, they would be 80 years old and the older ones
>past 100 years old.
>
>There is quite slim chanches that any youngster of today would be so
>committed to be fluent in such assembly language. How many youngsters
>learn assembler for some modern architectures e.g. to optimize
>computer game graphics, not many.
PDP-11 and 68K assembly is easy to learn, if you can understand bits and
addressing and logical operations and 2's comp math. *IF*.
The 68K is also an elegant machine. RISC processors are less friendly; they were
designed to make compiler happy, not to make people happy.
x86 doesn't make anything happy.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
Reply by ●June 23, 20132013-06-23
On Wed, 19 Jun 2013 14:34:51 -0700, John Larkin
<jlarkin@highlandtechnology.com> wrote:
That would require training a second generation of MACRO-11
programmers.
Not much assembly programming was than on PDP-11's in 1980's, so the
youngest PDP-11 assembly programmers would have been about 20 years
old in 1990. At 2050, they would be 80 years old and the older ones
past 100 years old.
There is quite slim chanches that any youngster of today would be so
committed to be fluent in such assembly language. How many youngsters
learn assembler for some modern architectures e.g. to optimize
computer game graphics, not many.
Originally nuclear power plants were designed for 30 years life time,
but when it became clear that getting a license for a new station is
almost impossible, extending the life time of current reactors to
50-60 years become quite appealing. In practice, this means that
practically every component is replaced, at least once during that
extended life time. This includes most of the tubes, turbines and
generators (to increase power) and control systems are gradually
updated. In a PWR, only the pressure vessel is not considered
replaceable. I do not know, if there are other consideration with
CANDU reactors (burning un-riched natural uranium), compared to
ordinary light water PWRs and BWRs.
In Finland, there are two Russian made VVER-440 reactors, originally
equipped with Ferranti control system, which during the last decade
has been gradually replaced by Siemens automation.
I have been working with control systems for various industrial
sectors for a long time and the usual contractual requirement is
software and hardware support for at least 10 years. In many cases,
our customers quite happily run 20-30 years old systems.
However, a 60 year support period would mean that we now would have to
support electronics from the early 1950's with firebotles (tubes).
Thus, the most natural technological update is the mid-life update for
a 50-60 year life cycle.
This plea for help to year 2050 sounds more like that they have lost
all requirement specifications, design documents and possibly
assembly/high level language source code. So now they want someone to
disassemble the binary code and document it :-).
In that case, if they still have relocatable object records produced
by MACRO-11 or Fortran IV Plus (i.e. input to TKB), I might even be
able to help them, if only if I could find a paper tape reader for
reading my object disassembler :-).
Reply by josephkk●June 23, 20132013-06-23
On Sat, 22 Jun 2013 03:28:19 +0000 (UTC), gregz <zekor@comcast.net> =
wrote:
>
>>> What's a "VAXen"? Are you thinking of the first-developed
>>> pre-commercial VAX systems? (I never saw one.)
>>>=20
>>> Jon
>>=20
>> That's the plural of VAX.
>>=20
>> Come to think of it, it was probably a PDP-10 that I remember as using=
a PDP-8
>> to start things up.
>>=20
>
>Seems to me, the 8 might have been the first computer in a box. Was not =
in
>a rack, but the 8I was. The PDP-15 was my first experience on a =
computer,
>but only worked with the 8I.
>
>Greg
I don't suggest being very sure about that one. In the very early 1970s =
i
was trained on and could use AN/UYK-4 systems that were built since the
mid 1960s. 30 bit computer including ram and IO in a single 38 inch wide
cabinet 6 feet tall. Then i trained on the AN/UYK-7, a 32 bit modular
computer with a minimal configuration of 28 inches wide and 36 inches =
tall
that did about 1 mips. Still early 1970s. Both used magnetic core
memory.
?-)
Reply by Lawrence Statton●June 22, 20132013-06-22
John Larkin <jjlarkin@highNOTlandTHIStechnologyPART.com> writes:
>
> Come to think of it, it was probably a PDP-10 that I remember as
> using a PDP-8 to start things up.
>
Well, the KA and KI were "PDP-10s in hardware" so they had no
front-end processor. "Everyone knows" the FEP on the KL was a Unibus
11/40.
I'm pretty sure the KS had a microproccesor based front-end much like
the 8085 one in the contemporaneous VAXen)
After the PDP-8 had pretty-much gone to pasture, the only thing at
Digital that was 8-ish was the WS78, which used a PDP-8-on-a-chip (I
now can't recall if it was the Intersil 6120, or an in-house Digital
part, or a Digital-fabbed 6120)
--
NK1G - Lawrence
echo 'lawrenabae@abaluon.abaom' | sed s/aba/c/g
Reply by Lawrence Statton●June 22, 20132013-06-22
John Larkin <jlarkin@highlandtechnology.com> writes:
> I think the first VAXen had a PDP-8 as the console. It actually booted
> the VAX engine and did some slow i/o.
>
The first commercial VAX (the /780) had a very small LSI-11 (11/03 I'm
pretty sure, but I can't swear to that in court) that managed the RX01
boot diskette and loaded the WCS as part of the cold-start procedure.
It was in the very bottom of the left half of the CPU cabinet, under
the space where the massbus adapters were, IIRC.
The most vivid thing I recall of cold-starting the 780 was that the
LSI-11 ROM would load the head, read some very small amount (a sector?
Maybe two), unload the head, wait a few revs, load the head, read,
unlaod, wait ... It took several minutes to load the microcode off the
floppy. There were different revs of the microcode for VMS or BSD
4.3. (Didn't the KL have the same kind of deal -- there were
microcode differences for ITS versus "everything else"? [Rich
Alderson, you're on deck here....])
I never owned a /750, so I'm not sure what it had -- but I do recall,
that was the only of the 11/7x0 machines that could load the WCS from
the hard disk ... I also have a vague memory that the /750 had some of
the microcode in mask roms, and that only ECO fixes were loaded during
cold-start, but that's highly dubious -- I could be mistaking it for
another machine of the vintage.
The /730 had The Tape Drive Whose Name We Don't Utter and an 8085
controlling it to load the WCS ... and took - f - o - r - e - v
......... e - ..... r .... to boot. It did have the "neat - in the
same way that shoving a dozen undergrads into a phone booth was neat -
totally useless, but people do it 'because they can'" feature that the
boot media drive was available under VMS as an ALLOCable device. In
case there was someone masochistic enough that they wanted to make a
backup on very low-density very unreliable tape at a couple hundred
bytes per second. The serial port that connected the tape drive to
the console processor ran at 9600 bps, but only transferred six (am I
remembering that right? Was it six or was it four) bits at a time.
The one time that the boot-media drive was *actually* useful was if
you had the microcode assembler and source (which I did (painstakingly
typed in by hand from photostat prints made from the microfiches)), so
you could add instructions and what-not. By VMS 4.5, there was
something like eight words of WCS free after all the ECOs were loaded.
One night around pizza and beer, we did a back-of-the-envelope
calculation about the feasability of turning the /730 into a Very
Heavy, Very Slow, Very Noisy, Very Hot, High Power Consumption 680x0.
The consensus was "Maybe - but it would be a VERY tight fit".
--
NK1G - Lawrence
echo 'lawrenabae@abaluon.abaom' | sed s/aba/c/g