🚀 go-pugleaf

In article <129...@sun.Eng.Sun.COM> linden@sun.UUCP (Peter van der Linden) writes:
>The PDP-8 was a peculiar system.  I first programmed one in 1977,
>and it was obsolete even then.   It has a 12-bit word length,
>leading to an astonishing 4K of memory.   This can be extended by
>bank switching to 12K or even more.

> [..]

On a somewhat related note, the DECMate I (and II and III, I think)
word processors, which DEC brought out in the early 80's, ran on a PDP-8
microprocessor.

The DECMate was also something of a peculiar system.  The DECMate I
consisted of a somewhat modified VT100 terminal, which in addition to
the usual VT100 innards was equipped with a slot for the PDP-8 motherboard.
The VT100 sat atop a compact wheeled box, which contained two to four
RX02 8-inch floppy drives (klunk! klunk!) and a couple of logic boards
to drive them.  The keyboard was supported by a removable metal "shelf"
that mounted to the front of the rolling case.  (The "grabbers" that held
the shelf onto the case were made of what appeared to be cheap pot-metal,
and the ones in use here were constantly breaking.  We finally had our
machine shop whip up a bunch of stainless steel ones for us when we ran
out of spares.)

Not a bad system to use, considering that DEC had more or less cobbled
it together as a quick first entry into the word processor market (although
I think there was an even older version, using a VT52 and a slightly
different disk format, and (I think) called something else), and I think
the price was pretty reasonable for the time.  However, despite the
fact that the monitor was a (full-screen) VT100, all editing occurred on
the bottom line of the screen.  This was a major annoyance to many, who
tended to flock to true full-screen products like WordPerfect when they
became available.

I always wondered what method had been used by DEC to optimize the
handling of all those 8-bit characters using a 12-bit machine, or if
in fact any optimization had been implemented at all.  Does anyone know
how this was handled?  Was 2/3 (the upper 4 bits of each word) of the
DECMate's memory "wasted", or were the characters packed into memory
across word boundaries?  Were the characters written to disk in 8 or
12 bit "bytes"?  It seems like a good use of the upper 4 bits might
have been to indicate things like whether the character was bold,
underlined, etc.  Anyone know?


A curious former DECMate repair-being

...who wonders, too, how many DEC LQP02 (El Kewpie :) printers out there
might today be leading more-or-less productive lives attached to IBM PC's,
if only the stupid things knew how to handle EIA flow control (they used
only XON/XOFF).  Then again, maybe it's the stupid IBM PC, whose standard
printer device is (or was) too dumb to know about XON/XOFF...
-- 

---
"Puir Airnie"

In article <34...@tahoe.unr.edu> ma...@tahoe.unr.edu (Vampire In The Sun) writes:

[Some stuff about DECmate I, II and III deleted]

>Not a bad system to use, considering that DEC had more or less cobbled
>it together as a quick first entry into the word processor market (although
>I think there was an even older version, using a VT52 and a slightly
>different disk format, and (I think) called something else), and I think
>the price was pretty reasonable for the time.

It was called VT78. I don't exactly know why, but I guess the
VT is for the VT52, and the 8 in 78 is for the CMOS pdp8.

>I always wondered what method had been used by DEC to optimize the
>handling of all those 8-bit characters using a 12-bit machine, or if
>in fact any optimization had been implemented at all.  Does anyone know
>how this was handled?

I don't know about the wordprocessing software, but OS/8, the
"general" operating system used a scheme like this.
The first character was stored in the first word. The second
character was stored in the second word, and the four
high bits of the third character was stored in the first
word's high bits, while the four low bits was stored in
the second words high bits. If you didn't want to bother,
you just cleared the high bits in both words, and you
had two characters/word, but all character oriented device
drivers supported the three chars per two words packing.
I would guess the same applied to the wordprocessing
software.

Some might not consider this to be folklore, I don't
know. However, it is a thing from the past though...

To make the picture complete, DEC also released
something called DECstation 78 and DECstation 88.
The DECstation 78 was a repacking of the VT78,
and the DECstation 88 was a pdp8/a with different
configurations.

While I'm at it, I'll shoot some at those anti-pdp8
fans as well. The pdp8 could handle 32kW (12bit),
except for the pdp8/a (the last model), which could
handle 128kW. (So much for those "Max 12kW" speakers... :-)

Since we're on the pdp8, I'll make a short "commercial"
about the pdp8-lovers mailing list. Request should
be sent to pdp8-love...@mc.lcs.mit.edu
and all kind of old DEC equipment is mumbled about, not
just the pdp-8, though it is the most common machine...

> Then again, maybe it's the stupid IBM PC, whose standard
>printer device is (or was) too dumb to know about XON/XOFF...

Got to be that stupid PC, since DEC once upon a time
only did good stuff... :-)

======================================================================
Everybody know that the DECstation is a pdp8, which is a RISC,
but where did MIPS computers get into it?

	- Johnny Billquist
======================================================================

In article <129...@sun.Eng.Sun.COM> linden@sun.UUCP (Peter van der Linden) writes:
>The PDP-8 was a peculiar system.  I first programmed one in 1977,
>There are three instructions bits, hence 8 basic instructions in
>the machine.  Nowadays it would be lauded as a revolutionary RISC
>architecture.  It wasn't quite that simple then, e.g. because I/O
>(instruction 6) didn't use the address portion of a word, and hence
>this was available for further refining or extending the instruction
>set.

But you could do such *cool* things with the microprogrammed
stuff with group 1 and group 2 instructions! Remember clearing
the accumulator *and* the link bit using a single instruction --
7300, wasn't it? Or doing an OSR and a SZA to do neat things with
the address register lights?

I miss that machine; a gift from Tek to a local high school I
hung out near.

>Peter van der Linden     lin...@SuN.cOm    (415) 336-6206


-- 
David Boyes       Life is a fatal, sexually transmitted disease.
dbo...@rice.edu   

David Boyes, re PDP-8 instruction set:
--But you could do such *cool* things with the microprogrammed
--stuff with group 1 and group 2 instructions! Remember clearing
--the accumulator *and* the link bit using a single instruction --
--7300, wasn't it? Or doing an OSR and a SZA to do neat things with
--the address register lights?
--
Yep, packing code was fun on the PDP-8 with its 128 word "pages".  You
could address page zero or the one you were in. Page zero was usually used
for indirect references to other pages.  I wrote a utility package including
DECTape transfers to the high field (only 4K could be seen by the DECTape
at first), as well as several other neat things, all in the last page
of low core.  The first address of the last page was 7600, which was
a variant of the "clear accumulator" command.  So I could use this instruction
in page zero as a referent to my package, as well as a -200 (128 decimal)
counter-initialization constant for the DECTape transfer package.

"Any program can be made one word shorter"
-- 
Martin Taylor (m...@zorac.dciem.dnd.ca ...!uunet!dciem!mmt) (416) 635-2048
If the universe transcends formal methods, it might be interesting.
     (Steven Ryan).

In 1986 I worked for a company that had a DEC wordprocessor which turned
out to be a PDP-8. One day a DEC-technician arrived to repair something
in the machine.

To my surprise the machine still had core in it. When the repair was
finished, the technician asked if we were interested in a special demo
program he had. Sure, we're always interested.

The technician took an AM-radio from his toolbox and installed it next
to the machine. He tuned in on a certain frequency indicated on the
radio. Next he removed some panels from the PDP-8, uncovering the core.
He started a program from floppy which provided a menu from which you
could choose a song. You also had to specify the kind of terminal and
what type of floppy-drives were connected to the machine.

Then, we could hear the song on the radio!! The two DX-type floppy drives
ware mistreated as drums (they make an awful noise when the head is
loaded on the floppy!). The program also transmitted CTRL-G's to the
terminal to have it behave as another instrument.

One of the most spectacular demo's I've ever seen.
Actually it is very simple: the core generates interference when it is
accessed. All your program has to do is to access the core in such a way
that the correct interference is generated. The only problem is that
your program also resides in the core, so any changes to the program may
change the core's access-pattern, and subsequently changing the notes.

Rob.

My friend Jack relates the story of the time he found a forlorn PDP-8
and powered it up. On an impluse he did a search of core memory, and
yup, the FOCAL interpreter was still there and could be run.

In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
>My friend Jack relates the story of the time he found a forlorn PDP-8
>and powered it up. On an impluse he did a search of core memory, and
>yup, the FOCAL interpreter was still there and could be run.


I had heard a similar story, one where a PDP-8 arrived with the diagnostic
program still in core.  The guy telling the story had claimed to just start
it back up running fine after being shipped across the country.  I didn't
believe it at the time.

A fellow I knew in college (late 70's) had a Nova computer, apparently from
his high school (seems nobody else could run the thing so, when he left, he
got to take it!).  He'd drag it out of his closet, hook it up to the ASCII
terminal (which was an RCA TV and a wire-wrapped board he'd built), power it
on, and there was his BASIC interpreter and everything else he'd had in the
machine the last time...

In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
>In article<1990Jan5.2...@cs.rochester.edu>k...@cs.rochester.edu writes:
>>and powered it up. On an impluse he did a search of core memory, and
>>yup, the FOCAL interpreter was still there and could be run.

>program still in core.  The guy telling the story had claimed to just start
>it back up running fine after being shipped across the country.  I didn't
>believe it at the time.
-- 
{decuac.dec.com,cucstud,sundc}!c3pe!charles	ex::!echo Boo:

In article <6...@ehvie0.tq.ine.philips.nl> huls...@tq.ine.philips.nl
(Rob Hulsebos) writes about song playing on a PDP-8 using an AM radio.

So, we've come to the musical department in our "hangar flying"
newsgroup.  Guess there will be lots of entries about this.

The old swedish DATASAAB D21 from SAAB (60's, transistors...) had a
speaker on the console connected to some bit on the bus. One of the
last machines of this kind was run by the LYSATOR computer club at
Linkoping University, Sweden. It had a program named QUANTAM, which
plaued "Quantanamera" on the speaker, and used the line printer (or
was it the eight 1-inch tape stations?) for rythm section...

-- 
Per Lindberg (The Mad Programmer)  !      __!__
Front Capital Systems              !  _____(_)_____  Ceci n'est pas une Piper
Linneg 5, 11447 Stockholm, Sweden  !     !  !  !

In article <34...@tahoe.unr.edu>, ma...@tahoe.unr.edu (Vampire In The Sun) writes:
> In article <129...@sun.Eng.Sun.COM> linden@sun.UUCP (Peter van der Linden) writes:
>>The PDP-8 was a peculiar system.  I first programmed one in 1977,
>>and it was obsolete even then.   It has a 12-bit word length,
>>leading to an astonishing 4K of memory.   This can be extended by
>>bank switching to 12K or even more.
> 
>> [..]
> 
> On a somewhat related note, the DECMate I (and II and III, I think)
> word processors, which DEC brought out in the early 80's, ran on a PDP-8
> microprocessor.
> 

Most of what I had to say abt PDP-8's and DECmates has already been covered.
The DECmate series uses the Harris 6100 microprocessors (I use present
tense because I happen to like my DECmate II very much).

My first encounter with a PDP-8 was in 1981. The machine in question was
a PDP-8/e with 32KW running about 6 users under an operating system called
ETOS. ETOS gave each user a virtual PDP-8 that looked to all the world
like it was running OS/8.

I discovered that 12 users was about the limit for the machine. If you
managed to log in the 13th user, the 8 would spend all of its time in
the idle loop and never do any actual real work.

There's a good 8 story that my roommate has told me. One day, the 
multi-user PDP-8 was running very slowly. My roommate complained to the
system manager who, WHILE THE MACHINE WAS RUNNING, inserted a pair of
non-insulated metal scissors between two boards and snipped a wire
connecting them. The machine thereafter ran at its normal speed.

Perhaps I should mention that the PDP-8 was the second DEC machine
I ever used. The first was a VAX 11/780. Until only a few years ago,
I despised the 8. Since then I have "seen the light", and now own
two of them. It's really a very interesting architecture.

===============================================================================
Roger Ivie

35 S 300 W
Logan, Ut.  84321
(801) 752-8633
===============================================================================

Our PDP11/10 had non-volatile core (Harvard School of Public Wealth,
er, Health.)

We used it (mostly) one user at a time, there was a sign-up sheet
hanging next to it. This was a 28KB machine and we ran either
mini-unix or RT-11 (some people were still running RT-11 V2 with paper
tape as their major I/O device well into the late 70's.)

The worst disasters were when someone would halt and power down the
machine after running mini-unix and someone else would come along,
power it up and put their RT-11 RK05 disk into the drive (a whopping
2.5MB disk!)

Mini-unix still being resident in low core would immediately on
power-up sync the unix super-block to their RT-11 hard disk. Ah well,
gone, caused much grief.

Someone once said the main problem with computers was that some things
they remembered long after it was useful and other things they forgot
long before they should.
-- 
        -Barry Shein

Software Tool & Die, Purveyors to the Trade         | b...@world.std.com
1330 Beacon St, Brookline, MA 02146, (617) 739-0202 | {xylogics,uunet}world!bzs

>(Rob Hulsebos) writes about song playing on a PDP-8 using an AM radio.
>
>So, we've come to the musical department in our "hangar flying"
>newsgroup.  Guess there will be lots of entries about this.
sure will...
>
>Linkoping University, Sweden. It had a program named QUANTAM, which
>plaued "Quantanamera" on the speaker, and used the line printer (or
>was it the eight 1-inch tape stations?) for rythm section...

The ICL 2900 series also came with speakers attached to the bus, and demo
tunes on the engineers disk. The speaker was _very_ useful to us ops.
You could tell _immediately_ you had blundered, and crashed the thing.
Different programs had quite distinctive noises, as did tight loops in
interactive programs, and nothing, not even hitting The Power Switch,
at Christmas, could match the eerie silence when the processor hung.
(relative silence that is).

Nothing as good as using the line printer for rhythm though (sigh).

(You can still use radios to listen to PCs bTW)

ps as a new reader to this group, its t the funniest i've found (better even
than rhf). The story about tennis in the machine room brought back fond
memories...
-- 
 ++

In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
}In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
}>My friend Jack relates the story of the time he found a forlorn PDP-8
}>and powered it up. On an impluse he did a search of core memory, and
}>yup, the FOCAL interpreter was still there and could be run.

}I had heard a similar story, one where a PDP-8 arrived with the diagnostic
}program still in core.  The guy telling the story had claimed to just start
}it back up running fine after being shipped across the country.  ...

The same is true of the AP-101 computers on the Space Shuttle -- a source
of much amusement to the techs (and much confusion to novices) at the
Rockwell Avionics Development Lab (ADL).  Once we spent three weeks
thinking we were running our version of the software.  In actuality, some
lines were down and our stuff never loaded.  We were only there a few days
a week and the computers were often powered down in between.  It sure
_looked_ like our software.  The machines were powered down, ferchrissake.
Oh well.

("We" shall remain nameless to protect the embarrassed. (-: ).

-- 
The Polymath (aka: Jerry Hollombe, holl...@ttidca.tti.com)  Illegitimis non
Citicorp(+)TTI                                                 Carborundum
3100 Ocean Park Blvd.   (213) 450-9111, x2483
Santa Monica, CA  90405 {csun | philabs | psivax}!ttidca!hollombe

In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
|I had heard a similar story, one where a PDP-8 arrived with the diagnostic
|program still in core.  The guy telling the story had claimed to just start
|it back up running fine after being shipped across the country.  I didn't
|believe it at the time.

This is nothing special; the original Arpanet (Internet) IMPs (packet
switches) were all core based (almost everything was in those days).
They also had good automatic power failure and recovery mechanisms.
We powered them down for shipping running either the diagnostic or the
operational software.  Upon uncrating one of these machines at its
site, the FE just had to plug it in and turn it on and it was up and
running.

I kind of miss core.

	/JBL
=
Nets: le...@bbn.com  |  "There were sweetheart roses on Yancey Wilmerding's
 or {...}!bbn!levin  |  bureau that morning.  Wide-eyed and distraught, she
POTS: (617)873-3463  |  stood with all her faculties rooted to the floor."

In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
>In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
>>My friend Jack relates the story of the time he found a forlorn PDP-8
>>and powered it up. On an impluse he did a search of core memory, and
>>yup, the FOCAL interpreter was still there and could be run.
>
>
>I had heard a similar story, one where a PDP-8 arrived with the diagnostic
>program still in core.  The guy telling the story had claimed to just start
>it back up running fine after being shipped across the country.  I didn't
>believe it at the time.

In the late sixties, the negative bus PDP-8/i used to control a PICKER X-ray
diffractometer in the UCLA Dept. of Chemistry showed up in Los Angeles, CA
with the BIN and RIM loaders still in memory after shipment from MA. Nothing
of use on the DF-32 32KW disk, however, so the 4K DMS had to be loaded from
the ASR-35.

Is there a prize for minimum time to toggle in RIM? You could slap the switches
a lot faster on a PDP-8/i than you could twiddle the little metal switchies on
a NOVA. What ever did happen to the pleasures of self-modifying code? Those
bootstrap loaders were short. (Yes, the RIM replacement was shorter :-)
--
Mike Murphy  Sceard Systems, Inc.  544 South Pacific St. San Marcos, CA  92069
m...@Sceard.COM        {hp-sdd,nosc,ucsd,uunet}!sceard!mrm      +1 619 471 0655

Back around '81 or '82 I was doing a pm on a 360/40 system.  The console,
a selectric typewriter, seemed to need an asjustment.  After tweaking
around for a while, I decided to take the cover off to get at something.
In removing the cover, I shorted a contact to ground and the whole machine
powered down.

Remembering that the machine had core memory, I set it to single step mode,
reset the breaker, and powered it back up, went to run mode and pressed
start.  Much to my relief, everything continued running from where it
left off.  The customer never knew what happened.

dale

-- 
Dale P. Smith
                          da...@ncoast.org
                   ncoast!da...@hal.cwru.edu
uunet!hal.cwru.edu!ncoast!dale

In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
>In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
>>My friend Jack relates the story of the time he found a forlorn PDP-8
>>and powered it up. On an impluse he did a search of core memory, and
>>yup, the FOCAL interpreter was still there and could be run.
>
>I had heard a similar story, one where a PDP-8 arrived with the diagnostic
>program still in core.  The guy telling the story had claimed to just start
>it back up running fine after being shipped across the country.  I didn't
>believe it at the time.

Damn young whippersnappers.  That was SOP.  You were EXPECTED to "continue" the
diagnostics when your core machine was powered up the first time on site ...

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Brian G. Gordon	bri...@Corp.Sun.COM (if you trust exotic mailers)     |
|			...!sun!briangordon (if you route it yourself)	      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Our CDC3600 (actually, an upgraded 1604 if I remember right) had a
program that would play the Stars and Stripes Forever on it.  After
booting a tape (this machine didn't have disks on it until much later),
it would use the vacuum flapper valves in up to three tape drives as the
drums, the line printer for a crash cymbal, and the console speaker
(top three bits of the accumulator) as a fife.

Just before we sold it, I made a film of it.  In extremely maudlin days,
I dig out the film and sing along with my old friend....
	- Brian

The old Microdata REALITY systems had core memory in them; we'd get
them ready for a show by setting them up completely with all the demos
running, then turn it off, lock the disk heads, drive it down to the
show in a pickup truck, unlock the disks, and turn it back on.  It
always came right up where it had left off, still running the demos and
everything.  Just had to set the clock.

Watching the competition trying to get their demo machines running the
night before the show opened was lots of fun - while we were kicking
back and sucking the suds down.
	- Brian

One of my more interesting core experiences involved multi-user PDP-8s
during a thunderstorm. There was an 8/e and an 8/m side-by-side in the
same rack. The 8/m had core, the 8/e didn't. Every time the lights blinked,
the 8/e crashed but the 8/m kept running merrily along.

Of course, now that we have those companies building "ferroelectric" RAM
chips (read: core on a chip), we will be able to do this again. I can't
wait to build a ferroelectric memory for my 8/e...

===============================================================================
Roger Ivie

35 S 300 W
Logan, Ut.  84321
(801) 752-8633
===============================================================================

In article <129...@sun.Eng.Sun.COM>, a...@grendal.Sun.COM (Andrew MacRae) writes:
> In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
>>My friend Jack relates the story of the time he found a forlorn PDP-8
>>and powered it up. On an impluse he did a search of core memory, and
>>yup, the FOCAL interpreter was still there and could be run.
> 
> 
> I had heard a similar story, one where a PDP-8 arrived with the diagnostic
> program still in core.  The guy telling the story had claimed to just start
> it back up running fine after being shipped across the country.  I didn't
> believe it at the time.

Of course, the greatest core story of all time has to be that DEC used
spacewars as the final diagnostic on the PDP-1 and shipped all the PDP-1s
with it still in core. When the machine was installed, the field tech would
turn on the machine and start the machine executing (it was still in core)
to verify that everything was okay.

Refer to the book "Hackers" for a more detailed description.

===============================================================================
Roger Ivie

35 S 300 W
Logan, Ut.  84321
(801) 752-8633
===============================================================================

In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
>My friend Jack relates the story of the time he found a forlorn PDP-8
>and powered it up. On an impluse he did a search of core memory, and
>yup, the FOCAL interpreter was still there and could be run.

We used to make use of this characteristic, when I was working for a
company which had LOTS of PDP-8's, often differently configured, and which
also built them into its product as OEM controllers.  If you needed to
use a particular machine, and its paper-tape reader wasn't working
(not uncommon) or had been 'borrowed' (not uncommon) you could take
out the memory cards, swap them into another machine, load your program
from paper tape, and then put them back into the machine you wanted to
use.  Used to have to do that fairly regularly.

-- 
Paul Smee, Univ of Bristol Comp Centre, Bristol BS8 1TW, Tel +44 272 303132
 Sm...@bristol.ac.uk  :-)  (..!uunet!ukc!gdr.bath.ac.uk!exspes if you MUST)

In article <15...@sceard.Sceard.COM> m...@Sceard.COM (M.R.Murphy) writes:
>Is there a prize for minimum time to toggle in RIM? You could slap the switches
>a lot faster on a PDP-8/i than you could twiddle the little metal switchies on
>a NOVA. What ever did happen to the pleasures of self-modifying code? Those
>bootstrap loaders were short. (Yes, the RIM replacement was shorter :-)

There was always a certain amount of pleasure in toggling in the boot code
on the PDP-ll/35 from memory.  But when I found that I no longer needed
the instruction reference card while patching programs via the front
panel on the fly I decided it was time to move on to something else.
(The first thirty or forty words of stack always made a nice place to
stick patches.  After all, you *knew* that the program would never
use up that much stack space, right?)

In article <10...@ucsd.Edu> br...@ucsd.edu (Brian Kantor) writes:
|Our CDC3600 (actually, an upgraded 1604 if I remember right) had a
|program that would play the Stars and Stripes Forever on it.  After
|booting a tape (this machine didn't have disks on it until much later),
|it would use the vacuum flapper valves in up to three tape drives as the
|drums, the line printer for a crash cymbal, and the console speaker
|(top three bits of the accumulator) as a fife.

I saw this on a CDC 3600 (or 3200?) ... this was the summer of 1965 in
Santa Barbara or Goleta or somewhere like that.  Actually the tape
drives did the Sousaphone or tuba part (I thought it was by varying
the lengthy of the records being written), and the line printer was
drums (like snares).  They also had the first demo I ever heard of
digitized voice; at least the same speaker played what sounded like a
chorus singing "America the Beautiful".

(Oh yes; I believe the 1604 was the front end to the 3600, a function
equivalent to the ubiquitous 1401s in 7070 or 7090 shops.)

	/JBL
Nets: le...@bbn.com  |  "There were sweetheart roses on Yancey Wilmerding's
 or {...}!bbn!levin  |  bureau that morning.  Wide-eyed and distraught, she
POTS: (617)873-3463  |  stood with all her faculties rooted to the floor."

In article <130...@sun.Eng.Sun.COM> briang@sun.UUCP (Brian Gordon) writes:
}In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
}>In article <1990Jan5.2...@cs.rochester.edu> k...@cs.rochester.edu writes:
}>>My friend Jack relates the story of the time he found a forlorn PDP-8
}>>and powered it up. On an impluse he did a search of core memory, and
}>>yup, the FOCAL interpreter was still there and could be run.
}>
}>I had heard a similar story, one where a PDP-8 arrived with the diagnostic
}>program still in core.  The guy telling the story had claimed to just start
}>it back up running fine after being shipped across the country.  ...
}
}Damn young whippersnappers.  That was SOP.  You were EXPECTED to "continue" the
}diagnostics when your core machine was powered up the first time on site ...

Lest anyone think this phenomenon is a thing of the past, I just took
delivery of two industrial robots.  The factory diagnostic and burn-in
programs were still alive and well in their battery-backed non-volatile
RAM.

Unfortunately, we thought they were demos.  One of the commands reset the
calibration registers.  Oh well.

(They came with floppies on which were files containing the correct values
for the registers.  Otherwise I'd be in about $30K worth of trouble right
now).

I politely suggested to the manufacturer (name withheld to protect the
guilty) that, in future, they clear memory before shipping.  They agreed.

-- 
The Polymath (aka: Jerry Hollombe, holl...@ttidca.tti.com)  Illegitimis non
Citicorp(+)TTI                                                 Carborundum
3100 Ocean Park Blvd.   (213) 450-9111, x2483
Santa Monica, CA  90405 {csun | philabs | psivax}!ttidca!hollombe

In article <1990Jan8.2...@isis.educ.lon.ac.uk>
           tee...@isis.educ.lon.ac.uk (Matthew Moore) writes:

>The ICL 2900 series also came with speakers attached to the bus...
> ...
>Different programs had quite distinctive noises, as did tight loops...

Some of the users of the DATASAAB D21 at LYSATOR were quite proficient
at using the speaker as a diagnotics tool.

<crackle> <cracle> <crackle>  "Aha, it's compiling"...
<rackarackarackarackaracka>   "Linking"...
<snap> <crackle> <pop>        "OK, running"...

Reminds me of the Good Old BASIC Manual for the Apple ][ by Jef Raskin:

"If you can determine what's on a tape by listening to it, you are a
mutant, and will go far in the computer world"

Ahh, those were the times...

-- 
Per Lindberg (The Mad Programmer)  !      __!__
Front Capital Systems              !  _____(_)_____  Ceci n'est pas un Piper
Linneg 5, 11447 Stockholm, Sweden  !     !  !  !

In article <9...@front.se>, p...@front.se (Per Lindberg) writes:
> In article <1990Jan8.2...@isis.educ.lon.ac.uk>
>            tee...@isis.educ.lon.ac.uk (Matthew Moore) writes:
> 
>>The ICL 2900 series also came with speakers attached to the bus...
>> ...
>>Different programs had quite distinctive noises, as did tight loops...
> 
> Some of the users of the DATASAAB D21 at LYSATOR were quite proficient
> at using the speaker as a diagnotics tool.
> 
> <crackle> <cracle> <crackle>  "Aha, it's compiling"...
> <rackarackarackarackaracka>   "Linking"...
> <snap> <crackle> <pop>        "OK, running"...
> 

I used to own a TI57 calculator. It's battery pack made different noises
depending upon how many segments were lit in the display. I used to program
it to play simple songs (you had to REALLY use your imagination to recognize
them).

We have an old CP/M machine hanging around the office built by Digilog. It
has a speaker whose amplifier picks up random noise from the machine. You
really can tell what the machine's doing by the noise it makes. One of
the guys at the office liked that feature so much that when he built his
own homebrew, he ran the speaker amp inputs all around the place so that
he had the same effect.

My roommate recently installed a large hard disk on his Sun 4. Because the
room was so noisy with fans, he tuned the radio to the drive so that he
could hear it seek. At other frequencies, you could hear the data going
to the drive.


===============================================================================
Roger Ivie

35 S 300 W
Logan, Ut.  84321
(801) 752-8633
===============================================================================

The most amazing thing I've seen in non-volatile memory was the 2114
static ram chip.  When they called them static they weren't kidding;
those things would hold onto programs for DAYS.  I've successfully run
a Basic interpreter that had been loaded into ram several days earlier
even though the machine had sat for more than 24 hours with the power
off.

Of course it helped that the system didn't run a memory test on power up.
	- Brian

In article <50...@bbn.COM> le...@bbn.com (Joel B Levin) writes:

   In article <129...@sun.Eng.Sun.COM> acm@sun.UUCP (Andrew MacRae) writes:
   |I had heard a similar story, one where a PDP-8 arrived with the diagnostic
   |program still in core.  The guy telling the story had claimed to just start
   |it back up running fine after being shipped across the country.  I didn't
   |believe it at the time.

   This is nothing special; the original Arpanet (Internet) IMPs (packet
   switches) were all core based (almost everything was in those days).
   They also had good automatic power failure and recovery mechanisms.
   We powered them down for shipping running either the diagnostic or the
   operational software.  Upon uncrating one of these machines at its
   site, the FE just had to plug it in and turn it on and it was up and
   running.

True story: Back when I worked for DEC, I was doing a field test installa-
tion at a customer site. One of the interesting tasks I inherited was to
make sure all the old kruftye comm gear still worked with our latest-and-
greatest OS release. They had this little ole PDP-8 hooked up with 96
anal-thermometers monitoring rabbits in an experiment (Yes, folks, a TRUE
case of online assholes! sorry). We (DEC) had de-supported this particular
comm frontend about a decade earlier, but TOPS-10 being TOPS-10, we liked
to claim that old stuff just kept on working. And it did. For, when I asked
them where the sources were (in case I needed to look at them to figure out
what wasn't working), they said they didn't have them anymore. OK, said I,
how about the binary (more out of curiousity - I had no intention of looking
at a binary dump of a PDP-8!). "Well, uh," they said, "we don't have that
anymore either". In fact, the last time it had been loaded was at least
five years earlier - no one then employed at the computer center even knew
how to load the PDP-8, assuming they had the code to load into it! I asked
them what happens if it ever breaks, and they said that field service comes
out and replaces whatever board is broken. They replaced the cpu board, the
power supply, the ..., everything BUT the core memory stack! And when that
goes? asked I. Oh, then I guess the experiment is over...

					-RDH

In article <16...@cc.usu.edu> SL...@cc.usu.edu (Roger Ivie) writes:
   My roommate recently installed a large hard disk on his Sun 4. Because the
   room was so noisy with fans, he tuned the radio to the drive so that he
   could hear it seek. At other frequencies, you could hear the data going
   to the drive.

We were trying to hack the timer circuitry on a Canon As-100 (really weird, like
an combination of a IBM-PC and a failed graphics workstation).  Since the system was
CP/M-86, it was a pain trying to figure out if the once-a-second timer was running or
not.  So, we tuned the radio onto "the frequency" of the machine, and suprise, suprise,
you could actually hear the timer going tick tick tick once a second.  We tried all
kinds of things, and finally got the timer acting the way we wanted it.

I'd call it a wireless in-circuit debugger :-)
--
* * * Otto J. Makela (ot...@jyu.fi, MAKELA_OTTO_@FINJYU.BITNET) * * * * * * *
* Phone: +358 41 613 847, BBS: +358 41 211 562 (CCITT, Bell 2400/1200/300) *
* Mail: Kauppakatu 1 B 18, SF-40100 Jyvaskyla, Finland, EUROPE             *
* * * freopen("/dev/null","r",stdflame); * * * * * * * * * * * * * * * * * *

In article <1990Jan9.1...@NCoast.ORG> da...@ncoast.ORG (Dale Smith) writes:
>Back around '81 or '82 I was doing a pm on a 360/40 system.  The console,
>a selectric typewriter, seemed to need an asjustment.  After tweaking
>around for a while, I decided to take the cover off to get at something.
>In removing the cover, I shorted a contact to ground and the whole machine
>powered down.
>
>Remembering that the machine had core memory, I set it to single step mode,
>reset the breaker, and powered it back up, went to run mode and pressed
>start.  Much to my relief, everything continued running from where it
>left off.  The customer never knew what happened.
>

Back in '78 I had just started a job programming in assembler on a 360/40.
The manager showed me a terminal where I could try out some of the on-line
applications.  All our work was done via punch cards, but amongst the
simple hangman and word search games I found an on-line debugger.  It would
let you enter 360 assembler code and would single step through each
instruction allowing you to dump and modify the values of the registers
and memory locations.  I thought that was pretty high tech and tried it
out at once.  After a few minutes, just as I was getting the hang of single
stepping through my code, the system manager came back to where I was and
asked if by chance I was using the debugger.  I said yes and began to show
him what I was doing with it but he interrupted and asked that I exit the
program *now*.  Afterwards he explained that the debugger had a small problem.
It seemed that when I put the debugger into single step mode, I was putting
the *entire* system into single step mode.  Forty terminals and half a dozen
batch jobs had all been single stepping for the last fifteen minutes.

						Andrew MacRae

People who just tell stories about core memory are all wimps. :-)  Real
programmers still have a machine running with it.  We do.  Actually, it's
sort of a museum case in several ways:  it runs the last functioning card
reader on campus!  The reader is used a few times a year for converting
"dusty decks" to more modern formats.  The interface is a real cheapie
job with no buffering, and it's very difficult for an operating system
that ever locks out interrupts to keep up with it.  So one of our old
pdp11/05s, core memory and all, is now a dedicated card-reader controller,
hooked into our production systems by serial line.  Naturally, I couldn't
resist the temptation to make the program power-fail proof, so we never
have to reload the thing.  (We *can* still do it if needed, however!)
-- 
1972: Saturn V #15 flight-ready|     Henry Spencer at U of Toronto Zoology
1990: birds nesting in engines | uunet!attcan!utzoo!henry he...@zoo.toronto.edu

In article <1990Jan16.181122.3747@utzoo.uucp>, henry@utzoo.uucp (Henry
Spencer) writes:
> People who just tell stories about core memory are all wimps. :-)  Real
> programmers still have a machine running with it.  ...

	On a related tangent, my nomination for "Least Wimpy
	Memory" was the core on a 360/91.  When UCLA's 91KK
	arrived in 1968, its 4 megabytes of core made it one
	of (I think) the 2 biggest machines in the world.

	This isn't "big" just in megabytes; the core sat in what
	we'd now think of as "giant" boxes at opposite ends of
	the CPU.  It used so much power that it was water-cooled,
	via copper plumbing under the machine room's raised floor.

	... And at least once the machine went down because
	the plumbing sprang a leak.


----------------
Paul Raveling
Rave...@isi.edu