blinking lights on computers
Michael Krell
Over in misc.writing.screenplays, someone wrote:
> The scientists who invented ENIAC felt that the computer was just not
> that exciting to tv viewers so they surrounded it with blinking
> lights, lights that had nothing to do with what was going on inside
> the computer. the media loved it and blinking lights have been a
> movie computer stable since.
Can anyone confirm this story? It sounds a bit urban-legendish to me.
Mike
Matthew P. Sayler
Michael Krell (mk...@jade.cs.Virginia.EDU) wrote one fine day:
I'm fairly sure read this in a book called _Bit by Bit_, author not
remembered at present. I picked it up at a dollar book sale at the
local library one day, and it provided at least a dollar's worth of
reading. I'm not sure, but I want to say that the blinkenlights were
ping pong balls with the digits 0-9 stenciled on them--the Real
Users had no use for such things, but they sure looked neat.
I'll try to corroborate this when I return to the dorm . .
m@
--
-- Matt Sayler -- University of Texas, Austin, Texas, USA -- . _ _ ,--.
-- I speak only for myself and my invisible friends -------- | | | | .-|
-- mpsa...@cs.utexas.edu, mpsa...@jove.acs.unt.edu ------- | | | | |_|
-- http://www.cs.utexas.edu/users/mpsayler (512)472 5064 --- `-__.`
Carl R. Friend
Matthew P. Sayler wrote in article Nr. <55jg3o$12...@opus.cs.utexas.edu>:
>
> Michael Krell (mk...@jade.cs.Virginia.EDU) wrote one fine day:
> > Over in misc.writing.screenplays, someone wrote:
>
> > > The scientists who invented ENIAC felt that the computer was just not
> > > that exciting to tv viewers so they surrounded it with blinking
> > > lights, lights that had nothing to do with what was going on inside
> > > the computer. the media loved it and blinking lights have been a
> > > movie computer stable since.
>
> > Can anyone confirm this story? It sounds a bit urban-legendish to me.
>
> I'm fairly sure read this in a book called _Bit by Bit_, author not
> remembered at present. I picked it up at a dollar book sale at the
> local library one day, and it provided at least a dollar's worth of
> reading. I'm not sure, but I want to say that the blinkenlights were
> ping pong balls with the digits 0-9 stenciled on them--the Real
> Users had no use for such things, but they sure looked neat.
True, the users of the machines didn't have a lot of use for all
those lights, but the maintenance staff sure did! The thing to remember
is that this was before the microprocessor arrived on the scene -
computers were frequently composed of hundreds, if not thousands, of
components and boards (some didn't even use PC boards) and the lights
gave _very_ important clues as to the locations of various problems.
It's very safe to say that the lights were _not_ placed there
solely for the edification of an occasional tour group. A good
technician could usually look at the lights and head right for the
problem area of the machine.
--
______________________________________________________________________
| | |
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:carl....@cliff.swec.com | |
| http://www.ultranet.com/~engelbrt/carl/museum | ICBM: N42:22 W71:47 |
|________________________________________________|_____________________|
Andrew
In article <55jg3o$12...@opus.cs.utexas.edu>, mpsa...@cs.utexas.edu
(Matthew P. Sayler) wrote:
> Michael Krell (mk...@jade.cs.Virginia.EDU) wrote one fine day:
> > > The scientists who invented ENIAC felt that the computer was just not
> > > that exciting to tv viewers so they surrounded it with blinking
> > > lights, lights that had nothing to do with what was going on inside
> > > the computer. the media loved it and blinking lights have been a
> > > movie computer stable since.
>
> > Can anyone confirm this story? It sounds a bit urban-legendish to me.
>
> I'm fairly sure read this in a book called _Bit by Bit_, author not
> remembered at present. I picked it up at a dollar book sale at the
> local library one day, and it provided at least a dollar's worth of
> reading. I'm not sure, but I want to say that the blinkenlights were
> ping pong balls with the digits 0-9 stenciled on them--the Real
> Users had no use for such things, but they sure looked neat.
I have never used an ENIAC, but have done work on 'blinking lights'
type systems.
Many older systems tied various registers and busses to lights
so that they could be monitored. When the machine 'crashed',
or was 'single stepped' the lights would indicate the actual
internal state of the machine.
Even when running at full speed the lights connected to the
program counter (PC) changed in a very nice pattern which was
easy to notice if it became abnormal. (At least running the
program my machine did 24 hours a day which was air defense
oriented)
I would bet a fair bit of money that the lights in ENIAC did
exactly the same thing. They simply exported important internal
state information usefull in testing, debugging, and maintaince.
Oh well....
--
Andrew Carol "Could be worse. Could be raining."
car...@apple.com ca...@woz.org
ed thelen
How time flies, and times change.
I got into the computer fixing game late (1960), but even then
the multitude of lights were very handy for program debugging
and hardware trouble shooting.
Its hard to believe, but there were no software debuggers.
A system programmer reserved time on one of these big
expensive machines, sat down, and single stepped through
looking at the lights for the program counter, the register
contents, and condition flags. (Fortran application programers
could add print statements and view the intermediate results.)
I remember how shocked I was to find that the DEC LSI-11
(a PDP-11 variant) had no lights, ah, saved, depressing the
halt switch put the micro-code into "ODT" mode and you
could single step, etc from your teletype.
How times have changed!!!
Michael Krell <mk...@jade.cs.Virginia.EDU> wrote in article
<E0BF9...@murdoch.acc.Virginia.EDU>...
> Over in misc.writing.screenplays, someone wrote:
>
> > The scientists who invented ENIAC felt that the computer was just not
> > that exciting to tv viewers so they surrounded it with blinking
> > lights, lights that had nothing to do with what was going on inside
> > the computer. the media loved it and blinking lights have been a
> > movie computer stable since.
>
> Can anyone confirm this story? It sounds a bit urban-legendish to me.
>
> Mike
>
>
>
Marvin E. Kurtti
In article <327E42...@corp2.mck.co.za>,
se...@corp2.mck.co.za says...
> Carl R. Friend wrote:
> >
> > It's very safe to say that the lights were _not_ placed there
> > solely for the edification of an occasional tour group. A good
> > technician could usually look at the lights and head right for the
> > problem area of the machine.
why their panels showed up for years in SF tv shows and
movies. The XD-1 at Lincoln labs didn't have as big a
panel but all the essential regs of the cou were shown.
One day when I was the duty programmer during a live
exercise with radars tracking live aircraft. They
started having problems with the speed reported by the
computer. I had been staring at the blinking neons
when I noticed something on the clock register. It
looked to me like one of lights that should be blinking
at 1/2 or 1/4 sec was stuck on. An IBM white shirt
looked at it and said he didn't see anything. I
insisted so we went back to the clock rack. He probed
around a little and found a flip-flop stuck on.
Years later when I got my first microprocessor (8080) I
said to myself how the hell are you supposed to debug
these things, you can't see the registers.
Marv
Huntsville Alabama
Mikie
Michael Krell wrote:
>
> Over in misc.writing.screenplays, someone wrote:
>
> > The scientists who invented ENIAC felt that the computer was just not
> > that exciting to tv viewers so they surrounded it with blinking
> > lights, lights that had nothing to do with what was going on inside
> > the computer. the media loved it and blinking lights have been a
> > movie computer stable since.
>
> Can anyone confirm this story? It sounds a bit urban-legendish to me.
>
> Mike
blinky lights. The lights were suppose to tell you: The PSW, the
register that was being accessed, and the real time clock. Of course
all this valuable information was useless if the system crashed because
little blinky lights got dead when dos computer goes dead.
Nick Spalding
.wri...@sabu.ebay.sun.com wrote:
> When ENIAC was designed, TV was many years in the future.
> The story is a legend.
TV broadcasting started in 1936 in England and 1939 in the USA (the
first broadcasts were from the New York Worlds Fair of that year),
many years before ENIAC.
--
Nick Spalding
Derek Peschel
In article <55l6ik$5...@male.EBay.Sun.COM>, <wri...@sabu.ebay.sun.com> wrote:
>In article B...@murdoch.acc.Virginia.EDU, mk...@jade.cs.Virginia.EDU (Michael Krell) writes:
>> Over in misc.writing.screenplays, someone wrote:
>> > The scientists who invented ENIAC felt that the computer was just not
>> > that exciting to tv viewers so they surrounded it with blinking
>> > lights, lights that had nothing to do with what was going on inside
>> > the computer. the media loved it and blinking lights have been a
>> > movie computer stable since.
>> Can anyone confirm this story? It sounds a bit urban-legendish to me.
>>
>> Mike
>When ENIAC was designed, TV was many years in the future.
>The story is a legend.
I disagree -- the version above isn't completely true, but it's certainly not
a legend either. From my recollection of "The Machine that Changed the World":
TV broadcasts weren't common in the days of ENIAC, but newsreels were.
At least one ENIAC story made it onto the news (probably more). The
programmers did want to make the machine look exciting (or comprehensible, at
least), and they did attach blinking lights to the machine. (Someone mentioned
ping-pong balls... yes, that's what the lights were.)
But the lights weren't useless. They showed the results of a program
(probably firing-table calculations, which the ENIAC did a lot of). I'm sure
that impressed people more than seeing punched-card output would have.
The lights were arranged in columns, something like this:
9 9 9 9 9 9 9 9
8 8 8 8 8 8 8 8
7 7 7 7 7 7 7 7
. . . . . . . . . .
1 1 1 1 1 1 1 1
0 0 0 0 0 0 0 0
and updated in a rather wierd sequential fashion -- a whole row of 0's would
appear, change to 1's, then 2's, etc. At each step, some of the lights stopped
changing. Eventually, the whole answer appeared. I'm sure that the
distributed design of the machine explains that kind of pattern.
-- Derek
John Everett
In article <327E65...@aol.com>, MBer...@aol.com says...
>
>Michael Krell wrote:
>>
>> Over in misc.writing.screenplays, someone wrote:
>>
>> > The scientists who invented ENIAC felt that the computer was just not
>> > that exciting to tv viewers so they surrounded it with blinking
>> > lights, lights that had nothing to do with what was going on inside
>> > the computer. the media loved it and blinking lights have been a
>> > movie computer stable since.
>>
>> Can anyone confirm this story? It sounds a bit urban-legendish to me.
>>
>> Mike
>blinky lights. The lights were suppose to tell you: The PSW, the
>register that was being accessed, and the real time clock. Of course
>all this valuable information was useless if the system crashed because
>little blinky lights got dead when dos computer goes dead.
This thread seems to be rapidly filling up with posts from people who
apparently never worked on computers with status lights. If all the lights on
a System/360 went dead it meant you just suffered a power failure. Otherwise,
you could read the status of the machine when it crashed. One installation I
worked at (a DECsystem-10 timesharing bureau) had Polaroid cameras in the
machine room. When a system crashed, the operator took a crash dump, filled
out a crash log, and attached a Polaroid of the CPU and Memory status lights
to the log entry. The systems programmer (me) found the photos invaluable
when trying to analyze the dump.
We had some really creative operators who found other uses for the Polaroid
cameras. One of the classics was a photo essay on the sex life of a dskrat.
This was eventually framed and AFAIK still resides in the office of a certain
ADP executive. Old TOPS-10 hands will know what a dskrat was.
--
jeve...@wwa.com (John V. Everett)
Carl R. Friend
Douglas W. Jones,201H MLH,3193350740,3193382879 wrote
in Aarticle Nr. <55js29$14...@flood.weeg.uiowa.edu>:
>
> From article <01bbc9f3$9a594c40$4693d6ce@ed>,
> by "ed thelen" <eth...@ix.netcom.com>:
>
> > I remember how shocked I was to find that the DEC LSI-11
> > (a PDP-11 variant) had no lights, ah, saved, depressing the
> > halt switch put the micro-code into "ODT" mode and you
> > could single step, etc from your teletype.
>
> blinking lights. Whether it's a mainframe or a micro, or even a modem,
> having the row of lights up front really did seem handy, and I miss them.
> (Of course, I've collected a few machines that have them, and as I write
> this, I'm watching the lights on my old US Robotics Courrier HST modem
> blink.)
I think that they stopped making computers when front panels
disappeared. Without blinking lights (some will say loud fans), they
look more loke heaters than anything else. Sniff, sniff, whimper. :-(
For what it's worth, I find looking at light patterns on panels
much more intuitive than decoding octal (or hex) on a Teletype. This
is especially so on panels that show more than one register
simultaneously.
Like Dr. Jones, I, too collect machines with _real panels_ on them
and wouldn't trade my old external modem for the world (it also works
with other members of my collection).
J. Chris Hausler
Derek Peschel <dpes...@u.washington.edu> writes:
>In article <55l6ik$5...@male.EBay.Sun.COM>, <wri...@sabu.ebay.sun.com> wrote:
>>In article B...@murdoch.acc.Virginia.EDU, mk...@jade.cs.Virginia.EDU (Michael Krell) writes:
>
>>> > The scientists who invented ENIAC felt that the computer was just not
>>> > that exciting to tv viewers so they surrounded it with blinking
>>> > lights, lights that had nothing to do with what was going on inside
>>> > the computer. the media loved it and blinking lights have been a
>>> > movie computer stable since.
>
>>> Can anyone confirm this story? It sounds a bit urban-legendish to me.
>>>
>>> Mike
>
>>The story is a legend.
>
>I disagree -- the version above isn't completely true, but it's certainly not
>a legend either. From my recollection of "The Machine that Changed the World":
>
>TV broadcasts weren't common in the days of ENIAC, but newsreels were.
>At least one ENIAC story made it onto the news (probably more). The
>programmers did want to make the machine look exciting (or comprehensible, at
>least), and they did attach blinking lights to the machine. (Someone mentioned
>ping-pong balls... yes, that's what the lights were.)
>
>But the lights weren't useless. They showed the results of a program
>(probably firing-table calculations, which the ENIAC did a lot of). I'm sure
>that impressed people more than seeing punched-card output would have.
>
>The lights were arranged in columns, something like this:
>
>9 9 9 9 9 9 9 9
>8 8 8 8 8 8 8 8
>7 7 7 7 7 7 7 7
> . . . . . . . . . .
>1 1 1 1 1 1 1 1
>0 0 0 0 0 0 0 0
>
>and updated in a rather wierd sequential fashion -- a whole row of 0's would
>appear, change to 1's, then 2's, etc. At each step, some of the lights stopped
>changing. Eventually, the whole answer appeared. I'm sure that the
>distributed design of the machine explains that kind of pattern.
>
>-- Derek
small to show up well on film so a display of brighter lamps surrounded
by ping pong balls with the digits painted on them was used for the film.
The use of lights on ENIAC was for diagnostic debugging of the machine,
the use of the lighted ping pong balls was just for the film. (I believe
this film was made to announce the "birth" of ENIAC as it did not
actually come on-line until after the war and thus there were no
security issues (ENIAC was principally designed to calculate shell
trajectories although it was quickly put to other uses too).
Max F Lang
Carl R. Friend wrote:
>
> Douglas W. Jones,201H MLH,3193350740,3193382879 wrote
> in Aarticle Nr. <55js29$14...@flood.weeg.uiowa.edu>:
> >
> >
> > I also remember my surprise when they phased out making machines with
> > blinking lights. Whether it's a mainframe or a micro, or even a modem,
> > having the row of lights up front really did seem handy, and I miss them.
> I think that they stopped making computers when front panels
> disappeared. Without blinking lights (some will say loud fans), they
> look more loke heaters than anything else. Sniff, sniff, whimper. :-(
>
I'm kind of a newbie with computers (only been using them since the
Apple ][
came out), but I read somewhere that operators at many installations
used to
keep an AM radio nearby the main system, and keep it tuned to the white
noise
made by the system. Supposedly they could tell alot about the state of
the
system from the various whistles and beeps that came out of the radio.
Of course, along time ago, I tried this with my first PC, but I don't
think I
ever heard one peep that had anything to do with the box.
/\/\ax
Sean Stanley-Adams
Carl R. Friend wrote:
>
> It's very safe to say that the lights were _not_ placed there
> solely for the edification of an occasional tour group. A good
> technician could usually look at the lights and head right for the
> problem area of the machine.
>
IBM 360/40 and a low-end 370, both of which were liberally festooned
with lights. Though I wasn't able to use the lights to the same extent
as the hardware engineers, it was possible with practise to identify a
looping program. The programmers could never understand how we did it
but then they didn't spend long nights with precious little else to do
but looking at the blinking lights!
Andrew Carol
In article <327E65...@aol.com>, MBer...@aol.com wrote:
> Michael Krell wrote:
> >
> > Over in misc.writing.screenplays, someone wrote:
> >
> > > The scientists who invented ENIAC felt that the computer was just not
> > > that exciting to tv viewers so they surrounded it with blinking
> > > lights, lights that had nothing to do with what was going on inside
> > > the computer. the media loved it and blinking lights have been a
> > > movie computer stable since.
> >
> > Can anyone confirm this story? It sounds a bit urban-legendish to me.
> >
> This one I suspect is true..... I worked on an IBM 360 with tons o'
> blinky lights. The lights were suppose to tell you: The PSW, the
> register that was being accessed, and the real time clock. Of course
> all this valuable information was useless if the system crashed because
> little blinky lights got dead when dos computer goes dead.
I think you underestimate the value of those lights!
Depending on the failure, those lights provided many clues to
the hardware people. Unless the failure was _total_ (ie, power)
most 'blinky-lights' systems maintined them.
They are also useful, after a reboot, when running hardware
tests because they allowed us to pull out the internal state
of the machine.
I could not have done my maintaince for our military systems
if it were not for the 'blinky-lights'.
Lisa Jeff
Per Sabu's post...
While TV wasn't widespread in the late 1940s, movie newsreels certainly
were.
According to the PBS documentary, the ENIAC had a display of pingpong
balls with numbers on them just for show, and it was a big hit.
However, when computers were programmed in assembler language, debugging
was often done stepping through instructions and viewing the
intermediate internal results on the console lamps by the programmers.
I think by S/360 this petered out, though I'm sure it was done often.
I believe in the S/370 console lights were phased out in favor of CRT
displays.
Console lights served two purposes---debugging and "for show". The
computer mfrs realized the computer was new and some blinking lights
added to it---gave it more mystique than a plain gray box.
Kedamono
In article <carol1-0411...@17.219.103.246>, car...@apple.com
(Andrew Carol) wrote:
> I think you underestimate the value of those lights!
>
> Depending on the failure, those lights provided many clues to
> the hardware people. Unless the failure was _total_ (ie, power)
> most 'blinky-lights' systems maintined them.
>
> They are also useful, after a reboot, when running hardware
> tests because they allowed us to pull out the internal state
> of the machine.
>
> I could not have done my maintaince for our military systems
> if it were not for the 'blinky-lights'.
>
> Oh well....
During my stint in the army, circa 1980 to 1984, I was assigned to a IHAWK
unit as a "computer mechanic". After working on the ADP system that ran
the IHAWK system, computer mechanic was about right.
We had two vans that housed identical systems, and up to 1983, were
programed by punched metal tape. It took 7 minutes to load and fire up the
main program that ran the radar consoles and firing solutions in the
command van, it took 14 minutes to load the diagnostic tape, a couple of
minutes to find the problem, and another 7 minutes to reload the main
program again during a 30 minute tactical eval. We always had a crew
outside ready to uncable the main van, and hook up the secondary just in
case.
The diagnostic program showed all it results as blinking lights or light
patterns, and could be stepped through until it fatally terminated on the
fault. The light patterns would correspond to different sequences and
tests, and hopefully point out the bad card or component. Unfortunately,
the diagnostic system on the main van was broken, we had to hot wire one
portion of the circuitry, to prevent a false error from coming up. No
matter what cards we tested, pulled, ohmed out, we couldn't find where the
fault was. They finally took the van to higher echelon maintance, and we
found out that the cabling was screwed up. They literally had to
disassemble the entire ADP unit down to the runners of the circuitry
drawers to find that problem.
Finally in 1983, we got eprom chips that would program the computer in
seconds. In fact they were to fast, so a delaying circuit had to be
retrofitted to slow down their clock, so the ADP could understand what
they were saying. This didn't stop us from having a crew ready in the
secondary van during an evaluation, it just meant we got to have more trys
at finding the problem.
When the eproms where finally given the full go ahead, we had a tape
buring party. The taps where made out of aluminum, and they burned real
good, with thermite. Interesting note: I took the diagnostic tape and
threw it up into the air, let the tape dangle and roll off the reel. I
gave the reel a good spin, and it spun fast enough, that the tape coming
off the reel gave it enough lift, that the whole thing hung in the air 7
feet up for about 2 seconds. We rewound the tape by hand and did it again
and again, it was amazing.
--
The Kedamono Dragon
I'm now on Concentric as well as on AOL
Two of me for the price of One!
Keda...@concentric.net or Keda...@aol.com
----------------------
Take a look at the Alternate History Travel Guides!
http://users.aol.com/kedamono/sliders/alterguides.html
Nick Spalding
.Max F Lang <mfl...@bellsouth.net> wrote:
> I'm kind of a newbie with computers (only been using them since the
> Apple ][
> came out), but I read somewhere that operators at many installations
> used to
> keep an AM radio nearby the main system, and keep it tuned to the white
> noise
> made by the system. Supposedly they could tell alot about the state of
> the
> system from the various whistles and beeps that came out of the radio.
> Of course, along time ago, I tried this with my first PC, but I don't
> think I
> ever heard one peep that had anything to do with the box.
>
They worked a great deal better on machines with mag. core memories,
and of course much larger machines with miles of wire to radiate from.
--
Nick Spalding
Joe Morris
hanc...@cpcn.com (Lisa Jeff) writes:
>However, when computers were programmed in assembler language, debugging
>was often done stepping through instructions and viewing the
>intermediate internal results on the console lamps by the programmers.
>I think by S/360 this petered out, though I'm sure it was done often.
On the larger computers the practice ended because of economic limits: the
machines were ruinously expensive (and time on the non-multiuser systems
was charged based on wall-clock time) and there were too many users
who had work to be done on the machines. In the days of big iron, the
only people who routinely used the console lights for debugging were
the system programmers (ahem) and the maintenance staff.
This was responsible for some of the singularly lousy design of many
programs, because the application programmer in many cases had not
the faintest idea of how the system really worked, and how program
design could wreck performance. (At a university I once worked for we
had a programmer who was assigned the task of extracting a number of
names at random from the student master record tape. He took the known
number of student names on the tape (somewhat over 35,000), generated
a random number in the interval [0,1], multiplied the two numbers,
spaced down the tape that many records, read the entry, rewound the
tape, and then repeated the process...and was totally clueless about
just *why* the operators killed the job.
The early S/360 boxes had walls of lights; in fact, the model 75 had
so many that the early serial number machines would blow the console
power supply if the "Lamp Test" button was pressed. The S/370 systems
began the use of video terminals to replace the console; you could still
display and change the machine registers but the impressive array of
lamps and switches were gone.
>Console lights served two purposes---debugging and "for show". The
>computer mfrs realized the computer was new and some blinking lights
>added to it---gave it more mystique than a plain gray box.
True; they *were* good for impressing the clueless. We had a 7040
at a PPOE; I discovered a way to use some "don't care" holes in
the hardware logic (along with a couple of pieces of wadded-up punched
cards to jam certain console pushbuttons) to produce an impressive but
totally meaningless display for our open houses.
Joe Morris / MITRE
Bill Thater
In article <55ni3a$b...@top.mitre.org>,
jcmo...@mwunix.mitre.org (Joe Morris) writes:
:>True; they *were* good for impressing the clueless. We had a 7040
:>at a PPOE; I discovered a way to use some "don't care" holes in
:>the hardware logic (along with a couple of pieces of wadded-up punched
:>cards to jam certain console pushbuttons) to produce an impressive but
:>totally meaningless display for our open houses.
i'm gld to see someone else did the "flash the pretty lights" hack. as
a "glass castle" operator for a state agency, i wrote a S/360 hack to
flash the lights in pretty patterns for "showing off" to the tours.
remember playing music on the old 1403s? ;-)
--
o.-.o Witch's Bear | o.-.o
((^)) Bill Thater, KiTA, BOfH | you can't stop the waves, ((^))
0)._.(0 DBA for NPAC | but you can learn to surf! ;-) 0) _ (0
O/ \O bi...@npac.syr.edu | O/ \O
Heinz W. Wiggeshoff
Joe Morris (jcmo...@mwunix.mitre.org) writes:
>
...snip...
> This was responsible for some of the singularly lousy design of many
> programs, because the application programmer in many cases had not
> the faintest idea of how the system really worked, and how program
> design could wreck performance. (At a university I once worked for we
> had a programmer who was assigned the task of extracting a number of
> names at random from the student master record tape. He took the known
> number of student names on the tape (somewhat over 35,000), generated
> a random number in the interval [0,1], multiplied the two numbers,
> spaced down the tape that many records, read the entry, rewound the
> tape, and then repeated the process...and was totally clueless about
> just *why* the operators killed the job.
Circa 1970, I consumed half a library of scratch tapes doing number
conversion in Fortran IV. To convert from one format (say, an EBCDIC
character string) to another: allocate a tape, (via JCL) then
1 WRITE the initial format to the tape
REWIND the tape
READ the number in the new representation
GO TO 1 until finished.
Just one more reason why I grew to love PL/1.
Misusing VSAM as a sort program could bring a system to its knees too.
Carl R. Friend
Bill Thater wrote in article Nr. <55nlth$p...@newstand.syr.edu>:
>
> i'm gld to see someone else did the "flash the pretty lights" hack. as
> a "glass castle" operator for a state agency, i wrote a S/360 hack to
> flash the lights in pretty patterns for "showing off" to the tours.
Does anyone here know who got the word 'TILT" to display on the
PDP-10 panels? I've heard this one a few times here, but never saw
it attributed to anybody.
I suspect it could have been done fairly easily on the center top
panel... If only I had a -10 to play with...
Jim Cassidy
In article <327FE886...@swec.com>,
Carl R. Friend <carl....@swec.com> wrote:
>Bill Thater wrote in article Nr. <55nlth$p...@newstand.syr.edu>:
>>
>> i'm gld to see someone else did the "flash the pretty lights" hack. as
>> a "glass castle" operator for a state agency, i wrote a S/360 hack to
>> flash the lights in pretty patterns for "showing off" to the tours.
>
> Does anyone here know who got the word 'TILT" to display on the
>PDP-10 panels? I've heard this one a few times here, but never saw
>it attributed to anybody.
>
on it that I took around 1978.
But I cheated... I turned on the Lamp Test switch, and just turned off
the bulbs I wanted to be off by pulling them out about 1/8". Still
looked pretty neat though.
Jim Cassidy / cas...@neosoft.com
wri...@sabu.ebay.sun.com
In article 3425...@news.iol.ie, spal...@iol.ie# (Nick Spalding) writes:
> ..wri...@sabu.ebay.sun.com wrote:
> > When ENIAC was designed, TV was many years in the future.
> > The story is a legend.
>
> first broadcasts were from the New York Worlds Fair of that year),
> many years before ENIAC.
> --
> Nick Spalding
---------------------------
The original post said that the ENIAC designers were thinking
about TV viewers. There was no commercial TV in the US until after
WWII. No one owned TV sets during the NY Worlds Fair.
Sabu
Tim Shoppa
In article <327FE886...@swec.com>,
Carl R. Friend <carl....@swec.com> wrote:
>Bill Thater wrote in article Nr. <55nlth$p...@newstand.syr.edu>:
>>
>> i'm gld to see someone else did the "flash the pretty lights" hack. as
>> a "glass castle" operator for a state agency, i wrote a S/360 hack to
>> flash the lights in pretty patterns for "showing off" to the tours.
>
> Does anyone here know who got the word 'TILT" to display on the
>PDP-10 panels? I've heard this one a few times here, but never saw
>it attributed to anybody.
I've been told that when the Intel Touchstone (a parallel machine
with 1 or a few status LED's per CPU) powers up, it blinks
"12:00" on the status array (ala VCR's). (Treat this purely as
urban-legend, friend-of-a-friend type folklore.)
Tim. (sho...@triumf.ca)
Gene Wirchenko
Max F Lang <mfl...@bellsouth.net> wrote:
[snip]
>I'm kind of a newbie with computers (only been using them since the
>Apple ][
>came out), but I read somewhere that operators at many installations
>used to
>keep an AM radio nearby the main system, and keep it tuned to the white
>noise
>made by the system. Supposedly they could tell alot about the state of
That's EM radiation from inadequate shielding.
>the
>system from the various whistles and beeps that came out of the radio.
>Of course, along time ago, I tried this with my first PC, but I don't
>think I
>ever heard one peep that had anything to do with the box.
It worked on the Radio Shack TRS-80 Model I. I think someone
even wrote a music program based on it.
The FCC has gotten fussier about shielding which is why it didn't
work on your PC.
Sincerely,
Gene Wirchenko
C Pronunciation Guide:
y=x++; "wye equals ex plus plus semicolon"
x=x++; "ex equals ex doublecross semicolon"
Johnny Billquist
> Michael Krell wrote:
> >
> > Over in misc.writing.screenplays, someone wrote:
> >
> > > The scientists who invented ENIAC felt that the computer was just not
> > > that exciting to tv viewers so they surrounded it with blinking
> > > lights, lights that had nothing to do with what was going on inside
> > > the computer. the media loved it and blinking lights have been a
> > > movie computer stable since.
> >
> > Can anyone confirm this story? It sounds a bit urban-legendish to me.
It is urban legend as for as the ENIAC goes. However, Thinking Machines CM5
got lot of lights added for customer satisfaction, unless I'm misinformed.
But that is a modern day machine.
Back in the age of ENIAC you really needed the blinking lights.
> This one I suspect is true..... I worked on an IBM 360 with tons o'
> blinky lights. The lights were suppose to tell you: The PSW, the
> register that was being accessed, and the real time clock. Of course
> all this valuable information was useless if the system crashed because
> little blinky lights got dead when dos computer goes dead.
No, you are wrong. First, the blinking lights can be used to debug a system
by looking at the front panel of a running system which hangs. Second,
the blinking lights tell the state of the machine at all time, including
when it has stopped, so you can see what PC, PSW and all other stuff
of the system when it has stopped. You don't loose that information
because software crashes. The blinking lights don't have anything to
do with your software. They are a part of the hardware!
Second, you seem to be ignorant of the fact that old hardware can
be programmed via their front panels. And when you write a program
that way, you also want to be able to check if you entered the program
correctly, which means you have to be able to read the memory.
After you have entered the program, a useful technique to debug programs
are to single step them. You didn't really think that this feature popped
up with gdb do you?
And when you single step a program, you definitely wants to see what
happens after each instruction, and viola, the blinken lights tell
you.
Actually, the whole thread is silly. Anyone with half a brain should
understand the value of those blinken lights when programming a bare
system. Both as diagnostic aids, program development aids, and
informational aids.
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@update.uu.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Ed Barrows
In article <327EEE...@bellsouth.net>, mfl...@bellsouth.net says...
>I'm kind of a newbie with computers (only been using them since the
>Apple ][
>came out), but I read somewhere that operators at many installations
>used to
>keep an AM radio nearby the main system, and keep it tuned to the white
>noise
>made by the system. Supposedly they could tell alot about the state of
>the
>system from the various whistles and beeps that came out of the radio.
>Of course, along time ago, I tried this with my first PC, but I don't
>think I
>ever heard one peep that had anything to do with the box.
>
We never monitored the computer with the portable radio but we did have a small
radio that we would put on the top of the 360-30 processor when we had
visitors. We had made up a bunch of card decks that we could read in and they
would produce tones in the radio. We had some old traditional tunes and at
Christmas we played Christmas carols on the thing. This was only to impress
visitors and not used for monitoring the operation of the computer. And yes,
coming from the "old" school of punch boards and wires, I still wonder about
the registers in my pentium 100 with 32 MB of memory.
Just my $0000.02
Ed
Brian Raiter
Gene Wirchenko <ge...@mindlink.bc.ca>:
> It worked on the Radio Shack TRS-80 Model I. I think someone
>even wrote a music program based on it.
Yep. I typed one of those in. I believe there were twenty-four
subroutines, one for each pitch. The author had figured out that an
empty BASIC statement took a certain amount of time to interpret,
characters in a REMark a certain less amount of time, and whitespace
less than that. With the proper combination of each in a FOR loop, you
had an instant one-timbre synthesizer. It worst drawback was that
there was no way to insert a rest.
> The FCC has gotten fussier about shielding which is why it didn't
>work on your PC.
The FCC has always been fussy; the Model I was a fluke that got
overlooked, IIRC. The Model III, which was what I had, was much better
shielded - the music program was barely audible (though it helped to
have the radio plugged into the same outlet).
The Model I's keyboard was the major source of the noise. If you put
it on top of a radio, you could fill a concert hall with the sound.
b
Lon Stowell
In article <55nlth$p...@newstand.syr.edu> bi...@npac.syr.edu (Bill Thater) writes:
>i'm gld to see someone else did the "flash the pretty lights" hack. as
>a "glass castle" operator for a state agency, i wrote a S/360 hack to
>flash the lights in pretty patterns for "showing off" to the tours.
There were many of these hacks for different reasons. At
SAC Headquarters, we had to hack their RCA Spectra message
switches to turn off the idle light on the front panel. Seems
that most operating systems that do message switching tend to
spend a lot of time in the idle loop....and when the Air
Farce General in charge of the project first got a demo of
the working system [switched messages from SAC bases around
the world as well as to such locations as the Looking Glass airborne
command post and the white house's comms center], he was annoyed
at paying millions of dollars for a computer system which
had its idle light on most of the time.
Suggestions that running batch jobs on the computer would be
detrimental to it being able to handle a peak load of message
switching on demand were obviously wasted on an uncomprehending
intellect...so the operating system was hacked to run a very
low priority task that did nothing more than turn the light out.
Lon Stowell
In article <55qgqm$b...@eve.speakeasy.org> bri...@speakeasy.org (Brian Raiter) writes:
>
>The FCC has always been fussy; the Model I was a fluke that got
>overlooked, IIRC.
You recall wrong. The FCC only got fussy as of September 1984
with two classes of electronic devices with clock speeds above
100 Mhz. The radiation and conducted energy limits were tighter
for residential devices than for commercial devices (which could
not be sold for residential use), but neither was quite as tight
as the VDE limits....or the current FCC limits.
Initial enforcement of the Part 15 emissions standards was widely
publicized within the engineering community as being at the retail
level. Non-conformant consumer products were simply siezed
(say, from Toys-R-Us in one case) and destroyed. If the device
didn't have the FCC Part/15 compliance label on it, it was by
definition non-conformant,,,but the penalties for falsely putting
a label on a non-conforming device were pretty draconian.
Commercial equipment enforcement was expected to be by your
competitors who had gone to the trouble of testing, modifying,
and making their equipment legal. Any competent engineer can
tell at a glance whether or not a given design is likely to
be able to pass compliance.
wri...@sabu.ebay.sun.com
----------------------------------------
The thread started about blinken lights that had nothing to do
with what was going on inside the computer. Who is ignorant?
Sabu
Gene Wirchenko
bri...@speakeasy.org (Brian Raiter) wrote:
>Gene Wirchenko <ge...@mindlink.bc.ca>:
>> It worked on the Radio Shack TRS-80 Model I. I think someone
>>even wrote a music program based on it.
>Yep. I typed one of those in. I believe there were twenty-four
>subroutines, one for each pitch. The author had figured out that an
>empty BASIC statement took a certain amount of time to interpret,
>characters in a REMark a certain less amount of time, and whitespace
>less than that. With the proper combination of each in a FOR loop, you
>had an instant one-timbre synthesizer. It worst drawback was that
>there was no way to insert a rest.
>> The FCC has gotten fussier about shielding which is why it didn't
>>work on your PC.
>The FCC has always been fussy; the Model I was a fluke that got
>overlooked, IIRC. The Model III, which was what I had, was much better
>shielded - the music program was barely audible (though it helped to
>have the radio plugged into the same outlet).
>The Model I's keyboard was the major source of the noise. If you put
>it on top of a radio, you could fill a concert hall with the sound.
Well, yes, of course. The keyboard was the computer. CPU and
16K of memory were inside.
Peter Liljenberg
ge...@mindlink.bc.ca (Gene Wirchenko) writes:
>
> Max F Lang <mfl...@bellsouth.net> wrote:
>
> [snip]
>
> >I'm kind of a newbie with computers (only been using them since the
> >Apple ][
> >came out), but I read somewhere that operators at many installations
> >used to
> >keep an AM radio nearby the main system, and keep it tuned to the white
> >noise
> >made by the system. Supposedly they could tell alot about the state of
>
> That's EM radiation from inadequate shielding.
>
> >the
> >system from the various whistles and beeps that came out of the radio.
>
> It worked on the Radio Shack TRS-80 Model I. I think someone
> even wrote a music program based on it.
In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was the
rise of the personal computer) they told about a guy who wrote a program to
an Altair which, when you placed a transistor radio on it, played
"Fool on the Hill", by the Beatles.
--
Good wishes,
Peter Liljenberg <c96p...@und.ida.liu.se>
Eric Fischer
Peter Liljenberg <c96p...@und.ida.liu.se> wrote:
> In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was the
> rise of the personal computer) they told about a guy who wrote a program to
> an Altair which, when you placed a transistor radio on it, played
> "Fool on the Hill", by the Beatles.
Steve Dompier was the programmer's name. His article describing
the effect appears in volume 1 (1976) of Dr. Dobb's Journal if
you want to read the full story.
[BTW, it's Robert Cringely, not Mike, and it's actually a pseudonym
that's been used by several InfoWorld writers over the years. This
particular Cringely was (a few months ago) actually being sued by
the magazine for using the name on TV without their permission]
Eric
Peter Liljenberg
er...@fudge.uchicago.edu (Eric Fischer) writes:
>
> [BTW, it's Robert Cringely, not Mike, and it's actually a pseudonym
> that's been used by several InfoWorld writers over the years. This
> particular Cringely was (a few months ago) actually being sued by
> the magazine for using the name on TV without their permission]
Well, he sure fooled me :) He put on a good show of being a failed hacker
in Silly Valley, though. What's the state of the world when you can't
trust TV? <manic grin>
Andrew Rogers
Peter Liljenberg wrote:
> In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was the
> rise of the personal computer) they told about a guy who wrote a program to
> an Altair which, when you placed a transistor radio on it, played
> "Fool on the Hill", by the Beatles.
Altair? I heard that hack on some IBM system (I don't remember the
model, but I do remember that it repeated about 10 seconds of Bach)
when my high school math class toured a computer center in 1965-66,
and I'll bet it wasn't exactly a new idea even then.
Andrew
Carl R. Friend
Gene Wirchenko wrote in article <55p5sl$5...@fountain.mindlink.net>:
>
> The FCC has gotten fussier about shielding which is why it
> didn't work on your PC.
Oddly enough, the "dirtiest" system in the house now (other than
my ISC 8001) is my wife's brand new 200MHz Gateway box. When that
machine's on, it's impossible to use my shortwave radio in the next
room. By contrast, my -11s and Novas are quite "quiet"; I have to
get a radio quite close to the systems - and they use core memory!
The ISC, by the way, can (and did, until I "shielded" it with
an application of aluminium foil on the case insides) jam commercial
television at about 25 feet.
Cheers.
Huw Davies
Johnny Billquist (b...@Zeke.Update.UU.SE) wrote:
: First, the blinking lights can be used to debug a system
: by looking at the front panel of a running system which hangs. Second,
: the blinking lights tell the state of the machine at all time, including
: when it has stopped, so you can see what PC, PSW and all other stuff
: of the system when it has stopped. You don't loose that information
: because software crashes. The blinking lights don't have anything to
: do with your software. They are a part of the hardware!
I remember that the lights on our dearly departed DECsystem-10 were useful
for a number of things, in particular the counter that was incremented by
the null job was displayed in one 36 bit (naturally) line of lights. You
could instantly tell how busy the system was by seeing how fast this number
was increasing....
Now onto a slightly different use for the lights (more particulaly a register
attached to lights on the console). I was involved with a group of students
employed as night-time operators/Help Desk support. Well, there wasn't too
much to do so we read everything we could - in particular, about the UUO that
set the console lights register, and the corresponding UUO to read it.
Given that our exclusive group was less than 36 strong, we allocated a bit
in this register to each of us, and had a program run at login time to
see who was logged in. This worked fine for a couple of years until we found
out that our on-site DEC engineer had been working many hours a week to
discover why the console lights display was 'playing up'. We quietly killed
the program and the problem went away....
--
Huw Davies | e-mail: Huw.D...@latrobe.edu.au
Information Technology Services | Phone: +61 3 9479 1550 Fax: +61 3 9479 1999
La Trobe University | "My Alfas keep me poor in a monetary
Melbourne Australia 3083 | sense, but rich in so many other ways"
Eric Werme
"Carl R. Friend" <carl....@swec.com> writes:
> Does anyone here know who got the word 'TILT" to display on the
>PDP-10 panels? I've heard this one a few times here, but never saw
>it attributed to anybody.
> I suspect it could have been done fairly easily on the center top
>panel... If only I had a -10 to play with...
I'm not so sure. I may well have considered that, but it would have
been a little tricky. As I recall it displayed the 64 bit math result,
the parity register which all code and data visited, and I forget what
else. I certainly never tried, so I won't say it was impossible (or
even difficult).
OTOH, I *did* write a program that displayed TILT on a PDP-10 peripheral.
That was pretty easy - it was a 16x16 bank of lights showing the state of
the audio processing box we built. My program was pretty short (was
it 57 instructions? Something near there.) Of course, that included
the system call to give me access to the I/O instructions I needed and
the code to make two 3-bit worms run around the lights at random until
the collided (whereupon I flashed TI
LT a few times and restarted).
Back on that post that claimed the IBM 360's lights were useless when the
system crashed, before TOPS-10 was called TOPS-10, it's crash code
was simply a HALT instruction. The PC was still in the lights and
was very valuable. We talked about hooking up a siren to the RUN
light to get the operator's attention (HALT turned off the RUN light).
Fortunately the STOPCD macro came out in the 5 series OS that ran the CTY
bell several times and printed the 3 character stop code.
--
<> Eric (Ric) Werme <> This space under reconstruction <>
<> <we...@zk3.dec.com> <> <>
Joe Morris
Andrew Rogers <rog...@hi.com> writes:
Nowhere near "new". By the 60s playing music through a nearby radio
was a standard open-house demo. There were also programs that would
play music on the line printer, using the patterns of hammer strikes
to provide the different tones.
And the TX-0 computer (of Lincoln Labs fame) included a hi-fi amplifier
built into the console, with the sign bit of the accumulator tied to
its input.
Joe Morris / MITRE
Jeffrey D Fox
On 5 Nov 1996 20:27:36 GMT, wri...@sabu.ebay.sun.com wrote:
>---------------------------
>The original post said that the ENIAC designers were thinking
>about TV viewers. There was no commercial TV in the US until after
>WWII. No one owned TV sets during the NY Worlds Fair.
>Sabu
Maybe it applied to movie newsreels too, which were around since WW I.
Jeff
Tom Watson
> Peter Liljenberg wrote:
> > In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was the
> > rise of the personal computer) they told about a guy who wrote a program to
> > an Altair which, when you placed a transistor radio on it, played
> > "Fool on the Hill", by the Beatles.
>
> Altair? I heard that hack on some IBM system (I don't remember the
> model, but I do remember that it repeated about 10 seconds of Bach)
> when my high school math class toured a computer center in 1965-66,
> and I'll bet it wasn't exactly a new idea even then.
>
> Andrew
This is OLD stuff. The IBM 1620 (c. 1965) had music programs that worked
quite well. Since the 1620's crystal clock was a whopping 1Mhz (divided
by 20 for memory cycle time) one could take an ordinary AM radio (guess
where it was tuned to) and with proper placement you could get music.
Later in my computing career, when micro processors came out, the 1620
(now later in life, the 70's) was used as a comparison point for intel's
first microprocessor (the 4004). The 1620 we played around with, we did
some probing around, and found out that direct connection to various lamps
on the front panel (which it had many) gave "interesting" renditions of
tunes, each a bit different in sound. Since we could now "pick" out
instrument, and wanted to display them, we wired up the "register select"
rotary switch to both light a light and select the proper light to
"observe". It worked out quite nicely.
Now for the fun part...
When intel wanted to make comparison photos for the 4004 and "something
else" it used this 1620 (I now own the machine) and yes, the pictures do
have the "music register" nicely lettered if you look hard. The pictures
appeared in Business Week in the early 70's. I don't know the exact
issues, but they were probalby in the 1972-73 time frame. If anyone has
the exact issue date, I'd like to know, as I've lost the exact issue (but
did see it in passing).
Wonderful folklore....
--
Tom Watson
t...@3do.com (Home: t...@johana.com)
Charlie Gibbs
In article <tsw-081196...@cypher.3do.com> t...@3do.com
(Tom Watson) writes:
>This is OLD stuff. The IBM 1620 (c. 1965) had music programs that
>worked quite well. Since the 1620's crystal clock was a whopping
>1Mhz (divided by 20 for memory cycle time) one could take an ordinary
>AM radio (guess where it was tuned to) and with proper placement you
>could get music.
Interesting. The one I saw asked to have the radio tuned to 580 kHz.
Also, this machine had a 2311 disk drive, and the program would step
the heads in time with the music.
Being the pack rat that I am, I probably still have the printout
around somewhere...
Charli...@mindlink.bc.ca
I used to be indecisive, but now I'm not so sure.
Carl R. Friend
Eric Werme wrote:
>
> Back on that post that claimed the IBM 360's lights were useless
> when the system crashed, before TOPS-10 was called TOPS-10, it's
> crash code was simply a HALT instruction. The PC was still in
> the lights and was very valuable. We talked about hooking up a
> siren to the RUN light to get the operator's attention (HALT turned
> off the RUN light). Fortunately the STOPCD macro came out in the
> 5 series OS that ran the CTY bell several times and printed the 3
> character stop code.
Back where I held employment at my first "day job", the monitor
level was such that we had "real" stopcodes, and an LA-36 bell
sequence that sounded remarkably like the Morse Code "SOS".
In recognition that light patterns left over after a crash, or
most especially, a hang, we also had a Polaroid camera that was used
to record the contents of all the lights (which, by the time I
resigned for other things, had been mostly replaced with LEDs).
The funniest part of it was that some wag had hooked the speaker
output to an old automobile-theft alarm that featured a pager. If
a system crashed, the pager went off. Of course false alarms were
a large problem with the system...
Charlie Gibbs
>Peter Liljenberg wrote:
>
>> In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic
>> was the rise of the personal computer) they told about a guy who
>> wrote a program to an Altair which, when you placed a transistor
>> radio on it, played "Fool on the Hill", by the Beatles.
>
>Altair? I heard that hack on some IBM system (I don't remember the
>model, but I do remember that it repeated about 10 seconds of Bach)
>when my high school math class toured a computer center in 1965-66,
>and I'll bet it wasn't exactly a new idea even then.
Indeed. I first saw it on an IBM 1620 in 1967, and wrote one of my
own for the Univac 9300 a few years later.
My favourite from the standpoint of a clever hack was the one that
ran on my IMSAI (it would work on an Altair too) where you fed the
"interrupts enabled" line to an audio amplifier through a suitable
shaping circuit. It wasn't radio pickup anymore, but the program
could toggle the line in fancy enough patterns to play three-part
harmony. Quite impressive.
Mike Umbricht
The first 'music recital' by a computer is generally attributed to Max Mathews
of Bell Labs in Murray Hill, NJ. Mathews was working on computer generated
speech for use in testing new telephone inventions by simulation.
In 1957 a computer at Bell 'sang' the song _Daisey Bell, A Bicycle Built for
Two_ which was later made famous in _2001_ and recreated by Dompier on the
Altair. Several of these early experiments were recorded on Decca DL 79103.
Does anyone know what hardware was used to play _Daisey_?
In the late '50s Mathews did a great deal of work on computer generated music.
He wrote a program called MUSIC IV for the IBM 7094 to synthesize sounds.
MUSIC V ran on a later IBM.
Another project at Bell was the GROOVE system which ran on a Honeywell DDP-224.
-mike
In article <55vmnk$i...@top.mitre.org>, jcmo...@mwunix.mitre.org (Joe Morris) writes:
>Andrew Rogers <rog...@hi.com> writes:
>
>>Peter Liljenberg wrote:
>>> an Altair which, when you placed a transistor radio on it, played
>>> "Fool on the Hill", by the Beatles.
>
>>Altair? I heard that hack on some IBM system (I don't remember the
>>model, but I do remember that it repeated about 10 seconds of Bach)
>>when my high school math class toured a computer center in 1965-66,
>
>Nowhere near "new". By the 60s playing music through a nearby radio
>
Dan Bernstein,Rehovot Israel
BTW, how about the LED digits on the front panels of Intel-based machines, which allegedly show you the speed the machine runs at (not that I understand why someone would want to run their machine at less than the highest
safe speed) - are those hard-wired or somewhat more sophisticated?
--Dan.
Ralph Wade Phillips
Hi, Dan!
In article <561pmh$o...@learnet.freenet.hut.fi>, da...@mail.freenet.hut.fi
(Dan Bernstein,Rehovot Israel) wrote:
On most machines (ALL that I've seen!), the display is set to
display one number on TURBO, another number on NON-TURBO. So, as you can
guess, the display has not much to do with what the processor actually
does.
I was setting up a system for one customer, where we were waking
up an old 286-10 to serve as a print server. The new case he ordered had
a VERY involved jumper setup for the display. We ended up with the case
saying "120MHz" ... <B-)
RwP
--
Ralph Wade Phillips, CET #LA-82
ral...@gcstation.net
Joe Morris
Charli...@mindlink.bc.ca (Charlie Gibbs) writes:
>My favourite from the standpoint of a clever hack was the one that
>ran on my IMSAI (it would work on an Altair too) where you fed the
>"interrupts enabled" line to an audio amplifier through a suitable
>shaping circuit. It wasn't radio pickup anymore, but the program
>could toggle the line in fancy enough patterns to play three-part
>harmony. Quite impressive.
The PDP-1 at MIT was wired to a Heath amplifier by taking the sense
light lines from the machine and running them through a homebrew
mixer. Pete Sampson wrote a *very* nice music compiler for it;
the input (via -ugh- Flexowriter tape) could almost be transcribed
directly from a score. His program could handle three parts, and
was not limited by memory (it would read the paper tape input as
required, without interrupting the output).
Not surprisingly, it couldn't produce the quality audio output that
comes from today's sound cards, but in its day (the early 1960s)
the result was quite impressive.
Joe Morris / MITRE
Tim Shoppa
In article <561pmh$o...@learnet.freenet.hut.fi>,
Dan Bernstein,Rehovot Israel <da...@mail.freenet.hut.fi> wrote:
>
>BTW, how about the LED digits on the front panels of Intel-based
>machines, which allegedly show you the speed the machine runs at
>(not that I understand why someone would want to run their
>machine at less than the highest
> safe speed) - are those hard-wired or somewhat more sophisticated?
>--Dan.
They are typically set by a bunch of jumpers on the back of
the display PC board. Sometimes the jumpers feed a BCD to 7-segment
decoder, giving you only the possibility of the ten decimal digits.
But others have a jumper for each of the seven segments, allowing
you to be creative in setting the display. For example, with
the very latest units (which allow 3, instead of 2 1/2 digits),
I like to sneak into cow orker's offices and spell out
_
|_ _ _
| |_| |_|
and
_
|_ |_| _
|_| | | |
for the low and high speeds, respectively.
Tim. (sho...@triumf.ca)
Scott Norwood
In article <561pmh$o...@learnet.freenet.hut.fi>,
Dan Bernstein,Rehovot Israel <da...@mail.freenet.hut.fi> wrote:
>
>BTW, how about the LED digits on the front panels of Intel-based machines,
>which allegedly show you the speed the machine runs at (not that I
>understand why someone would want to run their machine at less than the highest
>safe speed) - are those hard-wired or somewhat more sophisticated?
They're a holdover from when an 8MHz 80286 was considered fast. Early
copy-protection schemes got flaky on machines faster than abut 4.77 to 6MHz,
and (l)users needed a way to 'downshift' so that these programs would work.
Thus, cases were made with a 'turbo' switch and a little light which indicated
the current status of machine speed. Since the 486, though, these 'turbo'
switches generally don't do anything except just sit there. I think maybe
they look cool, kind of like the 'turbo' label on certain cars... Kind
of dumb, though.
-- Scott
Michael Hyman
In article <55o7vo$h...@male.EBay.Sun.COM>, wri...@sabu.ebay.sun.com says...
>
>In article 3425...@news.iol.ie, spal...@iol.ie# (Nick Spalding) writes:
>> ..wri...@sabu.ebay.sun.com wrote:
>> > When ENIAC was designed, TV was many years in the future.
>> > The story is a legend.
>>
>> TV broadcasting started in 1936 in England and 1939 in the USA (the
>> first broadcasts were from the New York Worlds Fair of that year),
>> many years before ENIAC.
>> --
>> Nick Spalding
>
>---------------------------
>The original post said that the ENIAC designers were thinking
>about TV viewers. There was no commercial TV in the US until after
>WWII. No one owned TV sets during the NY Worlds Fair.
>
>Sabu
Misinformation correction alert:
The blinking light, painted ping-pong balls on ENIAC were *filmed* for newsreel
news films. This 50-year-old newsreel footage actually exists, and was shown
in the PBS-TV series "The Machine That Changed The World".
--
Mike
mi...@op.net ENSONIQ Resources on the Internet:
http://www.op.net/~mikeh/ http://www.op.net/~mikeh/ensoniq.html
Lisa Jeff
> The original post said that the ENIAC designers were thinking
> about TV viewers. There was no commercial TV in the US until after
> WWII. No one owned TV sets during the NY Worlds Fair.
>
> Sabu
While TV wasn't widespread when the ENIAC came out, motion picture
newsreels were very widespread and a major news medium. (You can purchase
video tapes of newsreels from those years, which are pretty interesting.)
peda...@aol.com
>BTW, how about the LED digits on the front panels of Intel-based
machines, which allegedly >show you the speed the machine runs at (not
that I understand why someone would want to run >their machine at less
than the highest
> safe speed) - are those hard-wired or somewhat more sophisticated?
Somewhat more sophisticated, but not by much. My computer has a small
printed circuit board with jumpers, one jumper per LED segment. Being
of humorous bent, my machine now shows my initials :-)
Peter Smith
Michael Meissner
In article <327E7F84...@swec.com> "Carl R. Friend" <carl....@swec.com>
writes:
| I think that they stopped making computers when front panels
| disappeared. Without blinking lights (some will say loud fans), they
| look more loke heaters than anything else. Sniff, sniff, whimper. :-(
In reading the blinking light threads, I started thinking how many lights you
would need to fully represent the state of a pentium pro or a PowerPC 604e.
Both of these machines are heavily pipelined, can do out of order execution,
and in general can have many instructions in flight at the same time.
--
Michael Meissner, Cygnus Support (East Coast)
4th floor, 955 Massachusetts Avenue, Cambridge, MA 02139, USA
meis...@cygnus.com, 617-354-5416 (office), 617-354-7161 (fax)
Max ben-Aaron
High their:
In the early sixties there was am IBM computer called "Stretch". It
had a panel of lights that was (it seemed to me) about six feet by four
feet. The lights were about 1/4" in diameter and were of different
colours. The display must have been structured ...
When something went wrong, the first thing the service reps did was to
take a polaroid picture of the panel.
Max ben-Aaron
================================================================
In <327E42...@corp2.mck.co.za> Sean Stanley-Adams
<se...@corp2.mck.co.za> writes:
>
>Carl R. Friend wrote:
>>
>> It's very safe to say that the lights were _not_ placed there
>> solely for the edification of an occasional tour group. A good
>> technician could usually look at the lights and head right for the
>> problem area of the machine.
>>
>I joined the computer game in the early 70s as an operator on both an
>IBM 360/40 and a low-end 370, both of which were liberally festooned
>with lights. Though I wasn't able to use the lights to the same
extent
>as the hardware engineers, it was possible with practise to identify a
>looping program. The programmers could never understand how we did it
>but then they didn't spend long nights with precious little else to do
>but looking at the blinking lights!
Bill Bradford
Dan Bernstein,Rehovot Israel (da...@mail.freenet.hut.fi) wrote:
: BTW, how about the LED digits on the front panels of Intel-based machines, which allegedly show you the speed the machine runs at (not that I understand why someone would want to run their machine at less than the highest
: safe speed) - are those hard-wired or somewhat more sophisticated?
hardwired... and switchable between two displays
but an animated display would be cool... I had an Amiga expansion box for
my A1000 once that had a 'RAM ACCESS' light- it rocked. I could tell
what the machine was doing by the pulsing of the LED...
--
bill bradford system admin, unix geek, super hero, BOFH
mrb...@texas.net texas networking, inc. http://www.texas.net
--------------------------------------------------------------------------
"It's important to remember that the creatures of the night aren't simply
the people of the day staying up late because they think that makes them
cool and interesting. It takes a lot more than heavy makeup and a pale
complexion to cross the divide." - Terry Pratchett, _Soul Music_
jmf...@aol.com
In article <sy918as...@wogglebug.ziplink.net> meis...@cygnus.com
(Michael Meissner) writes:
<In article <327E7F84...@swec.com> "Carl R. Friend" <
<carl....@swec.com>writes:
<
<| I think that they stopped making computers when front panels
<| disappeared. Without blinking lights (some will say loud fans), they
<| look more loke heaters than anything else. Sniff, sniff, whimper. :-(
<
<In reading the blinking light threads, I started thinking how many lights
you
<would need to fully represent the state of a pentium pro or a PowerPC
604e.
<Both of these machines are heavily pipelined, can do out of order
execution,
<and in general can have many instructions in flight at the same time.
I don't think it matters. What matters is the pattern of blinks. When I
started working with PDP-10s and could see the lights (both on the CPU and
memory), one "learned" when the machine was in trouble (or going to be in
trouble). One didn't need to understand what the lights meant--one just
needed to learn (by experience) a pattern that indicated problems.
When I worked in Tape Prep we would have a bunch of women typing away at a
bank of TTY35s. They would turn in unison when a sound happened, look at
the lights, turn back around and hurriedly type EX$$ to save their work.
New employees being trained learned the same behaviour; this wasn't part
of the training plan; it just happened.
And yes. I really missed those lights when the KL appeared. Half of the
information went missing since we only had the memory lights to work with.
/BAH
Geoffrey Welsh
>Peter Liljenberg wrote:
>> In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was the
>> rise of the personal computer) they told about a guy who wrote a program to
>> an Altair which, when you placed a transistor radio on it, played
>> "Fool on the Hill", by the Beatles.
>Altair? I heard that hack on some IBM system (I don't remember the
>model, but I do remember that it repeated about 10 seconds of Bach)
>when my high school math class toured a computer center in 1965-66,
>and I'll bet it wasn't exactly a new idea even then.
In "Hackers", Stephen Levy reports that someone did this at MIT
(probably on the DEC TX-0).
--
Geoffrey Welsh, MIS Co-ordinator, InSystems Technologies (gwe...@insystems.com)
At home: xenitec.on.ca!zswamp!geoff; Temporary: crs...@inforamp.net
DO NOT SEND ANY FORM OF ADVERTISING TO ANY OF THESE ADDRESSES
Brian Raiter
Tim Shoppa <sho...@alph02.triumf.ca>:
>But others have a jumper for each of the seven segments, allowing
>you to be creative in setting the display. For example, with
>the very latest units (which allow 3, instead of 2 1/2 digits),
>I like to sneak into cow orker's offices and spell out
>
> _
>|_ _ _
>| |_| |_|
>
>and
> _
>|_ |_| _
>|_| | | |
>
>for the low and high speeds, respectively.
Okay, that's it. All else aside; my next computer must have a turbo
display with three digits.
b
Mr M P Stevens
In article <561pmh$o...@learnet.freenet.hut.fi>,
da...@mail.freenet.hut.fi (Dan Bernstein,Rehovot Israel) writes:
>
>BTW, how about the LED digits on the front panels of Intel-based machines,
>which allegedly show you the speed the machine runs at (not that I
>understand why someone would want to run their machine at less than the
>highest
> safe speed) - are those hard-wired or somewhat more sophisticated?
>--Dan.
digits, which control what led bars are on and which off, depending on the
position of the turbo switch, which on machines where it is fitted, will
switch the machine between normal full speed and a much slower speed.
You can set the LED digits to read anything you want, no matter
how fast the computer actually is, at least on the systems I've encountered.
My 386dx40 claims, if you read the front panel, to run at 70mhz.
People used to want to run their machines at slow speeds because
some hardware wouldn't work in faster computers, due to problems with
the bus speed, and some programs, such as mainly games, didn't realise they
were running on a faster computer, and ran way too fast....
--
--
Michael Stevens First Year Computer Systems Engineering Undergrad
G7SFA. es...@warwick.ac.uk. Home Page:http://www.warwick.ac.uk/~esuci
Ralph Wade Phillips
Hi there!
In article <5652ir$3...@news.inforamp.net>, crs...@inforamp.net (Geoffrey
Welsh) wrote:
> On Thu, 07 Nov 1996 13:12:51 -0500, in alt.folklore.computers, Andrew
> Rogers <rog...@hi.com> wrote:
>
> >Peter Liljenberg wrote:
> >> In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was the
> >> rise of the personal computer) they told about a guy who wrote a program to
> >> an Altair which, when you placed a transistor radio on it, played
> >> "Fool on the Hill", by the Beatles.
>
> >Altair? I heard that hack on some IBM system (I don't remember the
> >model, but I do remember that it repeated about 10 seconds of Bach)
> >when my high school math class toured a computer center in 1965-66,
> >and I'll bet it wasn't exactly a new idea even then.
>
> In "Hackers", Stephen Levy reports that someone did this at MIT
> (probably on the DEC TX-0).
>
Back in my old TRS-80 days ("Son, it wasn't a Model 1. They
weren't Model 1's until the Model II came out ... " <B-) ), there was at
least ONE program which generated the sound effects to be listened to by a
small AM radio near the computer. Also, that was one of the better hints
and tips - "Leave a small AM radio tuned to xxxx (forget the optimum
frequency), and you'll be able to tell if the computer hung during a
cassette load ... " <B-)
David E. Fox
In article <ralphp-0911...@du-00.gcstation.net>, Ralph Wade Phillips wrote:
>
> I was setting up a system for one customer, where we were waking
>up an old 286-10 to serve as a print server. The new case he ordered had
My brother's 486/56 (cyrix) was wired so that the speed indicator said
"JC" (jeezus christ) :)
>Ralph Wade Phillips, CET #LA-82
>ral...@gcstation.net
--
------------------------------------------------------------------------
David E. Fox Tax Thanks for lettimg me
df...@belvdere.vip.best.com the change magnetic patterns
ro...@belvedere.sbay.org churches on your hard disk.
-----------------------------------------------------------------------
John G Dobnick
From article <55lt2q$i...@kirin.wwa.com>, by jeve...@wwa.com (John Everett):
>
> This thread seems to be rapidly filling up with posts from people who
> apparently never worked on computers with status lights. If all the lights on
> a System/360 went dead it meant you just suffered a power failure. Otherwise,
> you could read the status of the machine when it crashed. One installation I
>
Amen! I worked on/with Univac 1100 machines, which had pretty
impressive maintenance panels, with *lots* of "blinkey-lights". (And
Oh, how I miss those lights! :-( )
After a while, one sort of "gets used to" certain light patterns
appearing when certain code ran -- one gets a "feel" for "normal".
More particularly, from my systems programming point of view, one
noticed when expected light patterns did *not* appear -- that usually
meant something _unusual_ was going on. In most instances, "unusual"
equated to "trouble".
The most serious problem with light panels, however, was determining
whether all the lamps actually _worked_! Burned out lamps were
indistinguishable from turned-off lamps, which raised merry hell when
troubleshooting. (And, unfortunately, the particular models of 1100 I
worked with lacked a "lamp test" button. Mega - :-( But that's another
story.)
--
John G Dobnick "Knowing how things work is the basis
Information & Media Technologies for appreciation, and is thus a
University of Wisconsin - Milwaukee source of civilized delight."
j...@csd.uwm.edu ATTnet: (414) 229-5727 -- William Safire
jmf...@aol.com
jeve...@wwa.com (John Everett) asked:
<Barb, were you around early enough to remember when you had to type
<PWEF$$ because EX hadn't been implemented yet?
Actually, that was around when I still worked at WMU. The requirement
seemed to go away when we went to Level D disk architecture.
Interestingly, that (PWEF$$) is what was taught in Tape Prep; they didn't
"advance" with the software technology very fast in those days. I think
it finally got "purged" as part of the "you have to do it this way"
attitude when I always managed to save my work by typing EX$$ faster than
they could type PWEF$$ during those wonderfully irritating crashes.
[ahhh, a happy emoticon here remember the good ol' days]
/BAH
mhmc...@aol.com
About RF Noise & the FCC:
Several years ago the local Fire & Ambulance dispatch center was nearly
put
completely out of service by video games in a nearby tavern.
The radio system used a frequency of 33.980 mhz and several of the games
managed to get that as either a fundamental or REALLY strong harmonic -
when
the tavern was open, the dispatchers couldn't hear anything but the white
noise
from the games.
After an extended search for the source, the tavern owner wasn't willing
to
shut down the games. The FCC didn't care either. Finally, the situation
was
resolved when an irate boro councilman wrecked the offending game.
Micheal H. McCabe
Mike Miller
> I could not have done my maintaince for our military systems
> if it were not for the 'blinky-lights'.
>
> Oh well....
When working as a temp for a large back as they moved their mainframe from
here to St. Louis, an intergral part of my training was summed up in these
few words:
This panel, we call it the 'winkie light' panel, show all of the modem
activity for the state. If you look up and you don't see any winkie lights,
well ya just better hit the Halon button on your way out the door.
--
Mike....@dot.state.co.us | mil...@henge.com
http://spot.elfwerks.com/~mjmiller | webm...@dot.state.co.us
I also respond well to "Hey, you!"
Tom Watson
> BTW, how about the LED digits on the front panels of Intel-based
machines, which allegedly show you the speed the machine runs at (not that
I understand why someone would want to run their machine at less than the
highest
> safe speed) - are those hard-wired or somewhat more sophisticated?
It is totally a gimmic. Designed to impress the luser who walks into a
store. Usually they are independent even of the motherboard. Segments
have one of 4 values: On, Off, On with turbo switch engaged, On with
turbo switch disengaged. How this switches from one group of segments to
another with the turbo switch is left as an exercise to the reader.
Suffice it to say, they can be set to anything, and one can play with them
wuite a bit. One could put the segments to "oFF" (three digits), or ones
initals (suggested in this thread), or something like 4.7 (speed of
original PC). Just a neat gimic. Like putting 120 MPH on the speedometer
of a Yugo, as if it will even get there!! :-)
--
Tom Watson
t...@3do.com (Home: t...@johana.com)
Robert Humphris
DoN. Nichols <dnic...@d-and-d.com> wrote in article
<563dfl$2...@windigo.d-and-d.com>...
> In article <88531-8...@mindlink.bc.ca>,
> Charlie Gibbs <Charli...@mindlink.bc.ca> wrote:
>
> [ ... ]
>
> >My favourite from the standpoint of a clever hack was the one that
> >ran on my IMSAI (it would work on an Altair too) where you fed the
> >"interrupts enabled" line to an audio amplifier through a suitable
> >shaping circuit. It wasn't radio pickup anymore, but the program
> >could toggle the line in fancy enough patterns to play three-part
> >harmony. Quite impressive.
>
> It would work on *some* Altairs. The 8800 and 8800B. However, it
> would not work on the 680b, which had the wrong CPU, and didn't (as I
> remember) have an external "interrupts enabled" line from the CPU.
>
A similar hack for the ZX81 was printed in Personal Computer World ( back
in the days when it was a real computer magazine )
The method was to write a series of bits to the cassette port, knowing that
the frequency was 1200 baud a stream of bits 101010101 would produce a
square wave at a fixed frequency, and a vile noise would eminated through a
small amp that I connected to it. Adjust the bit string and voila! variable
drones were available to upset your family and pets.
Rob
Bill Sudbrink
In article <32852F29...@swec.com> "Carl R. Friend" <carl....@swec.com> writes:
>peda...@aol.com wrote:
>>
>> My computer has a small printed circuit board with jumpers, one
>> jumper per LED segment. Being of humorous bent, my machine now
>> shows my initials :-)
>>
>> Peter Smith
>
>for Intel's P5 chip... <big old grin here!>
>
After changing the thing three or four times as I upgraded
CPUs, mine now says Hi or Lo for high or low speed. As the
case now contains a pentium, you only get to see the Hi
display. Most people think the thing is just being friendly
and cheerful.
Bad ASCII art to demonstrate segment arrangement:
I_I I _
I I I I_ I_I
>
>--
> ______________________________________________________________________
>| | |
>| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
>| Minicomputer Collector / Enthusiast | Massachusetts, USA |
>| mailto:carl....@cliff.swec.com | |
>| http://www.ultranet.com/~engelbrt/carl/museum | ICBM: N42:22 W71:47 |
>|________________________________________________|_____________________|
Bill Sudbrink
In article <563khs$m...@news2.texas.net> mrb...@texas.net (Bill Bradford) writes:
>Dan Bernstein,Rehovot Israel (da...@mail.freenet.hut.fi) wrote:
>: BTW, how about the LED digits on the front panels of Intel-based machines, which allegedly show you the speed the machine runs at (not that I understand why someone would want to run their machine at less than the highest
>: safe speed) - are those hard-wired or somewhat more sophisticated?
>
>hardwired... and switchable between two displays
>but an animated display would be cool... I had an Amiga expansion box for
>my A1000 once that had a 'RAM ACCESS' light- it rocked. I could tell
>what the machine was doing by the pulsing of the LED...
>
type machine with a front panel was the PCs Limited (later Dell)
286-12. At the time, this machine was a screemer. Had all static
RAM (640K upgradable to 2Megs). The panel could be set to show
all sorts of system info, It displayed POST codes during boot-up,
could show track/head/sector access info for the hard or floppy
disks, could show port (serial or parallel) activity and some
other stuff too.
John Everett
In article <5652ir$3...@news.inforamp.net>, crs...@inforamp.net says...
>
>On Thu, 07 Nov 1996 13:12:51 -0500, in alt.folklore.computers, Andrew
>Rogers <rog...@hi.com> wrote:
>
>>Peter Liljenberg wrote:
>>> In Mike Cringely's tv documentary "Triumph of the Nerds" (its topic was
the
>>> rise of the personal computer) they told about a guy who wrote a program
to
>>> an Altair which, when you placed a transistor radio on it, played
>>> "Fool on the Hill", by the Beatles.
>
>>Altair? I heard that hack on some IBM system (I don't remember the
>>model, but I do remember that it repeated about 10 seconds of Bach)
>>when my high school math class toured a computer center in 1965-66,
>>and I'll bet it wasn't exactly a new idea even then.
>
>In "Hackers", Stephen Levy reports that someone did this at MIT
>(probably on the DEC TX-0).
>
I haven't read "Hackers", but music was certainly produced on the PDP-6, one
of which was installed at MIT (Project MAC). As I recall the -6 could produce
six part harmony. Wires were hooked up to the lamp drivers on the front panel
and somehow fed through an amplifier to the speakers. One of the more complex
pieces programmed was Little Fugue in G Minor. In the mid '60s a tape of this
was regularly played in DEC's trade show booth. I remember working the booth
once and getting SO sick of hearing Little Fugue I wanted to scream.
Later the MK-10 (Music Kompiler-10) performed the same function on the
PDP-10, at least the KA10. MK-10 was the commonly used but unofficial name
for the (unofficial) music option.
One of the PDP-8 programmers (I don't remember his name but he wrote FOCAL;
perhaps it was Rick something) produced a music program for the -8. This one
played by putting a radio in front of the processor. The first thing he got
running (as I recall) was "Downtown".
--
jeve...@wwa.com (John V. Everett)
Brian Abernathy
Ron Hunsinger (hns...@sirius.com) wrote:
: One of my early programming endeavors was a ballot-counting program
: running on a Burroughs Medium Systems computer. Normal I/O was
: synchronous, at least from the point of view of normal user programs. That
: is, if you tried to read from the card reader, and the input buffer was
: empty, you'd hang at the read call until the buffer was full (or the
: operator cancelled the job).
Just to mutate the thread a bit more, I remember an election night at
college. We (the students) didn't have any jobs submitted that night,
because the election office commandeered the IBM 360/40 for the ballot
count. They allowed one operator on the floor that evening, and in
general ticked everybody off.
It made us feel better when we realized that they had stacks of ballot
cards on the top of the printer, and it was about to run out of paper ...
I've never seen so many bureaucrats move so fast when that cover began
opening by itself!
Ron Hunsinger
> After a while, one sort of "gets used to" certain light patterns
> appearing when certain code ran -- one gets a "feel" for "normal".
> More particularly, from my systems programming point of view, one
> noticed when expected light patterns did *not* appear -- that usually
> meant something _unusual_ was going on. In most instances, "unusual"
> equated to "trouble".
One of my early programming endeavors was a ballot-counting program
running on a Burroughs Medium Systems computer. Normal I/O was
synchronous, at least from the point of view of normal user programs. That
is, if you tried to read from the card reader, and the input buffer was
empty, you'd hang at the read call until the buffer was full (or the
operator cancelled the job).
For my purposes, I wanted the program to be able to overlap card reading,
card punching, and printing. I did this by opening buffered files to all
three devices, and then polling them, checking for the I/O complete status
in each buffer and only issuing an I/O call if I knew it wouldn't block.
Normally, this would be a very rude thing to do. All that polling sucks up
all the available processor time, starving other applications. But like I
said, this was a ballot-counting program. On election night, the machine
was reserved for just this one application. No way would the County Clerk
have allowed any other program to run on that night.
One consequence of this was that the machine spent all of its time in
NORMAL (i.e., user) mode. Usually, the machine spent almost all of its
time in supervisor (i.e., operating system) mode, waiting for interrupts.
There was a large NORMAL light on the console that, under normal usage,
would flicker briefly as applications got time, but spent most of its time
off. The only time you'd see the NORMAL light come on solid was if a
program had gone into an infinite loop.
The first time I cranked up the program at the customer's site, the NORMAL
light came on solid, and all the customer's programmers turned to look at
me with that smug "poor devil, your program crashed coming out of the box"
look. And then they heard the card reader reading, the printer printing,
everything working, and I saw their expression turn to puzzlement. They
kept looking back and forth between the console display and the
peripherals, not quite able to figure out which was lying to them.
I loved it!
(What I didn't tell them was that, although I knew how the program was
written, it hadn't occurred to what effect it would have on the NORMAL
light until I saw it come on. Fortunately, I knew the reason, so my heart
only stopped for a fraction of a second.)
-Ron Hunsinger
the Jabberwock
So who's going to come up with a circuit which will plug into an ISA slot,
with the other end going to these LED displays, and come up with a
corresponding Windows VxD (or whatever) so we can get those displays to be
showing what they by rights should -- the current value of the Program
Counter or your choice of register...
And then a set of switches which fit into a 5.25" floppy bezel so we can
get a real front panel for our Pentium...
Joe Morris
bbrey...@aol.com writes:
>Charlie Gibbs writes:
>>Also, this machine had a 2311 disk drive, and the program would step
>>the heads in time with the music.
>More likely a 1311 drive on the 1620...ever see one break a hydraulic line?
The 1620 could attach a 1311, but as far as I could tell the mechanical
design of the 2311 was pretty much the same -- including the hydraulic
system and its %^&*(_~!# oil. The IBM 2314 also used hydraulic actuators;
memory says that the 3330 drives were the first standard IBM disk with
removable media that used electromagnetic actuators for the head rack.
(Does anyone recall what type of actuator was used by the 1301?)
Joe Morris / MITRE
Nick Spalding
.jcmo...@mwunix.mitre.org (Joe Morris) wrote:
>
> (Does anyone recall what type of actuator was used by the 1301?)
>
adder and the piston adder. The glob adder consisted of a piston in a
cylinder fed with oil from a number of small cylinders with solenoid
actuated pistons which if energised pumped a metered quantity of oil
into the main cylinder, the volumes being binarily related, eg 1, 2,
4, 8 times the unit volume. The piston adder is more difficult to
visualise. It consisted of a number of small pistons loosely linked
end to end in a single cylinder with the amount of free play between
them again binarily related and oil pressure was either applied or not
to the gaps between them again by solenoid operated valves. The
Cylinder Address Register bits were directly coupled to the solenoid
drivers which caused the access arm to extend to the (approximate)
correct position, then as in the 1311/2311 a detent drove home into a
triangular toothed rack to lock it in place.
Now will someone describe the Ramac access, also used on the 1405 I
think? Pneumatic with one pair of heads which retracted, moved up or
down to the desired platter and then accessed the required track. I
never saw one so I don't know the details. Alleged to sound like a
steam engine.
--
Nick Spalding
Tom Watson
>
> After a while, one sort of "gets used to" certain light patterns
> appearing when certain code ran -- one gets a "feel" for "normal".
> More particularly, from my systems programming point of view, one
> noticed when expected light patterns did *not* appear -- that usually
> meant something _unusual_ was going on. In most instances, "unusual"
> equated to "trouble".
>
Blinky lights...
On one of the machines I worked on, the main 32 bit register that was
displayed on the console was the equivalent to the "next" instruction to
be executed. This was especially handy if the current instruction was a
"wait" (wait for interrupt) instruction, which was common in the idle
loop. Being the productive soul I was at the time, I decided to be
inventive and make the next instruction someting useful. Being that
internally the time of day was kept as number of seconds from midnight (in
binary) I decided that it would be a nice thing to display (it changed,
etc...). The problem was that I didn't want to play around with self
altering code. Knowing that the registers were mapped into the first 16
locations of memory and one could execute things, led me to code up in
registers the wait instruction, followed by a NOP with its low order bits
as the binary time of day.
The one thing I gained from this exercise was that when the time rolled
over to 0x10000, it was around 6:15 (exact time left as exercise to the
reader) in the evening, and it was getting late. Time to go home.
Other than that, it served as a nice "warm fuzzy" to see that the machine
was running, and not stalled forever (which the prevous loop indicated).
*SIGH*
p.s. It wasn't a DEC machine!!
gl...@glass.cv.lexington.ibm.com
The IBM PS/2 model 95 has an "Information Display Panel" on
the front of it that is an eight character LED matrix used
for diagnostic purposes (One of the uses of the model 95 is
as a display-less LAN server, and some means of displaying
the health of the system was needed). During boot-up, the
panel is used to provide diagnostic information relating to
the boot process. During normal operation, one of the
digits is used as a disk access light.
In any case, it only takes a small software routine to
display anything you want on that panel. For example, on
mine, I have a program which displays a digital clock.
I imagine it could also be used for things such as CPU
loading, processor speed, or debug information, or...
Dave
P.S. Add Standard Disclaimer here.
P.S. The software is left as an exercise for the student.
Alan Bowler
In article <19961111225...@ladder01.news.aol.com> bbrey...@aol.com writes:
>
>More likely a 1311 drive on the 1620...ever see one break a hydraulic
>line?
>
dripped on the drive belt. You could see that the pack was spinning
slower due to the slipping belt, but it continued to work just fine.
The electronics adjusted the read/write clocking to match the actual
spindle speed.
gl...@glass.cv.lexington.ibm.com
I want one! I want one now! I WANT ONE NOW!!!
Actually, I've even gone as far as to consider wire-wrapping a
card to do this myself, but I just haven't had the time yet.
It is possible (and I've actually done this) to use the
wire-wrap area on the old IBM Games Adapter Card (aka
Joystick Adapter) to add an output port by using a pair of
4-bit D latches (7474 or 7476? I can't remember). This
8-bit port can then be used to drive 8 LEDs (Ok, so that's
not enough by today's standards. It was reasonable, though,
for the XT I performed this on).
Of course, the IBM PS/2 model 95 has an 8 character LED
"Information Panel" display that can be used in this mode, but
it doesn't have the switches for input. And, the display is
character oriented instead of single lamps (It does make a
neat digital clock, though).
Dave
P.S. Hey, I mentioned wire-wrapping and XTs, so that makes
this post appropriate to this new-group, doesn't it? :-)
Pete Olson
Mr M P Stevens (es...@csv.warwick.ac.uk) wrote:
: In article <561pmh$o...@learnet.freenet.hut.fi>,
: da...@mail.freenet.hut.fi (Dan Bernstein,Rehovot Israel) writes:
: >
: >BTW, how about the LED digits on the front panels of Intel-based machines,
: >which allegedly show you the speed the machine runs at (not that I
: >understand why someone would want to run their machine at less than the
: >highest
: > safe speed) - are those hard-wired or somewhat more sophisticated?
: >--Dan.
Not universally. My PC has a setup where if you hold the reset button
for something like five seconds it starts counting through speeds. Hmm,
don't actually remember how to set the non-turbo number.
Also, an old Zenith 386SX I had a set of POST indicator lights inside
the box. They were all in a row and blinked out from the outside in,
like Jeopardy lights! It was K-Rad :-)
That machine also had an off-chip memory cache (4k or 16K). So, if
there is no L1 cache (386) is off chip still considered L2?
Brad King
There's one for Linux that runs off a parallel port. You can assign each
of the eight LED's a seperate meaning or have them chase each other like
Cylon eyes to indicate system load. Pretty cool. Not up to AS/400 or
PS/2 85-95 level, yet, but it shouldn't be terribly tough to adapt...
-= Brad =-
the Jabberwock (jab...@goodnet.com) wrote:
: So who's going to come up with a circuit which will plug into an ISA slot,
:
:
:
Brian Raiter
the Jabberwock <jab...@goodnet.com>:
>So who's going to come up with a circuit which will plug into an ISA slot,
>with the other end going to these LED displays, and come up with a
>corresponding Windows VxD (or whatever) so we can get those displays to be
>showing what they by rights should -- the current value of the Program
>Counter or your choice of register...
I would pay money for that.
b
Michael Krell
In article <55rdvp$r...@pyrtech.mis.pyramid.com>,
Lon Stowell <lsto...@pyrtech.mis.pyramid.com> wrote:
> There were many of these hacks for different reasons. At
> SAC Headquarters, we had to hack their RCA Spectra message
> switches to turn off the idle light on the front panel...
> ...when the Air
> Farce General in charge of the project first got a demo of
> the working system...
> he was annoyed
> at paying millions of dollars for a computer system which
> had its idle light on most of the time.
>
> Suggestions that running batch jobs on the computer would be
> detrimental to it being able to handle a peak load of message
> switching on demand were obviously wasted on an uncomprehending
> intellect...so the operating system was hacked to run a very
> low priority task that did nothing more than turn the light out.
Hmmm... I guess I would have just relabeled the light "targeting USSR"
and left it at that. :-)
Mike
Simon Slavin
In article <566fep$r...@uwm.edu>,
j...@alpha1.csd.uwm.edu (John G Dobnick) wrote:
> After a while, one sort of "gets used to" certain light patterns
> appearing when certain code ran -- one gets a "feel" for "normal".
> More particularly, from my systems programming point of view, one
> noticed when expected light patterns did *not* appear -- that usually
> meant something _unusual_ was going on. In most instances, "unusual"
> equated to "trouble".
I started working with computers when they had the types of light
panels described in this thread. Now days I work with micros which
lack the panels. Even if they had them, the lights would flash so
fast and so many different routines run concurrently that I doubt
that front-panel lights would be of much use.
Instead, I've developed the habit of listening to disk heads. With
a number of long processing jobs I can tell which phase of the
program is running (the rhythm), how widely spread the records are
(if the record is already in cache it skips a beat) and how many
errors are being noted (writing to the error log adds an extra
click to the beat).
Now I just have to figure-out how to convince my customers that I'm
working even while I'm reading old Sci-Fi magazines.
Simon.
--
... or, here's a thought, why don't we talk about SF for a while?
-- Simon van Dongen <sg...@pi.net> on rec.arts.sf.written
Simon Slavin -- Computer Contractor Ordinaire. Junk email not welcome here.
Will administer ISO 9000 and year 2000 certification tests for food.
C Sanjayan Rosenmund
Ralph Wade Phillips wrote:
>
> Back in my old TRS-80 days ("Son, it wasn't a Model 1. They
> weren't Model 1's until the Model II came out ... " <B-) ),
Reminds me of trying to expalin to folks that my first computer (ok so
I'm a young pup) was (and it still runs) an Apple ][. *Not* a ][+,
*not* a //e, *not* a //c . . . .
> "Leave a small AM radio tuned to xxxx (forget the optimum
> frequency), and you'll be able to tell if the computer hung during a
> cassette load ... " <B-)
>
Ah yes, cassette loading. Makes a 140K Floppy look good!
Sanjay
sa...@sinewave.com
Geoff Short
kents (ke...@fs.cei.net) wrote:
: I've never had the opportunity to work with (or on) a large computer
: with lots of blinking lights, but there are plenty of times when
: I wish that my PC had some blinking lights. (is it actually doing
: something or is it just hung up?)
On a machine with little memory, the hard drive light performs this
task as the machine swaps - and you get an audible confirmation :-)
Our Linux server has an 8-led display to show load and memory usage,
but we cheated and used a software driver via the parallel port.
Geoff
--
----------------------------------------------------------------------------
Ever sit and watch ants? They're always busy with grs...@york.ac.uk
something, never stop for a moment. I just ge...@kipper.york.ac.uk
can't identify with that kind of work ethic. http://kipper.york.ac.uk/~geoff
Phil Romig
Sean Stanley-Adams <se...@corp2.mck.co.za> writes:
>Carl R. Friend wrote:
>>
>> It's very safe to say that the lights were _not_ placed there
>> solely for the edification of an occasional tour group. A good
>> technician could usually look at the lights and head right for the
>> problem area of the machine.
>>
>I joined the computer game in the early 70s as an operator on both an
>IBM 360/40 and a low-end 370, both of which were liberally festooned
>with lights. Though I wasn't able to use the lights to the same extent
>as the hardware engineers, it was possible with practise to identify a
>looping program. The programmers could never understand how we did it
>but then they didn't spend long nights with precious little else to do
>but looking at the blinking lights!
I worked for a shop that had both a PDP-11/44 and PDP-11/84. The
11/44 had a smoth front with no lights at all. The 11/84 had a pair
of displays which would do nothing but count down from 100 as fast as they
could. The display was driven by the "null" job, a job started at boot
by the OS which did some housekeeping and such. The null job had the
lowest priory, and so as the machine got busy the speed at which
the LED would count would slow down.
I ask a Digital field service rep one time why they bothered with the
lights on the 84, and she told me that up to the 44, all the PDP's had
LEDs which actually did something (error codes, buss state etc..).
As Sean said, operators began to rely on the lights to help them know
if the machine was just very slow or actually stuck.
When the 44 came out with no lights, the operators complained that they
had lost a valuable diagnostic tool. So when the introduced the newer
models, the added the LED Displays, driven by software since there was
not need for then in hardware any more. The only reason for them was
to help operators/programmers tell at a glance that the machines were
running.
I will say that it worked. It was really nice to be able to just glance
up and tell if the system was running or not.
Robert Humphris
Brian Raiter <bri...@speakeasy.org> wrote in article
<56cv58$h...@eve.speakeasy.org>...
Why go for the hardware solution?
It would be simple to bang up a windows app that would interogate which
ever reister
you wanted. You could perhaps even get a bitmap of an old computers facia,
and blit the light changes directly over it.
I know that it is not the 'real-thing' but it would be good fun to annoy
people with.
Rob
Tom Watson
> Ah yes, cassette loading. Makes a 140K Floppy look good!
Look, you had 140k on a floppy. That was one of those new fangled "Double
Density" things. I built an interface for a 100k floppy (256
bytes/sector, 10 sectors/track, 40 tracks) but that was on the 40 track
drives. The original SA400's only had 35 tracks, so they were only 89.6k
floppies. Nowdays that is smaller than the cache memory INSIDE some CPU
chips. Times have changed...
Tim Shoppa
In article <tsw-151196...@cypher.3do.com>,
Tom Watson <t...@3do.com> wrote:
>In article <328AAC...@sinewave.com>, sa...@sinewave.com wrote:
>
>> Ah yes, cassette loading. Makes a 140K Floppy look good!
>
>Look, you had 140k on a floppy. That was one of those new fangled "Double
>Density" things.
The most common 140k mini-floppy format that I know of was the Apple II
GCR format. Technically, it's not double density; the number of flux
transitions matches that of regular single-density 5.25" FM floppies,
such as the SA390 and SA400.
Tim. (sho...@triumf.ca)
John G Dobnick
From article <19961111225...@ladder01.news.aol.com>, by bbrey...@aol.com:
>
> More likely a 1311 drive on the 1620...ever see one break a hydraulic
> line?
I remember those! Never saw a hydraulic line break, but I did
notice pools of oil on the floor every once in a while. :-)
I'm told the FE's would come in and "top off" the hydraulics
every month, too. (Whether they _changed_ the oil or not, I don't
know.)
--
John G Dobnick "Knowing how things work is the basis
Information & Media Technologies for appreciation, and is thus a
University of Wisconsin - Milwaukee source of civilized delight."
j...@csd.uwm.edu ATTnet: (414) 229-5727 -- William Safire
Mike Czaplinski
grs...@york.ac.uk (Geoff Short) wrote:
>
> kents (ke...@fs.cei.net) wrote:
>
> : I've never had the opportunity to work with (or on) a large computer
> : with lots of blinking lights, but there are plenty of times when
> : I wish that my PC had some blinking lights. (is it actually doing
> : something or is it just hung up?)
>
> On a machine with little memory, the hard drive light performs this
> task as the machine swaps - and you get an audible confirmation :-)
DEC hard drives are quite useful for this purpose.
My boss has the swap file for his NT box at home on a DEC 2gig drive
for just this purpose.
Mike "Thus, DEC is still producing excellent analog realtime diagnostic
tools" Czaplinski
mike.cz...@washingtondc.ncr.com
Alistair Young
In <562o3d$3...@nyx10.cs.du.edu>, snor...@nyx10.cs.du.edu (Scott Norwood) writes:
>Thus, cases were made with a 'turbo' switch and a little light which indicated
>the current status of machine speed. Since the 486, though, these 'turbo'
>switches generally don't do anything except just sit there. I think maybe
>they look cool, kind of like the 'turbo' label on certain cars... Kind
>of dumb, though.
If I'd bothered to connect mine properly, it would disable the cache
RAM on the motherboard, which I suppose would give some speed reduction.
As it is, I've turned it into an enable/disable switch for the internal
speaker.
Alistair
--
Arkane Systems Ltd. Sysimperator, dominus retis deusque machinarum.
e-mail: ava...@arkane.demon.co.uk Phone: +44 (1833) 638233 (10am-9pm GMT)
sl...@bofh.net WWW: http://www.arkane.demon.co.uk/
The opinions above ARE my company's, because I OWN it! [Team OS/2]
Mark Gonzales
In article <01bbd231$e55c9120$9734...@irvine.csl.co.uk>,
Robert Humphris <rhum...@csl.co.uk> wrote:
>It would be simple to bang up a windows app that would interogate which
>and blit the light changes directly over it.
That would make a *great* screen-saver. You could map the x86 registers
onto a bitmap of the front panel of an IBM 360, or a pdp-11. Maybe even
allow the mouse to toggle the front-panel switches. Single-step mode
might require some careful programming though. . .
--Mark
absolutely not speaking for Intel.
JOHN RYLAND
On 11 Nov 1996, David E. Fox wrote:
<
<My brother's 486/56 (cyrix) was wired so that the speed indicator said
<"JC" (jeezus christ) :)
A really good one I once saw was the speed indicator set up to say "ILL"
<--
<------------------------------------------------------------------------
<David E. Fox Tax Thanks for lettimg me
<df...@belvdere.vip.best.com the change magnetic patterns
<ro...@belvedere.sbay.org churches on your hard disk.
<-----------------------------------------------------------------------
Rich Hurd
Michael Krell <mk...@clove.cs.Virginia.EDU> wrote in article
<E0tr6...@murdoch.acc.Virginia.EDU>...
> In article <55rdvp$r...@pyrtech.mis.pyramid.com>,
> Lon Stowell <lsto...@pyrtech.mis.pyramid.com> wrote:
[snip]
> > ...so the operating system was hacked to run a very
> > low priority task that did nothing more than turn the light out.
Of course! The genesis of the "FLI" assembler instruction. <g>
--
r/h
rah...@poboxes.com / It is by caffeine alone I set my mind in motion, it
ICT Associates __/ is by the beans of Java that thoughts acquire speed,
Easton, PA USA / the hands acquire shaking, the shaking becomes a
/ warning, it is by caffeine alone I set my mind in motion.
----------------------------------------------------------------------
Remove the * from my address to reply. Foil the spambots!