Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.folklore.computers > #208394 > unrolled thread

Remington Rand punched card

Back to article view | Back to alt.folklore.computers

Contents

  Remington Rand punched card hancock4@bbs.cpcn.com - 2019-12-05 12:29 -0800
    Re: Remington Rand punched card Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2019-12-05 23:09 +0000
      Re: Remington Rand punched card hancock4@bbs.cpcn.com - 2019-12-06 13:30 -0800
        Re: Remington Rand punched card Quadibloc <jsavard@ecn.ab.ca> - 2019-12-06 13:45 -0800
          Re: Remington Rand punched card Quadibloc <jsavard@ecn.ab.ca> - 2019-12-06 13:49 -0800
          Re: Remington Rand punched card hancock4@bbs.cpcn.com - 2019-12-06 14:27 -0800
            Re: Remington Rand punched card Quadibloc <jsavard@ecn.ab.ca> - 2019-12-06 19:20 -0800
              Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-10 06:21 -0700
                Re: Remington Rand punched card Quadibloc <jsavard@ecn.ab.ca> - 2020-03-10 11:14 -0700
                  Re: Remington Rand punched card J. Clarke <jclarke.873638@gmail.com> - 2020-03-10 19:34 -0400
                    Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-10 17:58 -0700
                      Re: Remington Rand punched card J. Clarke <jclarke.873638@gmail.com> - 2020-03-10 22:50 -0400
                        Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-11 06:52 -0700
                          Re: Remington Rand punched card J. Clarke <jclarke.873638@gmail.com> - 2020-03-11 19:11 -0400
                            Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-11 18:19 -0700
                              Re: Remington Rand punched card J. Clarke <jclarke.873638@gmail.com> - 2020-03-11 21:38 -0400
                              Re: Remington Rand punched card Dan Espen <dan1espen@gmail.com> - 2020-03-11 22:21 -0400
                                Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-11 21:26 -0700
                            Re: Remington Rand punched card Dan Espen <dan1espen@gmail.com> - 2020-03-11 22:08 -0400
                              Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-11 21:21 -0700
                                Re: Remington Rand punched card Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-03-12 19:08 +0000
                  Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-10 17:56 -0700
            Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-10 06:17 -0700
              Re: Remington Rand punched card Dan Espen <dan1espen@gmail.com> - 2020-03-10 12:26 -0400
                Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-10 17:52 -0700
        Re: Remington Rand punched card Quadibloc <jsavard@ecn.ab.ca> - 2019-12-06 13:46 -0800
        Re: Remington Rand punched card scott@slp53.sl.home (Scott Lurndal) - 2019-12-08 17:08 +0000
          Re: Remington Rand punched card Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2019-12-08 18:59 +0000
          Re: Remington Rand punched card Quadibloc <jsavard@ecn.ab.ca> - 2019-12-08 20:57 -0800
    Re: Remington Rand punched card robin.vowels@gmail.com - 2020-03-10 06:07 -0700

Page 2 of 2 — ← Prev page 1 [2]

#210493

On 2020-03-12, robin.vowels@gmail.com <robin.vowels@gmail.com> wrote:

> On Thursday, March 12, 2020 at 1:08:36 PM UTC+11, Dan Espen wrote:
>
>> Not sure where this is going or if I want to get on the bus,
>> but I've never seen a card torn into more than a couple of pieces,
>> normally they just jammed together and got curled up or ripped a bit.

I suspect that the term "confetti" is an exaggeration.

>> But all my experience is with IBM card readers, and some EAM equipment.
>> They were all designed to sense jams and come to a rapid halt.
>
> That's right.
> Ours sensed that if the second reading station did not receive
> a card, the reader halted.  If it was a misfeed owing to a worn card,
> nothing was lost, and no damage done to the card.
> At worse, a "foreign" card might be torn slightly.

One of the best jams I saw resulted in two consecutive cards being
torn lengthwise about half their length, and winding up overlapped.
(Try pressing your hands towards each other with fingers extended
to see what I mean.)

One jam left a card in contact with a rapidly spinning roller,
which burned a hole about 3/4 of an inch in diameter in the card.
We saw smoke coming out of the machine.

> The card punch was another matter.  Occasionally the stacker
> mis-stacked a card, and every card after that was compressed
> into a fan about 3 mm wide.  You could get about 100 cards
> concertina'd into a small space until the punch began to shreik.

Yes, I've seen some nice (and amazingly neat) accordion pleats
put into cards.

> At one stage we had some cards printed to a special order, with green ink.
> The ink was blotchy, and occasionally two cards were stuck together
> at a blotch. That caused a jam.

One shop that went through a lot of cards bought them wherever they
could find them to save money.  Most of these were overruns from
other applications, which is how I built such an interesting
collection of differently-printed cards.  Sometimes the quality
of these cards was definitely dodgy.  I suspect that the thickness
of some of them wasn't that tightly controlled, which resulted in
one operator turning the air blue as a result of all the jams.
When you fanned and joggled a fresh deck of cards before putting
them into the punch, you had to work a bit; the cards often stuck
together.  I remember picking up a handful of cards and smacking
them against my other hand to break them apart.  I almost broke
my hand; they were like a block of wood.  Sometimes we'd stand
blocks of cards on edge on a machine or table top, where they'd
sit in a solid block with no rubber band holding them together.

> Not to be put off, I noticed that the card feeler that detected
> whether that a card had been fed, had been pushed up by the jam.
> There was a spare set of contacts on that feeler, which I wired
> up to the "not ready" circuit.  Presto, no jams at all, because
> as soon as a card began to buckle, the sensor activated the "not ready"
> circuit, and the card punch stopped without feeding a card (well,
> just a tiny bit past the throat).

Nice.  My best card reader fix was when a timing belt broke.
The first tihng I did was go to the CE cabinet and get out the
manuals for the reader.  I looked up the part number of the
broken belt, then called the service department.  They were
used to service requests of the form "the machine doesn't work,"
so when I told them a timing belt had broken they started the
game of Twenty Questions to figure out which belt it was.
("What does it look like?  Where in the machine is it?"
Et cetera.)  I read off the part number and I could hear
the person on the other end light up.  He quickly dispatched
a CE with the needed part.

Meanwhile, we had lots of work to do and didn't want to wait
until the CE arrived.  So I took a lot of rubber bands and
wrapped them around the pulleys in place of the missing belt.
It took a couple of seconds for everything to spin up, but by
loading the program and getting it ready to start reading,
I was able to use the manual feed button to spin up the card
reader and feed the first card, then start the program before
the card reader motor timed out and shut down again.  It worked
well enough for production to continue.

-- 
/~\  Charlie Gibbs                  |  Microsoft is a dictatorship.
\ /  <cgibbs@kltpzyxm.invalid>      |  Apple is a cult.
 X   I'm really at ac.dekanfrus     |  Linux is anarchy.
/ \  if you read it the right way.  |  Pick your poison.

[toc] | [prev] | [next] | [standalone]

#210433

On Wednesday, March 11, 2020 at 5:14:26 AM UTC+11, Quadibloc wrote:
> On Tuesday, March 10, 2020 at 7:21:33 AM UTC-6, robin...@gmail.com wrote:
> 
> > IBM cards were thinner than ICT/ICL cards.
> > One needed to adjust the throat of card reader to
> > suit the brand of card.
> 
> Partly, I am astonished to hear that, I would have expected everyone to make cards that were compatible - in conformance to the standard for punched cards.

The dimensions of the card, viz., width and height, were the same.

> However, I also encountered differing specifications for card thickness from IBM and from Control Data:
> 
> IBM's spec is .0065" plus or minus .0005".
> 
> Control Data's spec is .007" plus or minus .0004".
> 
> These ranges _overlap_, but Control Data's is definitely thicker.

We had thickness gauges to set the throuat width (like the feeler gauges
used for motor car engines).

[toc] | [prev] | [next] | [standalone]

#210393

On Saturday, December 7, 2019 at 9:27:06 AM UTC+11, hanc...@bbs.cpcn.com wrote:
> On Friday, December 6, 2019 at 4:45:53 PM UTC-5, Quadibloc wrote:
> > On Friday, December 6, 2019 at 2:30:25 PM UTC-7, hanc...@bbs.cpcn.com wrote:
> >  
> > > IBM cards used to have round holes and 45 columns.  But then
> > > around 1930 IBM developed the rectangular 80 column format.
> > > Univac kept the round holes, but went to a 90 column format.
> > 
> > This was, of course, because when they developed the 80 column card, they 
> > patented it. Their patent was broad enough to cover the use of smaller round 
> > holes than those used in 45 column cards as well.
> > 
> > This is why Remington Rand, in order to remain competitive, had to make more 
> > efficient use of the holes that were already there.
> > 
> > In Britain, though, apparently their patent office did not accept that making 
> > the round holes smaller was patentable - and so another maker of punched cards 
> > in that country came up with a 65-column card, also having the same dimensions 
> > as the 80-column card, but, because the holes were smaller, they were able to 
> > put the columns closer together and have 65 of them, if not 80 or 90.
> > 
> > One source appears to claim they went on to do what Remington Rand did as well, 
> > and make a 130-column card out of it, but I could never verify that.
> 
> I think IBM researched bigger cards to hold more information,
> but never marketed a tab product based on that.  Customers certainly
> would've liked storing more information to avoid, if possible,
> having multiple cards for a transaction.
> 
> Eventually, they did develop the S/3 96 column card.  But
> their history says the reason as to hold cost down as less
> hardware was required for the S/3 card than the 80-column
> card.

That sounds like a furphy.
To read the holes off a card required a cylindrical roller
and wire brushes  (one per card column).
Later, photoelectric sensors were used, but the old
wire brushes were just right for the job.

> 
> IBM punched cards went through a lot of trees over the years.
> 
> IBM made a lot of profit from the sale of punched card stock.
> Indeed, circa 1960 IBM advertised the care it took in designing
> and printing punched card stock.  bitsavers even as a
> booklet on how to design a punched card.
> http://bitsavers.org/pdf/ibm/generalInfo/C20-8078_Form_and_Card_Design_1961.pdf
> 
> Going back to the early 1950s, on the one hand, major IBM
> customers sought mag tape since they were drowning in
> punched card records and running out of space.  On the other
> hand, a punched card was readable while mag tape wasn't.  
> Would mag tape maintain records reliably?  Businesses
> weren't sure.  Indeed, the earliest tapes weren't that
> good.

[toc] | [prev] | [next] | [standalone]

#210404

robin.vowels@gmail.com writes:

> On Saturday, December 7, 2019 at 9:27:06 AM UTC+11, hanc...@bbs.cpcn.com wrote:
>> On Friday, December 6, 2019 at 4:45:53 PM UTC-5, Quadibloc wrote:
>> > On Friday, December 6, 2019 at 2:30:25 PM UTC-7, hanc...@bbs.cpcn.com wrote:
>> >  
>> > > IBM cards used to have round holes and 45 columns.  But then
>> > > around 1930 IBM developed the rectangular 80 column format.
>> > > Univac kept the round holes, but went to a 90 column format.
>> > 
>> > This was, of course, because when they developed the 80 column card, they 
>> > patented it. Their patent was broad enough to cover the use of smaller round 
>> > holes than those used in 45 column cards as well.
>> > 
>> > This is why Remington Rand, in order to remain competitive, had to make more 
>> > efficient use of the holes that were already there.
>> > 
>> > In Britain, though, apparently their patent office did not accept that making 
>> > the round holes smaller was patentable - and so another maker of punched cards 
>> > in that country came up with a 65-column card, also having the same dimensions 
>> > as the 80-column card, but, because the holes were smaller, they were able to 
>> > put the columns closer together and have 65 of them, if not 80 or 90.
>> > 
>> > One source appears to claim they went on to do what Remington Rand did as well, 
>> > and make a 130-column card out of it, but I could never verify that.
>> 
>> I think IBM researched bigger cards to hold more information,
>> but never marketed a tab product based on that.  Customers certainly
>> would've liked storing more information to avoid, if possible,
>> having multiple cards for a transaction.
>> 
>> Eventually, they did develop the S/3 96 column card.  But
>> their history says the reason as to hold cost down as less
>> hardware was required for the S/3 card than the 80-column
>> card.
>
> That sounds like a furphy.
> To read the holes off a card required a cylindrical roller
> and wire brushes  (one per card column).
> Later, photoelectric sensors were used, but the old
> wire brushes were just right for the job.

They must have been using optics on those 96 column cards,
brushes wouldn't make contact through those small holes.
The 80 column card had elongated holes for the brushes to work.

The 1442 read 80 column cards sideways, but used optics.

-- 
Dan Espen

[toc] | [prev] | [next] | [standalone]

#210432

On Wednesday, March 11, 2020 at 3:26:58 AM UTC+11, Dan Espen wrote:
> r.....@gmail.com writes:
> 
> > On Saturday, December 7, 2019 at 9:27:06 AM UTC+11, hanc...@bbs.cpcn.com wrote:
> >> On Friday, December 6, 2019 at 4:45:53 PM UTC-5, Quadibloc wrote:
> >> > On Friday, December 6, 2019 at 2:30:25 PM UTC-7, hanc...@bbs.cpcn.com wrote:
> >> >  
> >> > > IBM cards used to have round holes and 45 columns.  But then
> >> > > around 1930 IBM developed the rectangular 80 column format.
> >> > > Univac kept the round holes, but went to a 90 column format.
> >> > 
> >> > This was, of course, because when they developed the 80 column card, they 
> >> > patented it. Their patent was broad enough to cover the use of smaller round 
> >> > holes than those used in 45 column cards as well.
> >> > 
> >> > This is why Remington Rand, in order to remain competitive, had to make more 
> >> > efficient use of the holes that were already there.
> >> > 
> >> > In Britain, though, apparently their patent office did not accept that making 
> >> > the round holes smaller was patentable - and so another maker of punched cards 
> >> > in that country came up with a 65-column card, also having the same dimensions 
> >> > as the 80-column card, but, because the holes were smaller, they were able to 
> >> > put the columns closer together and have 65 of them, if not 80 or 90.
> >> > 
> >> > One source appears to claim they went on to do what Remington Rand did as well, 
> >> > and make a 130-column card out of it, but I could never verify that.
> >> 
> >> I think IBM researched bigger cards to hold more information,
> >> but never marketed a tab product based on that.  Customers certainly
> >> would've liked storing more information to avoid, if possible,
> >> having multiple cards for a transaction.
> >> 
> >> Eventually, they did develop the S/3 96 column card.  But
> >> their history says the reason as to hold cost down as less
> >> hardware was required for the S/3 card than the 80-column
> >> card.
> >
> > That sounds like a furphy.
> > To read the holes off a card required a cylindrical roller
> > and wire brushes  (one per card column).
> > Later, photoelectric sensors were used, but the old
> > wire brushes were just right for the job.
> 
> They must have been using optics on those 96 column cards,
> brushes wouldn't make contact through those small holes.

I was referring to the statement that the 96-column card was
introduced to keep costs down. as "less hardware was required..."

I pointed out that the hardware required for the 80-column card
was a cylinder and wire brushes.  That's about as simple as you can get.
and was used for some 40+ years.

> The 80 column card had elongated holes for the brushes to work.

Naturally.

> The 1442 read 80 column cards sideways, but used optics.

[toc] | [prev] | [next] | [standalone]

#208454

On Friday, December 6, 2019 at 2:30:25 PM UTC-7, hanc...@bbs.cpcn.com wrote:
> Indeed, it was said IBM quietly
> helped Remington Rand out to keep it in business to avoid
> a total monopoly.

Something like Microsoft and Apple.

John Savard

[toc] | [prev] | [next] | [standalone]

#208485

hancock4@bbs.cpcn.com writes:
>On Thursday, December 5, 2019 at 6:10:17 PM UTC-5, Charlie Gibbs wrote:
>
>> If you zoom in on the second one you can see the same 90-column
>> cards that I played with a bit in the late '60s.  The cards are
>> the same dimensions as IBM's 80-column cards, and have the same
>> 12 rows, but the 45 columns are divided into upper and lower tiers -
>> perfect for 6-bit codes.
>
>IBM cards used to have round holes and 45 columns.  But then
>around 1930 IBM developed the rectangular 80 column format.
>Univac kept the round holes, but went to a 90 column format.

I've got some blank Burroughs 90-column cards here in my hand.

Although I never saw any 90-column gear during my years with
Burroughs/Unisys, even after the acquisition of Sperry/Univac.

[toc] | [prev] | [next] | [standalone]

#208487

On 2019-12-08, Scott Lurndal <scott@slp53.sl.home> wrote:

> hancock4@bbs.cpcn.com writes:
>
>> On Thursday, December 5, 2019 at 6:10:17 PM UTC-5, Charlie Gibbs wrote:
>>
>>> If you zoom in on the second one you can see the same 90-column
>>> cards that I played with a bit in the late '60s.  The cards are
>>> the same dimensions as IBM's 80-column cards, and have the same
>>> 12 rows, but the 45 columns are divided into upper and lower tiers -
>>> perfect for 6-bit codes.
>>
>> IBM cards used to have round holes and 45 columns.  But then
>> around 1930 IBM developed the rectangular 80 column format.
>> Univac kept the round holes, but went to a 90 column format.
>
> I've got some blank Burroughs 90-column cards here in my hand.
>
> Although I never saw any 90-column gear during my years with
> Burroughs/Unisys, even after the acquisition of Sperry/Univac.

I have some punched 90-column cards.  I once tried to write a
program to read them on an 80-column reader.  The columns that
were more or less aligned with the 80-column positions came out
OK, but the ones that fell between the 80-column positions were,
as I expected, unreadable.  Too bad - I even had code working to
translate the 90-column cards' arcane encoding, which followed
no pattern that I could discern.

The Univac card readers of the day derived column timing from a
little drum attached to the feed rollers; one head wrote timing
marks to the drum and and another head read them back after a
suitable rotation of the drum to advance the card to the next
column.  The card reader on the Univac 1005 (a hopped-up electronic
answer to the IBM 407) had an extra head on this drum, and could
read both 80- and 90-column cards.

-- 
/~\  cgibbs@kltpzyxm.invalid (Charlie Gibbs)
\ /  I'm really at ac.dekanfrus if you read it the right way.
 X   Top-posted messages will probably be ignored.  See RFC1855.
/ \  "Alexa, define 'bugging'."

[toc] | [prev] | [next] | [standalone]

#208489

On Sunday, December 8, 2019 at 10:08:59 AM UTC-7, Scott Lurndal wrote:

> Although I never saw any 90-column gear during my years with
> Burroughs/Unisys, even after the acquisition of Sperry/Univac.

That doesn't surprise me. Soon after the expiry of the IBM patents, I'm sure that 
when they offered conventional 80-column equipment, that was what their customers 
preferred.

John Savard

[toc] | [prev] | [next] | [standalone]

#210392

On Friday, December 6, 2019 at 7:29:40 AM UTC+11, hanc...@bbs.cpcn.com wrote:
> IBM wasn't the only punched card business.  Remington
> Rand also had a line of punched card tabulating machines.
> 
> Here are two ads for it:
> 
> https://archive.org/details/Nations-Business-1934-01/page/n55
> 
> https://archive.org/details/Nations-Business-1934-08/page/n45
> 
> 
> 
> Supposedly, IBM integrated its punched card machines with
> its new computers, but Remington kept them separate.
> This hurt Remington's computers.
> 
> I don't know the detailed history of Remington Rand,
> but I can't help but suspect it was not a well run
> company after WW II.  I know they hired Gen. Leslie Groves
> basically as an expensive figurehead to go around and
> make semi-political speeches that the CEO liked.  One
> books cites that Univac and ERA were basically two
> separate competing divisions.
> 
> Maybe things got better when they merged with Sperry.

Don't forget Burroughs and ICT/ICL.

[toc] | [prev] | [standalone]

Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | alt.folklore.computers

csiph-web