🚀 go-pugleaf

On Tuesday, February 27, 2018 at 1:28:52 PM UTC-7, Andreas Kohlbach wrote:
> When you look at microprocessors up to the very late 70s you might notice
> many of them have a 6 and/or 8 in their names. Like the 6502, 8080, 8086,
> Z80, M680X, M680XY...

> Is this a coincidence or a pattern behind that?

Well, the MOS Technology 6502 was intended to be "similar" to the Motorola 6800, 
and thus a cheaper alternative to it.

The number 8 is considered a lucky number among the Chinese - but those 
processors acted on 8-bit bytes. And the Z80 was intended to be a replacement 
for the 8080, offering a bunch of new additional functions.

Also, back in earlier times, IBM made a number of series of computers...

7010: an expanded version of their popular 1401
7030: the later copies of the original STRETCH sent to Los Alamos
7040, 7044: a budget 36-bit scientific with some additional character instructions
7070: a computer with decimal arithmetic but a fixed word length, intended to be useful in both commercial and scientific applications
7090, 7094: their 36-bit scientific

and so that may have influenced people away from the famous Western lucky 
number.

John Savard

On 2/27/2018 3:25 PM, Quadibloc wrote:
> On Tuesday, February 27, 2018 at 1:28:52 PM UTC-7, Andreas Kohlbach wrote:
>> When you look at microprocessors up to the very late 70s you might notice
>> many of them have a 6 and/or 8 in their names. Like the 6502, 8080, 8086,
>> Z80, M680X, M680XY...
> 
>> Is this a coincidence or a pattern behind that?
> 
> Well, the MOS Technology 6502 was intended to be "similar" to the Motorola 6800,
> and thus a cheaper alternative to it.
> 
> The number 8 is considered a lucky number among the Chinese - but those
> processors acted on 8-bit bytes. And the Z80 was intended to be a replacement
> for the 8080, offering a bunch of new additional functions.
> 

Also, since Motorola began with the 6800, it was logical for another 
8-bit to be 6809.  Then the 16 adne 32 bit micros would be 68000 plus 
some other distinguishing number:  010, 020, 030, 040, 060, etc.

So with Intel:  4040, 8008, 8080, 8085, 8086, 8088...  then later the 
80... 286, 386, 486...  hen the Pentium I, II, III... and now to the 
myriad of processors with unordered names and multiple cores.

Intel also put out 8-bit micro-controllers like the 8051, 8048, 8042, 
etc...  ISTM the keyboard controller on the original IBM PC was an Intel 
8042.  Now of course there is a custom chip.  I saw a broken open Dec 
VT220 once... it was run rather slowly by an Intel 8048.

Let's not forget the Fairchild F-8, the Signetics 2650, and the National 
Semiconductor SC/MP "scamp".

-- 
numerist at aquaporin4 dot com

Quadibloc <jsa...@ecn.ab.ca> wrote:
>Also, back in earlier times, IBM made a number of series of computers...

>7010: an expanded version of their popular 1401
>7030: the later copies of the original STRETCH sent to Los Alamos
>7040, 7044: a budget 36-bit scientific with some additional character instructions
>7070: a computer with decimal arithmetic but a fixed word length, intended to be useful in both commercial and scientific applications
>7090, 7094: their 36-bit scientific

>and so that may have influenced people away from the famous Western lucky 
>number.

>John Savard

And the 709 a precusser of the 709X. Preceeding this was the IBM 700.
-- 
-- 
  O wad some Power the giftie gie us
  To see oursels as ithers see us!
  It wad frae mony a blunder free us,
  An' foolish notion:

In article <p7ptui$sgd$2...@reader1.panix.com>,
Walter Bushell  <pr...@panix.com> wrote:
>And the 709 a precusser of the 709X. Preceeding this was the IBM 700.

701 and 704, actually.

The 701 was IBM's first real computer in 1952.  The much improved 704 followed it
in 1954, and the 709 in 1957.

The 702 and 705 were character addressed commercial machines, same era
but quite different architectures.

I don't know why they all started with 7.

-- 
Regards,
John Levine, jo...@iecc.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

On Wednesday, March 7, 2018 at 9:19:17 PM UTC-5, John Levine wrote:
> In article <p7ptui$sgd$2...@reader1.panix.com>,
> Walter Bushell  <pr...@panix.com> wrote:
> >And the 709 a precusser of the 709X. Preceeding this was the IBM 700.
> 
> 701 and 704, actually.
> 
> The 701 was IBM's first real computer in 1952.  The much improved 704 followed it
> in 1954, and the 709 in 1957.
> 
> The 702 and 705 were character addressed commercial machines, same era
> but quite different architectures.
> 
> I don't know why they all started with 7.

IBM's calculator series were in the 600s, so I would guess that
7 was the next available series.

IBM planned to have the subsequent computers in the 800 or 8000
series, but this was cancelled and replaced by S/360.

I don't know where the "14" came from for the 1401 series.

As an aside, I believe IBM's most popular calculator was the 604
which used vacuum tubes.  But IBM also had the transistorized 
608 and 609.   They only sold a few models and I suspect more
done as an experimental platform for transistor logic as opposed
to a real product.

On Friday, March 16, 2018 at 1:45:05 PM UTC-4, han...@bbs.cpcn.com wrote:
> IBM's calculator series were in the 600s, so I would guess that
> 7 was the next available series.
> 
> IBM planned to have the subsequent computers in the 800 or 8000
> series, but this was cancelled and replaced by S/360.
> 
> I don't know where the "14" came from for the 1401 series.

The 14 came from the fact that the origial model, sold as an "elecronically programmable tabulator", had 1400 postions of internal storage.

Bob Netzlof

On Friday, March 16, 2018 at 8:27:21 PM UTC-4, rne...@gmail.com wrote:

> The 14 came from the fact that the origial model, sold as an "elecronically programmable tabulator", had 1400 postions of internal storage.

Mske that "original model"

Bob Netzlof

rnet...@gmail.com writes:

> On Friday, March 16, 2018 at 1:45:05 PM UTC-4, han...@bbs.cpcn.com wrote:
>> IBM's calculator series were in the 600s, so I would guess that 7 was
>> the next available series.
>> 
>> IBM planned to have the subsequent computers in the 800 or 8000
>> series, but this was cancelled and replaced by S/360.
>> 
>> I don't know where the "14" came from for the 1401 series.
>
> The 14 came from the fact that the origial model, sold as an
> "elecronically programmable tabulator", had 1400 postions of internal
> storage.
>
> Bob Netzlof

Never heard that before but that sounds plausible.

Leaving the mystery of how they chose:

1401
1410
1460
1440

I've always assumed model numbers and names are chosen by the
sales force and their reasoning is something like "that sounds nice".

-- 
Dan Espen

On Friday, March 16, 2018 at 9:18:29 PM UTC-4, Dan Espen wrote:
 
> > The 14 came from the fact that the origial model, sold as an
> > "elecronically programmable tabulator", had 1400 postions of internal
> > storage.

> Never heard that before but that sounds plausible.

I think, underscore think, that I read it in some piece of 1401 literature "way back when".
 
> Leaving the mystery of how they chose:
 
> 1410
> 1460
> 1440

As to those, I suppose the 14xx came from having an instruction set and general internal arrangements similar to the 1401. If I remember correctly, the 1440 was a 1401, less parallel I/O, plus serial I/O to 1442 card machine, 1443 printer. Where the 1401 had I/O areas in fixed locations in storage, the 1440 could read into and write from anywhere in storage.

The 1460 (hm. think hard) was a faster 1401, with I/O controls for a huge random access storage device.

The 1410 was a bigger 1401 which used (I think) 5 digit storage addresses, but was otherwise rather like a 1401.

Bob Netzlof

On Friday, March 16, 2018 at 9:18:29 PM UTC-4, Dan Espen wrote:
> rnet...@gmail.com writes:
> 
> > On Friday, March 16, 2018 at 1:45:05 PM UTC-4, han...@bbs.cpcn.com wrote:
> >> IBM's calculator series were in the 600s, so I would guess that 7 was
> >> the next available series.
> >> 
> >> IBM planned to have the subsequent computers in the 800 or 8000
> >> series, but this was cancelled and replaced by S/360.
> >> 
> >> I don't know where the "14" came from for the 1401 series.
> >
> > The 14 came from the fact that the origial model, sold as an
> > "elecronically programmable tabulator", had 1400 postions of internal
> > storage.
> >
> > Bob Netzlof
> 
> Never heard that before but that sounds plausible.
> 
> Leaving the mystery of how they chose:
> 
> 1401
> 1410
> 1460
> 1440
> 
> I've always assumed model numbers and names are chosen by the
> sales force and their reasoning is something like "that sounds nice".

In addition to the 1410, there was also the 7010.
https://www-03.ibm.com/ibm/history/exhibits/dpd50/dpd50_chronology2.html

I don't have my IBM Early Computers history handy, but I believe
it said the 14xx series was actually an outgrowth of several
different development efforts, for example, a super-tabulator
for European service and something called the "Wooden Wheel".
IBM eventually re-oriented those efforts into a common 
stored-program computer product line.

The book talks a little about the development history of the
other models.  I think at least one model, and maybe several,
were actually intended to be something else, and IBM had them
change their architecture to 14xx compatible. 

As mentioned, the 1410 was "sort of" 1401 compatible, while the
1460 was compatible.  I'm not sure of the differences of the 1410
and 1460--that is, why a customer would buy one over the other--
but I suspect that came out of the differing background.  The
unity of S/360 was supposed to eliminate that kind of confusion.

Some of the other 14xx models were higher end, but one was a
low-end.  I think the low-end model didn't sell too well because
it was too costly for the low performance.

As we know, Haanstra was not a supporter of S/360 and actually
quietly built a 'super' 1401 using new components.  IBM was
not pleased by this effort and he was punished for it.

As Campbell-Kelly explains in their book, the beauty of the 
IBM 1401 was not in its electronics (not a big deal) but rather
the 1403 printer, which gave high quality at a significantly
faster rate.  As mentioned, I remain amazed that for _years_,
customers tolerated wavy lines of other mainframe line printers,
even slow ones.

rnet...@gmail.com writes:

> On Friday, March 16, 2018 at 9:18:29 PM UTC-4, Dan Espen wrote:
>  
>> > The 14 came from the fact that the origial model, sold as an
>> > "elecronically programmable tabulator", had 1400 postions of
>> > internal storage.
>
>> Never heard that before but that sounds plausible.
>
> I think, underscore think, that I read it in some piece of 1401
> literature "way back when".

If I ever read that the bits in my brain must have flipped.
I only encountered a 1400 character 1401 after I'd worked on them
for a few years.  I was surprised they made them that small.
The small machine also had no index registers, but by that time
I'd stop using index registers anyway so no big deal.

>> Leaving the mystery of how they chose:
>  
>> 1410 1460 1440
>
> As to those, I suppose the 14xx came from having an instruction set
> and general internal arrangements similar to the 1401. If I remember
> correctly, the 1440 was a 1401, less parallel I/O, plus serial I/O to
> 1442 card machine, 1443 printer. Where the 1401 had I/O areas in fixed
> locations in storage, the 1440 could read into and write from anywhere
> in storage.

Yep.  Started in a shop with a 1401 and a 1460 then spent 2 years with
the 1440.  You could read/write unit record anywhere in storage but
habits die hard and I used the same locations the 1401/1460 used.

> The 1460 (hm. think hard) was a faster 1401, with I/O controls for a
> huge random access storage device.

No disk on the 1460 I used, but the 1440 had disk.
The 1311 disk wasn't exactly huge.  1 meg per drive.
Our Accounts Receivable ended up on 20 packs.  We only needed
10 packs for one generation of the data but I used split cylinder
so I could do updates by using the top/bottom of each cylinder for new/old
data during update.  We only had 2 drives.  The operator could mount the
next pack in sequence while updates where running on the other drive.

> The 1410 was a bigger 1401 which used (I think) 5 digit storage
> addresses, but was otherwise rather like a 1401.

Yep.  When I finally worked in a 1410 shop I could not tell the
difference.  The bulk of their code used the 3 character address,
only their largest programs needed the 5 character address.
Of course by that time they didn't actually have a 1410 in house
they were using emulation on a S/360.

-- 
Dan Espen

hanc...@bbs.cpcn.com writes:

> On Friday, March 16, 2018 at 9:18:29 PM UTC-4, Dan Espen wrote:
>> rnet...@gmail.com writes:
>> 
>> > On Friday, March 16, 2018 at 1:45:05 PM UTC-4, han...@bbs.cpcn.com wrote:
>> >> IBM's calculator series were in the 600s, so I would guess that 7 was
>> >> the next available series.
>> >> 
>> >> IBM planned to have the subsequent computers in the 800 or 8000
>> >> series, but this was cancelled and replaced by S/360.
>> >> 
>> >> I don't know where the "14" came from for the 1401 series.
>> >
>> > The 14 came from the fact that the origial model, sold as an
>> > "elecronically programmable tabulator", had 1400 postions of internal
>> > storage.
>> >
>> > Bob Netzlof
>> 
>> Never heard that before but that sounds plausible.
>> 
>> Leaving the mystery of how they chose:
>> 
>> 1401
>> 1410
>> 1460
>> 1440
>> 
>> I've always assumed model numbers and names are chosen by the
>> sales force and their reasoning is something like "that sounds nice".
>
> In addition to the 1410, there was also the 7010.
> https://www-03.ibm.com/ibm/history/exhibits/dpd50/dpd50_chronology2.html
>
> I don't have my IBM Early Computers history handy, but I believe
> it said the 14xx series was actually an outgrowth of several
> different development efforts, for example, a super-tabulator
> for European service and something called the "Wooden Wheel".
> IBM eventually re-oriented those efforts into a common 
> stored-program computer product line.
>
> The book talks a little about the development history of the
> other models.  I think at least one model, and maybe several,
> were actually intended to be something else, and IBM had them
> change their architecture to 14xx compatible. 
>
> As mentioned, the 1410 was "sort of" 1401 compatible, while the
> 1460 was compatible.  I'm not sure of the differences of the 1410
> and 1460--that is, why a customer would buy one over the other--
> but I suspect that came out of the differing background.  The
> unity of S/360 was supposed to eliminate that kind of confusion.

As mentioned elsewhere, the 1410 could use 5 characters for addressing
supporting a lot more storage.

> Some of the other 14xx models were higher end, but one was a
> low-end.  I think the low-end model didn't sell too well because
> it was too costly for the low performance.
>
> As we know, Haanstra was not a supporter of S/360 and actually
> quietly built a 'super' 1401 using new components.  IBM was
> not pleased by this effort and he was punished for it.
>
> As Campbell-Kelly explains in their book, the beauty of the 
> IBM 1401 was not in its electronics (not a big deal) but rather
> the 1403 printer, which gave high quality at a significantly
> faster rate.  As mentioned, I remain amazed that for _years_,
> customers tolerated wavy lines of other mainframe line printers,
> even slow ones.

Just to defend the slower IBM 1443 common on the 1440, it produced
pretty high quality output.

-- 
Dan Espen

On Saturday, March 17, 2018 at 8:15:27 PM UTC-4, Dan Espen wrote:
> If I ever read that the bits in my brain must have flipped.
> I only encountered a 1400 character 1401 after I'd worked on them
> for a few years.  I was surprised they made them that small.

Remember, "electronically programmable tabulator". I distinctly remember reading a sales brochure pushing that view. The idea was that (in computer terms) the thing would be programmed in machine language. Hence A for Add, S for Subtract, M for Move, and so on. Compared to the existing plug board programmable tabulators, 1400 characters of storage for data and program was huge.

> No disk on the 1460 I used, but the 1440 had disk.
> The 1311 disk wasn't exactly huge.  1 meg per drive.

Agreed. But there was a big fixed disk random access unit (I do not remember the model number) which had a total capacity of several megabytes, more than you could get by hanging 1311's on a 1401. While a 1401 could run 1311's, a 1460 was required to operate this big beast.
 
Bob Netzlof

On Saturday, March 17, 2018 at 8:48:56 PM UTC-4, rne...@gmail.com wrote:

> But there was a big fixed disk random access unit (I do not remember the 
> model number) which had a total capacity of several megabytes, more than you > could get by hanging 1311's on a 1401. While a 1401 could run 1311's, a 1460 > was required to operate this big beast.

At least, that is my recollection. My employer moved from a 1401 with several 1311's to a 1460 with a 1405. Could be a need for a faster processor drove the change from 1401 to 1460. I was under the impression, however, that the big disk needed the 1460.

Google is Our Friend. IBM 1405. Available in two models, 10 and 20 million characters of storage.

And harking back to the comments on 1401, 1402, 1403... There was a 1404 printer which also incorporated a card reader. Hence, 1405 was the next available 14xx number. 1406 was a cabinet of core storage (the base machine maxed at 4000 positions, the 1406 added 4k, 8k, or 12k), and 1407 was the typewriter. Don't believe there was a 1408 nor 1409.
 
Bob Netzlof

Dan Espen <dan1...@gmail.com> writes:

> hanc...@bbs.cpcn.com writes:

>> As Campbell-Kelly explains in their book, the beauty of the 
>> IBM 1401 was not in its electronics (not a big deal) but rather
>> the 1403 printer, which gave high quality at a significantly
>> faster rate.  As mentioned, I remain amazed that for _years_,
>> customers tolerated wavy lines of other mainframe line printers,
>> even slow ones.

> Just to defend the slower IBM 1443 common on the 1440, it produced
> pretty high quality output.

I'm going to disagree with this.  I learned to program on a 1401/1402/1403
system as a senior in high school.  When I was exposed to the 1442 attached
to the IBM 1800 at the CAI Laboratory at UTexas the next year, I was appalled
at the print quality.

-- 
Rich Alderson					  ne...@alderson.users.panix.com
      Audendum est, et veritas investiganda; quam etiamsi non assequamur,
	  omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
									--Galen

Rich Alderson <ne...@alderson.users.panix.com> writes:

> Dan Espen <dan1...@gmail.com> writes:
>
>> hanc...@bbs.cpcn.com writes:
>
>>> As Campbell-Kelly explains in their book, the beauty of the 
>>> IBM 1401 was not in its electronics (not a big deal) but rather
>>> the 1403 printer, which gave high quality at a significantly
>>> faster rate.  As mentioned, I remain amazed that for _years_,
>>> customers tolerated wavy lines of other mainframe line printers,
>>> even slow ones.
>
>> Just to defend the slower IBM 1443 common on the 1440, it produced
>> pretty high quality output.
>
> I'm going to disagree with this.  I learned to program on a 1401/1402/1403
> system as a senior in high school.  When I was exposed to the 1442 attached
> to the IBM 1800 at the CAI Laboratory at UTexas the next year, I was appalled
> at the print quality.

Of course you would be unhappy, the 1442 was a card reader punch.
:)
Yeah, I know typo.

Anyway I remember it as pretty good but I don't have any listings from
a 1443 and Google isn't helping.  Vertical alignment should have
been perfect.  Horizontal would be based on timing circuits, possibly
your printer was out of adjustment.  Print density should be similar to
a 1403, the hammer had to hit the paper pretty much the same way.

-- 
Dan Espen

Rich Alderson <ne...@alderson.users.panix.com> wrote:
> Dan Espen <dan1...@gmail.com> writes:
> 
>> hanc...@bbs.cpcn.com writes:
> 
>>> As Campbell-Kelly explains in their book, the beauty of the 
>>> IBM 1401 was not in its electronics (not a big deal) but rather
>>> the 1403 printer, which gave high quality at a significantly
>>> faster rate.  As mentioned, I remain amazed that for _years_,
>>> customers tolerated wavy lines of other mainframe line printers,
>>> even slow ones.
> 
>> Just to defend the slower IBM 1443 common on the 1440, it produced
>> pretty high quality output.
> 
> I'm going to disagree with this.  I learned to program on a 1401/1402/1403
> system as a senior in high school.  When I was exposed to the 1442 attached
> to the IBM 1800 at the CAI Laboratory at UTexas the next year, I was appalled
> at the print quality.
> 

I'm not sorprised, since the 1442 was a card reader/punch ;-)

-- 
Pete

On 2018-03-19, Dan Espen <dan1...@gmail.com> wrote:

> Rich Alderson <ne...@alderson.users.panix.com> writes:
>
>> Dan Espen <dan1...@gmail.com> writes:
>>
>>> hanc...@bbs.cpcn.com writes:
>>
>>>> As Campbell-Kelly explains in their book, the beauty of the 
>>>> IBM 1401 was not in its electronics (not a big deal) but rather
>>>> the 1403 printer, which gave high quality at a significantly
>>>> faster rate.  As mentioned, I remain amazed that for _years_,
>>>> customers tolerated wavy lines of other mainframe line printers,
>>>> even slow ones.
>>>
>>> Just to defend the slower IBM 1443 common on the 1440, it produced
>>> pretty high quality output.
>>
>> I'm going to disagree with this.  I learned to program on a 1401/1402/1403
>> system as a senior in high school.  When I was exposed to the 1442 attached
>> to the IBM 1800 at the CAI Laboratory at UTexas the next year, I was appalled
>> at the print quality.
>
> Of course you would be unhappy, the 1442 was a card reader punch.
> :)
> Yeah, I know typo.
>
> Anyway I remember it as pretty good but I don't have any listings from
> a 1443 and Google isn't helping.  Vertical alignment should have
> been perfect.  Horizontal would be based on timing circuits, possibly
> your printer was out of adjustment.  Print density should be similar to
> a 1403, the hammer had to hit the paper pretty much the same way.

I remember seeing printouts from a 2780 RJE station, whose print mechanism
was (IIRC) basically a 1443.  They were quite crisp and clear.  The printer
was much slower than a 1403, though.

-- 
/~\  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.
/ \  HTML will DEFINITELY be ignored.  Join the ASCII ribbon campaign!

I used a line printer on a Honeywell 516 which was a drum printer, but 
because it was little used, it was in perfect alignment, with better print 
quality than the heavily-used 1403s of the campus mainframe.

In general, of course, slight horizontal errors are less obtrusive 
than slight vertical ones, and that works to the advantage of chain and 
train printers, but also to bar and band printers, over drum printers.

Charlie Gibbs <cgibbs@kltpzyxm.invalid> writes:
> I remember seeing printouts from a 2780 RJE station, whose print mechanism
> was (IIRC) basically a 1443.  They were quite crisp and clear.  The printer
> was much slower than a 1403, though.

1443 had bar that moved back and forth ... 150 lines/min ... this
doesn't actually show 1443 printer
https://en.wikipedia.org/wiki/IBM_1443

univ. 360/67, running as 360/65 with MFT (up through release 14) and
then MVT (starting with release 15/16, combined release). Eventually Was
typing so much on 1052-7 operator's console ... that they got a 1443 for
all operator's console output ... and then selected subset for 1052-7.

wiki page has URL for 1443 document at bitsavers
http://bitsavers.org/pdf/ibm/1620/A26-5730-2_1443_printer_for_1620.pdf

shows 1443 "flying type bar" at computer history museum
http://www.computerhistory.org/collections/catalog/102640497

some past posts
http://www.garlic.com/~lynn/2004.html#42 40th anniversary of IBM System/360 on 7 Apr 2004
http://www.garlic.com/~lynn/2004d.html#44 who were the original fortran installations?
http://www.garlic.com/~lynn/2005s.html#21 MVCIN instruction
http://www.garlic.com/~lynn/2007q.html#34 what does xp do when system is copying
http://www.garlic.com/~lynn/2010h.html#55 IBM 029 service manual
http://www.garlic.com/~lynn/2017.html#38 Paper tape (was Re: Hidden Figures)

-- 
virtualization experience starting Jan1968, online at home since Mar1970

Rich Alderson <ne...@alderson.users.panix.com> writes:

> I'm going to disagree with this.  I learned to program on a 1401/1402/1403
> system as a senior in high school.  When I was exposed to the 1442 attached
> to the IBM 1800 at the CAI Laboratory at UTexas the next year, I was appalled
> at the print quality.

Typo:  That's "1443", not "1442"--although it had one of those, too.

-- 
Rich Alderson					  ne...@alderson.users.panix.com
      Audendum est, et veritas investiganda; quam etiamsi non assequamur,
	  omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
									--Galen

On Saturday, March 17, 2018 at 8:15:27 PM UTC-4, Dan Espen wrote:

> If I ever read that the bits in my brain must have flipped.
> I only encountered a 1400 character 1401 after I'd worked on them
> for a few years.  I was surprised they made them that small.
> The small machine also had no index registers, but by that time
> I'd stop using index registers anyway so no big deal.

The IBM Early Computers history talks about the development
of what eventually became the 1401.  One of the goals was to
develop kind of a 'super' 407 tabulator.  For that, not a
whole memory was required, though I would think only 1,400
characters was awfully low.  The 407 manual is on bitsavers
and it is amazing how much functionality that machine had.

I wish I remembered the exact numbers, but the history said
memory would be very expensive.  So, a lot of customers would
want the lowest amount they could get away with.

The 1401 evolved.  Tape and disk support were added, which
greatly enhanced its value.  I suspect that as IBM ramped
up its core production in that era (early 1960s--they came
up with better ways to make and string cores), the memory
costs went down.

[snip]

On Saturday, March 17, 2018 at 8:48:56 PM UTC-4, rne...@gmail.com wrote:
> On Saturday, March 17, 2018 at 8:15:27 PM UTC-4, Dan Espen wrote:
> > If I ever read that the bits in my brain must have flipped.
> > I only encountered a 1400 character 1401 after I'd worked on them
> > for a few years.  I was surprised they made them that small.
> 
> Remember, "electronically programmable tabulator". I distinctly remember reading a sales brochure pushing that view. The idea was that (in computer terms) the thing would be programmed in machine language. Hence A for Add, S for Subtract, M for Move, and so on. Compared to the existing plug board programmable tabulators, 1400 characters of storage for data and program was huge.

Part of the IBM's goals was to meet European needs for a more
sophisticated tabulator, due to tougher banking demands in Europe.

The IBM history also notes the development of RPG for the 1401,
the book says it made it possible for former plugboard operators
to easily program the computer, more easily than in Autocoder or
SPS.  (RPG needed more memory).  As we know, RPG in fancier
versions became very popular on S/3 machines and their
descendants.  (I don't know about today and the "i" series.)





> > No disk on the 1460 I used, but the 1440 had disk.
> > The 1311 disk wasn't exactly huge.  1 meg per drive.
> 
> Agreed. But there was a big fixed disk random access unit (I do not remember the model number) which had a total capacity of several megabytes, more than you could get by hanging 1311's on a 1401. While a 1401 could run 1311's, a 1460 was required to operate this big beast.

Ironically, I think the book said the 1440 got disk before the
1401 did for marketing reasons.  I it said the 1440 turned out to
be a poor seller due to poor price/performance.

I believe the 14xx series eventually could support the big disk
drive (50 meg) developed for RAMAC.

I still wonder when the last 1401 emulator was turned off (or if
any are still running.  I'm guessing 1990s for Y2K, which is
what we did.

hanc...@bbs.cpcn.com writes:
> The IBM Early Computers history talks about the development
> of what eventually became the 1401.  One of the goals was to
> develop kind of a 'super' 407 tabulator.  For that, not a
> whole memory was required, though I would think only 1,400
> characters was awfully low.  The 407 manual is on bitsavers
> and it is amazing how much functionality that machine had.
>
> I wish I remembered the exact numbers, but the history said
> memory would be very expensive.  So, a lot of customers would
> want the lowest amount they could get away with.
>
> The 1401 evolved.  Tape and disk support were added, which
> greatly enhanced its value.  I suspect that as IBM ramped
> up its core production in that era (early 1960s--they came
> up with better ways to make and string cores), the memory
> costs went down.

somebody at univ. wrote a 709 cobol program that emulated 407 "plug
board" application ... which was eventually coverted to 360 cobol. The
cobol program at the end would even print out the 407 sense switch
values. One day, the 360 program printed out some values that had never
been seen before. everything stopped while operations decided what to
do, they tried contacting the person in the administration responsible
for the program, to see if they knew what to do. after an hour or so
... when the 360 (360/67 running with os/360 in 360/65 mode) sitting
idle, they decided to rerun the program and see if it came out with the
same values. They then sent both outputs on to person in administration.


407 reference
https://www.revolvy.com/main/index.php?s=IBM+407&item_type=topic&nojs=1
has several URL references, ibm vintage
http://www-03.ibm.com/ibm/history/exhibits/vintage/vintage_4506VV4007.html
and bitsavers
http://www.bitsavers.org/pdf/ibm/punchedCard/
http://www.bitsavers.org/pdf/ibm/punchedCard/AccountingMachine/
http://www.bitsavers.org/pdf/ibm/punchedCard/Training/
this has ... wiring plug board
http://www.bitsavers.org/pdf/ibm/punchedCard/Training/A24-1007-1_IBM_Functional_Wiring_Principles_Feb66.pdf


past tales of cobol simulate 407
http://www.garlic.com/~lynn/2002d.html#21 Mainframers: Take back the light (spotlight, that is)
http://www.garlic.com/~lynn/2006b.html#5 IBM 610 workstation computer
http://www.garlic.com/~lynn/2007j.html#78 IBM 360 Model 20 Questions
http://www.garlic.com/~lynn/2009b.html#76 IBM tried to kill VM?
http://www.garlic.com/~lynn/2010f.html#23 history of RPG and other languages, was search engine history
http://www.garlic.com/~lynn/2010o.html#34 origin of 'fields'?
http://www.garlic.com/~lynn/2011b.html#75 If IBM Hadn't Bet the Company
http://www.garlic.com/~lynn/2012i.html#12 IEBPTPCH questions

-- 
virtualization experience starting Jan1968, online at home since Mar1970