Graphics on the IBM 2260?

[j***@ecn.ab.ca]

The IBM 2250, of course, was a vector graphics display terminal that
came in several configurations. One depended on the host computer for
all the work, another had an IBM 1130 as a local processor.

It came with a keyboard of round pushbuttons that could be labelled
with an overlay, thus presaging the original keyboard of the IBM PCjr.
But in addition, it had a sensor (presumably composed of microswitches)
that let it determine which of 16 overlays was in use. This principle
is, of course, familiar today from innumerable computerized toys for
small children.

The University where I went couldn't afford one of these. They used a
similar terminal from Control Data which used an embedded CDC 120A
computer for processing, called the GRID terminal (presumably for
GRaphics Input Device - it had a light pen). At least, that's what they
called it *there*, but that may not have been the name under which CDC
sold the device, since it appears as if the sole mention of this
terminal on the whole Web is the abstract of a paper authored by a
professor at the University of Alberta.

But the IBM 2260 was the little video terminal used for text purposes
with IBM 360 computers before the IBM 3270 came out. It was basically
an EBCDIC dumb terminal.

IBM 2260 terminals in combination with a slide projector formed the
specialized terminal used with the IBM 1500 educational system.

I recalled seeing photographs of graphics not on the projector screen,
but on the 2260 screen itself.

In a web search on this topic, I've found that these graphics
apparently were achieved by means of a programmable character set.
Another reference mentions the IBM 2260/65 as the graphics-capable
member of the 2260 family.

Incidentally, the 3270 could not have been used for graphics if it had
a programmable character set. Having one day briefly turned up the
brightness on one (not one in actual service, but one being sold as
government surplus, IIRC - those babies are apparently "half-duplex"
enough that you can type on the keyboard and see what you've typed on
the screen of one that isn't connected to anything), so that I could
see the raster, I discovered that instead of having blank scan lines
between lines of characters, it increased the rate of vertical sweep,
to allow generous space between lines (otherwise, characters in
80-column displays tended to be tall and thin) and yet keep the number
of scan lines to a minimum (hence minimizing flicker at a given level
of technology).

For those who are completely unfamiliar with the IBM 2260, it featured
prominently in the recent movie version of _1984_ which starred William
Hurt. (An earlier version was made in the 1950s.) Although they turned
some of the ones they used into TV monitors, IIRC, having graphic
display capabilities no real 2260 ever had.

John Savard

[Howard S Shubs]

Post by j***@ecn.ab.ca
The IBM 2250, of course, was a vector graphics display terminal that
came in several configurations. One depended on the host computer for
all the work, another had an IBM 1130 as a local processor.

Yes, as I documented
<http://www.ibm1130.net/functional/DisplayUnit.html>.

Post by j***@ecn.ab.ca
For those who are completely unfamiliar with the IBM 2260, it featured
prominently in the recent movie version of _1984_ which starred William
Hurt. (An earlier version was made in the 1950s.) Although they turned
some of the ones they used into TV monitors, IIRC, having graphic
display capabilities no real 2260 ever had.

I found <http://www.columbia.edu/acis/history/2260.html>, which doesn't
disagree with what you've said.

[j***@ecn.ab.ca]

Post by Howard S Shubs
I found <http://www.columbia.edu/acis/history/2260.html>, which doesn't
disagree with what you've said.

Although I had run across that page before, I didn't read the text on
it carefully; this answers several of my questions.

I remember once at University someone trying to explain to me how a
3270 worked, and I was doubtful... it turns out that his explanation
did not apply to the 3270, but it did apply to the 2260 - it gets a
video signal from a central controller.

And that site also explains the limitation of the graphics capabilities
of the 2260. The character set is programmable, but at the *controller*
level; this is fine if a project requires a special character set for a
whole classroom of students, but it does not permit an individual,
running his own application at one terminal, to use a special
programmed character set for that application.

John Savard

[Eric Smith]

Post by j***@ecn.ab.ca
The IBM 2250, of course, was a vector graphics display terminal that
came in several configurations. One depended on the host computer for
all the work, another had an IBM 1130 as a local processor.

In addition to the 2250 Model 4 which attaches to an 1130, there were at
least three different models for use on the System/360, and recently I
saw a reference to one being built into the 360/9x.

As I understand it (from only a very cursory review of the
documentation), the 2250 Model 1 had an integral control unit, which
does not appear to be an 1130. The Model 2 and Model 3 used the 2840
Display Control Model 1 and 2, respectively. The 2840 could support two
2250s (optionally expandable to four).

The 2250 Model 4 was used with the 1130. The manual primarily describes
operation of the 2250 on the 1130 locally. Brief mention is given to
the idea of using the combination as a remote terminal for a 360, but no
particular details of such an arrangement are provided.

Eric

[Anne & Lynn Wheeler]

Post by Eric Smith
In addition to the 2250 Model 4 which attaches to an 1130, there were at
least three different models for use on the System/360, and recently I
saw a reference to one being built into the 360/9x.
As I understand it (from only a very cursory review of the
documentation), the 2250 Model 1 had an integral control unit, which
does not appear to be an 1130. The Model 2 and Model 3 used the 2840
Display Control Model 1 and 2, respectively. The 2840 could support two
2250s (optionally expandable to four).
The 2250 Model 4 was used with the 1130. The manual primarily describes
operation of the 2250 on the 1130 locally. Brief mention is given to
the idea of using the combination as a remote terminal for a 360, but no
particular details of such an arrangement are provided.

the university i was at had a 2250m1 (direct channel attach)
... lincoln labs had done a cms fortran graphics labrary for the
2250m1. i took the lincoln labs library and integrated it into cms
edit ... to create an early full-screen editor.

the science center had a 2250m4 (i.e. with 1130). somebody at the
science center had ported space wars to the 1130 and you could play
two-person space war on the thing .... bascially the keyboard was
divided in half (with each plaayer getting their half of the keyboard)
which had the various controls.

[Joe Morris]

Post by j***@ecn.ab.ca
But the IBM 2260 was the little video terminal used for text purposes
with IBM 360 computers before the IBM 3270 came out. It was basically
an EBCDIC dumb terminal.

Not-quite-a-nitpick: the 2260 display terminal knew nothing
about character sets. It was essentially a picture tube with an
attached keyboard; the display showed the (analog) image sent to
it over a coax line from the controller.

The 2260s my PPOE used were attached to the S/360 channel via a
2848 control unit, which (surprise!) spoke EBCDIC.

In the 2848, the data for each attached terminal was contained in
a dedicated mechanical (torsion) delay line. The data was converted
into bit patterns for the video display through a matrix of magnetic
cores (you could eyeball many of the letterforms in the matrix; cores
were located where illuminated phosphor dots were needed and missing
where they weren't wanted).

Post by j***@ecn.ab.ca
IBM 2260 terminals in combination with a slide projector formed the
specialized terminal used with the IBM 1500 educational system.

I don't know the 1500; was it by any chance based on the 14xx
architecture? If it's a pre-S/360 design then it would be unlikely
to be using EBCDIC.

Post by j***@ecn.ab.ca
In a web search on this topic, I've found that these graphics
apparently were achieved by means of a programmable character set.
Another reference mentions the IBM 2260/65 as the graphics-capable
member of the 2260 family.

H'mmm...At least for the 2260 in my shop I have no recollection of
having heard of a programmable character set. At the same time, we
used it as a HASP console (this was on a very memory-constrained
MVT system where we couldn't dedicate the memory required to use MCS)
so there wouldn't have been any reason for us to have investigated
the availability of a programmable character set.

Joe Morris

[CBFalconer]

Joe Morris wrote:
... snip ...

Post by Joe Morris
In the 2848, the data for each attached terminal was contained in
a dedicated mechanical (torsion) delay line. The data was converted
into bit patterns for the video display through a matrix of magnetic
cores (you could eyeball many of the letterforms in the matrix; cores
were located where illuminated phosphor dots were needed and missing
where they weren't wanted).

How old was this? Sounds like using rope memory for the ROM char.
generator, and preceding the invention of PMOS shift registers.
i.e. the whole thing should be earlier than about 1970, maybe early
'60s.

[Tim Shoppa]

Post by CBFalconer
How old was this? Sounds like using rope memory
for the ROM char. generator [...]

Just to jump in on an even older technology, my favorite terminal/CRT
character generators is the mask character generator, with separate
steering for gun to mask and then mask to CRT. The result looks
remarkably typewriter-like, especially as small misalignments of
characters are repeated every time that character is used.

Did IBM use this technology in any of its terminals? I think I saw it
in a RCA terminal hooked up to a Spectra.

Tim.

[Eric Smith]

Post by Tim Shoppa
Just to jump in on an even older technology, my favorite terminal/CRT
character generators is the mask character generator, with separate
steering for gun to mask and then mask to CRT. The result looks
remarkably typewriter-like, especially as small misalignments of
characters are repeated every time that character is used.

That was the RCA "Charactron" tube.

[Joe Morris]

Post by CBFalconer

Post by Joe Morris
In the 2848, the data for each attached terminal was contained in
a dedicated mechanical (torsion) delay line.

How old was this? Sounds like using rope memory for the ROM char.
generator, and preceding the invention of PMOS shift registers.
i.e. the whole thing should be earlier than about 1970, maybe early
'60s.

Almost. IIRC we replaced the 2848/2260 displays with new 327x
displays and controllers in ~1973; certainly before 1975 because
we moved into a new building in '75 and I remember fighting a major
channel interface bug in the 3272 controller in the old building.
We had the 2260s for two or three years, so they would have been
initially installed in '70 or '71.

Of course, these dates say nothing about the age of the technology,
but as a state university we couldn't always go with the latest toys
that IBM had in its sales manual.

Joe Morris

[Rich Alderson]

Post by Joe Morris
I don't know the 1500; was it by any chance based on the 14xx
architecture? If it's a pre-S/360 design then it would be unlikely
to be using EBCDIC.

It's an 1800 (the process-control relative of the 1130); the one at the CAI Lab
at the University of Texas had a 1442 and a 1443 attached to it, as well as the
1500-related terminals. Provided Coursewriter II services.

The original Coursewriter software *did* run on a 14xx of some kind.

[j***@ecn.ab.ca]

Post by Rich Alderson

Post by Joe Morris
I don't know the 1500;

It's an 1800 (the process-control relative of the 1130); the one at the CAI Lab
at the University of Texas had a 1442 and a 1443 attached to it, as well as the
1500-related terminals. Provided Coursewriter II services.

I see on the web that some 1500 systems were built around 1800
computers; the 1800 ran faster than the 1130 as well as having a few
extra features. But 1500 systems were also sold with an 1130 as their
central processor; the one at the University of Alberta was of this
kind.

I suppose it depended on how many students one wished to serve.

John Savard

[Joe Morris]

Post by Rich Alderson
The original Coursewriter software *did* run on a 14xx of some kind.

Not only that, but the port of Coursewriter to the S/360 didn't change
the data, which was kept in 1401 format internally. The communications
interface to the BTAM r/w modules translated it between 1401 and line
code, making dump reading an interesting exercise.

Joe Morris

[j***@ecn.ab.ca]

Post by Joe Morris
I don't know the 1500;

The IBM 1500 was an educational system which included an IBM 1130
computer and specialized peripherals for computer-assisted instruction.

APL/1500 was interesting in that it had a domino function used to
control the multimedia hardware, as well as providing some of the
functions handled by system variables in later forms of APL.

I see from the APL/1130 manual on Al Kossow's site that that version of
APL did not do that; apparently, the operator console on an 1130 was in
the keypunch-style arrangement instead of a typewriter-style
arrangement, so a very unusual keyboard arrangement was provided for
using APL from the operator console.

Although unavoidably quite unusual, it was a very well thought-out
response to an unfortunate situation.

John Savard

[David Wade]

Post by Joe Morris

Post by j***@ecn.ab.ca
But the IBM 2260 was the little video terminal used for text purposes
with IBM 360 computers before the IBM 3270 came out. It was basically
an EBCDIC dumb terminal.

Not-quite-a-nitpick: the 2260 display terminal knew nothing
about character sets. It was essentially a picture tube with an
attached keyboard; the display showed the (analog) image sent to
it over a coax line from the controller.
The 2260s my PPOE used were attached to the S/360 channel via a
2848 control unit, which (surprise!) spoke EBCDIC.

I remeber using these (well I assume they were 2260) attached to the 360/67
at newcastle (england)

Post by Joe Morris
In the 2848, the data for each attached terminal was contained in
a dedicated mechanical (torsion) delay line. The data was converted
into bit patterns for the video display through a matrix of magnetic
cores (you could eyeball many of the letterforms in the matrix; cores
were located where illuminated phosphor dots were needed and missing
where they weren't wanted).

Post by j***@ecn.ab.ca
IBM 2260 terminals in combination with a slide projector formed the
specialized terminal used with the IBM 1500 educational system.

I don't know the 1500; was it by any chance based on the 14xx
architecture? If it's a pre-S/360 design then it would be unlikely
to be using EBCDIC.

Post by j***@ecn.ab.ca
In a web search on this topic, I've found that these graphics
apparently were achieved by means of a programmable character set.
Another reference mentions the IBM 2260/65 as the graphics-capable
member of the 2260 family.

H'mmm...At least for the 2260 in my shop I have no recollection of
having heard of a programmable character set. At the same time, we
used it as a HASP console (this was on a very memory-constrained
MVT system where we couldn't dedicate the memory required to use MCS)
so there wouldn't have been any reason for us to have investigated
the availability of a programmable character set.

I do recall the controller was "clever". I seem to recall that by typing
some odd command (the % character rings a bell) you could scroll back and
forwards in the terminal buffer...

Post by Joe Morris
Joe Morris

[Rich Alderson]

Post by j***@ecn.ab.ca
IBM 2260 terminals in combination with a slide projector formed the
specialized terminal used with the IBM 1500 educational system.
I recalled seeing photographs of graphics not on the projector screen,
but on the 2260 screen itself.

As I recall, the terminals on the 1500 system at the UT School of Education
CAI Lab had a 15xx number, but they might have been 2260-based for all that.

They had some kind of programmable character sets, because one of the courses
using the system was on Russian perfective vs. imperfective verbs, and used
Cyrillic on the screen.