What is this 8-knobbed IBM human interface device in a 1980's Boeing photo?

[rs20]

What is the 8-knob device that's being manipulated by this fellow's left hand?:



(also can be seen here in case the image link breaks)

It's clearly IBM-branded to some extent, and my best guess is that it's a bank of 8 encoders that could be used for whatever, but used here to control the 6 degrees of freedom in a CAD system. But any specific details on what the device is called or anything else about it would be really interesting!

[Halcyon]

Nipple array?

[RAPo]

A controller for 3D Cad design, see here

[westfw]

Just dials.  Vector graphics displays were common in those days, and frequently had various attached dials for doing data input, rotating the screen image about various axes, zoom, etc (all of which a vector display could do in hardware.)
You can see similar dials in use in the famous (?)  Larry Cuba Star Wars Computer Graphics documentary:

[Roehrenonkel]

Hi rs20,
 
HP used these as well, called it just "Knob-box". :-))
Was used to adjust settings or scroll through menus.

Ciao4now

[Alex Eisenhut]

https://www.ardent-tool.com/6094/Dials.html

[Calambres]

Was used with CATIA

[janoc]

Here is a variant sold by SGI:

https://pbs.twimg.com/media/Fiwen67XgAA-yjK?format=jpg&name=large

http://bukosek.si/hardware/collection/sgi-dials-and-buttons.html

These serial button boxes (gamers have reinvented these as "macropads" these days) and dial pads were very common with Unix CAD workstations from major manufacturers until the late 90s.

Coincidentally, if you have ever seen the SGI Inventor UI/viewer or later the early VRML viewers, they had these "wheels" to control camera:
https://web.cs.wpi.edu/~matt/courses/cs563/talks/inv_images/roller.gif

These could look & feel strange if you had to use a mouse (this predates the concept of "mouselook" in games and scroll wheel mice!) but the idea was that you have turned those dials and they were mapped directly to the motion of these wheels and thus camera rotation around the principal axes. So the wheel UI metaphor comes from there.

And the last piece of CAD puzzle was the Spaceball where you had the 6 degrees of freedom (3x rotation, 3x translation) in a single controller instead of 6-8 dials:

http://spacemice.org/gallery/Spaceball_1003.jpg

This was originally invented by the German company Magellan in the late 1980s, then sold to SGI, licensed and rebranded multiple times (e.g. by HP, IBM and other workstation manufacturers) until it ended up with Logitech and 3DConnexion where the same concept using pretty much the same protocol and almost identical mechanism (just over USB instead of RS232) is still used in their SpaceMice today.

The dials were more versatile, though - both the dials and the buttons could be dynamically remapped to different functions, whenever some analog value needed to be smoothly and finely adjusted, not only to control the camera on the screen. SpaceBall/Mouse are difficult to use for other things and they don't allow as precise control.

[coppice]

That equipment is from a different time, and different way of thinking. To modern eyes the device on the right is a mouse tablet. When it was made it was very much a digitizing tablet, primarily to get designs from paper into the machines. Hence, the magnifier with cross hairs. Those tablets were expected to have extreme linearity in both the X and Y directions, and a precise step count per centimetre. These days such precision is rarely necessary. Any ability to use the tablet as a pointer device was secondary, and often strongly discouraged.

[janoc]

That equipment is from a different time, and different way of thinking. To modern eyes the device on the right is a mouse tablet. When it was made it was very much a digitizing tablet, primarily to get designs from paper into the machines. Hence, the magnifier with cross hairs. Those tablets were expected to have extreme linearity in both the X and Y directions, and a precise step count per centimetre. These days such precision is rarely necessary. Any ability to use the tablet as a pointer device was secondary, and often strongly discouraged.

Actually those tablets often came with different "mice"/pointing devices. Wacom used to make these back in the day too. We had stylus/pen for it, the digitizer with the crosshair and also a regular mouse which didn't have a ball but an inductive sensor (optical mice weren't invented/common yet).

The only difference between the mouse and the digitizer was that the reference point was under the crosshair and not at the top of the mouse (about 2cm or so offset). Our digitizer also didn't have a magnifier, it was just a mouse-like body with a bit of plexiglass attached in front with a scribed crosshair in the middle. As you said - the idea was to digitize paper drawings with that - it was less fiddly and more accurate than using the pen for it.

BTW, those digitizing tablets are still made and sold, mostly for working with paper maps or architectural drawings.

[soldar]

The modern version of the Roman dodecahedronm which date from the 2nd to 4th centuries AD and their purpose remains unknown

https://en.wikipedia.org/wiki/Roman_dodecahedron

[Alex Eisenhut]

I don't know if youtube checks google searches and it somehow knew to queue up this in my recommendations ...
https://youtu.be/J_Q-wjRxAYs?t=589

[tszaboo]

It's the knobs to reduce the structural integrity of your plane, until it falls apart on it's own.

[coppice]

That equipment is from a different time, and different way of thinking. To modern eyes the device on the right is a mouse tablet. When it was made it was very much a digitizing tablet, primarily to get designs from paper into the machines. Hence, the magnifier with cross hairs. Those tablets were expected to have extreme linearity in both the X and Y directions, and a precise step count per centimetre. These days such precision is rarely necessary. Any ability to use the tablet as a pointer device was secondary, and often strongly discouraged.

Actually those tablets often came with different "mice"/pointing devices.

The king of the digitising tablets, Summagraphics, actually became an early king of actual mice. Sun and other workstation makers used their mice. They required a special pad with matrix of lines that was scanned optically by the mouse. Much cruder than today's correlation based optical mouse, but great for their time.

Wacom used to make these back in the day too. We had stylus/pen for it, the digitizer with the crosshair and also a regular mouse which didn't have a ball but an inductive sensor (optical mice weren't invented/common yet).

Wacom came late in the day, and was always focussed on its pen input, with mice and other gadgets being a secondary focus. They never really offered the high resolution of a Summagraphics, but they lead offer wireless handheld devices. That was their big innovation.

The only difference between the mouse and the digitizer was that the reference point was under the crosshair and not at the top of the mouse (about 2cm or so offset). Our digitizer also didn't have a magnifier, it was just a mouse-like body with a bit of plexiglass attached in front with a scribed crosshair in the middle. As you said - the idea was to digitize paper drawings with that - it was less fiddly and more accurate than using the pen for it.

That reflect that digitizing was a secondary goal for Wacom. They made a weak input device compared to a Summagraphics tablet, with its easy to precision locate crosshairs.

Strangely Summa still exists, but if you go to the "about" page on their web site you'd never guess that they were once mostly known for their digitizers and mice.

[tooki]

Wacom used to make these back in the day too. We had stylus/pen for it, the digitizer with the crosshair and also a regular mouse which didn't have a ball but an inductive sensor (optical mice weren't invented/common yet).

Wacom came late in the day, and was always focussed on its pen input, with mice and other gadgets being a secondary focus. They never really offered the high resolution of a Summagraphics, but they lead offer wireless handheld devices. That was their big innovation.

How do you figure? The Wacom tablet I bought in the mid-90s had the same 2540lpi resolution of the SummaGraphics, and as far as I know that was in no way new in that generation of Wacom tablets. The manual specifies 0.15mm accuracy with the puck, 0.25mm with the pen.

[abeyer]

The king of the digitising tablets, Summagraphics, actually became an early king of actual mice. Sun and other workstation makers used their mice. They required a special pad with matrix of lines that was scanned optically by the mouse. Much cruder than today's correlation based optical mouse, but great for their time.

I'd put a huge asterisk after that "great"... they were pretty advanced technologically, but imho were actually a significantly worse experience to use compared to simpler but well built electromechanical ones.

[coppice]

The king of the digitising tablets, Summagraphics, actually became an early king of actual mice. Sun and other workstation makers used their mice. They required a special pad with matrix of lines that was scanned optically by the mouse. Much cruder than today's correlation based optical mouse, but great for their time.

I'd put a huge asterisk after that "great"... they were pretty advanced technologically, but imho were actually a significantly worse experience to use compared to simpler but well built electromechanical ones.

I did have problems with some room lighting. I think reflections from the shiny pad confused the mouse if the angles were just right. When I avoided that situation I was very happy with the Summagraphics mice. I started using them because I was fed up with all the skipping I suffered with mechanical mice.

[coppice]

Wacom used to make these back in the day too. We had stylus/pen for it, the digitizer with the crosshair and also a regular mouse which didn't have a ball but an inductive sensor (optical mice weren't invented/common yet).

Wacom came late in the day, and was always focussed on its pen input, with mice and other gadgets being a secondary focus. They never really offered the high resolution of a Summagraphics, but they lead offer wireless handheld devices. That was their big innovation.

How do you figure? The Wacom tablet I bought in the mid-90s had the same 2540lpi resolution of the SummaGraphics, and as far as I know that was in no way new in that generation of Wacom tablets. The manual specifies 0.15mm accuracy with the puck, 0.25mm with the pen.

It might have been just the lack of a puck with fine crosshairs and a lens for the Wacom tablets, but I found them less effective for digitizing. Now you mention it, the actual steps per centimetre must have been competitive, so achieve their smooth fine pen drawing performance.

[AG6QR]

That's an IBM 5080 graphics terminal.  Here's some documentation on it, with that array of dials illustrated on the cover page:

https://archive.org/details/bitsavers_ibm5080GA2csSystemsPrinciplesofOperationMar1984_30030908/mode/2up

It was introduced in approximately 1984 if I recall correctly.  I remember where I was working when I first saw it, and I only worked there from '83 to '86. 

Interestingly, the 5080 was the generic name for the entire system, consisting of various parts, all with part numbers of the form 508x.  IBM recycled part numbers.  The display unit, consisting of the CRT screen, was given the number 5081.  That is also the part number of the IBM 80 column Hollerith Punch Card, used for programming in FORTRAN and COBOL in the 1960s and '70s.

I used the 5080 system from shortly after its introduction, when I worked at Pratt & Whitney Aircraft, makers of jet engines.  We used it as a display unit for ANVIL-4000, an early CAD/CAM system, which we ran on an IBM 3090 mainframe under VM/CMS.  The knobs were used to rotate 3-d CAD models about the X, Y, and Z axes, as well as translate along those axes, and also to zoom the display in and out.  But 3 axes of rotation, plus three axes of translation, plus zoom, only accounts for the use of 7 out of those 8 knobs.  I don't recall what other function might have been assigned to the last knob.

As I recall, the Evans & Sutherland Picture System 2, an early vector graphics system used for CAD/CAM, had a very similar array of knobs, and that predated the IBM 5080.  But the E&S knobs, while housed in a very similar array, had more conventional cylindrical shapes, less reminiscent of a cow's udder.

[AG6QR]

That equipment is from a different time, and different way of thinking. To modern eyes the device on the right is a mouse tablet. When it was made it was very much a digitizing tablet, primarily to get designs from paper into the machines. Hence, the magnifier with cross hairs. Those tablets were expected to have extreme linearity in both the X and Y directions, and a precise step count per centimetre. These days such precision is rarely necessary. Any ability to use the tablet as a pointer device was secondary, and often strongly discouraged.

We used those digitizing tablets on CAD/CAM systems in the way you describe, for accurately digitizing paper drawings, as well as for guiding a pointer around the screen for general purposes, like selecting entities on the screen.

But there was another common use case.  CAD/CAM systems of the day had various ways for the user to issue commands to do different operations, by typing commands on a keyboard, or by going through menu systems.  There were many specialized and esoteric operations available, and a particular operator would often have a favorite subset of the commands that were very frequently used.  As an alternative to using the keyboard to type in commands, you could produce a grid of commands, and make a plot of it on paper, then tape that paper onto the digitizing tablet, and then use the puck to select your desired command by pointing at it on the paper.  The software in ComputerVision and ANVIL-4000 (two CAD/CAM systems I was most familiar with) made it reasonably easy to create custom templates of commands and macros this way. Some operators made their own, or groups of operators would share them.  Putting your most used commands in big boxes near the center, with less commonly used commands in smaller boxes around the edges, was a common strategy to improve productivity.

[tooki]

Wacom used to make these back in the day too. We had stylus/pen for it, the digitizer with the crosshair and also a regular mouse which didn't have a ball but an inductive sensor (optical mice weren't invented/common yet).

Wacom came late in the day, and was always focussed on its pen input, with mice and other gadgets being a secondary focus. They never really offered the high resolution of a Summagraphics, but they lead offer wireless handheld devices. That was their big innovation.

How do you figure? The Wacom tablet I bought in the mid-90s had the same 2540lpi resolution of the SummaGraphics, and as far as I know that was in no way new in that generation of Wacom tablets. The manual specifies 0.15mm accuracy with the puck, 0.25mm with the pen.

It might have been just the lack of a puck with fine crosshairs and a lens for the Wacom tablets, but I found them less effective for digitizing. Now you mention it, the actual steps per centimetre must have been competitive, so achieve their smooth fine pen drawing performance.

Can you be more specific? Are you saying that Wacom’s lens and crosshairs were less effective in some way than SummaGraphics’? Or that they didn’t offer a puck with lens and crosshairs at all? (Because they most certainly did.)

[coppice]

Wacom used to make these back in the day too. We had stylus/pen for it, the digitizer with the crosshair and also a regular mouse which didn't have a ball but an inductive sensor (optical mice weren't invented/common yet).

Wacom came late in the day, and was always focussed on its pen input, with mice and other gadgets being a secondary focus. They never really offered the high resolution of a Summagraphics, but they lead offer wireless handheld devices. That was their big innovation.

How do you figure? The Wacom tablet I bought in the mid-90s had the same 2540lpi resolution of the SummaGraphics, and as far as I know that was in no way new in that generation of Wacom tablets. The manual specifies 0.15mm accuracy with the puck, 0.25mm with the pen.

It might have been just the lack of a puck with fine crosshairs and a lens for the Wacom tablets, but I found them less effective for digitizing. Now you mention it, the actual steps per centimetre must have been competitive, so achieve their smooth fine pen drawing performance.

Can you be more specific? Are you saying that Wacom’s lens and crosshairs were less effective in some way than SummaGraphics’? Or that they didn’t offer a puck with lens and crosshairs at all? (Because they most certainly did.)

A Summagraphics tablet had a magnifying lens with a a hair line cross on it. I had a Wacom puck with a clear piece of plastic with a fairly thick black cross on it. I never saw an option like the Summagraphics one. Wacom never made anything big enough to be a serious digitizing tablet, anyway.

[tooki]

18x25” active area doesn’t qualify? (Honest question, not being snarky.) Dunno if that’s the biggest they ever made, just the largest version of the tablet I had back then.

[coppice]

18x25” active area doesn’t qualify? (Honest question, not being snarky.) Dunno if that’s the biggest they ever made, just the largest version of the tablet I had back then.

That's entry level for a serious technical digitizer. If you don't have a product that's at least A1 sized, you aren't really trying for the engineering, map, and other technical digitizer markets. A1 was the minimum size for any serious engineering drawing before we all went electronic. Wacom was more about allowing artists to digitize their drawings. High resolution matters there, but high dimensional accuracy isn't usually a strong requirement.

[D. Head]

I have found a video about this:


See at 09:09