NOSTALGIA: Remember when flatscreen monitors appeared ?

I was just remembering when the first LCD monitors appeared around 1990 (and a few flatscreen CRT monitors during the transition)...

Do you remember how it took a few days to get comfortable with the flat vs curved display surface, and the true corner to corner image ?

Now nobody thinks twice about it !
Time flies.

CRT raster scan monitors had quite a short lifetime. They replaced green storage tube displays which were used with low ambient light. They had the big advantage of being able to erase and draw lines without refreshing the entire display, they could be viewed in normal lighting, and were cheaper. It was not long though before LCD displays came along.

ardly:
CRT raster scan monitors had quite a short lifetime.

Perhaps for storage screens, but the humble CRT was (is) around for over 50 years ?

lastchancename:
Perhaps for storage screens, but the humble CRT was (is) around for over 50 years ?

CRTs had been around for a long time but to display computer graphics people used storage tube displays. I am not sure why but probably there were technical reasons why raster scan CRTs at the time were not suitable, perhaps they could not display flicker free images at the required resolution. As a result technology deviated down the side track of using storage tubes.

Then CRT technology improved. Being cheaper, because of mass production for TVs, and being able to erase lines (which storagage tubes cannot do), CRTs came to the fore in computer graphics displacing storage tubes. It was short lived though, only a few years, and then CRT displays were replaced by LED displays.

It is probably like transistors and valves. Transistors came first but for practical reasons vacuum tubes became dominant. Once transistor technology improved they rightly supplanted vacuum tubes. So far though nothing has replaced electronic transistors though in future perhaps some kind of optical technology might.

CRTs had been around for a long time but to display computer graphics people used storage tube displays.

Nonsense.

Most of the CRTs used on computers are normal CRTs, and early (70s) computers used the same video monitors as video studios or converted TVs, or TVs with RF modulators (This was also pre-videogame and pre VCR, so the average TV did not have a direct video input.)

A fair number of the actual "computer monitors" had high-persistence phosphors, enabling them to use higher resolutions at the same sort of horizontal scan frequencies and dot-rates that were implementable, but that is not at all the same thing as a "storage tube."

There were also "vector" displays that permitted the electron beam to be moved by separate X and Y values rather than the usual raster-based display. These were good for graphics (and also usually mixed with high-persistence phosphors.)

The only "storage tube" terminal I'm familiar with were the Tektronics 40xx graphics terminals, which were early high-res graphics terminals (also vector-based, IIIRC)

(Hmm. There were also the PLATO displays, which were dot-matrix plasma displays. ~1972 (!))

Monitors were horribly expensive, even compared with other computer components, for quite a long time. I think the original IBM PC Monitor/MGA card combo (a high-persistence phosphor) was close to $1000, as was a 14" color CRT in the Mac-IIci timeframe (1990) The advent of cheap PC clones and the "Hercules Graphics Adapter" in the mid-1980s resulted in major improvements in price to small monochrome monitors, but it was quite a while before color monitors came down in price.

I've actually used numerous ancient computer "terminals" ranging from Hazeltine-2000s to ADM-3s to (real) VT-100s, plus a bunch of others, including a kit-built terminal that displayed on a 9inch surplus Airline reservation video monitor. Also A Vector General 4096x4096 vector display, IBM MGA and CGA adapters with genuine IBM monitors. Assorted display boards for the S100 bus, an Alto, assorted early SUN workstations, and a SAIL "Datadisk" terminal. Also a Tek 4010 graphics display with a storage tube - the ONLY ONE of the bunch with a storage tube... All in the ~1975-current timeframe.

Flat screens are great. I bought my first (14"?) when they dropped down to about $300, and they were worth every penny for the space they saved. The 20inch color CRT at work got replaced with a 24inch dell flatscreen, and I got back an enormous amount of desk space, and no long needed help to move the monitor (exaggeration, but not much of one!)

westfw:
Nonsense.

....
The only "storage tube" terminal I'm familiar with were the Tektronics 40xx graphics terminals, which were early high-res graphics terminals (also vector-based, IIIRC)
.....

For text based applications CRTs were used but for CAD/CAM graphics vector based high-res storage tube terminals such as the expensive Tektronics range were used.

Probably one of the reasons things went down the storage tube route was that the display itself stored the image without the need for external memory. The software could go through the steps of drawing the vectors but had no need to store a representation of the drawing in memory. A drawback of storage tubes was that once you drew a vector you could not erase it, the whole screen had to be cleared.

As RAM increased in size and dropped in price high resolution raster scan CRTS came to the fore in CAD/CAM, initially in monochrome and then colour. They had the big advantage of allowing individual lines to be erased. Many of these CRT based terminals implemented the Tektronics protocols so that they could seamlessly replace existing storage tube displays.

X... please!

lastchancename:
X... please!

I am assuming you are looking for an example. I had not researched this before making earlier posts but looking at Wikipedia now I can see this;

What I had not realised at all until now was that initially Williams Tubes were actually used as main computer memory pre-dating core memory!!!

The jigsaw starts to fall in place. The computer did not have enough memory to store an image, so it drew vectors. However it also did not have enough processing power to do much more than keep refreshing the vectors to keep the image visible. So Tecktronics made storage tube displays which you could send instructions to to draw vectors. The display must have then kept replaying the vector list to keep the image visible thus reducing the burden on the computer processor.

Once memory became cheap graphics cards with high res raster scan CRTs took over. Then LED displays came along.

Here is another link confirming memory was the driver;

Even well into the second half of the 1980s, the dual-port RAM for a 24 bit low resolution (512x512 pixel) could occupy three, double height Euro cards.

Vector displays were useful because the display file only needed to record the endpoints of the vectors, and the display generated the two ramps necessary to draw the lines.
(Can remember playing a simple plan+view tank battle game on a PDP-8 and storage display, using the front panel toggle switches as controls)

Sorry @ardly, I was being OCD for TektroniX... not cs !

BTW, in a previous life, I was the SME for these monsters, we thought they were special... expensive, heavy and nasty!
http://www.bitsavers.org/pdf/aed/990003-01A_AED_512_767_Color_Graphics_Terminals_Users_Manual_Apr1983.pdf
(Sorry, I couldn’t upload the image for some reason)

but for CAD/CAM graphics vector based high-res storage tube terminals such as the expensive Tektronics range were used.

So just how much CAD/CAM do you think was done on $5k terminals (Hmm. Actually a lot cheaper than I would have thought) connected to multi-million-dollar mainframes (that mainly did batch processing, charged for by the CPU-second) over slow comm links (our Tek-4013 (big on APL in those days!) was on a 2400bps UART link (which was as fast as the mainframe's front-end would go)? Graphics of any kind was quite rare.

Probably one of the reasons things went down the storage tube route

I'm mostly arguing with the "went down the route" idea. Storage tube displays clearly existed, but they were a TINY NICHE in the display market, and never had anything to do with the whole "personal computer" or "personal workstation" revolution.

You found ONE vendor of Storage tubes. How many vendors for CRT raster displays?

memory was the driver

The Stanford SAIL DataDisk display used a multi-fixed-head hard drive, generating raster video directly off the platter (more or less) and piping video all over the building (~32 displays worth?) (until this was replaced ~1980 with RAM based frame buffers.)

But the same technology innovations that started to put CPUs on desks made RAM get cheaper really fast (comparatively speaking.)

Even well into the second half of the 1980s, the dual-port RAM for a 24 bit low resolution (512x512 pixel) could occupy three, double height Euro cards.

Most displays made do with less complex schemes, doing away with the dual-port, and color, and so on. Hercules Graphics adapters were short ISA cards and did 720×350, and the original Mac did 512×342, and VGA (1987, and color!) at 640x480 was a single PC expansion board. (and 512x512 wasn't "low resolution" for those days, especially not for color.)

Can remember playing a simple plan+view tank battle game on a PDP-8 and storage display, using the front panel toggle switches as controls

"Vector" is not the same as "storage." Storage displays tended to suck for games, since they couldn't erase except a slow "whole page" erase. (Still, there was a storage display sold by DEC that attached to a PDP/8 (and through the 8, to other systems.)

I found this interesting reference. It's a bit jumped WRT to the way it presents a timeline ("displays were CRTs. Let's talk about Crookes, who invented the electrical discharge tube"), but it seems to contain a lot of the details...
http://what-when-how.com/Tutorial/topic-203v31/The-History-of-Visual-Magic-in-Computers-327.html

westfw:
So just how much CAD/CAM do you think was done on $5k terminals (Hmm. Actually a lot cheaper than I would have thought) connected to multi-million-dollar mainframes (that mainly did batch processing, charged for by the CPU-second) over slow comm links (our Tek-4013 (big on APL in those days!) was on a 2400bps UART link (which was as fast as the mainframe's front-end would go)? Graphics of any kind was quite rare.

I'm mostly arguing with the "went down the route" idea. Storage tube displays clearly existed, but they were a TINY NICHE in the display market, and never had anything to do with the whole "personal computer" or "personal workstation" revolution.

You found ONE vendor of Storage tubes. How many vendors for CRT raster displays?
....

The OP was writing about the transition from CRT monitors to LCD monitors in the 1990s - those were graphical displays.
Originally, although raster scan CRTs existed, computer graphics went down the storage tube route. With cheap memory they came back to raster scan CRT but only briefly because LCD monitors came along.

There was actually quite a lot CAD/CAM done using the expensive TektroniX storage tube displays. TektroniX was the market leader and graphics libraries used their protocols which were implemented by other manufacturers. It was not mainframes that were used but minicomputers such as Prime or PDPs, often as timeshare via dialup. The graphics though were normally monochrome line drawings, not 3D rendered images.

The replacement raster scan CRT monitors were much cheaper but still bulky and you had lug around a paper copy of @lastchancename's 252 page manual - no internet then. If you take a look at the manual it says "The AED 512/767 emulates the Tektronix 4010-4014 series terminals and is, therefore, compatible with host software written for them".

"Vector" is not the same as "storage." Storage displays tended to suck for games, since they couldn't erase except a slow "whole page" erase. (Still, there was a storage display sold by DEC that attached to a PDP/8 (and through the 8, to other systems.)

I think the display was probably a Tektronix 611 (portrait), with the intensity (or persistence?) control under the hinged flap on the right-hand side turned down, usually with a paperclip jammed into the split in the pot.

Originally, although raster scan CRTs existed, computer graphics went down the storage tube route.

I still disagree. The Tek was probaly the first "affordable" graphics terminal, and therefore ended up defining a standard graphics-over-serial protocol, but the storage tube technology was essentially a short-lived dead end, not significant in the future development of graphics technology (Evans and Sutherland, Ramtec, Vector General, Xerox, Symbolics, LISP Machines Inc, IBM, SUN, Apple, NEXT... All based on non-storage CRT technology, and mostly bit-mapped graphics.)

LCD monitors are still raster scan devices based on bit-mapped graphics. Each cell is vaguely like DRAM, needing periodic refreshing, and I guess the industry is pretty fixed on those rasters...

(Hmm. I'm not exactly sure why an AMLCD needs to be refreshed raster-wise. Has there ever been a vector LCD that only refreshes line segments? (I guess displays that only draw lines are relatively uninteresting any more, now that graphics as all 3D polygons and textures and stuff.))

... storage tube technology was essentially a short-lived dead end....

That was my point. I made the comparison between storage tube displays and valves.
Transistors pre-dated valves but for technical/practical reasons we went down the (largely) dead end of using valves before coming back to transistors.

The same with CRTs we wanted computer graphics but for technical/practical reasons we really could not use raster scan CRT at the time so we went down the dead end of storage tube displays before coming back to CRTs and then moving on to LCDs.

It was a surprise to learn today though that storage tubes were used as memory before magnetic cores.

ardly:
Transistors pre-dated valves

Did not.

...R

Semiconductors did, but not transistors.

Slightly OT, but while I was looking for the name of the storage tube device attached to the PDP-8, I found this from December 1967, about the Tek 611, and on page 7, describing how it could be used with the "SRI mouse".

According to Wikipedia, The Mother of All Demos wasn't until 1968, and Engelbart only applied for the mouse patent in 1967 - this was really cutting-edge stuff!

(The paper also describes how the display had a "write-through cursor", which explains how it could display moving graphics without the blink/erase)

This turned into far more ‘sport’ than I imagined !
Just BTW, in my sphere, we were driving the AED 512 displays with PDP-11s, and we also had some SGI and E&Sgear later... laughably at the time, the 100MIPs based machines were ‘cutting edge’, and we were hardware hacking PDPs up from 64K to 256K of RAM !