Looking for the First Computer Font.

July 17, 2011 9:51 AM

I would like to see the character set from the first computer typeface.

I've poked around online for a while, but couldn't find it. I want to see the first human-readable Roman character set designed for entry on a computer screen.

Specifically, I'm looking for the first that was displayed on a monitor in response to letters typed on a keyboard, with a computer somewhere in the middle there. The closest I've been able to figure is that it might have been on a terminal hooked up to a MULTICS system.

posted by lore to computers & internet (14 answers total)

The 5x7 uppercase ASCII font built into the Apple ][+ would be very close; there's not a huge amount you can do differently with 5x7 uppercase ASCII and still have a readable character set. Early terminals were uppercase-only, so you should ignore all glyphs with code points outside the range 0x20 - 0x5F.
posted by flabdablet at 10:23 AM on July 17, 2011

To the contrary, the very first interactive computer-controlled text-display CRTs were probably not rasters like the Apple II, but vector graphics such as the original SAGE system. Maybe that's farther back than the OP intends?
posted by rlk at 10:35 AM on July 17, 2011 [2 favorites]

This brochure has some screenshots.
posted by flabdablet at 10:38 AM on July 17, 2011

The Datapoint 3300 whose 5x7 font I'm talking about came out in 1969, and I think it was probably the first commercial device featuring both a video display and a keyboard.

There are closeups of a SAGE display starting from about 5:15 in On Guard! from 1956. It doesn't appear to have a keyboard.
posted by flabdablet at 10:52 AM on July 17, 2011

You might want to look at Hershy's fonts. They are vector format that predeate bitmap fonts and were developed by Dr. A.V. Hershey in the 60's.
posted by Ookseer at 12:32 PM on July 17, 2011

The Tektronix T-4002 DVST terminal dates back to the late 1960's. The character set in it was pixels rather than stroke.

Unfortunately, I can't find any pictures of it. But it used the same font as the Tek 4010, which looked like this. (Neither of them supported lower case; that was added in the 4014.)
posted by Chocolate Pickle at 1:12 PM on July 17, 2011

(Oops. Actually, lower case was added in the 4012.)
posted by Chocolate Pickle at 1:14 PM on July 17, 2011

This brochure shows the font from the UNIVAC Uniscope 300 terminal (scroll down to about page 8 for a nice screen shot).

According to wikipedia, "Screen size of the original Uniscope 100 was 12 X 80 or 16 x 64 characters. All letters were in capital. Each character was individually drawn as a series of splines using technology developed for displays in military cockpits." The Uniscope 300 (in the brochure linked above) seems to have the same spline-type character set and upper-case only--most likely just the same as the Uniscope 100.

Wikipedia seems to say these were made starting in about 1969. So maybe not the earliest of all (one of the pages above says it was "one of the first cost-effective CRT terminals"), but pretty early on.
posted by flug at 2:05 PM on July 17, 2011 [1 favorite]

This youtube video shows a DEC VT05 terminal at work (specs). Again, not the very earliest, but early on--about 1970.
posted by flug at 2:35 PM on July 17, 2011

Ok, here are some more early ones:

• IBM 740/IBM 780 CRT Recorder and Display (1954)
• DEC PDP-1 Display (1961)
• IBM 2260 (1964) - wiki - specs
• IBM 2250 (1965)

FWIW, the 2260 seems to be the earliest example (that I've been able to find) of what you're looking for--a keyboard and screen where what you type would end up on the screen, and the whole thing used to interface with a computer. It's a bitmapped type raster character display (but not exactly what . The 740/780 were a vector CRT recorder (to 35mm photographic film) and display--basically intended as a printer-type interface, and no keyboard of the type you're interested in. The PDP-1 was a round vector display and didn't have a keyboard, either. The 2250 was also a vector display, did have a keyboard (as well as light pens and other fancy stuff) and is generally much fancier but also perhaps a bit later than the 2260.

An interesting point: According to this the display part of the 2260 was just a dumb video screen and all the character generation etc was done via the 2848 control unit which the 2260 (actually several of them, usually) was attached to via a simple video cable. So the font was generated by the 2848 and:

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).

I haven't been able to find any really clear screen shots of any of these displays, but maybe with these model numbers as reference, someone else can.
posted by flug at 3:50 PM on July 17, 2011

FYI a few sources (this and this, for example) seem to place the Uniscope 300 to 1964 as well.
posted by flug at 4:22 PM on July 17, 2011

This guy developed a video terminal for the SAGE system in 1959. He "just needed a good terminal and didn't realize other computers didn't have them".

In 1959 I was an engineer on the IBM AN-FSQ 7 computer which cost about $25 million and performed Air Defense activities for the Air Force in the Nevada desert. The large CRTs we used to display tracks of aircraft, radar data and had a capability of displaying any of 64 characters from a matrix inside the tube. The data was intended to be placed next to vectors indicating flight information. However, as a maintenance engineer I wrote the software to actually make the tube a very interactive device from which we would run our diagnostics. It also had decent keyboard. I wrote one line at a time and moved everything up a line each time I wrote. There was room for about 64 lines as I recall. I just simulated a line printer and card reader for keyboard. My idea allowed us to run our diagnostics in half the time (4 hours instead of 8 hours) because we weren't printing on a the military version of the IBM 407 printer. My boss invented the Selectric Typewriter at the same time I was working on the display consoles. I just needed a good terminal and didn't realize other computers didn't have them. The SAGE was the only machine I had ever seen due to secrecy. We switched our giant classified mainframes to the Selectric after that which was a mistake because they were so unreliable. The 7030 Stretch for NSA and AEC (prior to DOE) is an example of that foolishness. I didn't see my back room engineering idea used again in IBM but we did a similar thing in 1966 while I was at RAND helping on the development of the RAND tablet. Yes, the guys in my department developed ARPANET. We never knew. John Bowen

posted by reynaert at 4:40 PM on July 17, 2011 [1 favorite]

This is all great info, folks. Thanks!
posted by lore at 5:26 PM on July 17, 2011

FWIW, Teletypes preceded CRTs. I know you said "screen,' but...
posted by ZenMasterThis at 5:37 PM on July 17, 2011