A history page. Nothing here is interactive beyond scrolling — the lights panel is decorative, not a control surface.
Computing History · 1956–1975

TX-0

The Transistorized Experimental computer, built to prove a wild idea: that transistors — not vacuum tubes — could run a real computer. It worked. And a few years later, at MIT, the people who stayed up all night playing with it invented the word "hacker."

CONSOLE — REGISTER DISPLAYSIM.
18-bit accumulator, simulated

Origins at Lincoln Laboratory

TX-0 — short for "Transistorized Experimental computer zero" — was built at MIT's Lincoln Laboratory starting in 1955, under the direction of Wesley Clark and Ken Olsen. It was a proof of concept for a much larger machine, the TX-2, and for a bet that seemed risky at the time: that transistors, still an unproven technology outside the lab, could replace the vacuum tubes running every serious computer of the era.

Vacuum tubes were bulky, power-hungry, and failed constantly — a large tube-based machine might have thousands of them, and keeping all of them alive at once was a genuine engineering feat. Transistors promised something smaller, cooler, and far more reliable. TX-0 was where that promise got tested against reality.

It worked. TX-0 became one of the first transistorized computers ever built, running reliably enough to prove the architecture that would eventually replace tube-based computing entirely.

The machine

By the standards of 1956, TX-0 was strikingly modern in how it felt to use — even though by modern standards it was almost unimaginably limited.

Word length18 bits
Memory4,096 words of magnetic core
Logic~3,600 transistors
DisplayCRT screen with light pen input
I/OPaper tape reader/punch, typewriter
Cycle time~5 microseconds per instruction

The CRT display and light pen were unusual for the time — most computers of the era communicated purely through printouts and punch cards. TX-0 let a user see output directly on a screen and point at it, which made it feel less like operating machinery and more like having a conversation with the machine.

The move to MIT

Once TX-2 was ready to take over TX-0's role at Lincoln Laboratory, the older machine was transferred to MIT's campus in 1958, installed in Building 26 alongside the Institute's other major computing resource, the enormous Whirlwind-derived IBM machines used for official coursework and research.

What made TX-0 different wasn't its specs — it was already outdated compared to newer machines by then — it was access. Unlike the tightly scheduled mainframes, TX-0 time was made available to students, including undergraduates, often late at night when demand was low. For the first time, a computer at MIT was something students could simply sit down and use, experiment with, and — crucially — break.

Where "hacker" was born

The students who gravitated toward TX-0's late-night open hours were largely drawn from the Tech Model Railroad Club (TMRC), a student group obsessed with the intricate switching and signaling systems behind their model railroad layout. TMRC already used the word "hack" for an elegant, technically impressive trick — and when its members got their hands on TX-0, they started applying it to code.

"A hack is a project undertaken or a product built not solely to fulfill some constructive goal, but with some wild pleasure taken in mere involvement."— TMRC glossary, as recalled by early TX-0 users

The culture that formed around TX-0's terminal — write code all night, chase elegance over correctness, share what you'd learned freely, distrust unnecessary authority over how a machine should be used — became what Steven Levy would later describe, in his 1984 book Hackers, as the foundation of "the hacker ethic." Many of the people who cut their teeth on TX-0 went on to found MIT's Artificial Intelligence Laboratory.

What people actually built on it

  • Mouse in the Maze — an early interactive program where a light pen let the user place walls and cheese in a maze, and watch a simulated mouse solve it. One of the first programs built purely to be fun rather than useful.
  • Munching Squares — a mesmerizing recursive graphics display, often cited as one of the earliest examples of generative computer art, born entirely from students exploring what the CRT and light pen could do.
  • Music synthesis experiments — TX-0 was coaxed into playing crude tunes by driving its speaker output at audio frequencies, a party trick that showed up again and again on early hacker-built machines.
  • Debugging tools built by the users, not the vendor — because there was no vendor. If you wanted better tools for working with TX-0, you wrote them yourself and left them for the next person.

Legacy

TX-0 was decommissioned in the 1970s, its core ideas long since absorbed into the machines that followed it — including the PDP-1, which inherited both its transistor-based design philosophy and no small number of its former users. The PDP-1 would go on to host Spacewar!, widely considered the first true video game, built by exactly the kind of late-night, access-driven tinkering culture that TX-0 had incubated a few years earlier.

The machine itself is gone. What it started — the idea that computers were something you could stay up all night playing with, not just formally scheduled to use — didn't go anywhere at all.