18-Bit Family Timeline
1960 PDP-1, Digitals first 18-bit computer
1963 PDP-4
1964 PDP-7, uses flip chip modules; used by Ritchie and
         Thompson to develop UNIX
1966 PDP-9, program compatible with PDP-7

1969 PDP-15 replaces PDP-9
1972 MUMPS-15 (Massachusetts General Hospital
         Utility Microprogramming System), PDP-15 -based
         timesharing system designed to handle medical
         records, still in use
1979 PDP-1 with working Spacewar! game installed at
         The Computer Museum, Boston
1988 A PDP-1 system (serial no. 44) is saved from a barn
         in Wichita, Kansas, and donated to the Digital
         Historical Collection

Presentation at decworld2001 (Computer History Museum) June 16, 2001

the name PDP-2 was reserved for a 20 bit machine - never produced the name PDP-3 was to be a 36 bit TX-2 - sold to the AirForce - never produced - sold the AirForce two PDP-1 machines instead does an 18 bit machine and an 18 bit machine add up to a 36 bit machine?

From "Digital at Work" , Digital Press, copyright 1992, page 33

A distinguished Lineage Interactive Computing with the PDP-1 Edited by Bob Hofmann from interviews with Edward Fredkin Digital's PDP-1 was the direct descendant of the TX-0 computer, developed along with the larger TX-2 system at Lincoln Laboratory in the late 1950s. Designed by Wesley Clark, the TX-0 and TX-2 were among the first transistor driven computers in the world. With many of the features pioneered on the massive Whirlwind system CRTs, consoles,paper-tape input and output-the TX-0 and TX-2 effected many advances in interactive computing. The TX-2 was planned as a large-scale 36-bit system for advanced graphic display research. The smaller TX-0 was built first, to test the transistor circuitry and complex core memory of the larger system. That process produced the TX-0 Direct Input Utility System, a set of programs that made it possible to communicate directly with the computer with an online typewriter. This primitive operating system was the first ever developed for a real-time computer. Soon the power of the TX-0 was applied to other tasks. A program to analyze electroencephalograph recordings for sleep research utilized the first moving window display ever to appear on a CRT. To create scientific characters on the display for mathematical equations, team member (and later PDP-1 architect) Ben Gurley developed a light pen. "The TX-0 could do everything a personal computer does today," remembers Ken Olsen, another member of the team, "limited only by the fact that the memory was small." The much larger TX-2 was less limited, especially after 65 K of memory from the TX-0 was added to it. At the first official meeting on interactive graphics, a young graduate student named Ivan Sutherland introduced his TX-2-based Sketchpad system, a new kind of simulation language that enabled the computer to translate abstractions into concrete visual forms. Many of Sketchpad's capabilities were sophisticated even by the workstation standards of the1980s."If had known how hard it was to do," Sutherland said later, "I probably wouldn't have done it." Stripped of most of its memory, the TX-0 was sent on a long-term loan to MIT, where it inspired a new generation of graphics pioneers. Assistant Professor Jack Dennis gave access to the system to members of the Signals and Power Subcommittee of the MIT Model Railroad Club, students who were to win fame as the first computer "hackers." Dennis himself designed a symbolic debugger for the system. Called FLIT, after a popular insecticide, this Flexowriter Interrogation Tape was the first of many playfully named but useful programs created by the group. The TX-0 was considered the ultimate in interactivity by the hackers-until the first PDP-1 arrived a few years later. ...