TX-2 Project

Re-creating the historically important TX-2 computer

TX-2 Documentation

Contents

Overview

The main categories of documentation are:

  1. A description of the TX-2 for modern readers.
  2. Documentation produced by the TX-2 project itself. For example explanation of the planned design, reference information for maintenance and documentation intended for TX-2 programmers.
  3. Papers published by people who wrote software on and for the TX-2; Ivan Sutherland’s thesis, for example.
  4. Proposals, progress reports, and other administrative documentation relating to the TX-2 project and Lincoln Lab. This mostly does not contain information of directly technical use, but can help establish the timing of changes and the motivation for changes.
  5. Retrospectives of the TX-2 project and programs such as Sketchpad.
  6. Project documentation, including commentary on the primary sources.

There are also some related photographs and videos.

How You Can Help

If you know of additional material that’s not listed here, please let us know (for example by email to james@youngman.org). Even if you can’t give it to us directly (e.g. because Lincoln Lab would need to release it), it will be useful for us to know that it exists.

Documentation Highlights

The TX-2 Technical Manual is the key reference work on the workings of the TX2, while the User Handbook is the first place to look for the kind of information a programmer would need to know.

Lincoln Laboratory TX-2 Project Documentation

Lincoln Laboratory Division 6 was responsible for the development of the TX-2. Their memos normally begin with the digit 6.

Introductory and Overview Material

  1. Papers presented at the February 1957 Western Joint Computer Conference (Los Angeles, California) by members of the TX-2 project.
  2. TX-2 Introductory Notes, A. Vanderburgh, 24 February 1959.
  3. Some Examples of TX-2 Programming, H. Philip Peterson, July 1958. Also known as Memorandum 6M-5780. These appear to have been prepared before some significant parts of the design (such as the address of the A register) were changed.
  4. User Program Measurement in a Time-Shared Environment. Alan G. Nemeth and Paul D. Rovner. Communications of the ACM, October 1971, Volume 14 Number 10; this describes the use of the sync system with the TX-2’s APEX time-sharing system. Includes some details of the implementation of the sync system.

TX-2 Users Handbook

The handbook describes the system’s instruction set, peripherals and some operational procedures and details (e.g. console controls and indicators). It also describes the system assembler, includes a listing of the customary set-up of the boot code in the A and B plug boards.

  1. TX-2 Users Handbook Chapter 3 - Operation Code, August 1963.
  2. TX-2 Users Handbook, Alexander Vanderburgh, July 1961 / October 1961 / August 1963.

TX-2 Technical Manual

Also known as Lincoln Manual No. 44. This was a three-volume work.

  1. TX-2 Technical Manual Volume 1 (Chapters 1 to 7)
  2. TX-2 Technical Manual Volume 2 (Chapters 8 to 15)
  3. TX-2 Technical Manual Volume 3 (Chapter 16)

The TX-2 Technical Manual is also available as a combined document at archive.org

The chapters of the Technical Manual are

  • Volume 1
    • Chapter 1: Introductory Description
    • Chapter 2: Functional Description of TX-2
    • Chapter 3: Circuit Logic Elements
    • Chapter 4: Memories
    • Chapter 5: Timing and Control
    • Chapter 6: Funcitonal Organization of the Control Element
    • Chapter 7: Operation Codes
  • Volume 2
    • Chapter 8: Pulse and Level Notation
    • Chapter 9: Computer Dynamics
    • Chapter 10: Control Element
    • Chapter 11: Memory Element
    • Chapter 12: Program Element
    • Chapter 13: Exchange Element
    • Chapter 14: Arithmetic Element
    • Chapter 15: In-Out Element
  • Volume 3
    • Chapter 16: Timing Charts

Programming

The Users Handbook contains a lot of information useful to programming for the TX-2.

  1. The New Skip-on-Index Instruction, J. M. Frankovitch, February 4th, 1960.

APEX

APEX was introduced in 1964 and was a system for time-sharing on the TX-2.

Sketchpad pre-dates APEX. Sketchpad is our primary focus for the simulator. The BCPL compiler used APEX, but we don’t have details of APEX or a copy of the BCPL compiler.

  1. A Time- and Memory-Sharing Executive Program for Quick-Response On-Line Applications, James W. Forgie. Fall Joint Computer Conference, 1965. AFIPS Conference Proceedings, Volume 27 Part 1.

TX-2 Hardware

The Technical Manual and Users Handbook contain a lot of useful information. Also see the Introductory and Overview Material.

Peripherals

The Users Handbook contains a lot of information about the TX-2’s peripherals.

Magnetic Tape Storage
  1. A Computer-Integrated Rapid-Access Magnetic Tape System with Fixed Address. IRE-ACM-AIEE ‘58 (Western): Proceedings of the Western Joint Computer Conference: Contrasts in Computers May 6-8, 1958, Pages 42–46
Lincoln Writer
  1. The Lincoln Keyboard - a typewriter keyboard designed for computers imput flexibility. A. Vanderburgh. Communications of the ACM, Volume 1, Issue 7, July 1958.
  2. The Lincoln Writer. J. T. Glmore, Jr., R. E. Sewell. Lincoln Laboratory Group report 51-8. October 6, 1959. (archive.org)
  3. Lincoln Lab Memo M-5001-11 “New Flexowriter Type Face” of 1959-08-27, by J. B. Dennis describes a flexowriter with a TX-2 typeface for use with the TX-0. The memo compares the “standard” and Lincoln type faces.

Memory

  1. Final Report: Development of a 10⁷ Bit Magnetic Film Memory. J. I. Raffel, A. H. Anderson, R. Berger, T. S. Crowther, T. O. Herndon, M. L.Naiman, C. E. Woodward, 1971-06-10.
    • Describes a memory unit they call LCM-II.
    • This report also mentions that in July 1968 a million-bit memory (LCM I) was installed in the TX-2.

Other

  1. A Discussion of the Circuitry Used in the Lincoln TX-2 Computer. Jonathan R. Fadiman. Lincoln Laboratory Memo 6D-2631, 1 October, 1958. Available from bitsavers.
  2. 6M-5661, Toggle Switch Storage System TX-2, Leopold Neumann, April 21, 1958.

Papers on TX-2 Software

See also the page on the software described by these papers.

Sketchpad

  1. Sketchpad, A Man-Machine Graphical Communication System, Ph.D thesis of Ivan Edward Sutherland, January 7, 1963.
  2. Sketchpad: A Man-Machine Graphical Communication System, Lincoln Laboratory Technical Report 296, January 30, 1963.
  3. Sketchpad listings and memoranda pertaining to TX-2 computer and programming. Computer History Museum; catalog number 102726903.
    • The listings come in two documents. Large sections of these are illegible. These listings appear to have been printed on the TX-2’s Xerox printer and then photocopied. Many of the key details of the listing are unrecognizable blobs. TX-2 assembly language makes extensive use of superscripts and subscripts, and the small size of these has meant that the photocopying did not always come out well.
  4. In Specialized Computer Equipment for Generation and Display of Three Dimensional Curar Figures, March 1963 (MIT Electronic Systems Laboratory report ESL-TM-167), Robert H. Stots states that Sutherland’s estimates for his TX-2 programs are that, “of the time they spend on generation of the display list about 10% is spent on orienting the window to the total picture, about 10% on deciding which parts of the picture are within the view of this window, and 80% on generating the point list.”

See also Sketchpad videos.

Sketchpad-III

  1. Sketchpad-III, Three-Dimensional Graphical Communication With A Computer, Ph.D thesis of Timothy Edward Johnson, May 1963.
  2. Sketchpad III: a computer program for drawing in three dimensions. Timothy E. Johnson. AFIPS ‘63 (Spring): Proceedings of the May 21-23, 1963, Spring Joint Computer Conference, May 1963. Pages 347–353.

See also Sketchpad-III videos.

Languages

BCPL

  1. BCPL Reference Manual, Martin Richards (M.I.T. Project MAC), Henry Ancona (Lincoln Laboratory), 6 May 1969.

LEAP

LEAP was a language similar to Algol.

  1. P. D. Rovner and J. A. Feldman, “An Associative Processing System for Conventional Digital Computers, Lincoln Laboratory Technical Note 1967-19. 1967-04-21.
  2. Jerome A. Feldman and Paul D. Rovner. 1969. An ALGOL-based associative language. Commun. ACM 12, 8 (Aug. 1969), 439–449. (pdf)
  3. The LEAP User’s Manual (DTIC AD0713221). P. D. Rovner, Lincoln Laboratory Group 23, 1970-09-11.

Other TX-2 Software

  1. LO - A Text Formatting Program A. Evans Jr, Lincoln Laboratory, 21 February 1975.

Work Done with the TX-2

Human/Computer Interaction

  1. The On-Line Graphical Specification of Computer Procedures. William Robert Sutherland, January, 1966.
  2. On-Line Graphical Specification of Computer Procedures. W. R. Sutherland, 23 May 1966, Lincoln Laboratory Technical Report 405.
  3. Graphical Communication and Control Languages. L. G. Roberts. Lincoln Laboratory, M.I.T.
    • This describes a list processing system which it states is similar to but not identical to those used in Sketchpad and Sketchpad-III.
    • Available from researchgate.net
  4. Graphical Manipulation Techniques using the Lincoln TX-2, H. H. Loomis Jr.
  5. Morfield, Maxwell A., Raymond A. Wiesen, Mitchell Grossberg, and Douwe B. Yntema. Initial experiments on the effects of system delay on on-line problem solving. Lincoln Lab Technical Note 1969-5, DTIC AD0691815. 1969.

Graphics

  1. Machine Perception of Three-Dimensional Solids. Lawrence Gilman Roberts, June 1963. Available from MIT libraries

Roberts’ work used a “ring list” representation for drawings. This is similar to but not identical to the methods used in Sketchpad and Sketchpad-III (see page 59 of Roberts’ thesis). We don’t have a listing of the program on which the thesis is based, and the thesis does not include code.

See also the Semiannual Technical Summary reports on Graphics by Lincoln Laboratory.

Image Processing

  1. L. D. Earnest worked on Machine Recognition of Recursive Writing (MITRE Corp.) (1962)
  2. James E. Cunningham worked on Image Correction on the TX-2; see page 244 of the Quarterly Report of MIT. Research Laboratory of Electronics For The Three month period ending 31 May 1963

Networking

  1. T. Marill and L.G. Roberts, “Toward a Cooperative Network of Time-Shared Computers”, Proc. AFIPS Fall Joint Computer Conference, pp . 425-431 (1966). (pdf). This describes some work to connect the TX-2 to the System Development Corporation’s AN/FSQ-32 machine.
  2. Copmuter Corporation of America, “An Experimental Computer Network” (DTIC AD0694055) describes the connection of the Lincoln Lab TX-2, System Development Corporation’s AN/FSQ-32 and some small DEC machines to an experimental network. The protocol seems similar to that used in the 1966 paper by Marill and Roberts.
  3. The ARPANET Completion Report, by F. Heart, A. McKenzie, J. McQuillian, and D. Walden, BBN Report 4799, January 4, 1978. Available from walden-family.org and archive.org.
  4. The Lincoln Lab report An Interface to the ARPA Network for the CP/CMS Time-Sharing System. Volume 1 (DTIC AD0773831) shows the TX-2 attached to the ARPAnet.
  5. Internet RFC 694 assigns a socket number (65) for the TX-2’s Speech Data Base.
  6. Forgie, J., C. McElwain, and C. Weinstein. “Network Measurement Facility on TX-2”, Masachussetts Institute of Technology, ESD-TR-74-218, May 1974.

Speech Recognition and Audio Processing

  1. Gold, B., Pitch Extraction on the TX-2 Computer. J. Acoust. Soc. Am., 33, 1664-1665 (A) (1961)
  2. James W. Forgie and Carma D. Forgie worked on speech recognition using the TX-2; see page 287 of Current Research and Development in Scientific Documentation No, 11, 1962-11-01. Edwin B. Newman states in “Paracomputers in Psychological Research”, on page 241 of Harvard Symposium on Digital Computers and Their Applications (1961 : Brookline, Mass.) that these programs were initially written for the Whirlwind computer and were later adapted for the TX-2.
  3. The Network Voice Protocol was developed by ISI and used between ISI and the Lincoln Laboratory (see ARPA report “A Research Program in Computer Technology (September 1975) page 55”. The experiments completed in October 1974 and were the first use of a packet-switched network for digital voice communication (page 61). The July 1976 report contains some further details.

See also the Semiannual Technocal Summary reports on Speech by Lincoln Laboratory.

Brain and Nerual Research

Wes Clark did more work in this field on other machines including the LINC.

  1. Activity in Networks of Neuron-Like Elements, R. G. Farley and W. A Clark. Symposium on Information Theory, Royal Institution, London, 1960-08-29 to 1960-09-02 (Cilin Cherry, Ed.).

Other Topics

  1. Aspects of Associative Processing, J. A. Feldman, MIT Lincoln Laboratory Group 23 Technnical Note 1965-13, 21 April 1965.

Conferences

  1. The 1957 Western Joint Computer Conference (Los Angeles, California) included multiple presentations about the TX-2. These papers are available from bitsavers.org. For the papers themselves see “Introductory and Overview Material” above.
  2. The 1959 Eastern Joint Computer Conference featured a trip (1959-12-02) featuring an inspection and a demonstration of the TX-2 computer (from Datamation, November-December 1959, page 46). The trip was hosted by A. Vanderburgh (per the list of personnel in the conference proceedings).
  3. 1965 Fall Joint Computer Conference
    • A Time- and Memory-Sharing Executive Program for Quick-Response On-Line Applications

News Coverage

1957

  1. Computers and Automation, Volume 6 Issue 1 (January 1957): mentions the TX-2 related talks at the upcoming Western Joint Computer Conference, scheduled for the morning of Feb 28.
  2. Digital Computer Newsletter, Volume 9 No. 2 (April 1957) (page 4): describes construction of the TX-2 and outlines its size, speed and multi-sequence operation, decribing it as a “general-purpose parallel binary machine”.
  3. Computers and Automation, Volume 6 Issue 4 (April 1957) (page 28): briefly decribes the talks given at the Western Joint Computer Conference, Statler Hotel, Los Angeles, 1957-02-26 to 1957-02-28.

1959

  1. MIT Technology Review, March 1959: briefly mentions the TX-2’s multiplication speed and work on speech and pattern recognition, in a general article on the work of Lincoln Lab.
  2. Control Engineering, September 1959: Digital components by both Digital Equipment Corp. and Harvey-Well Electronics Inc. are both derived from work on computers at the Lincoln Laboratory.
  3. Datamation, Sptember-October 1959, Volume 5 Issue 5: Magnetic Film Memory In Operation at Lincoln.
  4. Computers and Automation, October 1959, Volume 8 Issue 10: John A. Kessler, First successful operation of a Practical Magnetic Film Memory.
  5. Electronic Industries, October 1959, Voolume 18 Issue 10: What’s New … Magnetic Film Memory
  6. Automatic Data Processing,October 1959, Volume 1 Issue 9: High-Speed Magnetic Film Memory.
  7. MIT Technology Review, November 1959: Magnetic-Film Computer Memory.
  8. The Computer Bulletin Index, Volume 3 (December 1959): Mentions the TX-2’s magnetic-film memory.

1960

  1. Digital Computer Newsletter, Volume 12 No. 1 (Jan 1960) (page 29): high-speed magnetic-film memory.

Progress Reports

The periodic status reports of several of the divisions of Lincoln Laboratory (e.g. 2, 5, 6) provide some useful background information about the stages in the build of the TX-2 computer. See Lincoln Laboratory Progress Reports.

Retrospectives, Oral History, etc.

  1. An Interview with Ivan Sutherland, OH 171, Conducted by William Aspray, 1 May 1989. Charles Babbage Institute.
  2. The TX-2 Computer and Sketchpad, Lincoln Laboratory Journal, Volume 19, Number 1, 2012.
  3. Graphics in Time-Sharing: A Summary of the TX-2 Experience William R. Sutherland, James W. Forgie, Marie V. Morello.
  4. Interaction at Lincoln laboratory in the 1960’s: looking forward – looking back William Buxton, Ron Baecker, Wesley Clark, Fontaine Richardson, Ivan Sutherland. CHI EA ‘05: CHI ‘05 Extended Abstracts on Human Factors in Computing Systems. April 2005, Pages 1162–1167.
  5. The LINC was Early and Small. Wesley A. Clark. HPW ‘86: Proceedings of the ACM Conference on The history of personal workstations. January 1986 Pages 133–155. Video.

Changes to the TX-2

There are some discrepancies between the primary documents listed above. The primary cause of this appears to be the fact that the TX-2 computer was changed, both during the design and build process and also while in regular operation (routine changes occurred on Wednesdays).

Primary sources will likely reflect these changes. Here are some examples:

Plans to Shut the TX-2 Down