tX If ,@8s~~~I

. L K - ! vaTg Dc' 'n'_ !-41 2M IT 6u i ~~6:ill n -jW- ~ 4 1 2

The RESEARCH LABORATORYof

ELECTRONICSat the

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

CAMBRIDGE, MASSACHUSETTS 02139

TX-O Computer History

John A. McKenzie

RLE Technical Report No. 627

June 1999

1 7A I-:· .:~-

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

RESEARCH LABORATORY OF ELECTRONICS

CAMBRIDGE, MASSACHUSETTS 02139

TX-O COMPUTER HISTORY

John A. McKenzie

October 1, 1974

TX-O COMPUTER HISTORY

OUTLINE

ABSTRA CTPART I (at LINCOLN ABORATORY)

INTRODUCTION 1DESCRIPTION 2LOGIC 4CIRCUITRY 5MARGINAL CHECKING 6TRANSISTORSMEMORY (S Memory)SOFTWARE (Initial) 9TX-2 10TRANSISTORIZED MEMORY (T Memory) 11POWER CONTROL 12IN-OUT RACK 13CONSOLE 13

PART II (at CAMBRIDGE)INTRODUCTION 14MOVE to CAMBRIDGE 15EXTENDED INPUT/OUTPUT FACILITY, Addition of 16FIRST YEAR at CAMBRIDGE 18MODE of OPERATION 19MACHINE EXPANSION PHASE 20T-MEMORY EXPANSION 21ORDER CODE ENLARGEMENT 22DIGITAL MAGNETIC TAPE SYSTEM 24SOFTWARE DEVELOPMENT 25APPLICATIONS 29TIMESHARING (PDP-1) 34CONCLUSION 34A CKNOWLEDGEMENT 35

BIBLIOGRAPHY

TX-O COMPUTER HISTORY

A BSTRA CT

The TX-O Computer (meaning the Zeroth TransistorizedComputer) was designed and constructed, in 1956, by theLincoln Laboratory of the Massachusetts Institute of Tech-nology, with two purposes in mind. One objective was to testand evaluate the use of transistors as the logical elementsof a high-speed, 5 MHz, general-purpose, stored-program,parallel, digital computer. The second purpose was toprovide means for testing a large capacity (65,536 word)magnetic-core memory.

The TX-O was so sucessful, including the memory checkout,that construction was begun on the TX-2. During the springof 1958, the large memory was transferred to the TX-2 andthe TX-O was outfitted with a 4096 word, transistor-drivencore memory.

At that time the computer was moved to the MassachusettsInstitute of Technology Campus in Cambridge, on a long termloan basis, to the Electrical Engineering Department. Hereit was Jointly supported by the Research Laboratory ofElectronics and the Electronic Systems Laboratory, and madeavailable as a do-it-yourself facility where both research-ers and students could work on-line, having direct access tothe machine.

The desirable features of this, at that time, unusualmanner of operation were soon evident and this mode of usagewas enhanced by the development of good utility programs0Further attraction was the very excellent input/outputcapability allowing for ease of communication with externalequipment.

The transistorized computer proved to be so reliable thattime, normally allocated to scheduled maintenance, wasutilized to modify the machine. The core memory was doubledto 8192 words and the order code expanded. This included theaddition of an index register. The modifications to thecomputer continued until 1963, at which time the efforts andfunds of the group were directed towards the implementationof a time-sharing system on the PDP-i Computer, which hadbeen donated to the Electrical Engineering Department by theDigital Equipment Corporation. Several of the researchgroups, on the basis of demonstrated results on the TX-O,were able to negotiate funds to acquire their own computers,and the usage began to taper off.

The running-time clock now logs over 49,000 hours. Earlyin 1976 the TX-O will be moved to the Computer Museum of theDigital Equipment Corporation in Marlborough, Mass,

PART I (at LINCOLN LABORATORY)

INTRODUCTION

The TX-O Computer (meaning the Zeroth TransistorizedComputer) is a 5 MHz, parallel, 18 binary digits, general-purpose, stored-program, digital machine with a cycle timeof six (6) microseconds and capable of performing betterthan 80,000 additions per second. It was designed andconstructed at the Lincoln Laboratory of the MassachusettsInstitute of Technology by Group 63 of Division 6. The GroupLeader was William Papian, with Wesly Clark responsible forthe logical design and Kenneth Olsen heading the circuitdesign and construction aspects. The latter phase of theproject was completed by Benjamin Gurley. The memory designand fabrication were the responsibility of Richard Best andJack Mitchell.

The TX-O was conceived with two purposes in mind. Thefirst was to test and evaluate the use of transistorcircuitry as the logical elements in a high-speed computer.The second objective was to provide means to test a 65,536word, 18 binary digits plus parity, vacuum tube, switch-driven, magnetic-core memory.

The machine was the third in a succession of transistor-ized experimental digital systems, the first of which was a100 transistor, double-rank shift register, The second was asmall, high-speed, error-detecting multiplier.

After some thought about the possible minimal machine, adesign was completed in which the word length would be 18bits, just a half of the final projected form. This computerwas referred to as the TX-0 and the projected machine as theTX-2. There was no TX-1i The TX-O became operational inApril 1956

1

DESCRIPTION

In the original configuration, the TX-O used 16 bits foraddressing and two (2) bits were decoded for the four (4)instructions. The addressable instructions were StoreAccumulator (sto x), Add Memory to the Accumulator add x),Transfer on Negative Accumulator (trn x), The fourthinstruction, the Operate Command, had the feature of provid-ing the facility to microprogram via time-pulses, thusallowing for the possibility of more than one function orregister transfer, including input/output transfers, withinthe two (fetch and execute) cycles. Since all of the MemoryAddress bits were decoded in this instruction, it had manypossibilities and because of this powerful feature, the TX-Oproved to be useful, even with the simple order code, as anextremely versatile measuring device. Previous to the timethat the TX-O was moved to Cambridge, it was used by CharlesMolner, of the Communications Biophysics Laboratory underProfessor Walter Rosenblith, to aid in the analysis ofbrain-wave data gathered from the auditory cortex of a cat'ssbrain.

The original TX-O included a Memory Buffer Register (MBR,18 bits plus parity), and an Accumulator (AC, 18 bits),arranged as a ring adder. The one's complement arithmeticand logical operations are done between the MBR and AC. TheMemory Address Register (MAR, 16 bits) and the ProgramCounter (PC 16 bits) along with the Instruction Register(IR, 2 bits) and the Live Register (LR, 18 bits) completesthe list of active registers. The LR, at that time, wasconsidered as just another storage register, using flip-flops rather than magnetic cores, and was especially useful,in conjunction with the microprogramming, for temporarystorage.

Provided as a method of manual intervention, in both thetest and normal modes, is sixteen (16) words of ToggleSwitch Storage (TSS, 18 bits) and the single word registers,Toggle Buffer Register (TBR, 18 bits) and Toggle SwitchAccumulator (TAC, 18 bits).

The machine has three (3) operating modes, Normal, Testand Read-In. The TX-O is a synchronous machine with thetime-pulses generated by ten (10) adjustable delays, forminga chain, with the last time-pulse connected back to retrig-ger the timing chain.

Initially the peripheral equipment consisted of a 250lines/minute Ferranti Photo-Electric Paper Tape Reader thatwas modified to solid state circuitry, and a modifiedFlexowriter used to type-in, type-out and to punch-out papertape.

2

__ _ ill __111 __·1_ �1��111141�11----�

In 1957, a 10 inch, electro-static deflection, cathode-ray tube, having 512 by 512 addressable locations, in a 7 by7 inch raster, point by point display system was installed.

In 1958, a Light-Pen, a solid state version of an ideabeing developed for the Sage System, was added to the TX-Oo

3

LOGIC

All of the high speed logic is performed with transis-tors, there are no diodes used as logic gates. The TX-O usesRC coupled, negative logic, a "ONE being -3 volts and a"ZERO" being ground level. The AND and OR functions areobtained by mixing and gating levels. The levels aregenerated by flip-flops which are set or cleared by a gatedtime pulse.

Timing pulses are negative with an amplitude of from 2.5to 3.0 volts and a width of 100 nanoseconds. These aregenerated by vacuum tube circuits and stepped down by pulsetransformers to supply the necessary drive currents. Thepulse lines are kept as short as possible with the strobepulses generated directly at the register where they areused.

The flip-flop outputs are fanned out, by means of openwiring, within the Central Processing Unit Rack or viacoaxial cable or twisted-pairs, when conveyed any greaterdistance. All of the logical interconnections, other thanthe few of those made in coaxial cable, are terminated intaper-pin connections. The flexiblity of this method is verydesirable in an experimental machine and along with themodular concept adopted for construction, allowed for thepossiblity of the extensive changes which were later under-taken.

Except for the flip-flop module, the transistors arepackaged one, two or three in a plastic, plug-in bottle,with the transistor leads brought out to the base. Theimportance of this type of construction will be evidentlater, when the ideas of transistor life history arediscussed.

4

CIRCUITRY

The logical elements of the TX-O use surface-barriertransistors and are formed from saturated inverter andsaturated emitter-follower combinations. A two transistorcascode configuration, with fast rise and fall times, isused as a power amplifier. This is similiar to the TTLcircuit used today in medium speed integrated circuits.

The TX-O flip-flop module, an Eccles-Jordan flip-flopcircuit, is capable of 5 MHz operation, has a built-inlogical delay and uses 10 transistors including the cascodebuffered outputs which give it good driving ability. Thebuilt-in delay in flip-flops allows the information clockedinto a flip-flop to be a function of that flip-flop'soutputs.

A dozen different types of plug-in-units are employed.Power supply voltages are +10, -3 and -10 volts. Theinverter units have positive bias, calculated so that thereis a safety margin in both the cut-off and the saturatedconditions.

The careful, conservative circuit design, allowed for awide variation in the transistor parameters and operatingconditions and effort was made to minimize the effects ofthem.

5

_1_1_ ·11_1_ _1 _ ___IIIIYIII·___1111_ ^.-� �-_I -·lm�- .--- 1 --·11I.(.-_-- II�L-.^-·--- I I -

MARGINAL CHECKING

The transistors in the computer are checked, whilerunning a diagnostic type program in the Normal mode, by theuse of the Marginal Check Panel. This selection panelprovides a means to select any one of thirty (30) differentlines, or sections, in the computer. In most cases, marginalchecking of these lines allows for a +10 or -10 voltexcursion on the +10 volt bias to the base of the inverters,including the inverters in the flip-flops, without causingthe unit to malfunction.

In practice, from a preventative maintenance point ofview, after the machine had been in operation for severalhundred hours, and except when modifications were made toit, the greatest value of the marginal checking was todiscover deterioration in the amplitude of the time pulsesgenerated by the 100 vacuum tubes, Of course the system wasinvaluable as an aid in inducing the infrequent type offailure and assists in finding machine faults before theybecome serious enough to decrease reliability.

6

_ ·_�_ I�_ _ _ 1�1� I· Il--_I�U --·-̂Y^ ^CIL-�---_I-�___L--I-�-LI�II �I�-..----�

TRANS ISTORS

The type L-5122 transistors used in the TX-0 weredeveloped by the Philco Research Division under a subcon-tract, to determine a set of specifications that wouldguarantee the switching properties of the surface-barriertransistor (SBT) and to develop the necessary test facili-ties to measure their essential parameters. These transis-tors were later commercially available as the 2N240.

Each transistor was carefully tested before being accept-ed for the TX-0 and a record of the measured parameters wasmaintained.

In 1958, after 6000 hours of operation, some 800 transis-tors were removed and retested. The most serious change wasfound to be a slight decrease in current gain.

In 1959, after 10,000 hours of operation, these sametransistors were again tested. The changes in the last 4000hours were smaller than those of the first 6000 hours. ALincoln Laboratory Technical Report was published, withdetailed presentation of the test data, for each of thetests. The summary of the second report stated, that after10,000 hours of operation, the surface-barrier transistor isshown to be a reliable computer transistor if used, withinlimitations of power, in circuits with adequate margins.

This was the last time that the extensive testing wasdone. At the time that this is written in 1974 , there isclose to 49,000 hours of operation on these originaltransistors, with fewer than a dozen failures.

The transistor selection and evaluation was directed atLincoln Laboratory by Donald J. Eckl and Robert J. Burke.

7

_ s_ _I_ Ir _·(^·�·�·II�_� II___ _ _I_ _LLI_---.��-1111 1- I ·I

MEMORY (S Memory)

The TX-O first used a high-speed, random access, mag-netic-core memory with a storage capacity of 65,536 words,19 bits long, including a parity bit. The words were read inparallel with a cycle time of five (5) microseconds. The 80mils outside diameter, 50 mils inside diameter, ferritecores, switched with an 820 milliampere current pulse andwere manufactured at Lincoln Laboratory. The memory systemcontained 425 dual triodes and 625 transistors. The coreswere fabricated into 64 by 64 subassemblies, each of whichwas a complete memory plane, and could be tested at thisstage of construction. Sixteen (16) 64 by 64 subassemblieswere then assembled into an array to form a 256 by 256plane. Following the initial checkout period, the memoryword lengh was doubled to 38 bits.

8

___1_1 I __ �_ II _ __ _1_1_1___ ___ 11_1 11______ Illrm^_----------C·LI�--I

SOFTWARE (initial)

The availability of the high-speed, large capacity corememory opened the possibility for new techniques and philos-ophies in planning and programming computer applications.One technique was to use the bulk of memory as a secondarystorage medium instead of using a drum or magnetic tape Aconversion program using this idea was written by WesleyClark. This was a fast translation program, that allowed theuse of symbolic language in address tags, address sectionsof instructions and constants.

Another valuable technique was the utility program writ-ten by Jack T. Gilmore, Jr. to coexist in core memory alongwith the user's program. This approach allowed the user tocommunicate with the computer via the on-line typewriter.Numbered among the features were the ability to examine aregister, modify the contents of a register or a storagelocation in memory, search for all occurances of a specificword or address, the ability to list all or any part of aprogram, and the punchout routine that provided the userwith the means to obtain an up to date binary tape of hisprogram as it exists in core.

It is interesting to note the lengthy conversationalapproach that was used. For instance, when a printout wasrequested, the program responded with the following message,"DO YOU WANT A VERTICAL COLUMN LAYOUT". The required answerwas always equally verbose. However there was a BE BRIEFfeature that reduced each question to one or two words.

At that time, it was felt that only in certain criticalsituations, would one be able to justify the inefficient useof computer time, by working on-line. Of course that wasthinking in terms of computers of that era, only largemachines that rented for approximately three hundred dollarsan hour. The TX-O was a forerunner in helping to change muchof that philosophy.

9

� I__�_ I I 1^C--.l --·�I�--� LI�II^L--_--�I_- _....._l __.- I_-_-�-_ I

TX-2

TX-O was most successful, having fulfilled the objectivesfor which it was built. Transistors had been proven to bepractical from a circuitry point of view and to be reliable.The memory checkout likewise was successful. In 1957 thedecision was made to make the next step in the developmentprogram be the TX-2 Computer. The TX-1 was not constructed.The TX-2 incorporated several new ideas in general charac-teristics, logical operation, memories and circuits. Ini-tially it had in the order of 22,000 transistors compared tothe 3600 in the TX-O.

10

__I II___ 1___· ______II ·I I �_I·_ ·_· I __1_1 __I�^__I____I_^·_·__ _I^_LII__IYIII__�II-----

TRANSISTORIZED MEMORY (T Memory)

During the spring of 1958, the 65,536 word memory wastransferred to the TX-2 A pluggable, 4096 word, 18 bit plusparity, entirely transistor driven memory was installed onthe TX-O. This memory, expandable to 38 bits, was designedand constructed at Lincoln Laboratory, and was earmarked foruse on the TX-2.

Memory planes, using ferrite cores, 50 mils outsidediameter and 30 mils inside diameter, needing a full selectcurrent of 450 milliamperes, were developed at LincolnLaboratory. Later there were development contracts withRadio Corporation of America, Needham, Massachusetts andwith the General Ceramics Company of New Jersey, in order torelease Lincoln Laboratory from the manufacturing phase, andto allow these memories to be commercially available.

The cycle time of the T Memory is 5.5 microseconds, withan access time of 2.4 microseconds. The system uses thefour-wire, coincident current scheme of operation. Read andwrite currents are obtained from the Read/Write Drivers,returned in series with a variable, large resistor, to the -150 and +150 volt power supplies respectively, serving as acurrent source. The Read/Write Drivers use fast transistors.The Core Drivers are slower and are set-up earlier to gatethe fast Read/Write pulse to the selected line.

The inhibit current is supplied from a -30 volt powersupply via a gated Digit Plane Driver for each plane.

The Sense Amplifier is a nine (9) transistor module,including the preamplifier with a gain of 22. This istransformer coupled to a rectifying slicer. The slice levelis individually adjustable, by a trimpot mounted on eachmodule. The output is a cascode circuit which drives theSense Amplifier level output to the Memory Buffer Register.Here it is strobed in by the strobe time pulse.

Marginal check voltage is provided to vary the slicelevel, so that the clipping level on each module , for eachdigit, may be optimized.

All of the T Memory system is mounted in the same rack asthe Central Processing Unit.

The addition of the new memory represented an addition of1460 transistors and 64 diodes, excluding the control. Inall references to transistor testing and failure rate, thesetransistors are not included. Uncaused failures in thissystem have been somewhat greater, in the order of twenty(20), most of which were in the high current transistors inthe selection line or Core Drivers. Since this memory wasdesignated for the TX-2, the packaging is accomplished inTX-2 style modules.

11

POWER CONTROL

At the time of the changeover to the new, T Memory, thepower control and power supply situation was reworked, inpreparation for the forthcoming move of the TX-O fromLincoln Laboratory to the Massachusetts Institute of Tech-nology Campus in Cambridge. Originally some of the DC powerwas generated from motor generator sets and all of thatsystem, along with it's power control, was transferred tothe TX-2, which was approaching the stage for preliminarytesting.

A second rack was furnished for the TX-O to house thenecessary power supplies for the Central Processing Unitlogic, the newly added T Memory and various peripheraldevices. Considerable space was required for the highvoltage supplies associated with the display system.

In addition, a two-stage power control was provided forturn-on and turn-off. At turn-on, a 60 second warm-up wasnecessary for the filaments of the vacuum tubes of the timepulse generators. At the end of that time, first stage DCpower was available and the time pulse chain had to bemanually started. At the same time, CPU logic power wasapplied. After the second stage, power is applied to theRead/Write and Inhibit current power supplies for thememory o

DC power is distributed via a fuse panel, mounted on thePower Control Rack, to the various sections of the machine.

DC voltages are sensed by a Voltage Monitor Circuit whichhas provision for adjusting the trip level of the allowableexcursion of each voltage. Cycle I voltages are monitoredseparately from cycle 2 and the later can drop-out or beheld off independently. The fuses are of the grasshoppertype, with an alarm actuating tail that crowbars theaffected power supply, via a resistor, to a ground bus whenany one of the fuses blows, thus tripping the VoltageMonitor. A short delay at the end of each cycle is providedto hold off the monitor until the voltages have been giventime to stabilize,

12

IN-OUT RACK

A third rack was also provided to house the time pulseequipment and the planned for Extended Input/Output Facil-ity, which will be discussed later, under that title.

CONSOLE

The TX-O Console is L-shaped, each leg eight (8) feetlong. This includes space for the Control Panel, ToggleSwitch Storage Panel, Marginal Check Panel and DisplayPanel. Table space is ample for the Flexowriter, the Photo-Electric Paper Tape Reader, common program paper tape filesand for the user's paper work.

13

PART II (at CAMBRIDGE)

INTRODUCTION

Late in 1957, the possibility of moving the TX-O to theMIT Campus was discussed and early in 1958, Prof. J. F.ReinJes, Chairman of the Ad Hoc Committee on Computation,with the support of the committee, took direct responsibil-ity for the future operation of the TX-0. Earl W. Pughe, Jr.was appointed to have the immediate responsibility for theinstallation and operation, assisted by John A. McKenzie,who at that time was transferred, full time, to LincolnLaboratory to gain experience with the computer and partici-pate in the transfer.

The TX-O was to be housed on the second floor of Building26, (the newly completed Compton Laboratory), in 9000 sq.ft. of area with expansion possibilities. This was primespace, on the same floor as the Research Laboratory ofElectronics Administrative Offices.

The agreement was to transfer the TX-O, on a long termloan basis, to the Electrical Engineering Department. Thefacility was initially ointly supported by the ResearchLaboratory of Electronics, the Electronic Systems Laboratoryand the Electrical Engineering Department.

Lincoln Laboratory furnished funds for the purchase andinstallation of a fifteen (15) ton air conditioning systemin the new location.

14

MOVE TO CAMBRIDGE

The decision at Lincoln Laboratory, to cooperate andfurnish support, set the tone for the transfer, which wentvery smoothly due to the excellent cooperation of everyonethere, whether directly or indirectly involved. Rememberthat this was a laboratory built machine, not constructedwith the idea that it would be portable, or even moved.Interconnections were not pluggable. Racks were bolted tobases, which in turn were bolted to the floor.

On July 1, 1958, the TX-O was shut down, and work beguncutting free all of the interconnections, which numbered inthe hundreds, of individually tagged wires.

By the end of the month the machine had been installed atCambridge, under the guidance of Robert Hudson of LincolnLaboratory. Complete checkout was delayed, first by the lackof cooling and second by the need to shut down while the airconditioner was being installed. Operation started to becomemarginal whenever the temperature exceeded 80 degrees F.,and there was a caution not to run the machine under thoseconditions.

However by the end of August, everything had been checkedout, and the users started to use the machine. The groupfrom the Communications Biophysics Laboratory, having previ-ously used the TX-O at Lincoln, had working programs and, ofcourse, were anxious to get back to reducing their data. Thegoal, to be ready by the start of the fall term, had beenmet.

15

EXTENDED INPUT/OUTPUT FACILITY- Addition of

Soon after operation began at Cambridge, the TX-O wasprovided with an extensive, Extended Input/Output System,designed by Benjamin Gurly, built at Lincoln Laboratory, andinstalled in the newly provided In-Out Rack. The idea was tomake available a very flexible input-output arrangement.Actually, there was so much flexibility inherent in thesystem, that a series of adapter panels were built, includ-ing switches and patchcords, to defeat some of the logicaloptions, and to make it easier for the user, requiring onlythe simplest kind of setup, to attach his equipment to themachine. At the same time, none of the defeat logic was hardwired, and the user having need of the full capability ofthe system, was still accomodated.

Data is exchanged, in both directions, between the users'external equipment and the Live Register (LR) of the TX-O.Data outputs are available from all eighteen (18) bits ofthe LR. Up to eight (8) External Commands or trigger pulsesare available to control the external equipment and/or toselect which device, attached to one of the six possibleinput connections, is to be read in. The configuration ofthe way that it is read in is programmable if desired.

An Epsco Datrac, eleven (11ii) bit A/D Converter, with asample and hold input, is mounted in the In-Out Rack. It'sinput is sampled and the digitized result read into the LR,under program control. The digital output can be patched, inany format, to the LR. A D/A Converter, looking at nine (9)bits of the LR, is permanently hooked up.

Since there is no interrupt system, external interventionis accomplished via various means. One, a Transfer on Level(tlv) instruction is available which senses the signal levelon it's input. Two, there are inputs provided where anexternal pulse may be used to set either the Light Pen orLight Pen 2 flip-flopso These are sensed by the light pen(pen) instruction, Three, the external instructions (exOthru ex7) cause the computer to wait for a completion pulse.Normally the output pulse is patched back in, to form it'sown completion, but it may be patched to an external deviceor delay, which in turn will be used to provide thecompletion pulse.

Time pulses are available for use, for example, tocontrol external synchronizers.

A complement of eighty (80) Digital Equipment Corporation(DEC) Building Blocks, whose logic levels are directlycompatible with TX-O levels, is available to allow the userto extend the computer logic, to suit his own needs.

The ease with which a user may append external equipment,such as his own special hardware, to the TX-O, with aminimum of hardware necessary on his part, has proven to be

16

__II___ II

a very powerful and popular attraction to the researchgroups. This aspect will be mentioned again, when thevariety of applications is discussed.

17

___ll______lpl____lI_-········�--Plil�· �_·�-__-_·YI__I-.llt-��·ll____ll.- _I_-__·pll- -----· _I I ---· I

FIRST YEAR AT CAMBRIDGE

The installation of the 4096 word core memory obsoletedthe old utility routines. John C Gilmore, Jr. and CharlesWoodward wrote a new utility program, titled Utility Tape-3(UT-3), with subroutines to do register examination andmodification, to do a memory word search, to print out alisting and to punch out paper tape. It occupied 1184registers, residing in the top quarter of core memory, Newprograms were typed in, on-line, using UT-3, since there wasno assembly program.

Late in 1958, a conversion program, Golux, was finishedby Lawrence Gitteno The users then prepared the symboliclanguage, alphanumeric coded paper tape on the off-lineFlexowriters, and using Golux, produced a binary tape.

Early in 1959, the new and much improved assemblyprogram, Macro, was completed by Prof. J. B. Dennis. Thisprogram was a major step forward, and will be discussedlater under software development.

Dating back to the start of operation at Cambridge, manyof the staff members of the Electronic Systems Laboratorywere interested in becoming familiar with the machine. Themost ambitious project of that nature was a demonstrationprogram, written by John E. Ward and Douglas Ross, simulat-ing a mouse, searching through a maze, hunting for a cheese.Other programs were documented and submitted as subroutines.

Later the Speech Group of the Research Laboratory ofElectronics, under Prof. Kenneth Stevens, became active onthe machine, studying ways to characterize speech. Among thenew users were groups working in picture processing, trans-mission bandwidth simulation, pattern and character recogni-tion. Several graduate students used the computer in connec-tion with their theses. (See applications section later, fora more detailed account,)

18

---- I I ~ _1 I --I Y·l l--U IIII_.--~-^I--·--- -- P--_-l

MODE OF OPERATION

The TX-O Facility was set up as a do-it-yourself opera-tion, with the user doing his own programming, tape prepara-tion and computer operation. The machine was available, toqualified users, on a round the clock, seven days a weekbasis0 The computer proved to be so reliable, as has beenstated before, that this type of operation was feasible, andthe users seldom experienced catastrophic type machinefailures during off-hours operation. If trouble occured andif time was tightly scheduled, the staff would respond atany reasonable hour. At worst, the malfunction was correctedearly the next day,

The idea of working on-line caught on immediatly. It waseasily demonstrated that a researcher could get a workingprogram off the ground in much less time, and secondly, theoutput could be monitored so that a useless amount of datawas not generated when it was initially apparent that theprogram was not fully debugged, or that some unforseen eventhad not been taken into account.

In addition to the Flexowriter, a large number ofinteractive facilities, such as, Toggle Switch Storage,Toggle Buffer Register, Toggle Accumulator and the Light Penprovides good means to change parameters in, or to direct, arunning program.

The manner of operation is very informal, no exact numberof hours is billed to a user. Records are kept only as a wayof knowing what percentage of time is utilized by thevarious groups or affiliations. That data is then used asbargaining power to obtain funds for capital expenditures,when the computer facility is expanded.

19

I� �_� __�__1_ _111_111_1_·1�_ -11111�--11_-_- � -I_.��- ----�·- - 11

MACHINE EXPANSION PHASE

During the summer of 1959, when Earl W. Pughe, Jroresigned, Prof. J B Dennis, of the Research Laboratory ofElectronics, accepted the appointment, by the Ad Hoc Commit-tee on Computation, to be in charge of the TX-O Facility.The usefulness and versatility of the machine was evident,and the Committee was receptive to his proposals for theexpansion of the computer. The only restriction being thatthe changes had to be accomplished with a minimum ofdowntime or inconvenience to the users. In order to helpexpedite the modifications, the group was increased tothree, by the addition of John T. Connolly, who transferredto the TX-O when the Whirlwind Computer was phased out.

One day a week had been reserved for scheduled mainte-nance, Since only a minimum of time was necessary for thatpurpose, that day was usually devoted to modifications. Dueto the modular, physical structure of the TX-O, i.e., eachgate was pluggable and each flip-flop used as a register bitor in control, along with its associated gates, was pluggedinto a removable panel interconnected with taper pin connec-tions, it was feasible to rework the control and registers,bit by bit, and have the computer usable at the end of eachday. In only a few cases was it necessary to take more time,on a scheduled shutdown, to cut-in new logic0

20

I�____�ICI_ III 1 _� -�-�LIIII.·I�-XCI�--·I� �I-----1II --�-^-·---·L·-·- II-I I --

T-MEMORY EXPANSION

The first major enlargement, in 1959, was to double thesize of the T-Memory from 4096 to 8192 words. Another MemoryAddress bit was used so that all of core memory is directlyaddressable. The T-Memory had been designed and packaged asa 64 by 64 array, 38 bits long, for the TX-2. It had beenfabricated for the TX-O with only 19 planes installed. Theadditional planes were purchased from RCA, Needham, and werewired into the stack by the Memory Development Group atLincoln Laboratory.

Another set of input gates was added to the Memory BufferRegister (MBR) to accomodate the Sense Amplifier outputs ofthe additional planes.

The new arrangement is to read out both the top andbottom halves of core memory using the common set of CoreDrivers. The wanted (addressed) half is gated into theMemory Buffer Register as before. The unwanted half is gatedinto the newly constructed Memory Buffer Auxiliary register(MBA), used only to provide the necessary information to theInhibit Drivers during the rewrite cycle.

The MBA was constructed using TX-2 type modules.

21

_�II� �_ _I II_ ______(_____I__L__I___l__�__l___ ·_*__-- - _

ORDER CODE ENLARGEMENT

The installation of the smaller capacity T Memory leftthree (3) unused Memory Address Register (MAR) bits aval-able. These were added to the Instruction Register (IR),increasing its length from two (2) to five (5) bits, toallow for the future enlargement of the order code. The newaddressable commands, expandable up to a possible twenty-four (24) instructions, were grouped into the store class,add/load class, and transfer class. The operate classcommands remained, decoded as a single instruction, toperform the majority of the arithmetic and logical opera-tions.

The first instructions, added in 1959, were reasonablyeasy to implement, in that they required no additionalregister transfer gates and only slight changes in control.The orders added were, Store Live Register (slr), Load LiveRegister (llr), and the Unconditional Transfer (tra), givinga new total of six (6) addressable commands.

The large scale changes, required to realize the FutureTX-O System Organization, were implemented piecemeal overthe period from 1960 to 1962. However the logical design, ofwhat was close to being the final system, was completed byProf. J. B. Dennis in 1960, and the new logic was simulatedon the TX-O with a program written by Robert Wagner.

The procedure chosen to implement the proposed changeswas to update any register bit, or control, plug-in-unitmounting panel completely, even though only a fraction ofthe changes would be used at the time of the rework. Thatmeant providing, at that time, all of the components andwiring necessary to achieve the future machine organization.

The design considerations were more than deciding whatwould be a desireable order code. There were many physicalconstraints imposed by hardware and space limitations. Insome cases, new type plug-in-units were designed in order togain a tighter density of packaging. The TX-O Group con-structed all of the additional CPU logic modules, since thatstyle was not comercially available.

22

_I _ _� _ __ _Lll___ _I_ _1__��1__ __111^_�_�� _�_�1_�·�_-------�--L-II ----I C- I

The most important addition was an Index Register (XR)consisting of a sign and thirteen (13) binary digits Inorder to perform effective address calculation and indexaddition, a new Program Counter (PC) with a full addingcircuit, was constructed, using TX-2 type modules. The oldprogram counter was then reworked to become the IndexRegister (XR). Some of the indexed instructions requiredthree machine cycles, so this necessitated changes in themachine cycle control. Input transfer gates were added tothe Memory Buffer Register (MBR) for the XR-)MBR data path.

Shift right circuitry and new input gates were added tothe Live Register (LR) to accomodate the future MagneticTape System. Also added, were new gates to transfer the LR,zero sides, to the MBR, to allow for a Jam transfer.

A Program Flag Register (PFR), six bits long and expand-able to ten bits, was installed. This is used as a senseregister, setable and readable under program control, viathe Memory Buffer Register and the Accumulator. Inputs tothe PFR are brought out to the In-Out Rack for use withexternal equipment.

The Operate Class Commands, operate micro-orders, re-quired some reordering of bit combinations and the timepulses on which the events occurred. This change onlyaffected nine basic orders, although used in many morecombinations, and was necessary in order to work the IndexRegister transfer to and from the Memory Buffer Registerinto the system, A restore tape, providing the new defini-tions, was available to the users for use in reconvertingtheir tapes. All other changes were additions to the ordercode and did not affect existing programs.

23

-· I --- -------- ·I ----------------�II �-- - -- �- -.-- �--

DIGITAL MAGNETIC TAPE SYSTEM

A Potter M 906, MK II Tape Transport System, along withthe associated Drive Electronics and Record-Playback Ampli-fier System, was added to the TX-0 in 1961. The controllogic, capable of handling up to three transports, wasdesigned by C Gordon Bell, who was with the TX-0 Group atthat time. The system was constructed using DigitalEquipment Corporation (DEC), 4000 Series System Modules.

The information is stored on the tape in a format whichis compatible with the IBM seven track system, two hundredcharacters per inch, which was then the standard. The non-return to zero system of recording is used, and informationmay be written in the binary or binary-code-decimal (BCD)mode. Information is transferred to and from the TX-O, viathe Live Register. Since there is only a six bit read/writebuffer in the control logic, the 18-bit word assembly isdone in the Live Register. The transfer takes place duringthe execution of a copy (cpy) instruction, which is used tosynchronize the transfer. Program timing is automatic sincea copy (cpy) is held up until the previous (cpy) has beenexecuted. Four binary digits of the Program Flag Registerare set by the tape control logic, and sensed under programcontrol, for tape error conditions,

Final checout and debugging of the tape system wasstarted by Robert Spinrad and completed by NatalioKerllenevich, who participated while working as researchassistants.

24

SOFTWARE DEVELOPMENT

Starting with the early work by Gilmore, covered in PartI, there has been an interesting history of softwaredevelopment on the TX-O. The utility program (UT-3), byWoodward of Lincoln Laboratory, was discussed in thesections First Year at Cambridge.

The new assembly program, Macro, written by Prof. J. B.Dennis in 1959 as an extra curricular activity before hebecame officially associated with the TX-O, was consideredto be an outstanding piece of work, it being at least asflexible as assembly programs written for computers havingmuch more storage. At that time the TX-O had only 4096 wordsof core memory and four decoded instructions. In addition tothe use of symbols to represent addresses and the use of themnemonic form for all instructions, Macro provided twospecial features. The user could define, as a macro-instruction, any sequence of instructions which appearedseveral places in his program, thus saving typing effort.Provision was made for automatically storing constants, thussaving storage space. Macro handled integers in either octalor decimal form. Symbols could be up to three characterslong and either fixed or symbolic addresses were handled.Symbols could be defined and symbol values could bereassigned. A total of nine (9) pseudo-instructions wereincluded. An informative printout, at the time of conver-sion, listed all error stops and indicated the type of errorthat was encountered. Following the punching of the binarytape, a Macro Symbol Punch could be requested, whichproduced a binary symbol tape, of the symbol table from theconversion.

Macro underwent several revisions, including Macro IIAand Macro III. New features included the ability to nestmacro-instructions and to nest constants within constants.The new pseudo-instruction "variables" , automatically allo-cated storage for each distinct variable. The pseudo-instruction, of the form, "repeat n,x", caused the quantityx to be repeated n times in the program. The pseudo-instruction "text" gave the user the ability to quote anarbitrary string of characters as text. The pseudo-instruc-tions, relocatable", "entry" and "exit", were used inconnection with programs to be assembled in relocatableformat. The work on Macro III was done by Robert Saunders.

25

A utility program, Flexowriter Interrogation Tape (Flit),for use as an aid in on-line debugging, was written by ProfoThomas G Stockham and Jack B. Dennis in 1959. This was doneas a voluntary contribution since neither was on the TX-Ostaff° Flit was programmed for use after the memory expan-sion to 8192 words of core memory, and occupied the upper2500 registers. The largest single advantage gained was thatthe user could now type in and reference memory using,whenever he desired, any of the three character tags in hisMacro symbol table as well as any additional ones that hemight wish to define. In addition to the usual registerexamination and modification, the extra feature of colorcode was used to indicate what was typed in by the user(red) and what was typed out by the computer (black). Thismeant that any modification, in a long list, was easilyspotted. The feature of doing breakpoint testing was intro-duced, This allowed the user to insert up to four break-points in his program, At the time a breakpoint is encoun-tered, the state of the Accumulator and of the Live Registeris typed out. Then the user continues by doing a "proceed".This idea proved to be a great convenience in debuggingprograms and a tremendous improvement over the previousscheme of stepping through a routine at the console, asingle cycle, or at most a single instruction, at a time.Flit incorporated the idea of using some single characteroperators or commands. Also provided was the ability tocontrol the print mode and the radix mode. Flit interpreteda vocabulary of twenty-four (24) pseudo-instructions to setthe mode and twenty-nine (29) single character controlcommands determined the action,

26

In 1962, when most of the proposed improvements werecompleted, new software was written making use of theenlarged order code and the magnetic tape system.

A new and improved assembly program, Midas, was writtenby Robert Saunders. He had been active on the TX-O as anundergraduate and joined the TX-O staff after graduation°Among the new features, Midas recognized symbols up to sixcharacters in length. Much more flexibility in the use ofmacro-instructions was provided, such as, the use of recur-sive macro definitions, the use of pseudo-instructionswithin macros, the ability to create symbols to solve theproblem of address tags within macro definitions, and theuse of the "garbage collect" feature to recover space when amacro is redefined. Midas stores macro definitions byremembering the exact representation in text form. When amacro is called, arguments are "plugged" into the macro bodyin text form. The entire macro expansion is then supplied tothe assembler proper as if it had occurred in the sourceprogram. The pseudo-instructions "lif" "Oif", and thequalifiers, "vp" for value positive and vz' for value zeroare provided, useful particulary in macro-instructions, totest an expression and to condition part of an assembly onthe result of the test. The location counter may be set, oroffset, during an assembly, to the value of an expression.The pseudo-instructions "equals and "opsyn" are used to givea symbol a logical or operation value. The Midas on-linefeature enables the user to make simple corrections, via theon-line Flexowriter, to an assembly without preparing addi-tional tapes off-line. With all the features, Midas inter-preted twenty-eight (28) pseudo-instructions and over onehundred (103) instructions, i.e., including the commonlyused combinations of the operate class commands. Much of thetesting and debugging of Midas was done by Robert Wagner.

Doctor (Midas Debugger) is a symbolic debugging programoccupying the upper one-quarter of core memory, written byAlan Kotok while an undergraduate, and used as a replacementfor Flit, when Midas was introduced. The symbolic syntax isthe same as that of Midas, and Doctor uses the symbol tapepunched by Midas. Doctor uses single character operatorsexclusively, for control commands and the ever presentarguments prevailed as to, what constituted a reasonablecharacter set and meaningful operators.

27

With the completion of the new software and the magnetictape system, Alan Kotok and David Gross wrote a UtilitySystem that has a copy of Midas and Doctor, residing betweentwo loadpoints, at the front end of the reel of magnetictape. Either of these programs may be called by reading in ashort (4 instruction) paper call' tape. This system has thefeature of requiring only a single pass of the paper tapeduring an assembly. The second pass makes use of theinformation stored on the magnetic tape during pass 1. Atthe time of assembly, the binary copy of the program may beread into core memory from the magnetic tape or punched outonto paper tape. Several options are available to the userby the various settings of the Test Accumulator(TAC).

This system was updated by John Currano in 1966, with thefacility to store "dump" any area of core onto magnetic tapeand read it back at a later time, under control of Doctor.The information is stored on the tape in a format that canbe read by the IBM tape units, thus providing a convenientmethod to exchange data.

An English Text Editor (Acorn) was written for the TX-Oin 1967 by Eric Jensen. This is a TX-O version of theExpensive Typewriter (ET) used on the PDP-1,

It should be noted that practically all of the systemsoftware mentioned was done on a voluntary basis by usersnot on the TX-O staff at the time their programs werewrit ten.

28

A PPLICATIONS

There have been a wide variety of applications on the TX-Oo As mentioned earlier, the first research group to use theTX-0 was the Communications Biophysics Laboratory (CBL),under Profo Walter Rosenblith, of the Research Laboratory ofElectronics (RLE). The group used the computer to aid in theprocessing of electrophysiological data gathered from theauditory cortex of a cat's brain. The computer processedoutput was usually in the form of a histogram plotted on thedisplay scope and recorded by a Polaroid camera. Active inthis work were Prof. Thomas Weiss, Prof. William Peake,George Gerstein and Charles Molnar.

The Speech Group of RLE, under Prof. Kenneth Stevens,used the TX-O for several years exploring efficient ways ofcharacterizing, and the computer recognition of, speechsounds. The initial work, done by Prof. John Heinz, was toget the frequency spectra of speech sounds into the TX-O indigital form and to plot this information as a curve on thedisplay. Programs were written by C. Gordon Bell to investi-gate the poles of the frequency function and to providemanual and automatic means of having the computer fit acurve to the given curve, leading to automatic identifica-tion of speech sounds.

The Cognitive Information Processing Group (CIPG) of RLE,under Profo Samuel Mason, began using the TX-O in 1958 Anumber of theses on transmission bandwidth simulation andpicture coding were completed. The binary data tape had sixbit reflectance values of the sample points along scan linesof the picture. The transmission was simulated using one ortwo bits of information and image enhancement techniqueswere developed to upgrade the quality of the processedpicture. The Polaroid camera attachment to the display madeit possible to record and easily evaluate the results of thevarious methods tried. The earliest work was done by JamesCunningham

The Sensory Aids for the Blind Section of CIPG was activeon the TX-O over the years 1964 to 1968. The objective wasto develop a reading machine for the blind that could bebuilt for a modest sum. The first phase of the effort was aDoctoral thesis by Jon Clements, under the direction ofProf. Donald Troxel, studying various algorithms for recog-nizing the sample characters scanned when interposed betweenthe display and a photomultiplier tube viewing the raster.The initial output was spoken letters or spelled speechgenerated from stored data. This was the work of Dr. KennethIngham. Later Thomas Barnwell, under the supervision ofProf. Francis Lee, investigated computer synthesized speechas an output for the machine. On the basis of the progressmade on the TX-O a scanner was constructed, and operatedunder control of the PDP-1, that handled and could read awhole page of text. The activity was transferred from theTX-O when the group obtained their own computer.

29

The Laboratory for Nuclear Science (LNS) had a studyunder the direction of Prof. Martin Deutsch, in 1959, to usea computer to analyze bubble chamber data pictures, Thedevelopment was begun by Adam Boyarski on the TX-O. The datainput was recorded on a roll of 35mm film that was passed,initially about 15 or 20 frames per second, between thedisplay and a potomultiplier tube, called the light gun,which viewed the full raster of the display. By displayingall points and using the light gun to check the transmissionof light through the film, data could be read into thecomputer and processed using pattern recognition techniques.One of the methods used was a Master's thesis by PaulMermelstein in 1960, who progranmmed the TX-O to detect andlocate several well-defined patterns in pictures of theparticle track. Previous to that time, operators hadmanually reduced the data, a slow and tedious Job in whichthe event of interest sometimes occured only once in athousand frames. Following the successful start on thiswork, the grouped purchased their own PDP-1 computer andcontinued to develop more highly sophisticated methods offilm reading.

30

~- · ^1~1 - 1

Lawrence Roberts in 1958 as a seminar project, modeled anadaption of Rcsenblatt's perceptron, which he had never beenable to test experimentally. The simulation, to have thecomputer "learn" to recognize the difference between two ormore characters, showed only 60 per cent recognition.However with modifications developed by Roberts, the per-centage increased to better than 90 per cent. The TX-O wasideally suited for this type problem because the inputcapability of the light pen was unique at that time.

The frequency analysis of a human postural reflex wasundertaken as a Doctoral thesis by Avery Johnson of the RLENeurophysiology Laboratory in 1960. The torque applied bythe flexor muscles tending to hold the wrist angle fixed andthe resultant angle, were measured and recorded on magnetictape. The TX-O was employed to produce a cross-correlationfunction between the two signals and an auto-correlation ofeach separately.

William Daly, in 1960 as part of an MS thesis,programmed the TX-O to play Cubic, a three dimensional Tic-Tac-Toe. At the conclusion of the game the player was ratedand the computer strategy for the next game was based on theusers current rating. When the program lcok-ahead discoveredthat the computer had won, the play was taken away from theplayer and the computer listed sufgested moves for him.Anything other than the suggested moves would allow thecomputer to win sooner.

A Masterts thesis by Peter Katona in 1961 involed theexperimental determination of the relationship between heartrate and blood pressure, and was titled "Analysis of BloodPressure Regulation Using Correlation Techniques". The data,recorded on magnetic tape, was easily digitized and inputedby the TX-O.

31

_ _i I r___I _II II __1__1_ IPI____Y�L�_IIII__^------ -l I · II�-�---

In 1962 Henry Ernst worked on a Doctoral thesis using thecomputer to control a mechanical hand that was equipped withhalf a dozen sensors and made use of a large portion of theinput-output capability, including the D/A and A/D convert-ers in the closed loop of the computer and the hand.Operated by commands typed in via the on-line typewriter,the hand would grope along the platform hunting for a blockor a box. It could differentiate between them, could stackthe blocks upon each other or it could deposit the blocks ina box. The program incorporated an awareness concept so thatif the hand experienced a situation that it had notencountered before, or that had not been specified in thecommands, it could initiate some reasonable alternativeaction.

During the period, 1960-61, several theses in connectionwith handwriting analysis were done on the TX-O. Oneapproach was an attempt to recognize the handwriting bymeans of cross-correlation with a set of standard stokes.

One thesis involed measuring aspects of previously pre-pared handwriting samples of subjects known to have muscularcoordination disorders. This data was read in using thelight gun or the light pen. The objective was to study thepossibility of evaluating the signatures on life insuranceapplications.

Peter Sampson worked with various means of giving thecomputer the description of a piece of music. The computer" layed back" the music by changing the state of a flip-flopat the frequency of the desired pitch of each note. Anamplifier was attached to the flip-flop as output, The musicprogram was first written to handle one voice, then later tohandle three voices and three flip-flops whose outputs weremixed as the input to the audio amplifier.

The TX-O has been seen on television several times. Thefirst appearance was a local show on WGBH-TV showingcomputer demonstration programs. Later the same sort ofthing was done for the Canadian Broadcasting Company withthe commentary done in French. The big show was a sponsoredprogram on the national CBS Network with the TX-O writingthe script for a TV Western. The program, called Saga, setthe scene and detailed the actions of the robber and thesheriff, kept track of the gun, the shots fired, hits andmisses, and governed the rationality of the actors action bythe number of drinks that each had consumed from the bottleof liquor on the table. Each computer run was a completelynew and unpredictable skit, Saga was programmed by DouglasRoss and Harrison Morse of the Electronic SystemsLaboratory.

32

_I__ ·_ __11__1_ 1_^·^�1 _�_l^__ll__llll__·g____-.·--�YIL^···IPII I--�-�_-.. l --- -

The Radio Astronomy Group of RLE used the TX-O in thespring of 1970 to aid in the processing of a large amount ofpulsar data stored on magnetic tape. The procedure is a wayof getting an early view of the data by displaying it in aseries of intensity coded histograms, recorded on film, forevalution to determine what might be of interest for furtherprocessing.

Most of the applications mentioned have made use of someof the unique properties of the TX-0. During the same periodmany other projects and theses were handled in areas such asinformation theory, computer simulation, network synthesis,control systems, and in 1959, the simulation of a servosystem with the driving and error function computed digit-ally.

During the same period the TX-O was used in connectionwith introductory computer courses and many students usedthe machine for formal class work.

The TX-O history would not be complete without somereference to the computer "hackers"t These were students whowere highly motivated in learning all that they could withrespect to the computer, and in the process, spent longnights and weekend sessions on the machine, exploring theirpet projects, usually not connected with anything foracademic credit. Much of the good software and innovativeideas were contributed by the "hackers".

33

- -I I- I ~ -·- ·---- ------ -^1·~ . 1- · llll~r~r------~. ------~L--- I---·W- I

TIMESHARING (PDP-1)

The advantages gained by providing a researcher with thefacility to work on-line had been clearly demonstrated.While this procedure greatly reduced the delay usuallyencountered from the conception of a programming idea to thecompletion of the operating program, the inefficient use fthe computer's resources in that mode was also apparent. In1961 Prof. J. B, Dennis wrote a proposal describing atimesharing operating system for the TX-O.

However in September of 1961, the Digital EquipmentCorporation (DEC) presented a PDP-1, serial number 3, to theElectrical Engineering Department and the machine was oper-ated as a part of the TX-O facility. The need for acormmercial computer, oriented to scientific applications,was shown by the ever expanding range of activities pio-neered on the TX-O. The people at DEC, having built the TX-Oat Lincoln Laboratory, started with a strong background ofexperience in this field.

The PDP-1 was better suited as a base to use to implementa mixed hardware/software timesharing system. A Master'sthesis by John Yates in 1962, under the supervision of Prof.Dennis, became the starting point for the timesharingsystems that have continued to evolve on the machine sincethat time.

The TX-O has a parallel 18 bit, two-way data link withthe PDP-1 and information may be exchanged while timeshar-ing.

CONCLUSION

Since 1962, just about all of the group's efforts andfunds for capital expenditures have been devoted to enlarg-ing the capability of the PDP-1.

The TX-O continues to serve, has over 49,000 hours ofrunning time, requires a minimum of maintenance, and isoperated as a part of the Electrical Engineering /ResearchLaboratory of Electronics PDP-1/TX-O Computer Facility.

Early in 1975 the TX-O is going to be moved to a ComputerMuseum being set up by the Digital Equipment Corporation inMarlborough, Massachusetts.

34

__ �L� 1___1__� _ 1_1 _� __ 11 _ _

A CKNOWLEDGErENT

In this condensed form it has not been possible to morethan briefly sketch the topics included. The fact that manygroups and the majority of the users have not been mentionedin no way reflects upon their contribution to the activity,but was necessitated by lack of space. Names have been usedchiefly to lend authenticity, so that interested partiesmight delve further into the developments covered.

Whatever success the TX-O has achieved during its yearsat MIT is due in no small part to the ready support andencouragement given to the TX-O group by the administrativepeople directly involved with providing the means to improveand maintain the installation. Equally important was theirwilliness to sponser the very informal open-shop type ofoperation, making the computer easily accessible, on justabout the best conceivable terms, to researchers andstudents.

Ralph A. Sayers, the Assistant Director of the ResearchLaboratory of Electronics, was always the person to call onwhen advice and action were needed concerning administrativepolicy decisions and problems, and he in turn was supportedby the Director, Prof. Henry Zimmerman.

In the Electronic Systems Laboratory the Deputy DirectorJohn E. Ward and the Director Prof. J F Reinjes wereespecially active during the early years in setting thedirection and pattern that the new type facility was tosucessfully follow.

During most of the time covered, the direct contact withthe Electrical Engineering Department was the ExecutiveOfficer John A. Tucker, who, along with a succession ofDepartment Heads, Prof. Gordon S. Brown, Prof. Jerome EoWiesner, Prof, Peter Elias, and Prof. Louis D. Smullin, werealways sensitive to the group's interests.

35

_ _ � _ �I 1__�1_ � _X__ _I�·__ �I I_ ·_ 11�--�1�---· --1- - - �- I _-_

BIBLIOGRAPHY

Gilmore, J T. Jr., and Peterson, H P, "A Functional Descriptionof the TX-O Computer", L L Memo. 6M-4789-1, (Oct. 3, 1958)

Jeffry, R C., "Transistor Logic in TX-O", L. L Memo. 6M-4571,(Sept. 5, 1956)

Fadiman, J. R., "TX-O Circuitry", L. L. Memo. 6M-4561, (Oct. 22, 1956

Gilmore, J. T Jr., "TX-O Direct Input Utility System", L. L. Memo.6M-5097-1, (Oct, 3, 1958)

Mitchell, J. L., and Olsen, K. H., "TX-O, a Transistor Computer witha 256 by 256 Memory", Proc E J. C C, Special Publication T-92

Eckl, D. J., and Fergus, P. A., and Burke, R. L, "Transistor LifeExperience, 1954-1958", L. L. Technical Report No. 199, (Mar. 26, 19 =

Eckl, D J., and Burke, R. L., "Transistor Life in the TX-0 Computerafter 10,000 Hours of Operation", L. L. Technical Report No. 221,(Apr. 14, 1960)

Clark, W. A. et al, "The Lincoln TX-2 Computer", L. L. Memo. 6-4968,(Apr. , 1957)

Pughe, E W. Jr., "Preliminary Programming for TX-O", TX-O Memo.M-5001-1, (July 23, 1958)

Pughe, E. W Jr., "Words Recognized by UT-3", TX-O Memo. M-5001-2,(Oct. 16, 1958)

Pughe, E. W Jr., "Operation of the TX-O Computer and the Use of GOLLa Symbolic Conversion Program", TX-O Memo. M-5001-5, (Dec. 4, 1958)

Pughe, E. W Jr., and Dennis, J B, Use of the New ConversionProgram, MA CRO", TX-O Memo, M-5001-5, (Mar, 26, 1959)

Dennis, J B and Spinrad, R. J, "MA CRO IIA", TX-O Memo. M-5001-5-1(Apr. 7, 19613

Dennis, J. B "Programming for the TX-Ot", TX-O Memo, M-5001-13-1,(Oct.26, 19605

Roberts, L. G., "Some Useful Micro- and Macro- Instructions",TX-O Memo. M-5001-19-1, (Aug. 23,1961)

Dennis, J. B., "A More Powerful Order Code for the TX-O", TX-O Memo.M-5001-22, (May 23, 1960)

__ __�___ L ·1I·-I·--1I__I_ LII_.._�_·LIPI·_CII-X···II�-�XY i.- - ~ ~ ~ ~

Stockham, T G Jr., "FLIT- Flexowriter Interrogation Tape, A SymbolicUtility Program for TX-O", TX-O Memo. M-5001-23, (July 25, 1960)

Dennis, J B,, "The TX-O Instruction Code", TX-O Memo. M-5001-27-4,(Sep. 13, 1965)

Bell, C G, "Programming for the Magnetic Tape System", TX-OMemo. M-5001-28-1, (Nov. 16, 1962)

Dennis, J. B., and Hall, M., "TX-O Subroutine Library", TX-O Memo.M-5001-31, (Dec. 28, 1960)

Dennis, J. B., and Wagner, R. A., "Utility Program Flit Jr.", TX-OMemo. M-5001-32, (Mar. 24, 1961)

Dennis J. B., and Saunders, R. A., "Preliminary Operating Instructionsfor 4MACRO III", TX-O Memo. M-5001-19-1, (July 24, 1961)

Saunders, R. A., "The Midas Assembly Program", TX-O Memo. VM-5001-39-1,(Aug. 22, 19664)

Kotok, A. , "Doctor", TX-O Memo. M-5001-40, (Feb. 26, 1964)

Jensen, 0 , "A corn , TX-O Memo. M-5001-44, (Aug. 22, 1967)

�______i�_l 1 �__11_11 1-.11�11�-- ·-L�-·l -· --- ---- -- L -