Computer Pioneers: Pioneer Computers Part 2

you hi I'm Gordon bell here again at the computer Museum in Boston in part one of this series we looked at the pre-war machines that led to the modern stored-program computer now in part two we'll look at a remarkable 10-year period it started with one of a kind laboratory computers it ended in the birth of a computer industry presper eckert and John McLean met at the Moore School of Electrical Engineering in Philadelphia and their collaborated on the design and construction of a large-scale digital computing machine that machine was the electronic numerical integrator and computer ENIAC

and it changed the history of the world as far as computing is concerned the story begins with presper eckert recalling those days anything you can do in software you can also do in hardware or you can do in some intermediate round called firmware or by micro steps or by whatever it's a it's a problem of the designer just to find out what blend of those things gives the best overall mix we had a terrible problem with the same type here which is what caused me to think of the internal programming idea I thought of this

idea proposed it long before of annoyance or damage to stuff by the way the reason I thought of this is that I said well if we ever build another one of these machines and and when we don't just have the ballistics problems and the internal ballistics problem in a few specialized problems that we knew about to do how do we decide how many plugs to put over here for programming how many counters or flip-flops or something to put in the memory how many switches to put over here with resistors because that's going to be different

for each class of problems there are some narrow classes where it's pretty much the same and those are the ones we were catering to in this machine but how do I do it when the problems are wildly different well I said I've got to find a way of generalizing this I said everything in science that ever amounted anything was because somebody figured out a way of taking some specific problem and generalizing it so that what I said was that we need a memory device that's cheap enough that we can use the memory device for everything

at least at one speed level we may need punch cards or tapes or something at some other speed level but at least at the main speed level of machine we need a memory device which can hold instructions which can hold constants which can hold variables which can hold information on what has to happen next and so on I wrote a memo saying how we could put these information all on these magnetic always spinning disks with magnetic edges then I decided spending discipline magnetic etches are really too slow to match the speed of electronics and I

had invented for a radar purpose for some other purpose they make a mercury tank device which are useful for timing purposes and for some other purposes in radar a cou sztyc device in which you can store information for a limited period of time but it dies out after a millisecond or whatever the length of the tank determines and I thought of repeating that the reason I'm size the Machine grew was originally it was uh supposed to be a numerical integrator to solve just ballistic trajectory nothing else and Carl Gillan who was the contracting officer in

the Pentagon in his wisdom when he renamed this thing called it an electronic numerical integrator and computer and he said he put the word and computer in here because he knew we might want to make extensions to this and he didn't want the General Accounting Office say why did you allow extensions on this thing so if you put it in the name that it's going to include more than the base thing to start with you avoid that problem politically if you understand and so his wisdom he did that and it turned out that that it

happened this way the ballistics were Tibor Tory except we we want to not only interested in what happens to a shell after it's come out of the big rifle or a gun but we want to know how it's doing as its traveling down the barrel which is called the internal ballistics problem and we have some Montecarlo problems involved in our work and we have some other problems that we want to look at so some of those we were able to do and some of those we couldn't do but we tried to amend the machine in

such a way as to take on these additional requests for ability as time went on this panel looks pretty much as I remembered except these lights that were added since I last saw it I don't know what they're for but just normally to start the Machine one press this button here and that turned on the heaters of the machine and it also turned on the DC voltages now we could also turn the DC voltages off and have only the heaters on which we frequently did during testing because turning the heaters off and on do the

thermal expansion of the tubes frequently Blewett tubes in fact if you turn the Machine off and on once you could almost count on it having one blown tube where otherwise the machine would go for one or two days at a time without a blown tube this button over here is the initiating switch this this is the button that gave you the first pulse that started the program into operation now normally we didn't do it by pressing this button there was a button on a cord a portable button they could be plugged into one of the

trays that carried wires around the machine and you could actually work this initiation from anywheres around the machine and that's because lots of times you're over testing a particular panel you didn't wanna have to run halfway across the room which is 50 feet long and press this so you just press the plug listing in where you were and press it there to start the machine again we designed this circuit first before anything else I've we tried six different circuits out and finally modified one of them and picked this one is the best one we could

find it had far more ability to count accurately and withstand changes in voltage and park values than any other circuit we could find we nevertheless found it was critical of the shape of pulses that were fed into it and so we designed a special circuit called the pulse shapers which was found in these next three tubes here now going to the other end going to the other end there are six tubes down here which comprise a transmitting device for plus numbers and a transmitting device for negative numbers which are controlled by this counter there are

a number of tubes as you see intervening here we have I think there are 28 to band we've only accounted for 13 and 6 or 19 of them the rest of these tubes in here are primarily concerned with carry over mechanism in other words with remembering when two numbers add up to more than the nine and we have to carry it over and put it into the next channel and to do this in such a way that the process isn't slowed down the ENIAC project concluded with the design and eventual construction of edvac the draft

edvac report by john von neumann is the seminal description of the first stored-program computer in 1946 penn held a summer school about it back it was attended by many who would build computers this story is about three of the branches forming the story programmed computers that came from the ENIAC project the first branch is Cambridge University's ed Zack Morris Wilkes created the world's first operational computer it helps stimulate the formation of a British computing industry the second branch is the Institute for Advanced Studies architecture computers a dozen or so machines were made at various laboratories

using their basic design others including IBM built variants the final branch is the Eckert mockolate computer corporation that ended up as the Remington Rand UNIVAC division of Sperry Rand it produced univac one the first successful u.s. commercial computer to be sure Eenie act influenced other efforts but this is about the direct descendants Arthur Burks who worked on any act narrates a film compiled from clippings left over from a newscast pres Eckert is on the left and John Mark Lee on the right K McNulty has just brought them a paper I'm on the left and Herman

goldstine is on the right in front of the three racks of the high-speed multiplier from the top down you see wiring the digit trunk the control panels and the program trunk the diagonal lines are the outputs of the shifter circuits for each successive digit of the multiplier the partial products were shifted one more position to the left after the program was worked out on paper it took a couple of days to set up the machine wherever possible we mounted circuits on removable modules we called plug-in units Homer Spence has removed a defective plug in unit

he replaces it with one known to work correctly Goldstein is reading numbers and Spence is setting them on the switches this is where the resistance table for a shell was store that table gave the resistance of the air to the shell as a function of the shells velocity a problem was started at the initiating unit but most of the equipment on this unit was for testing these meters are for reading any of the approximately 70 direct current voltages here are some display lights which are operated by the vacuum tubes below the display panel was made

for this film since little neon lights on the machine itself were not bright enough to be photographed these lights seemed important to the film crew they felt the public could understand visual communication between man and machine better than communication by punch cards all electronic counters had small neon bulbs to show their state and the same was true of flip-flops this display board shows how an accumulators lights might look during addition or subtraction when the ENIAC computer the trajectory one could see the abscissa value gradually increase as the shell traveled out and the ordinate value go

up and then come back down as the shell rose and then fell in the public demonstration of the any act we computed the trajectory of a shell that took 30 seconds to reach its target a young woman whose mechanical desk calculator spent one or two days calculating such a trajectory the ENIAC computed the trajectory in only 20 seconds faster than the shell itself traveled this convinced the Colonel's in generals present that electronic computers were important ENIAC was moved from pan to Aberdeen's ballistic research laboratory and operated until 1955 clearly ENIAC was an amazing engineering accomplishment

it was dramatically larger than any other system using vacuum tubes this Rack was one of 40 panels ENIAC was designed and built by a dozen engineers working two years this figure gives the memory size plotted against operation rate for these machines Wilkes stated when the design work of ENIAC was finished and the construction was in progress Eckert and Mauchly had much time to think about future developments it did not take them long to realize that the potential existed for building much more powerful electronic computers they would also be much smaller in scale their thinking was

advanced when John von Neumann acting as a consultant to brl began to pay visits to the group so now we begin the story of where the stored-program computer we know and love God invented the ENIAC team as well as other machine builders recognized that the time-consuming process of setting up problems to control computation had to be solved it's what we now call programming any act was more general than a numerical integrator its function tables could store instructions and emit control pulses to the function units it had conditional branching subsequently a special function table memory was

built that could also take in instructions from punched cards in September 1944 John von Neumann of the Institute for Advanced Studies at Princeton was appointed consultant to the ENIAC project he had learned about the control problem a year earlier in England one task he took on was to select a permanent code for the function tables this became known as the von Neumann code for the ENIAC however Eckert made it very clear that the name was not chosen because von Neumann discovered the stored program concept but only because he had selected the instructions a second but

related problem with the early calculators was that the instructions were held by paper tape punch cards and plug boards recall both Aiken and Zeus control their calculators with these tapes these slow devices were a poor match for electronic speeds at a 1948 conference ma CLE commented that calculations can be performed at high speed only if instructions are supplied at high speed Eckert and MOC Lee's new machine was called edvac standing for electronic discrete variable automatic computer it was to have an electronic memory that could keep up with the arithmetic circuits and solve the control problem

it had only 4000 vacuum tubes and a 44 bit word length it vac was built at Penn and moved to Aberdeen's brl in 1950 but the physical machine is incidental to computing john von neumann published the first draft of a report on the edvac on june 30th 1945 this report outlined the computer architecture that remained pretty constant for the next 50 years von neumann's edvac report describes five elements of a stored-program computer central arithmetic see a central control CC memory M input I and output Oh interestingly the report used human analogues like neuron and memory

to describe parts and processes the central idea of the report was to use a single high speed memory to hold both numbers and instructions Wilkes reading of the evidence was that Eckert had arrived at this idea before one Neumann visited the group von neumann's name gave the clear credibility to crystalize Carrie and perhaps implicitly claimed the concept Wilkes further states that Eckert began his conviction that the ENIAC was more complicated than it needed to be and that it could have been simplified by more rationalization of function in particular he identifies three quite different kinds of

memory flip-flops in accumulators function tables or read-only memory and interconnecting cables with their associated switches used for setting up the program in September 1945 Eckert and Mauchly proposed edvac based on the stored program set their report stated that an important feature of this device was that operating instructions and function tables would be stored exactly in the same sort of memory device as that used for numbers at the 1976 international research conference on the history of computing held in Los Alamos MOC Lee wrote Eckert and I were planning on stored programs long before von Neumann had

heard of the edvac project here's MOC Lee at Los Alamos describing their effort the ENIAC was the only existing electronic fast large digital computer why it was not just that it had an advantage over things it had a practically in a an incomparable advantage because of how else would you get many of these problems done and so it was quite a shock to me the other night in New York to hear dick Clevenger talk about what a terrible experience they had after the machine been moved to Everdene and the downtime was oh well over 50%

each morning when we came in and turned the machine on a lot of things went wrong we had long before decided from our own experience in more school that you shouldn't turn it off and then turn it on you should keep the filaments at any rate the heaters in those tubes all of that constantly burning and not go that through that disastrous heating cooling cycle which not only might burn out cathodes heaters but might interfere and destroy with insulation and cathode sleeves and other things which any one of which would cause bad problems and by

the way we were doing these things they we would use less tubes if we use the decimal system than if you used a binary system was that the reason we use the decimal system no it wasn't the that was just a secondary reason the primary reason was we knew the world wasn't ready for the binary system yet it wasn't that we were ignorant of the binary system because we knew the whole machine was binary anyway inside it's just where do you convert and so the indicator lamps and things that the operator saw were decimal one

of the wise was why did we program it the way we did it's pretty stupid way you know to have to run around and plug wires in have to turn switches and so on to convey the instructions of a problem to that computer well depends on how cheap storage is of course we realized that storage comes at different prices for different things if you want it fast it was more expensive you want slow you can have lots of it and so some people ask well what was the long-term storage now how big was it the

long-term storage of Christmas is infinite if you punch out cards why nobody cared except somebody who's paying for the warehouse put back to the ENIAC I want to say a little more about that programming I say that stored-program would be nice we thought so all along started program wasn't in our vocabulary nobody heard the word but for that matter nobody had heard the word of an island machine then either and nobody had heard the word head back up to a certain time so for simplicity we did it with these plug boards and switches and things

that's implicitly of course bought us time we were really really trying to finish that machine before the war was over because we didn't know when the war would really be resolved and here we were in 43 trying our best to get that machine going and useful that old monster ENIAC had a start program if you wish to call it that if you can really reorient your thinking and your translation of language what we call a master programmer was essential to putting any program that was any putting on any problem that was worth putting on an

electronic computer much attention could be paid to Penn's efforts once the project was Declassified researchers from around the world came to visit in the summer of 1946 the more school offered a series of summer lectures together everyone interested in computing Wilkes and Sir Frederick Williams came from England Howard Aiken came from Harvard to speak but the focus was plans for the edvac Williams and Tom Kilburn had the first operational stored-program computer at the University of Manchester with only 32 words of 31 bits it was only built to test the electrostatic Williams tube memory for their

mark 1 however it ran a 17 word program for 52 minutes On June 21st 1948 in February 1951 their mark 1 designed along the line of the IAS parallel architecture was introduced as a product by the Ferranti corporation but that's another story maurice wilkes returned to Cambridge with the burning desire to build an ED vac like machine the project got started in October 1946 EDSAC became the first full-scale stored-program computer to operate it first ran on May 6th 1949 edvac constructions had four addresses stored in a 44 bit word thus an instruction could point to

the optimum memory locations EDSAC instructions had one address and restored in 17 bit word so edsac traded off potential performance for a memory efficiency wilkes EDSAC movie shown at the 1951 joint computer conference in philadelphia provides us with a great view of what computing was like at that time the mathematician explains his problem to a committee of experts the first thing to do is to make a list of the library subroutines that will be needed a subroutine for quadrature one for the exponential function a member of the committee interrupts to point out that there is

a more suitable exponential routine available a read routine and a print routine the committee proceeds to discuss the programming of the problem the programmer gets to work referring to the Library catalogue as required here is the program sheet the program is now checked and it is ready for punching this is Margaret Hartree who was my secretary at the time she is in the course of punching the tape and at the appropriate moment she will go over to get one of the library tapes this she copies mechanically onto the tape that she is preparing the recommended

practice not always followed was for a program tape to be punched twice and the resulting tapes to be compared mechanically she finishes off the tape and takes it over to a comparator to compare it with the other tape as long as the two tapes are identical the comparator runs but if there were a discrepancy it would stop she takes the tape to the computer room attaches a tape ticket and puts it on the job queue meanwhile the EDSAC is running on some other problem and the operator is waiting for the results to come out output

was not quite as voluminous in those days as it is now she puts it in the rack takes our program tape and puts it in the photoelectric tape reader the instructions are being read and going into the store the next tank now numbers are being read and here you see the numbers in the store changing as the calculation proceeds these are all short numbers occupying half a word while the program is running we're shown some shots of the EDSAC the storage tanks are in the thermostatically-controlled oven panels showing some of the vacuum tubes back of

panel wiring a shot from above the results are now coming out this is one of the other programmers coming in for her results but she must have made a mistake we used to use for debugging a trace routine that would print to the function letter of each instruction as it was executed our programmer collects his results he is joined by the mathematician and apparently all is well Ed sacks notoriety attracted attention with the possibilities of commercialization the president of the Lyons company a tea shop concern in England realized that a computer could be a valuable

tool for his business Lyons formed a division to commercialize edsac here's a film about their machine leo despite the large numbers of Clarks employed today sufficient clocks are still hard to find with full employment the security of clerical work does not offer the old attraction but trade is becoming more competitive so Clarks are in even greater demand to provide statistics from a mass of data so that management can grasp the changing factors and act accordingly to fulfill this modern need came Leo the first automatic office in the world electronic computers are not new but Leo

with the first design for office work since 1953 it had be employed regularly on accounting stock and cost control statistics and of course payroll Leo is fast and flexible it can test the feasibility of the information that is fed into it and check the accuracy of its own results following Orthodox accounting principles Leo can be installed anywhere it does not require air conditioning having its own ventilation system it is supplied complete with equipment for stabilizing the mains voltage leo mark 2 has four channels for input of information and four more for output this particular installation

is using three input channels two couples to punch card readers and one to a punched tape reader its output can be routed to card punches for machine reading or any of the normal printing devices any other suitable form of input or output can be coupled leo is said an operation from a console it is here that it's performance can be monitored jay lions besides payroll require their leo to do several other routine clerical jobs a job done every afternoon concerns deliveries to their hundred and fifty tea shops in London there are hundreds of items of

food bakery goods of all kinds kitchen goods in a wide variety for the breakfast lunch tea and supper trade and for take-home sales all these in a very in quantity each day are delivered to a precise timetable to the tea shops under stocking leads to lost sales but with food overstocking soon becomes intolerably wasteful each manager s has a standing order depending on the day of the week after lunch each day she considers her stock ways up local conditions and decides what variations up or down she will make to her order she speaks by telephone

to head office where her variations are taken down directly onto cards there is no written record what the girl hears she punches at the same time a short paper tape puts in last minute management decisions such as occur with changes in the weather thus his flexibility provided again the program is fed first laying down the sequence for the multiplicity of calculations Leo will perform next the standing orders and the telephone provisions tea shop by tea shop are fed in with the overriding variations on the paper tape immediately packing notes begin to print out 10 Shops

at a time at the same time charges to tea shops and sales statistics are being after further electronic processing these cards provide the statistics for the use of the management by means of discriminants built into the program Lia will examine all statistics but only print the ones that require action managers are in this way given precise up-to-the-minute information enabling decisions to be more closely related to trading conditions the packing notes which were printed by Leo tend to a sheet are separated guillotine clipped to a Packers board and sent to the dispatch sub turtles of the

different items have been worked out for bulk movement to the several loading bays although the last revision is not telephoned until 3:30 by 4:30 lil has printed for 150 tea shops and 40,000 items exactly what is wanted at each tea shop in the right order for the different loading bays they are also in the right order for the Commons calls so that the goods at the front of the lorry convictive at last and the first call is just inside the doors these are only a few examples of the wide range of work undertaken by leo

building each automatic office is the result of skilled investigation and design each application similarly calls for the experience and know-how of using automatic offices leo computers limited undertake all this in conjunction with the users staff leo is a machine that does routine clerical work more quickly and more accurately than clocks the clocks are freed for more rewarding and productive work as the use of Leo expands let's now go to the Institute for Advanced Studies at Princeton and the IAS parallel machines the design was described in a series of papers by Arthur Burks lieutenant Herman goldstine

and John von Neumann a paper called the preliminary discussion of the logical design of an electronic computing instrument was published June 28 1946 exactly one year after the draft edvac report Brian Randall points out that although the IES computer was not finished until 1952 the series of reports that were issued by the project were widely circulated and served many people as textbooks on logical design and programming the plan was to use an electrostatic storage tube as an alternative to mercury delay lines this provided random access rather than cyclic access with each word being read in

parallel rather than serially as a result of the papers many parallel binary machines or von Neumann machines as they came to be known were started up one such project resulted in the IBM 701 forerunner of a whole series of machines which within a few years became the dominant large-scale scientific computers the IAS design specified a 40-bit word that held two instructions the memory was held in a bank of 40 cathode ray tubes 12 bits were used to address memory and six bits specified the operation code their design uses a function table register F R to

address instructions retaining the ENIAC nomenclature thus the parallel IES machine the edvac report describing the stored program concept and the von Neumann computer all became synonymous parallelism gave the IES more than a factor of 40 in performance or the edvac altogether the is had about a dozen direct descendants the first to operate was argon Zavod ACK in 1951 Oak Ridge had them build a copy called Oracle the University of Illinois built or vac for B RL and a modified version iliac for themselves silly axe Iraq why zack cyclone mystique and computers at Iowa State and

Michigan State were created in LAX image here's johniac it was built by the RAND Corporation to test transistor logic it also acquired a core memory which was placed at the top of the Machine maniac the Los Alamos version of the IAS became operational in early 1952 our next film is an excellent description of the IAS computers even fifty years later it is a wonderful introduction to the five classical boxes often used to define a computer the fundamental operation of the maniac is illustrated by a block diagram of its basic components a problem is fed into

the input section a little at a time as soon as the input is loaded to capacity the control is notified and responds by sending back to the input in order to deposit its contents in the memory making room for more of the problem in the input this goes on rapidly until all of the problem is in the memory the solution of the problem then begins a button is pushed and a control tells the memory to send some material down into the earth medic unit so it can be worked on when an operation is completed the

arithmetic unit notifies the control intermediate and final answers are stored in the memory finally the control instructs the memory to deliver the answer to the output and orders the output to phrase the answer in usable form the maniac can be useful only when it is told precisely what to do in a language it understands a problem submitted to it for a solution must contain two kinds of information the actual numerical quantities involved in directions indicating exactly how they are to be handled this material is logically interwoven in a preliminary coding process where it becomes a

collection of symbols depicting the operations in the computer required to accomplish the solution of the problem the coder works from a flow diagram which has been drawn up by the author of the problem this diagram is essentially a picture of the path to be followed by the computer in the solution of the problem the diagram consists of directional flow lines broken at appropriate points for the insertion of boxes indicating the computation to be performed locally represented here are various necessary logical steps and decisions as well as purely mathematical operations also memory information were necessary a

problem in its final coded form is a sequence of instructions each instruction is made up of an order and an address an order is a command to the machine to perform a specific operation in code it is a combination of two letters in the range A through F there are about 30 such combinations in use problems of commonly used types are accumulated in a tape library the use of the basic portions of these problems in the taping of similar problems results in a considerable saving of time a taped problem enters the computer through a photoelectric

Reader which feeds inwards at the rate of about 20 per second the end of the tape is placed in the reader the load switch is flipped and the problem goes into the machine an average problem is loaded in perhaps 15 seconds in the reader the problem becomes the series of electrical signals with which the computer does its work the signals are sent first into one of the six registers of the earth medic unit a register consists externally of a horizontal row of 40 neon lamps set in a narrow strip of black bakelite vertical white stripes

divide the main registers into groups of four lamps representing tetrads each lamp is connected with an electronic flip-flop circuit each register thus contains 40 flip-flops a lighted lamp results from a hole in the tape and means a binary one no light is a blank or a binary zero thus a register may be regarded as an indicator of events occurring in the earth medic unit the problem is next transferred from the register to the memory the Maineiacs principle memory is of the electrostatic type utilizing 40 standard 2 inch cathode ray tubes monitored in individual metal cases

about the arithmetic unit 20 at the front and 20 at the rear a tape can be designed to write almost anything in the memory the register and memory are connected in a so-called parallel fashion one flip-flop of the register communicates through an electronic gate with one and only one storage tube in the memory for convenience all material of a particular type used in the computer is stored in its own arbitrary block of addresses in the memory the blocks are filled with words in a definite sequence from top to bottom and from right to left a

spot in any tube is the residence of one of the 40 binary characters of a word all characters of the same word resided the same address in their respective tubes the memory when completely filled contains a total of ten hundred and twenty four times forty or forty thousand nine hundred and sixty binary characters the array is about an inch and a half square these spots on the tube faces are being regenerated continuously if they were not they would fade away in a short time an important information would be lost an auxiliary storage device a rotating

magnetic drum with 200 heads increases the memory capacity of the machine by a factor of ten words are handled in blocks of 50 the relatively slow operation of this system makes it useful only as a supplementary memory to prevent excessive loss of time and information in case of power or component failure blocks of material are brought out from the memory and put on magnetic tape to be referred to if necessary this feature is especially useful in the case of unusually long problems requiring several hours of computation a problem may reside in the memory indefinitely but

it can't be solved there the instructions of which it is composed must be withdrawn in proper sequence interpreted and carried out the solution of a problem starts when a button on the operating desk is pushed instructions are transferred from the memory to the top or control register of the earth medic unit orders and addresses are separated in a set of function circuits and orders are sent to a diode matrix for interpretation while addresses go to the memory control circuits for consultation of the specified spots in the array The Matrix is a sorting device each of

the 30 odd orders which might enter it is recognized and sent to an electronic programmer which arranges for the execution of the order the active elements of the matrix are crystal diodes which are mounted in sets of six on octopus the memory circuits act to question the address involved in the instruction to determine what information it contains this information is sent to join the order information in the programmer the required operation is then performed and thus an instruction is carried out in March of 1946 presper eckert and jon moxley left the More school they formed

Eckert Mok Li computer company in December 1948 their first product by nack was shipped to Northrop for onboard missile control in August 1949 by knack was really a circuit prototype for the ambitious univac one that was accepted by the US Census Bureau in March 1951 UNIVAC 1 used the layline memory to store a thousand 12 digit words each word held to six digits single address instructions it was decimal like in yak and was designed for data processing univac had extensive checking circuitry and a complete IO system including tape printers and offline data conversion in order

to deliver their million-dollar machines Eckert and Mauchly had to get funding to survive Remington Rand purchased them with capitalization about 20 univ acts were delivered by 1954 in 1952 Remington Rand also purchased the engineering Research Associates making scientific drum computers these evolved to the 1103 that competed with IBM 701 both used a 36 bit word and parallel architecture like the IAS design this November 1952 newscast shows the UNIVAC in action predicting the election UNIVAC called for a landslide for General Eisenhower after looking at just a few returns but his keepers wouldn't let it speak out

and waited for the returns to be tabulated manually the story I've told here covers a decade from 1943 to 1953 at first no computers existed ten years later many laboratory built computers are in operation several computer companies including IBM and UNIVAC were designing and shipping scientific and business computers this was the beginning of the computer industry from the computer museum in Boston I'm Gordon Bell