Dette er hentet fra kapittel fire - Neumann János i boka "Turing's Cathedral: The Origins of the Digital Universe" av George Dyson. Alt som ikke er direkte avskrift fra boka, og mine bemerkninger er satt i trebuchet, mens avskrift er satt i Georgia. Alle lenker er naturligvis mine.
Den siste delen av kapitlet handler om John von Neumanns opphold i Los Alamos dit han ankom 21. september 1943. Dagen etter beskrev han stedet i et brev til kona Klári: "an odd combination of an Army Post, a Western National Park with Lodge, and a few other things". Han konkluderte om prosjektet: "...is worth meditating about, although one should probably not sell one's soul to it", og la til "...computers are, as you suspected, quite in demand here, too".
Laboratoriet i Los Alamos var en del av Manhattan-prosjektet hvis mål var å produsere de første atombomber.
Til dette var det nødvendig å gjøre beregninger.
As Nicholas Metropolis, who became director of computing at Los Alamos, put it: "the very nature of the Laboratory's objective - an atomic bomb - precludes extensive field testing".
At a time when even one shock wave at a time was poorly understood, predicting the behavior of an implosion weapon accurately enough to build one that stood a reasonable chance of working on the first try was out of reach of the small computing group. To follow the process from start to finish required modeling the initial propagation of a detonation wave through the high explosive, the transmission of the resulting shock wave through the tamper and into the fissile material (including the reflection of that shock wave as it reached the center), the propagation of another shock wave (followed by an equally violent rarefaction wave) outward through the remnants of the previous explosion and into the atmosphere, and finally the resulting blast wave's reflection if the bomb was at or near the ground. Von Neumann had arrived just in time.
A set of punched-card and tabulating machines were requisitioned from IBM, which could not be told where the machines were going, or why. The machines --- three 601 multipliers, a 402 tabulator, a reproducer, a verifier, a sorter, and a collator --- arrived, in huge wooden crates, without documentation or an installation crew. IBM was asked for the names of their best technicials who had been drafted into the army, who was immediately given a security clearance and reassigned to Los Alamos, but it took time. In the interrim, Stanley Frankel, a Berkeley student of Oppenheimer's who had been put in charge of the hand computing group, and Richard Feynman, a graduate student (and amateur safecracker) from Princeton who was game for any unauthorized challenge, managed to uncreate the machines and get them to work.
Maskinparken var etter vår målestokk svært primitiv, men til gjengjeld kunne menneskene tenke.
Feynman and Frankel were hooked. "Mr. Frankel, who started this program, began to suffer from the computer disease that anybody who works with computers now know about," Feynman later explained. "The trouble with computers is you play with them." Feynman and Frankel, joined by Nicholas Metropolis, adapted the IBM Machines to accelerate the work of the hand computing group. "If we got enough of these machines in a room, we coud take the cards and put them through a cycle," explained Feynman. "Everybody who does numerical calculations now knows exactly what I am talking about, but this was kind of a new thing then --- mass production with machines."
The strategy was to start from a prescribed initial state and model the progress of the explosion from point to point in space and from step to step in time. A single, initial punched card was established for each point in space, with a deck of these cards representing the state of the explosion at a given instant in time. "Processing a deck of cards through one cycle in calculation effectively integrated the differential equations ahead one step in the time dimension," explains Metropolis. "This one cycle required processing of the cards through about a dozen separate machines with each card spending 1 to 5 seconds at each machine." The result was a new deck of cards used as the input for the next time step. The process was tedious, repetitive, intolerant for errors, and quickly bogged down.
"The real trouble was that no one had ever told these fellows anything", explains Feynman. "The Army had selected them from all over the country for a thing called Special Engineer Detachment --- clever boys from high-school who had engineering ability. They sent them up to Los Alamos. They put them in barracks. And they would tell them nothing." Feynman secured permissions from Oppenheimer to give a lecture to the recruits. "They were all excited: 'We're fighting a War! We see what it is!' They knew what the numbers meant. If the pressure came out higher, that meant there was more energy released. Complete transformation! They began to invent ways of doing it better. They improved the scheme. They worked at night." Productivity went up by a factor of ten.
Von Neumann found himself back among the punched cards he remembered from when his father brought parts of a Jacquard loom-control system home from work. "In March or April 1944," says Metropolis, "von Neumann spent two weeks working in the punched-card-machine operation, pushing cards through the various machines, learning how to wire plugboards and design card layouts, and becoming thoroughly familiar with the machines operations."
Mer om Manhattan-prosjektet finnes hos Columbia University Computing History
Noen andre kuriositeter som jeg ble oppmerksom på i Dysons bok:
Computron-røret - integrerte radiorør
Selectron-røret - minnerør
Iconoscope - videokamera