| |
KF5JRV > TECH 15.08.16 12:32l 172 Lines 8504 Bytes #999 (0) @ WW
BID : 296_KF5JRV
Read: GUEST
Subj: SEAC Computer
Path: IW8PGT<IZ3LSV<IK6ZDE<F1OYP<ZL2BAU<LU4ECL<N0KFQ<KF5JRV
Sent: 160815/1124Z 296@KF5JRV.#NWAR.AR.USA.NA BPQK1.4.65
SEAC Boots Up
SEAC, was the first electronic computer with an internally stored
program in the United States Government. It was the first of three
computers built at NBS. It was designed, built, and operated at
NBS by engineers, scientists, and mathematicians. A later machine,
the SWAC, built after SEAC but before DYSEAC.
Originally, SEAC was developed as an interim facility while NBS was
awaiting the delivery of the first commercial computer to be used
by the Bureau of the Census and subsequently by the Army and the
Air Force. Its successful completion made computation possible at
NBS both for the sponsors and many other government agencies a
year before any commercial computer was delivered. Notwithstanding
the original interim nature of SEAC, this successful resource,
with all its subsequent enhancements, continued to function
usefully for the government for over 13 years.
The National Bureau of Standards completed the SEAC in April 1950.
Some of the design of the SEAC (also known as the "Standards
Electronic Automatic Computer") was based on the EDVAC computer,
which was built at the Moore School of Engineering at the
University of Pennsylvania.
The SEAC utilized sixty-four mercury filled glass tubes with a
quartz crystal at each end. One crystal was used as a transmitter
and one as a receiver in this acoustic delay line memory storage
unit. The acoustic delay line had a capacity of eight words.
Information traveled in the form of sound waves through the
mercury in the tubes.
The SEAC was the only fully functional computer available to the
NBS and other government agencies in 1950. Demands on it were very
high, and it was in operation 24 hours a day, 7 days per week. In
a given month, the SEAC would be used to solve more than 50
different unrelated scientific problems for a variety of users.
In addition to NBS staff and staff from other government agencies,
private universities and laboratories would also make use of the
SEAC's computing ability. In order to monitor the operation of
the computer, its operators connected an amplifier and speakers
to one of the machine's registers.
The SEAC operators could tell from any change in the SEAC's familiar
operating sounds if there was a malfunction in the computer.
Samuel N. Alexander was chief of the NBS section that produced
the SEAC. He later became Chief of the NBS Information Technology
Division. John Todd was chief of the Computation Section when the
SEAC came on line.
There were many engineering innovations introduced by SEAC. These
included the use of new input-output mechanisms, an early example
both of time sharing and the interconnection of two computers
(the SEAC and the DYSEAC) in 1954, the development of marginal
checking, new memory mechanisms, and a graphical display. Many
such innovations are described in, a collection surveying
work until 1953 and published in 1955 . An outstanding group of
mathematicians also made major contributions to the field of
numerical analysis. Much of this work was later collected in
the very popular Handbook of Mathematical Functions.
Many problems of science, mathematics, engineering and government
operation lent themselves naturally and obviously to assistance
by rapid computational methods. The diversity of these uses is
indicated in a list of representative examples of actual
computations done on SEAC.
Mathematics and Statistics
Physics and Chemistry
Engineering
Business and Economics
Air Force program planning
Social Security accounting
But the most important contributions made with SEAC were in
the development of entirely new uses for computers. There
was a fortunate correlation between an expanding economy
and an expanding technology that made it possible to develop
uses that were previously unknown. Furthermore, the unique
status of NBS enabled a kind of exploratory operation free
from commercial constraints. Many of the new applications
of SEAC to problems of science, mathematics, engineering,
and government operation were neither anticipated nor,indeed,
requested by the intended beneficiaries. It was usual for
the NBS engineers to understand new goals for the computer
and the technical way to achieve these goals. Only after
NBS developed prototype applications for these agencies,
on the SEAC, was there a demand for "more of the same,
whatever that is."
In the early days of electronic computation, a primary question,
and one which posed considerable anxiety, was whether computers
could be made to function flawlessly, sufficiently long, to do
useful computation. The ENIAC provided some encouragement, and,
by 1952, sufficient experience had accumulated at NBS to bolster
that encouragement with a measure of confidence. Consequently,
NBS was able to encourage the use of computers in areas for
which there was no previous experience with automation. As a
scientific and engineering agency, within the government, of
high reputation for integrity, NBS made proposals for
innovative uses of the SEAC that were received with a more
generous response by other agencies than would have been
possible in a more competitive commercial environment. In
fact, there was a widely held feeling that "nothing will be
restrained from them which they have imagined to do."
There were special conditions associated with the operation of
the SEAC which encouraged such evangelism. Everybody on the
staff shared in the onerous task of nursemaiding the computer
twenty four hours each day. There was a nominal maintenance
crew which was led by P.D.Shupe and which included Russell
Kirsch and was augmented by the rest of the staff. The
maintenance crew had the power to take the machine away
from the so-called "productive" users based entirely on
their own judgment. The rationale was they needed to keep
the computer in almost flawless operation to be able to
use it as a self diagnostic tool.
Such a diagnostic tool would be of no use unless you could
make the assumption that there was at most one fault in the
machine at any one time. Consequently, any time that the
crew suspected there was a fault, they would take the computer
away from the users to track the symptoms down, for fear
another fault might occur before the first was located.
Typically, they would find a memory error and would use
one half of the memory for the diagnostic program while
they checked the other half. If there were two errors
in the memory at the same time, they couldn't reliably
load the diagnostic program and then the problem of finding
the trouble would be very hard. So they were very scrupulous
about keeping the computer fault-free.
The Maintenance Crew was prepared to claim several years of
fault-free operation of the SEAC when they moved the computer
to a different part of the NBS campus. This required
disassembling the hardware and reconnecting it according
to the design diagrams (and the excellent memory of some
of the staff). Upon reconnecting the computer, it was
discovered there was a wiring error in the original design
which had survived about a decade of scrupulous diagnostic
checking and would result in a definitive error every time
that a suitable program was run. It had never occurred in
a decade of round-the-clock operation! From this, one may
conclude, throughout history, there probably has never
been a faultless computer built, since all computers have
always had known malfunctions with associated "workarounds"
in the software, if not in the hardware.
Such assiduousness in tracking down malfunctions is seldom
practiced, if ever, today. But in the process of testing
the computer after a suspicious fault had been detected,
the crew would confirm the correct operation by running
some of its own research programs. It turns out that many
people in the early days of computing had this same
practice. For example, Arthur Samuel at IBM, who wrote
the first artificial intelligence program for checker
playing, was the manager for the Poughkeepsie plant. He
had masses of IBM 701 computers playing checkers every
night supposedly in order to check out the computers,
but of course, really to accumulate learning experience
on his program. This also was the source of similar
computer time that Kirsch used to accumulate his own
artificial intelligence research results.
73, Scott kf5jrv
KF5JRV @ KF5JRV.#NWAR.AR.USA.NA
Read previous mail | Read next mail
| |
