| |
KF5JRV > TECH 11.10.16 13:36l 92 Lines 5042 Bytes #-3124 (0) @ WW
BID : 3427_KF5JRV
Read: GUEST DK3UZ DK5SG OE7FMI
Subj: Handheld Electronic Calculators
Path: DB0FHN<OE2XZR<OE5XBL<F1OYP<IZ3LSV<I0OJJ<VE2PKT<ZL2BAU<LU4ECL<N0KFQ<
KF5JRV
Sent: 161011/1122Z 3427@KF5JRV.#NWAR.AR.USA.NA BPQK6.0.13
Handheld electronic calculators
Until the 1970s, most calculating devices were constrained by either the
limited number of tasks that they could perform or by their extravagant size
and cost. Advances in integrated circuit research would prove to be the
solution to both these problems, enabling the manufacture of miniaturised
electronic calculators that were both flexible computing machines and, within
a relatively short period of time, affordable to most.
Making microchips
In 1968, Hewlett-Packard (HP) released its HP 9100A, the first fully
electronic desktop calculator: a limited yet powerful computer for its time.
About the size of a typewriter and costing a whopping $4,900, it found its way
into the pages of tech visionary Stewart Brand's Whole Earth Catalog,
advertised next to beads and moccasins as the 'machine of the future'. Bill
Hewlett congratulated his calculator design team on their achievement, but
immediately set them to work on a model that was affordable and could fit in
a shirt pocket.
At the same time, Texas Instruments (TI) and Sharp Electronics had also jumped
into the race to make a miniaturised calculator using only four or five
'integrated circuits'. These 'microchips' (as they are now known) are small
plates of semiconductor material composed of transistors and other tiny
components which replaced discrete circuits made of large vacuum tubes and
resistors.
A new electrical engineering company, Intel, was commissioned to make a
'microchip' for calculators manufactured by another Japanese company,
Busicom. Intel bought back the rights to this chip in 1971 and began selling
the Intel 4004, the world's first commercially available microprocessor,
which launched a great number of developments in microelectronics that quickly
swept through the computing industry.
Competing for efficiency
Numerous firms competed to exploit the microchip for miniaturised calculation.
The first commercial device to truly fit the 'shirt-pocket' was the Bowmar
901B. Bowmar was an LED company that bought its circuits and keypad from TI,
and its device was a simple four-function machine, capable of addition,
subtraction, multiplication, and division.
The device was bought and sold by competitors like Craig and Commodore, and it
rapidly travelled overseas. However, its market lead was quickly stripped away
later in the same year by Busicom's Handy LE-120A, prompting a wave of
competitive engineering and marketing. By 1972, it was fairly clear how to
manufacture and program four-function calculators, so competitors focused on
design and pushed for miniaturisation.
More interesting was the commercial viability of scientific calculators like
the HP 9100A, capable of computing transcendental functions such as logarithms
and square roots. Despite initial scepticism that a market existed for pocket
devices that were only affordable to laboratories and firms, the first
available pocket scientific calculator, the HP-35, was a commercial hit at the
price of $395 in 1972.
Few individuals could afford such a device: a problem British entrepreneur
Clive Sinclair, aimed to overcome with the Sinclair Scientific. Interested in
producing an inexpensive scientific calculator, Sinclair formed an agreement
with TI in 1974 to produce a device with a single chip that could perform
four-function calculations.
Holed up in a Texas hotel room with mathematics PhD Nigel Searle, Sinclair had
to write a program for scientific functions that would use only 320
instructions. The two created a program using an efficient notation known as
Reverse Polish Notation that simplified number storage and order of
operations, repeated addition and subtraction for multiplication and addition,
and used simplified logarithms and trigonometric functions.
Cutting costs
While highly desirable, most of these early devices were simply too expensive
for most users. For example, Sharp's QT-8B cost $495 in 1970 and the Bowmar
901B cost $240 in 1971. But prices soon began to fall rapidly.
Whilst the HP-35 cost $395 in 1972, within three years of its release the
retail price had halved to $195. The Sinclair Scientific cost just £49.95
($99.95) as a kit upon its release in 1974, and within two years it was
possible to purchase the same model for £7. Advances in integrated circuit
engineering, coupled with better programming, drove these decreases in
cost - as did competition between manufacturers.
This probably contributed to a rapid change in perception: soon all but the
best calculators were seen as disposable, mundane objects. Decreasing profit
margins led to the eventual bankruptcy of companies such as Bowmar, or a move
to other markets, as in the case of Sinclair. Only the most popular devices
could enable a stable business. The TI-30, one of the most commonly used
scientific calculators ever, cost only $25 at its introduction to the market
in 1976, and the brand is so recognisable that TI has continued to update it
to this day.
Read previous mail | Read next mail
| |
