[tt] BBC: Future Directions in Computing (several articles)

[Premise Checker]

Future Directions in Computing (several articles)
http://news.bbc.co.uk/2/hi/technology/7085019.stm
Last Updated: Tuesday, 13 November 2007, 16:46 GMT
[Thanks to Sarah for these.]

Silicon electronics are a staple of the computing industry, but
researchers are now exploring other techniques to deliver powerful
computers.

A quantum computer is a theoretical device that would make use of
the properties of quantum mechanics, the realm of physics that deals
with energy and matter at atomic scales.

In a quantum computer data is not processed by electrons passing
through transistors, as is the case in today's computers, but by
caged atoms known as quantum bits or Qubits.

"It is a new paradigm for computation," said Professor Artur Ekert
of the University of Oxford. "It's doing computation differently."

A bit is a simple unit of information that is represented by a "1"
or a "0" in a conventional electronic computer.

A qubit can also represent a "1" or a "0" but crucially can be both
at the same time - known as a superposition.

This allows a quantum computer to work through many problems and
arrive at their solutions simultaneously.

"It is like massively parallel processing but in one piece of
hardware," said Professor Ekert.

'Complex systems'

This has significant advantages, particularly for solving problems
with a large amount of data or variables.

"With quantum computing you are able to attack some problems on the
time scales of seconds, which might take an almost infinite amount
of time with classical computers," Professor David Awschalom of the
University of California, Santa Barbara told the BBC News website
recently.

In February 2007, the Canadian company D-Wave systems claimed to
have demonstrated a working quantum computer.

At the time, Herb Martin, chief executive officer of the company
said that the display represented a "substantial step forward in
solving commercial and scientific problems which, until now, were
considered intractable."

But many in the quantum computing world have remained sceptical,
primarily because the company released very little information about
the machine.

The display also failed to impress.

"It was not quite what we understand as quantum computing," said
Professor Ekert. "The demonstrations they showed could have been
solved by conventional computers."

However, Professor Ekert believes that quantum computing will
eventually come of age.

Then, he said, they will not be used in run-of-the-mill desktop
applications but specialist uses such as searching vast databases,
creating uncrackable ciphers or simulating the atomic structures of
substances.

"The really killer application will probably be in designing new
materials or complex systems," he said.

FUTURE OF COMPUTING VIDEO

[72]HP's Nano Chip [73]Nano chip developer
[74]Multi-core silicon chip [75]Multi-core chips
[76]Face reflected in chip wafer [77]Multi-core 'myth'
[78]Intel co-founder, Gordon Moore [79]Intel co-founder

RELATED INTERNET LINKS

[80]Moore's Law
[81]Artur Ekert
[82]D-wave

References

72. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7070000/newsid_7075000?redirect=7075050.stm&news=1&nbwm=1&nbram=1&bbwm=1&bbram=1&asb=1
73. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7070000/newsid_7075000?redirect=7075050.stm&news=1&nbwm=1&nbram=1&bbwm=1&bbram=1&asb=1
74. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7090000/newsid_7091700?redirect=7091718.stm&news=1&bbwm=1&nbwm=1&bbram=1&nbram=1&asb=1
75. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7090000/newsid_7091700?redirect=7091718.stm&news=1&bbwm=1&nbwm=1&bbram=1&nbram=1&asb=1
76. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7090000/newsid_7091800?redirect=7091841.stm&news=1&bbram=1&nbram=1&bbwm=1&nbwm=1&asb=1
77. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7090000/newsid_7091800?redirect=7091841.stm&news=1&bbram=1&nbram=1&bbwm=1&nbwm=1&asb=1
78. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7080000/newsid_7084200?redirect=7084268.stm&news=1&bbwm=1&nbwm=1&bbram=1&nbram=1&asb=1
79. 
http://www.bbc.co.uk/mediaselector/check/player/nol/newsid_7080000/newsid_7084200?redirect=7084268.stm&news=1&bbwm=1&nbwm=1&bbram=1&nbram=1&asb=1
80. http://www.intel.com/research/silicon/mooreslaw.htm
81. http://cam.qubit.org/users/artur/
82. http://www.dwavesys.com/


Getting more from Moore's Law
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/7080772.stm

By Jonathan Fildes
Science and technology reporter, BBC News

For more than 40 years the silicon industry has delivered ever
faster, cheaper chips.

The advances have underpinned everything from the rise of mobile
phones to digital photography and portable music players.

Chip-makers have been able to deliver many of these advances by
shrinking the components on a chip.

By making these building blocks, such as transistors, smaller they
have become faster and firms have been able to pack more of them
into the same area.

But according to many industry insiders this miniaturisation cannot
continue forever.

MOORE'S LAW

The number of transistors it is possible to squeeze in to a chip for
a fixed cost doubles every two years
First outlined by Gordon Moore, co-founder of Intel
Published in Electronics Magazine on 19 April, 1965

"The consensus in the industry is that we can do that shrink for
about another ten years and then after that we have to figure out
new ways to bring higher capability to our chips," said Professor
Stanley Williams of Hewlett Packard.

Even Gordon Moore, the founder of Intel and the man that gave his
name to the law that dictates the industry's progression, admits
that it can only go on for a few more years.

"Moore's Law should continue for at least another decade," he
recently told the BBC News website. "That's about as far as I can
see."

Tiny tubes

As a result, researchers around the world are engaged in efforts to
allow the industry to continue delivering the advances that computer
users have come to expect.

Key areas include advanced fabrication techniques, building new
components and finding new materials to augment silicon.

Already new materials are creeping into modern chips.

As components have shrunk critical elements of the transistors,
known as gate dielectrics, do not perform as well allowing currents
passing through the transistors to leak, reducing the effectiveness
of the chip.

To overcome this, companies have replaced the gate dielectrics,
previously made from silicon dioxide, with an oxide based on the
metal hafnium.

The material's development and integration into working components
has been described by Dr Moore as "the biggest change in transistor
technology" since the late 1960s.

But IBM researchers are working on materials that they believe offer
even bigger advances.

"Carbon nanotubes are a step beyond [hafnium]," explained Dr Phaedon
Avouris of the company.

'Superior' design

CARBON NANOTUBES
Sheets of carbon atoms folded into a cylinder
Unusual strength and electrical properties
Promise to revolutionise electronics, computers, chemistry and
materials science

Carbon nanotubes are tiny straw-like molecules less than 2
nanometres (billionths of a metre) in diameter, 50,000 times thinner
than a strand of a human hair.

"They are a more drastic change but still preserve the basic
architecture of field effect transistors."

These transistors are the basic building blocks of most silicon
chips.

Dr Avouris believes they can be used to replace a critical element
of the chip, known as the channel.

Today this is commonly made of silicon and is the area of the
transistor through which electrons flow.

Chip makers are constantly battling to make the channel length in
transistors smaller and smaller, to increase the performance of the
devices.

Carbon nanotube's small size and "superior" electrical properties
should be able to deliver this, said Dr Avouris.

Crucially, he also believes the molecules can be integrated with
traditional silicon manufacturing processes, meaning the technology
would more likely be accepted by an industry that has spent billions
perfecting manufacturing techniques.

The team have already shown off working transistors and are
currently working on optimising their production and integration
into working devices.

Tiny improvement

Professor Williams, at Hewlett Packard is also working on technology
that could be incorporated into the future generations of chips.

As well as exploring optical computing - using particles of light
instead of electrons to significantly increase the speed of today's
computers - he is building new electronic components for chips
called memristors.

Nano chip developer Multi-core chips Multi-core 'myth'

He says it would be the "fourth" basic element to build circuits
with, after capacitors, resistors and inductors.

"Now we have this type of device we have a broader palette with
which to paint our circuits," said Professor Williams.

Professor Williams and his team have shown that by putting two of
these devices together - a configuration called a crossbar latch -
it could do the job of a transistor.

"A cross bar latch has the type of functionality you want from a
transistor but it's working with very different physics," he
explained.

Crucially, these devices can also be made much smaller than a
transistor.

"And as they get smaller they get better," he said.

Professor Williams and his team are currently making prototype
hybrid circuits - built of memristors and transistors - in a
fabrication plant in North America.

"We want to keep the functional equivalent of Moore's Law going for
many decades into the future," said Professor Williams.


Meeting the man behind Moore's Law
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/7080646.stm

By Jonathan Fildes
Science and technology reporter, BBC News

In April 1965 a 36-year old electronics buff jotted down his
thoughts on the future of the juvenile silicon chip industry.

Writing in a "throw-away" journal, Gordon Moore accurately imagined
a future filled with mobile phones, home computers, and even
intelligent cars.

But it was a much more prosaic prediction that has come to dominate
his life and the industry that he helped found.

"I could see a change coming that the electronics were going to get
significantly cheaper," says the co-founder of Intel, the largest
maker of computer chips.

In the article in Electronics Magazine, he predicted that the number
of transistors on a silicon chip would double every year for ten
years.

He later revised the forecast to doubling every two years or so, as
the initial breakneck speed of development and shrinkage waned. It
was a prediction that became known as Moore's Law and it has helped
drive the computer revolution over the last four decades.

Modest growth

At first glance it is not the kind of observation that would
catapult a person to fame.

But the day I meet him, a now silver haired Dr Moore has just given
a talk to a packed auditorium of people and he is surrounded by
crowds of autograph hunters who were not even born when he made his
off the cuff observations.

Moore's Law has become shorthand for the pace of technological
change. It set a standard for the chip industry's phenomenal growth
and has in turn underpinned the world's digital awakening.

TRANSISTORS

Transistors are basic electronic switches found in silicon chips
Each transistor can be switched on or off, representing a "1" or
"0", known as binary code
All computation is done using different combinations of these two
outputs to do calculations
The number of transistors on a chip determine its speed
Modern chips contain millions of transistors allowing them to
execute millions of calculations per second
The tiny devices consist of a source, drain and gate
A voltage applied to the gate and drain turns the device on
Removing the gate voltage switches it off again

"It was an exciting technology in the beginning. It had so much
potential, we just had no idea how much potential," he says.

"When Intel was formed [in 1968] the total semiconductor industry
was only a couple of billion dollars worldwide - today it is 300
billion."

And as the industry has delivered each successive generation of
faster, smaller, cheaper chips, it has opened up rafts of new
possibilities for silicon that have ultimately delivered the
technologies he predicted more than 40 years ago.

Reflecting on his prophesies today, a retired and quietly-spoken Dr
Moore is characteristically modest.

"When I went back and read that I was amazed that I predicted all of
those things," he says.

But, as a young engineer, he was at least uniquely placed to make
his key observation, having co-founded Fairchild Semiconductors,
maker of the first commercial integrated circuit, or chip.

"I was directing a lab where we were trying to advance the
technology and from my perspective I could see some of these things
coming that weren't generally visible to the rest of the
population," he says.

Force for change

At the time, computers were mainly used by the military and PCs were
unheard of, he says.

"Computers were in glass rooms tended to by a core of monks that
knew how to do the proper incantations."

But as the silicon chip industry took hold and computer makers
learnt how to exploit the technology everything changed.

"Shortly after that the commercial market just completely dwarfed
anything in the military," he says.

And what had originally been just a prediction by Dr Moore became a
self-fulfilling prophesy.

"It has become a driving force for the industry," he says.
"Competitors have realised that if they don't move at least that
rate they are going to fall behind."

So far silicon producers have managed to keep on or ahead of the
curve for more than four decades by continually shrinking the
technology and packing more and more components inside a chip.

MOORE'S LAW

The number of transistors on a chip doubles every two years
First outlined by Gordon Moore, co-founder of Intel
Published in Electronics Magazine on 19 April, 1965

"It's a peculiar feature of this technology that by making things
smaller everything gets better," says Dr Moore. "The transistors get
faster, you can put more of a system on a chip."

But more importantly, and perhaps more curiously, the chips also
become cheaper.

And this is the key point of his 1965 paper, he says. Moore's Law is
an economic law and would probably have driven the industry
regardless of whether or not he had made his prediction.

"I am not sure that having Moore's Law held up there as a yardstick
increases the pressure [on chip manufacturers] because the need to
remain competitive is so strong."

Chip future

As a result, the industry has grown "far beyond" what he could have
imagined in 1965, he says.

Moore interview

"It is surprising that any of the things we predicted are still
valid."

He is most impressed with the industry's inventiveness, he says,
allowing it to overcome a series of seemingly insurmountable
technical hurdles as it grew.

"The industry has succeeded in getting around all of the ones that
have been thrown in front of it," he says. "It has been much more
successful than I probably would have predicted."

But, Dr Moore says, the industry can only go on shrinking
transistors for so long.

Eventually, the features will become so small that the atomic
structure of the materials will be a limitation, possibly spelling
the end of Moore's Law

So what does he think will happen in the next 40 years?

"I'm through with making predictions," he chuckles. "Get it right
once and quit."


Green pigment spins chip promise
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/4776479.stm

An unpopular pigment used by artists in the 18th Century could lead
to more energy efficient, faster computers.

Cobalt green, as it is known, has been tested by a US team who
believe it could be used in "spintronic" devices.

Spintronics involves manipulating the magnetic properties of
electrons to do useful computational work.

Cobalt green may be useful for building working devices as it can be
used at room temperature, unlike many other materials which must be
supercooled.

"The big challenge is to develop materials that can perform these
kinds of functions not just at cryogenic temperatures but at
practical temperatures," said Professor Daniel Gamelin of the
University of Washington in Seattle, one of the team that carried
out the work.

Magnetic force

Cobalt green, sometimes known as Rinmann's green, is a mixture of
zinc oxide and cobalt. The semi-transparent pigment was developed by
the Swedish chemist Sven Rinmann in 1780.

The colouring never found favour with artists as it was expensive
and created relatively weak colours.

The breakthrough with the materials we tested is that they exhibit
their magnetic properties at room temperature
Professor Gamelin

However, the material could be useful in spintronic devices because
of its magnetic properties.

Conventional electronics rely on the movement and accumulation of
electrons to carry out calculations or to store data.

In addition, spintronics uses the spin of electrons - detected as a
weak magnetic force - to increase the computational power of a
device.

Spintronic devices, in theory, should be much quicker than
conventional electronics and require much less power.

The technology is already used in some hard discs and could
eventually be used for sensors and computer memory.

Antique promise

A goal of spintronics is to develop a semiconductor that can
manipulate an electron's magnetism.

Semiconductors are used to manufacture microchips, at the heart of
all computers and many other electronic devices.

As computers have become faster, chips have become smaller.

But manufacturers recognise that conventional chip fabrications
techniques will eventually reach a physical limit.

Spintronics offers a potential solution.

To create a useful spintronic device the semiconductor must be
magnetised and remain stable at room temperature.

Researchers have proposed that the best way to create material like
this is to incorporate a magnetic element into a traditional
semiconductor such as silicon.

Until now, most attempts to do this have only shown useful
spintronic properties at low temperature, typically around minus
200C. This is much too cold to be useful in a working device.

Cobalt green appears to work at more useful temperatures.

"The breakthrough with the materials we tested is that they exhibit
their magnetic properties at room temperature," said Professor
Gamelin.

Chip makers

To test the antique pigment the researchers processed zinc oxide, a
simple semiconductor material, so that some of the zinc ions were
replaced with ions of magnetic cobalt. The process is known as
"doping".

This work shows there is a real effect here, and there is promise
for these materials
Professor Gamelin

The cobalt ions were then aligned by exposing the semiconductor to a
zinc metal vapour. The alignment causes the material to become
magnetic.

The magnetism continued when the material was warmed to room
temperature and when the exposure to the zinc metal vapour was
stopped.

When the material was heated further the magnetic properties
disappeared.

"This work shows there is a real effect here, and there is promise
for these materials," said Professor Gamelin.

However, the research is still in its early stages. To be of use to
chip-makers the technique must be shown to work using common
semiconductor material.

"The next step is to try to get these materials to interface with
silicon semiconductors."


Shrinking chips use novel recipe
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/7085480.stm

The chip industry's unrelenting quest to build smaller, faster
microchips has taken another step forward.

Chip-maker Intel has launched a range of processors, known as
Penryn, which will power the next generation of PCs.

The tiny chips contain a novel material and have features just 45
nanometres (billionths of a metre) wide.

The only PC processor in the line-up of 16 chips packs 820 million
of the tiny switches into an area little bigger than a postage
stamp.

"Had we used the same transistors that we used in our chips 15 to 20
years ago, the chip would be about the size of a two-storey
building," said Bill Kircos of Intel.

Paul Otellini, head of Intel, described the challenge of building
the chips as "awe-inspiring".

Although the chip-maker is the first company to make microprocessors
with such tiny features, other companies, such as Taiwan
Semiconductor Manufacturing Company (TSMC), are producing other
types of silicon chip.

"We have 45nm designs in production," said Chuck Byers of the firm.

TSMC manufactures chips on behalf of other companies.

Material world

The launch of the new multi-core chips comes nearly 60 years after
the transistor was invented.

The brainchild of three scientists working at the research labs of
the US Bell System telephone company, the tiny switches have gone on
to underpin the silicon age.

The first transistors were crude devices made of several different
materials and many centimetres tall.

Over the last six decades, scientists have refined the devices and
can now pack millions of them on to a square of silicon.

The onward progression of the silicon industry is known as Moore's
Law, and states that the numbers of transistors on a chip will
double every two years.

However, as the industry devices have shrunk, researchers have been
forced to confront major technical obstacles.

In the latest generation of Intel chips, critical elements of the
transistors, known as gate dielectrics, do not perform as well.

As a result, currents passing through the transistors leak, reducing
the effectiveness of the chip.

To overcome this, Intel has replaced the gate dielectrics,
previously made from silicon dioxide, with a material based on the
metal hafnium.

The hafnium-based compound is a so-called high-K material, which
refers to its dielectric constant, and has a greater ability to
store electrical charge than silicon dioxide.

The exact recipe for the new material has not been revealed but
Intel says that it offers greater performance at such tiny scales.

Moore interview

Intel co-founder Gordon Moore has described the inclusion of hafnium
as "one of the biggest transistor advancements in 40 years".

Tiny tweaks

In contrast, TSMC has said that its chips do not use the new
material.

"We have an alternative process that we believe has even more
performance value," said Mr Byers.

"One of the strategies we employ is to make it [a manufacturing
process] accessible as possible and one of the ways you do that is
to change the process as little as possible."

Rather than changing the design of the chips, TSMC has tweaked the
manufacturing process to produce the tiny features.

"We work very hard to maintain performance levels with existing
materials," said Mr Byers. "For instance, at 45nm there are only a
couple or three changes."

"There are several ways to skin this cat," added Gareth Jones, also
of TSMC.

However, other companies have signalled their intention to start
production of microchips using similar hafnium technology.

IBM, which has developed rival technology with Toshiba, Sony and
AMD, intends to incorporate the transistors into its chips in 2008.

Hewlett Packard, Lenovo and Dell have already said that they will
use the new Penryn processors in top of the range PCs.

TRANSISTORS COMPARED

Transistors are tiny switches and consist of a source, drain and
gate
When a voltage is applied to the gate and drain electrons flow from
the source to the drain and the transistor is switched on
If the voltage to the gate is removed the transistor is switched off
As transistors have shrunk the insulating layer has got so thin that
electrons leak, making chips inefficient
High-K materials such as hafnium have a greater ability to store
charge and prevent leaks
Using high-K materials allows chips to be made smaller and more
efficient


Teraflops chip points to future
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/6354225.stm

A chip with 80 processing cores and capable of more than a trillion
calculations per second (teraflops) has been unveiled by Intel.

The Teraflops chip is not a commercial release but could point the
way to more powerful processors, said the firm.

The chip achieves performance on a piece of silicon no bigger than a
fingernail that 11 years ago required a machine with 10,000 chips
inside it.

The challenge is to find a way to program the many cores
simultaneously.

Current desktop machines have up to four separate cores, while the
Cell processor inside the PlayStation 3 has eight (seven of them
useable). Each core is effectively a programmable chip in its own
right.

But to take advantage of the extra processing power, programmers
need to gives instructions to each core that work in parallel with
one another.

There are already specialist chips with multiple cores - such as
those used in router hardware and graphics cards - but Dr Mark Bull,
at the Edinburgh Parallel Computing Centre, said multi-core chips
were forcing a sea-change in the programming of desktop
applications.

"It's not too difficult to find two or four independent things you
can do concurrently, finding 80 or more things is more difficult,
especially for desktop applications.

"It is going to require quite a revolution in software programming.

"Massive parallelism has been the preserve of the minority - a few
people doing high-performance scientific computing.

"But that sort of thing is going to have to find its way into the
mainstream."

The first time teraflop performance was achieved was 11 years ago on
the ASCI Red Supercomputer built by Intel for the Sandia National
Laboratory.

It is going to require quite a revolution in software programming
Dr Mark Bull

That machine took up more than 2,000 square feet, was powered by
almost 10,000 Pentium Pro processors, and consumed more than 500
kilowatts of electricity.

"Our researchers have achieved a wonderful and key milestone in
terms of being able to drive multi-core and parallel computing
performance forward," said Justin Rattner, Intel Senior Fellow and
chief technology officer.

"It points the way to the near future when teraflops-capable designs
will be commonplace and reshape what we can all expect from our
computers and the internet at home and in the office."

The Teraflops chip uses less electricity than many current high-end
processors, making the design attractive for use in home computers.

It consumes 62 watts, and the cores can power on and off
independently, making it more energy efficient.

Intel says that commercial spin-offs of the chip could see it being
used in high-definition entertainment PCs, servers and handheld
devices.

Possible uses include artificial intelligence, instant video
communications, photo-realistic games and real-time speech
recognition, said the firm.


Chilly chip shatters speed record
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/5099584.stm

The world's fastest silicon-based microchip has been demonstrated by
scientists in the US.

The prototype operates at speeds up to 500 gigahertz (GHz), more
than 100 times faster than desktop PC chips.

To break the world record, the researchers from IBM and the Georgia
Institute of Technology had to super-cool the chip with liquid
helium.

The team believes the device could eventually speed up wireless
networks and develop cheaper mobile phones.

"Faster and faster chips open up new applications and reduce costs
for existing products," said Professor David Ahlgren of IBM.

Exotic chips

At the moment, most microchips are made from silicon.

But in recent years, there has been a realisation that silicon
cannot match other materials in terms of processing speed.

For applications that require huge amounts of calculations every
second, like collision-warning systems in cars and trucks, companies
use exotic materials to produce the necessary power.

Materials like gallium arsenide are commonly used, but are expensive
and difficult to fabricate.

However, the chip industry would like to continue to use proven
silicon manufacturing technology because it is reliable and cheap.

The new experiments were part of a project to explore the speed
limits of devices made of silicon and germanium.

Super cooled

Germanium is already added to the silicon chips used in mobile
phones to make them operate more efficiently.

A decade ago we couldn't even envisage being able to run at these
speeds
Professor David Ahlgren, IBM

Adding the element allows chips to run faster and use less power.
Importantly, they can also be fabricated using existing silicon
techniques.

These chips are already known to operate at faster and faster speeds
as they are cooled.

To break the speed record, the researchers super-cooled an IBM
prototype of a new "high frequency" device to -268.5C, using liquid
helium.

This temperature is just above absolute zero, the theoretical
minimum temperature possible. When cooled, the chips were able to
perform half a trillion calculations every second, a speed of 500
GHz.

By comparison, a powerful desktop PC is capable of about five
billion calculations per second.

"A decade ago we couldn't even envisage being able to run at these
speeds," said Professor Ahlgren.

Next generation

At room temperature the experimental chips still managed to
outperform standard silicon chips, running at about 350 billion
calculations per second.

However, the researchers say they can push them even further.

"We observe effects in these devices at cryogenic temperatures which
potentially make them faster than simple theory would suggest," said
Professor John Cressler of the Georgia Institute of Technology.

The team believes it is possible to make chips run at 1,000 Ghz, or
one Terahertz, at room temperature.

"Understanding the basic physics of these advanced transistors arms
us with knowledge that could make the next generation of
silicon-based integrated circuits even better," said Professor
Cressler.


Antique engines inspire nano chip
http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/6912023.stm

By Jonathan Fildes
Science and technology reporter, BBC News

The blueprint for a tiny, ultra-robust mechanical computer has been
outlined by US researchers.

The energy-efficient nano computer is inspired by ideas about
computing first put forward nearly 200 years ago.

Writing in the New Journal of Physics, the scientists say the
machine would be built from nanometre sized components, just
billionths of a metre across.

Chips based on the design could be used in places, such as car
engines, where silicon can be too delicate, they said.

"What we are proposing is a new type of computing architecture that
is only based on nano mechanical elements," said Professor Robert
Blick of the University of Wisconsin-Madison and one of the authors
of the paper.

"We are not going to compete with high-speed silicon, but where we
are competitive is for all of those mundane applications where you
need microprocessors which can be slow and cheap as well."

In addition to high-temperature automotive applications, Professor
Blick envisaged nano mechanical chips being used in everything from
toys to domestic appliances.

Tiny thinking

Mechanical computers are nothing new. The remains of a 2,000 year
old analogue computer known as the Antikythera mechanism were
discovered in Greece in 1902.

And during the nineteenth century, English mathematician and
engineer Charles Babbage designed various steam powered mechanical
computers.

We are quite confident that in a couple of years this work will lead
to commercial applications
Robert Blick

His "difference engine", for example, consisted of more than 25,000
individual levers, ratchets and cogs and weighed more than 13 tons.

Although none of his designs were ever finished, recent
reconstructions by London's Science Museum show they were capable of
carrying out complex calculations.

The US team's proposal owes a debt to these early concepts.

"It's inspired by Babbage's ideas but these days we can scale it
down," Professor Blick told the BBC News website.

"Now, we are able to process devices on the nano scale."

Work in progress

The team's tiny, hypothetical number-cruncher could be built out of
ultra-hard materials such as diamond or piezoelectric materials,
which change shape when an electric current is applied.

Unlike today's computers, which are based on the movement of
electrons around circuits to do useful calculations, the nano
mechanical computer would use the push and pull of each tiny part to
carry out calculations.

The researchers are currently building the first elements needed for
the computer, focusing initially on transistors, the basic switches
at the heart of all computers.

"We have demonstrated that a single element of these transistors
work," said Professor Blick.

Simulations also suggest that it should be possible to build working
circuits, he said.

"We are quite confident that in a couple of years this work will
lead to commercial applications," said Professor Blick.

The American military is interested in a working device because
unlike traditional chips, nano mechanical devices are not
susceptible to electromagnetic pulses, which could be used by an
enemy to knock out computing systems.

Design principles

However, the team believes the chip's key advantage could be in
keeping up with the chip industry's relentless pursuit of Moore's
Law.

The proposition, articulated by Intel co-founder Gordon Moore in
1965, states that the number of transistors on a chip could double
every 24 months.

"The smaller you make your devices, the more electrons rush through
and the hotter your device gets," said Professor Blick.

"In the not too distant future we will reach a thermal limit."

At that point, the heat will interfere with the function of the
circuits and offset any efficiency gained by packing more
transistors on to the chip.

The mechanical devices run much cooler than silicon and should
therefore not suffer from the same problem.

But Dr Michael Kraft of the University of Southampton believes it
will take a lot to persuade the silicon industry to abandon more
traditional chip designs, such as the common circuit pattern known
as CMOS.

"The industry has been working with CMOS for almost 40 years and
there is already so much expertise and infrastructure," he told the
BBC News website.

"I think, commercially, industry will continue with CMOS up to the
point where it is absolutely not possible to push it any further."

However, he said, the nano mechanical approach did offer certain
benefits that could one day lead to the development of hybrid chips.

"They consume less power which is more and more important because
most computers are mobile these days," he said.

"The battery is the big bottleneck, so anything that reduces the
power consumption is a real advantage."


Enhanced video for global users
http://news.bbc.co.uk/2/hi/6277422.stm
By Steve Herrmann
Editor, BBC News Interactive

BBC News Player

High quality video is available to all BBC website users

Service explained

If you are looking at the BBC News website from outside the UK there
are some new developments to tell you about.

We have upgraded the technical quality of the video on the site for
users overseas so that anyone anywhere in the world with a high
speed internet connection can watch the BBC's news reports in
broadband quality.

Until now this has only been possible for users in the UK - server
costs have restricted us from providing the same level of service
internationally.

We did not want the BBC's UK licence fee payers meeting this cost
and in effect subsidising the service for people outside the UK.

Now, with the help of our partners in BBC World, the BBC's
commercially funded international TV news channel, we are making our
broadband video news service available internationally.

BBC World is funded by advertising and subscription revenue and the
cost of the improved video service will be met by advertising around
the broadband news clips.

So if you are viewing video from outside the UK you may see a short
commercial before your clip plays.

If you want to watch clips ad-free, you can still choose to keep the
same narrowband video service we have offered you internationally up
to now.

media selector

The media selector is where you can choose how to watch BBC video

If you click on a video link and you are outside the UK you will see
a 'media selector' pop-up window with boxes allowing you to choose
Windows or Real media, high quality (broadband) or standard quality
(narrowband).

You can use the preferences link on the News Player to change your
choices at any time.

If you have chosen broadband video, the News Player will launch and
your clip will play, sometimes with a short commercial before it.

Some video items - including programmes, live events and reports
from the UK regions - will not be available in broadband for the
time being.

For the first time on the website we will also provide a regularly
updated video summary of international news from our colleagues in
the newsroom of BBC World television.

There is a link to the summary on the front page of the
international edition of the website.

A decade ago watching video online was the preserve of a small,
dedicated and patient group of internet pioneers. Today millions use
the web daily to watch and listen to all kinds of audio and video
content.

Traffic to audio and video on this website has continued to grow as
more people switch to broadband.

We want to make sure we can continue to provide all our users around
the world with top quality video news, and making these changes will
allow us to do that.

We hope that if you are outside the UK you will appreciate the
improved quality of our news video.

Give it a try and if you do have comments or questions, we would
like to hear them.