Editorial - The efficiency of miniaturisation


Welcome to Visions 97. This year we decided to go against the trend - no don't worry, the online version does still exist - the trend we're talking about is miniaturisation. As many things get smaller in our increasingly digital world, Visions has got bigger. This year we have twenty pages of reports, reviews and visions for you, our longest edition so far!

And yet throughout these twenty pages, one subject comes up again and again - miniaturisation. For a mobile PC company, you may not think that a surprise. After all, how are you going to get the latest developments inside an A4 notebook if you don't miniaturise?

But look around. Small is popular everywhere. One of the most widely-reported recent science news items was the announcement that a team of scientists had finally created a laser beam just one atom thick. Smaller is not possible. The fact that this was done at a horribly cold temperature (around -250°C) and that the "laser" was therefore more of a steady "drip" than a power beam was irrelevant. Here is miniaturisation at the limit.

The atom, once considered the smallest building block on earth, is back in fashion. Science magazines are full of reports on "nano-technology", a branch of science dedicated to the investigation of controlling our world at the atomic level. Nano-technologists dream of manufacturing using atomic components. Their tools would be lathes and jigs just a few atoms thick.

In the medical world, the control and make up of genes just a few hundred atoms long is considered vital to our future understanding of the body and illnesses. Genetic engineering aims to manufacture medicines and modify plants and perhaps even animals in order to provide a batter healthier world.

Quite clearly smaller is sexier. The phrase "Bigger is better" belongs to yesteryear. And nowhere is this more evident than in computing and telecommunications - those companies that trek to CeBIT every year. Challenged by a digital world that offers ever bigger mountains of data to process and transmit, the answer it seems is to miniaturise.

Whether it's hard disks packing more data per square centimetre, DVDs with shorter lasers and tinier pits, or memory chips with ever smaller gateways, the trend is clear: only miniaturisation can master the data masses. And don't believe that appearances mean anything either. When Toshiba expands from notebooks to desktops for the first time in ten years, we haven't abandoned the small is beautiful school of thought. Those desktops are full of miniaturised technology too.

Take the latest Intel Pentium processors as an example. They are manufactured to a 0.35 micron process - meaning that the halls and gangways for the flow of data are just 0.00035 mm thick. Compare that with the Intel 4004 which used a 10 micron process - 2857% thicker! Here at the heart of the modern computer it is clear that miniaturisation is the key to progress.

Miniaturisation also gave the Pentium Pro 5.5 million transistors, compared to the 2300 of the 4004. Miniaturisation will also bring us the 256 Mbit memory chip in coming years and predictions are that a 0.18 micron process will then lead the microscopic pack.

But just as there is a limit to how big something gets, there must be a limit to how small. If we carry on at the same rate, the 1 gigabit chip may reach us in 2010. In the 21st century the computing industry will need all the ingenuity of the nano-engineers.

Will memory chips require atom-sized gateways? Is molecular manufacturing the goal for future processor processes? Scientists believe these goals can be reached. So if next year you think Visions has got rather small - don't worry we'll just be back on trend again.