blank
blank
only search Cheshire Henbury

Cheshire Henbury's website is structured around several sub-sites to accommodate the large amount of content. Please pick a topic of interest from the above menu and begin to explore and learn. Or use the Google Search box to the left.

Paul T Kidd's Embedded Microprocessors Pages

Main Home > Legacy Content Home > Embedded Microprocessors Home > Article on Moore's Law

Article on Moore's Law - Will Moore's Law Make Your Application Dreams Come True?

Article on Moore's Law - Will Moore's Law Make Your Application Dreams Come True?

 

Introduction

Moore's law states that the increase in complexity of microprocessors and other types of semiconductor integrated circuits, will double every 18 months. History shows that so far this has been the case and in fact the observation made by Gordon Moore has become a self-fulfilling prophecy. This doubling in complexity has resulted in a phenomenal growth in the amount of computing power available on a single microprocessor chip. However, the cost of developing these chips and providing production facilities is also growing, and the profits from one generation of chips are needed to finance the development of the next generation.

This article addresses a number of issues surrounding Moore's Law. One of these is the extent to which Moore's Law is driving the development of applications. Another issue is the importance of Moore's Law in different industrial sectors. The question of whether in some sectors there are more important issues other than available computing power is considered. The article also addresses what will come after the microtechnologies of silicon, when it is no longer feasible to reduce the size of a silicon transistor. The article deals with the topic of nanotechnologies, seen by some as the step that eventually lies beyond the micro.

Main Issues

A new era is emerging in which computing is becoming invisible - microprocessors are embedded in many objects and interactions with these devices needs to become more user friendly and less reliant on keyboards and other traditional interaction devices. Achieving a situation where computers are ubiquitous and interaction is natural and effortless requires a massive research effort.

Applications centred around broadband residential applications include home entertainment, home automation and learning. The vision here is one of more applications and communications with silicon technology everywhere and always switched on. Moore's Law appears to be fundamental in achieving this vision, by providing increased computational power for signal processing and applications. The goal is a system on a chip, and one of the main drivers for this is cost reduction. Achieving a system on a chip requires more complexity on chips, in other words Moore's Law is an enabler.

Applications in the automotive industry are however different. In the automotive sector there are many application specific components, with up to 50 per cent of the electronic content in cars falling within this category. There are also many different types on applications in cars, including motion sensing, ignition, in-car entertainment, navigation, and environment sensing. Many applications have the potential to either improve safety or fuel consumption. However one of the main challenges is reducing warranty costs. Cars represented a harsh physical environment and there is a significant need to reduce component failure. There is a problem referred to as a design gap. Design productivity is increasing more slowly than the complexity and capabilities of the technology. Moore's Law is not the main issue in automotive applications, there being other more significant problems to address.

Smart card applications provide the basis for secure transactions. Smart card technology provides the personal enabling factor for a networked society. Smart card technology, however, is about three to five years behind the technology that is used in personal computers and other advanced applications. The risk of damage to the smart card electronics means that the usable space on a smart card is limited. There is a need therefore to increase the complexity of the usable space to provide new functionality and better security. Therefore, Moore's Law is relevant, even though the smart card industry is still working with earlier generations of microprocessor technology.

What will happen when the physical limits of silicon technology have been reached? One possible answer lies in nanotechnology. The physics were different on the nanoscale. Surface effects predominated over bulk effects and the materials are also less stable. Nevertheless, there is significant application potential, for example in creating drag free materials for aerospace, or carriers for other materials, or super strong materials. There are already companies selling products based on nanotechnology. Manufacture of nanomaterials is a significant challenge, as is the scaling up of processes for mass production.

The cost of developing nanomaterials is a possible. The development of new generations of silicon technology is paid for from the profits generated by earlier generations of technology. This raised the issue of how the development of nanotechnology would be funded. This may not be a major problem as many different industries are interested in the technology and what needs to happen is that different contributors should work together. The important point about nanotechnology is that it is relevant to industry in general and is not just an issue for the semiconductor industry.

Conclusions and Future Directions

In the area of broadband residential access, the key issues are achieving improved signal processing and development of systems on a chip. System on a chip implies increasing chip complexity and also better design methods, especially at the system level and achieving hardware and software co-design. Moore's Law is therefore important in this area. In the automotive field some of the main challenges are better design methods and improved packaging. For this industry, Moore's Law is not the main concern. In the smart card industry Moore's Law is of relevance as there is a need to add more to a limited area on the card. Looking towards the end of Moore's Law, when the physical limits of silicon have been reached, nanotechnologies may provide a way of enabling the continuation of this law. It is likely however that silicon will still be used in conjunction with nanotechnologies. It is certainly the case however, that many applications still need Moore's law, and will continue to do so for some time to come.

 

Some of Paul T Kidd's Books

Book Covers

Legal Notice: The information posted on the web site is designed to provide accurate and authoritative information in regard to the subject matter covered. Every effort has been made to ensure the accuracy of the information contained in the web site. The information is believed to be correct at the time of publication. Cheshire Henbury cannot however accept any responsibility for the completeness, accuracy and relevance of the information. Information is published with the understanding that publication does not represent the rendering of advice, consulting or other professional services. Specific application in a particular organisation is the sole responsibility of the representatives of that organisation. If expert advice is needed, the services of a competent person should be sought. Please read our terms and conditions (opens in a new window) for use of this web site.

Cheshire Henbury

Address and Phone Details (opens in new window)

Email: Contact form (opens in a new window)

Web address: www.cheshirehenbury.com