CORDIS Archive
This page has been archived. It will no longer be updated.
by Horst Forster, head of unit, Technologies for Components and Subsystems, DG III-F3.
Presented at Micro and Nano Engineering 95, Aix-en-Provence, September 1995
Six years ago European governments were concerned about a "technology gap" in microelectronics. That gap was measured in terms of the delay with which submicron technology was introduced by European compared to Japanese and American companies. The delay was measured in years. Today, after heavy investment in R&D, engineering and production facilities, the gap does not exist any more. Those European companies - in total seven - who participated in Esprit and in national programmes under the JESSI umbrella, are at par in terms of technology with their counterparts from other parts of the world. If in some cases a gap does still exist, it has to be expressed in months rather than years.
Moreover, the European companies are profitable.
Who would have dared to predict such development only a few years ago? That is why people now say, Europe has its second chance in microelectronics - assuming that the first one in the 1980s, which was driven by consumer electronics in Japan and computers in the US, passed by unexploited in Europe.
Having closed the technology gap, let's see which issues are at stake for Europe now.
First: technological progress in the field of integration is, in the first place, measured by the minimum feature size, and we are heading for 0.25 microns, even 0.18 microns now. The maximum number of wiring levels for logic application is at a level of 5 today, and the number of I/O s is increasing, reaching 1000 for advanced applications, and having a major influence on the packaging technology. Power supply voltage is decreasing as well for desktop as for battery applications, maximum allowed power is substantially increasing, and so on and so forth.
In order to identify these challenges, the Semiconductor Industry Association (SIA) in the US has set-up a "National Technological Roadmap for Semiconductors" presenting a way forward and pinpointing the technological difficulties to be overcome. This is nowadays considered as the reference, as the goals are set and the problems recognised. One of the main problems to be overcome by the technology and component providers is the increasing cost associated with each next generation of technology. as well as the development and production cost. To address the issues, increasingly strategic alliances are being set-up, sharing R&D and production ramp-up cost between different partners.
Similar to the SIA roadmap, a roadmap was established in Europe for CMOS technology in the context of a major Esprit/JESSI project. It was based on the inputs of three major European semiconductor producers. This European roadmap does not show any significant difference to the one from SIA. There are some details different in performance parameters which are due to a focus on complex logic technology and components rather than memory. There is still a slight difference in timing of introduction of a logic oriented technology compared to a memory oriented technology. The European roadmap furthermore has to take account of a number of specific technologies where European companies have a special interest and position, for example, high speed silicon technologies, BICMOS technology, low power glue logic, high speed glue logic and high voltage technologies.
Second: the use of semiconductor technologies.
What matters for European economy and consequently welfare and employment is the value added by semiconductors to systems and the competitive edge they provide in a large variety of products in communication systems, cars and other transport systems, consumer products - audio-visual as well as household equipment - industrial automation, measurement and control and, last but not least, of course, computers.
The semiconductor content of different systems varies largely. And so do the expectations for growth.
For computers the penetration is still the highest, the semiconductor content ranging from 20-35% of the systems value. Communications equipment has been strongly penetrated by semiconductors in the past years, notably on the terminal side; overall, the semiconductor content is on average still around 13% of the systems value. Most industrial applications as well as automotive applications are still much further away from such penetration figures and from what could eventually be expected as percentage of saturation.
The relative strength of the semiconductor markets in different regions reflects on the one hand the strength in the different applications sectors and on the other hand the penetration of the different technologies and components in this applications. Europe has 25% of the electronic equipment market and consumes only 18 % of the worldwide semiconductor production. This is, in particular, due to the weak position in computer production, and notably to the on average lower level of semiconductor content of the systems produced. The world-wide semiconductor consumption (1994) was more than 100 Bio$ in systems with a value of 670 Bio$. This represents 15%. In Europe the consumption was 18.3 Bio$ in systems worth 167 Bio$. This represents only 11%!
If we look at the portfolio of the companies we detect that European semiconductor companies are focused on the application markets where European system houses are strong. That makes them business-wise well suited to provide the right priority, technology and components to sustain the system houses in their way for further technology and component integration in their products. This gives the possibility to exploit more and more intense user-supplier relationships, which serve as a springboard for business which is global.
Relative investment in capacity compared to the competition and future market share are obviously directly related. At a moment where the demand is higher than the offer, the level of capacity offered may be of prime importance and issues as competitive technologies, time to market or service to the customer which are considered important may be only of secondary order with respect to gaining market share. J-P. Dauvin has compared the actual investments by semiconductor industries with the investments required to always fulfil the market demand. This calculation resulted in an installed undercapacity equivalent to 22.4 Bio$ for a ten-year period until 1995, or two times the 1994 Intel sales of missed opportunities !
European technology and component providers are investing in extended manufacturing capabilities or in new fabs. There is, however, a quite diverse attitude between different companies. Some companies invest more cautiously than others. Alain Dutheil, corporate vice-president strategic planning of SGS-Thomson, stated at the last Dataquest European semiconductor conference that "to satisfy the market needs the industry should invest 160 Bio$ from now until the year 2000" and "to maintain its position among the competition a broad range semiconductor company must build one IC megafab every 15 to 18 months". This means an investment of 700 Mio$ each year. A 15-20% growth of the market of today (130 Bio$) represents 20 Bio$ per year. If the major European companies want to maintain their position they would have to invest 1 Bio$ per year. Not all of them have reached these levels yet.
Moreover it is essential for European companies to further invest in the necessary R&D to deliver competitive technology at the right time at the right cost. The role of strategic alliances to reach the right critical mass to execute the R&D and the right mix of competencies necessary to reach the result and to lower the risk of the development is very important and European companies are encouraged to take part in global alliances in order to safeguard their competence and access to future technologies and components.
The performance of a system is often defined or decisively influenced by the interconnecting and assembly technology used. These technologies also may present a significant cost in the system. Packaging technology forecasts major technological challenges necessary to overcome to meet future product requirements and to make best use of the performances offered by the semiconductor technologies and semiconductor components.
The competitivity of a product is often defined by a small detail which however may be a decisive factor to differentiate its product from a competing product. Access to those technologies that provide the necessary competitive edge to a product may be one of the most important factors to increase competitiveness. Technologies combining sensing, actuating and processing in a clever way are of that type. Access to microsystem technologies may be for certain applications one important if not the decisive differentiating factor. What is a car without an airbag nowadays? Or what will a coffee machine be without intelligent sensors which provide you with the coffee brewed exactly as you like it?
Display technology is becoming more and more important, not only because it is a way of communicating between the equipment and the outside and decisive for the quality of use of the equipment, but also its value as a component in the overall system becomes more and more important. The control panels of future cars may well be flat displays. The time will come when you read your newspapers from a flat, flexible panel which fits into your handbag.
Easy access to all these technologies in a coherent way and being able to optimise the use of these technologies together to provide a competitive product and system in the market, will be the future challenge of the system companies.
It is not only about big companies and high volume.
Big users with high volume requirements are normally readily served by the technology suppliers. They have early access to technology needed for their systems. They call these relations "strategic alliances". They may, occasionally, as the case may be, purchase whole companies that offer them access to strategic technologies, functions or components.
However one should not forget the customers with lower volume applications. Their cumulative demand corresponds to a non-negligible market volume. Moreover, they often show high levels of innovation potential. The industrial structure in Europe is such that economy is relying strongly on smaller companies. Big companies subcontract tasks and subtasks of their system developments to smaller companies. Major systems companies rely on the quality and performance offered by subassemblies or subsystems and components provided by smaller companies. It is essential that these smaller companies have access to new technologies and components. Only if all elements in the food chain to deliver a competitive product are competitive, the final product may be competitive. The weakest links in the chain define the quality and competitiveness of a product.
System suppliers should make more and better use of the innovation potential, and innovative technologies, components and concepts localised in small innovative companies. They should further develop a culture of subcontracting work to these companies which offer the best expertise in the field.
Technology and component suppliers should provide access to new technologies and competitive components for small and medium sized innovative companies in a timely and cost efficient manner in order to allow these to develop competitive systems and subsystems even to the extend that this may mean sharing know-how and/or licensing the rights of exploitation of the results from both sides.
Only in such a way the whole industrial web will stay competitive, resulting in a large portfolio of competitive products.
I would like now to address another issue of concern: equipment and materials supply.
I cannot believe that in the long-term there can be a viable semiconductor industry in any region of the world, including Europe, without close links to a nearby (I do not say "indigenous") equipment and materials supply industry.
If Europe wants to safeguard the access to future competitive technologies and components in a timely manner, it will have to also safeguard the bottom of the food chain of producing technology and components. Advanced developments in this area may surpass the economic strength of individual companies involved.
Equipment and materials research will therefore require continued support from national programmes and from Esprit.
It is a business for "entrepreneurs" who know their customers.
Broaden the use of new technology and components will mean in many cases providing a service to the users of this new technology. Concurrent development between the system house and the technology and component supplier of key functions in the system is becoming a necessity to optimise system performances in particular with an increased emphasis to system integration in the future. The technology and component provider will have to be involved in the system aspects of the development, while the system house will have an influence on the performances of the technology.
The request for variety is growing fast. Customers are implementing changes in their products and product mix at a very fast rate in order to maintain their competitively. Time to market is key. The technology provider will have to deliver against this varying demand and will have to cope with these changing requests in a flexible, efficient and fast manner. He must be able to deliver at the right time with a constant speed in this changing environment. This again will need close contact with the customer.
This will allow some companies to pioneer in and to become champions in time to market or customer satisfaction rather than in offering the lowest cost or in the most advanced technology first on the market. This will allow them also to achieve a higher margin per processes silicon or even better a higher margin per die delivered.
Broaden the use of technology: a recipe for growth?
However bright the future looks like and however big the opportunities are, market share and business will not come by itself.
Technology suppliers need to invest heavily in the necessary R&D and manufacturing competence to safeguard their future technological capabilities and in extra capacity to fulfil future customer requirements. They should actively promote their capabilities and performances and should continue to pursue strategic alliances with key system houses.
System houses will have to make more and better use of the innovation potential and flexibility that small companies may offer and technology providers will have to provide small companies which supply mundane but potentially vital parts for complex assemblies access to their technologies in a competitive manner.
European and national industrial support programmes can stimulate progress in that direction. However they will not be able nor should they aim to replace the initiative and responsibilities of industry. Within Esprit 'broadening the use of technology' is becoming a common thread and a priority in all domains. We are using the above considerations to align our activities to this general goal.
Implications for European -funded programmes
Concretely, in the Components and Subsystems programme in Esprit, which involves funding of 500 MECU for four years, we intend, in particular, to:
I believe Europe has a new - a second - chance in microelectronics. I am convinced Europe will be able to exploit it vigorously.
Thank you for your attention.
For more information contact:
Horst Forster
European Commission
DG III F3
N-105 2/25
Rue de la Loi 200 Wetstraat
B-1049 Brussels
tel + 32 / 2-296-8063 - fax + 32 / 2-296-8389
e-mail horst.forster@dg3.cec.be
The URL for this page is /esprit/src/tcshf_1.htm
It was last updated on 18 October 1996, and is maintained by John Magan - john.magan@dg3.cec.be