On the two or three occasions when we have had the opportunity to meet before, it was during your tenure as CEO of Intersil Corporation. Back then you were in charge of a medium-sized, but still highly profitable, semiconductor firm with aspirations of getting bigger. Now the situation is somewhat reversed; you are head of one of the largest companies in the industry, but one that seems to struggle when it comes to keeping its balance sheet in the black. What do you feel are the main reasons for Freescale Semiconductor's rather poor performance in recent years, and now that you are in charge, how do you plan to go about changing things for the better?

First of all, let me just qualify your statement about the current health of Freescale’s business. Our financial results continue to demonstrate solid execution with revenues and gross margins exceeding the prior quarter and the prior year. Many people are still unclear about how purchase price accounting rules requiring intangible assets, like brand value and customer relationships, are amortized over time. These charges and other non-cash items associated with our Leveraged Buy Out (LBO) do not reflect on our ongoing operational performance. When these charges are excluded, Freescale is quite profitable and we continue to generate positive cash flow. We continue to produce strong gross margins, operating margins and positive Earnings Before Interest, Taxes, Depreciation and Amortization (EBITDA) figures. We have more than doubled our cash position since the buy out, ending the second quarter of this year with a total of $1.2 billion in cash. We have a great product portfolio that is aligned with some of the industry’s fastest growing business segments. Unfortunately, some of the companies that make up our customer base in automotive and wireless have been caught on the short side of shifting market dynamics. We are confident that they will adjust and rebound, and in the meantime, we have continued to grow our customer base elsewhere in order to make up the difference.

Rumours are rife that you are trying to refocus Freescale's business on to the analogue sector, where the margins are higher, and away from the more commoditised semiconductor business areas. At the moment analogue only makes up a small proportion of the company's total revenue. How much do you think it will be contributing in 5 to 10 years time?

We certainly see great opportunities to expand our analogue business, and are continuing to focus on areas such as these where we feel can provide differentiated solutions to the market. Our power management portfolio is gaining traction in consumer electronics from portable devices like digital still cameras and media players to LED backlighting for large screen displays. However, Freescale is a leader in embedded processors and we will continue to build and invest in the markets for these products as our main focus.

Thanks to your acquisition of SigmaTel you have been able to greatly increase your presence in the mobile media application processor arena. However, wouldn't you say this was going to be a tough gig, given the strength of Texas Instruments' OMAP products with many of the major cell phone manufacturers?

Well, Freescale has had a longstanding position within the multimedia application processor arena thanks to our i.MX architecture. We have a number of customers in the portable media player segment, and i.MX technology is at the heart of the Microsoft-Ford Sync solution for automotive designs. In addition to an expanded customer base, the SigmaTel acquisition has provided us with new mixed-signal IP and design methodologies that complement our existing offerings for the mobile consumer market, expand our product and software portfolio and add scale to our system-level solutions.

Has Freescale's close ties with its mother company Motorola limited its ability to engage with other major wireless OEMs, and if so, can this still be turned around?

This may have been an issue in the past, but since we have become an independent company, we see a pretty level playing field in front of us. The challenge moving forward is coming up with a differentiated solution that can unseat the incumbent chip suppliers at other major handset OEMs. We see Long-Term Evolution (LTE) as the next inflection point, and are currently ahead of the pack on architecture development and implementation. To my knowledge, Freescale is the only semiconductor company to have publicly demonstrated working LTE technology in both handset and network platforms. We expect to be in field trials with infrastructure providers in early 2009, with field deployment in selected markets shortly thereafter.

The industry has been dictated by Moore's Law for four decades now, and has ensured greater and greater proliferation of semiconductor devices, but even without the increasing technical challenges, the financial constraints of this model are now becoming enormous. The cost of building new fabs means that even the biggest of companies will need to partner together, at the same time the volumes of chips produced requires increasingly less essential applications need to be found (or created) to use them. Have we now reached a point where it no longer makes good business sense to continue with it?

As technology scaling continues to become more expensive and technically challenging, technology development partnerships and foundry outsourcing appear to be the most logical solutions. The foundry model can be very cost-competitive because the R&D and capital investment is distributed across a large and diverse customer base. However, the challenge with this model is that you begin to separate from the technology drivers of innovation. With fewer companies focusing on process innovation, semiconductor designers can lose their edge and sacrifice differentiation. We’ve learned that it is not possible to develop truly disruptive technology without intimate collaboration with our customers and end users. We feel our participation in partnerships like the IBM Alliance provides improved research and development and options for manufacturing flexibility to supplement the traditional foundry model.

But is it now time for semiconductor firms to look beyond scaling down processes and trying to concentrate on improving chip performance through better design, and enhanced software instead?

There is always beauty in simplicity and elegance of design. That said, it is extremely difficult to escape the economies of scale. I think we need more discipline across the board. Market expectations for ever-decreasing form factors demand smaller, more integrated solutions, whether that is systems-on-chip or systems-in-package. We believe our core competency lies beyond pure silicon, and our customers expect us to provide the fundamental software that helps them launch their designs more quickly and efficiently.

Freescale has spun out its Magnetoresistive Random Access Memory (MRAM) business unit, to create Everspin Technology, in order to give it a free reign in exploiting this memory technology. However, if the company truly believes that there is a future for MRAM, wouldn't it be better to have this added directly to your bottom line? Is it simply the case now that none of the semiconductor industry's top firms really want to be in memories (whether its in new emerging technologies or traditional product areas) if they can help it, with the likes of Infineon, STMicroelectronics, Intel, AMD, and more, all selling off their interests here or setting up very separate joint ventures?

We feel MRAM is an extremely exciting, non-volatile memory solution, with a great deal of potential, and it deserves the opportunity to benefit from a business model which is focused on its unique market opportunities. The decision to form a new company was intended to accelerate the adoption of MRAM across an entire suite of new applications. We continue to hold an equity position in EverSpin Technology, and we fully expect to benefit from MRAM technology as we develop new embedded products. I think the other examples you cited were subject primarily to commodity pricing pressures and shrinking margins compared to their other product investment opportunities.

The company (as Motorola Semiconductor) saw the value of contributing to the motor industry at a very early stage, and has been the recognised leader in automotive electronics for several decades now. But with the spiralling cost of fuel and a deep recession looking to be on the way in both Europe and North America, do you think this market (which has been particularly strong in recent years) is due for a slump? If so, how are you looking to avoid this impacting Freescale's revenues?

There is no doubt that increasing fuel costs have refocused buyer attention on smaller, more fuel efficient vehicles, and the companies that have developed these types of models will therefore benefit. While growth in some of the more mature markets may be slowing, the market is seeing significant expansion in emerging markets such as Brazil, Russia, India and China for example. Around the world, automotive manufacturers are looking for more efficient solutions, and they see semiconductor devices as one of the keys to overcoming those design challenges. We expect significant growth in plug-in hybrid vehicles to not only meet market demand for better fuel economy, but also meet increasingly stringent emissions standards. It often takes a forcing function like rising oil prices to create new business opportunities, and I believe we will soon see a new generation of personal transportation that is not only going to be good for the consumer but also good for the environment.

Although the number or cars shipped in developing economies is clearly going to increase quite substantially, isn't the fact that the relative amount of electronics involved in these models will be far less than that in the more luxurious cars that tend to be used here in the West. As a result, will there continue to be enough call for the higher end solutions?

There have always been different pricing levels associated with vehicle feature sets, but the interesting thing is that what was once considered a luxury is now a more economical solution. An electric window lift is actually a more cost-effective solution than a stamped, metal manual lift. Many of the domestic OEMs in these emerging markets are looking at exporting their lower-cost vehicles to markets in the West. That means they will need to meet the required safety standards that only electronic systems can provide, and as their markets mature, these features will gradually be incorporated.

Last year the company introduced its Flexis Microcontroller Unit (MCU) platform, which allowed design engineers to migrate between 8 and 32-bit system designs, in order to improve performance or likewise scale their system down in order to save cost. Are there plans to bring a 16-bit version in here, as I would imagine there would be even greater demand for movement between 8 and 16-bit or 16 and 32-bit?

The truth is that our entry-level 32-bit Flexis devices are actually able to deliver 32-bit performance at, or even below, normal 16-bit prices. We are increasingly seeing that the number of bits becomes irrelevant as designers select from a controller performance continuum with a peripheral portfolio that scales to their specific application. As microcontroller-based applications require more connectivity and functionality, designers moving up from 8-bit can immediately benefit from the increased performance of 32-bit.

Company: Freescale Semiconductor Established: 2004 (spun out of Motorola)
Headquartered: Austin, Texas, USA
Employees: 24,000
Business: Microcontrollers, DSPs, Analogue ICs, Communication Chips
CEO: Rich Beyer
Revenue: $5.7 billion (FY 2007)