RMG and Associates

Semiconductor & Patent Expert Consulting

(408) 446-3040 / Ron@Maltiel-consulting.com

Semiconductor Technology Consulting

Insightful, Timely, and Accurate


Q&A: Analogic CEO Likes Chaos, Shuns Foundries /EEtimes



For years, analog has been a steady business. Then, suddenly, the bottom has fallen out of the analog sector due to the downturn. ADI, Fairchild, Linear, Maxim, National, TI and others are seeing an unprecedented drop in business.

To explain the current issues in analog, EE Times caught up with Richard Williams, president, CEO and chief technology officer of Advanced Analogic Technologies Inc. (Santa Clara, Calif.), a fabless analog chip vendor. Formed in 1997, the company--sometimes called Analogic--is a supplier of power management semiconductors for consumer, communications and computing electronic devices.

Williams, a candid executive who refuses to pull any punches, explains the changes in the analog market and the current business climate. He also outlines Analogic's strategy, including its push towards integration and its concept of being fabless without foundries.

The company's sales are somewhat tied to the handset, but it is expanding into new markets. LG Electronics is its largest direct customer. Sales to Samsung accounted for 20 percent of its net revenue in 2008.

It has also seen its share of challenges. Revenue for the fiscal year ended Dec. 31, 2008, was $90.3 million, down from revenue of $109.6 million for 2007. Net loss for fiscal year 2008 was $18.4 million, compared to net income of $1.9 million for fiscal year 2007.

Despite the results and downturn, Williams is upbeat. In fact, he says ''chaos'' is good for the company.

EE Times: Where does Analogic fit in the analog world?

Williams: We are primarily focusing on the power management portion, which is a subset of the whole analog space. You have analog in RF. That industry has figured out how to make no money and how to do the most difficult technology on earth. How three suppliers can take all their margins away is beyond me. We don't want to go there. Then, there is analog for signal conditioning and data conversion. Analog Devices is the mother of that portion of the market. We are not doing that other than possibly some interface functions.

Where we come in is the power space. We look at power as four very broad areas. Display and lighting solutions. Battery management. The third category is the interface and audio. The last one is voltage regulation and DC-to-DC conversion. What we set to do is have a total power management solution.

EE Times: Where are we in the IC downturn in relation to analog?

Williams: There is more of a coupling to the economic cycle in analog than there has been in the past. In the past, people took general purpose analog components and they built a battery charger or voltage regulator or an audio amplifier. And they used that for a bunch of different products. That same product served all the markets. And as a result, even if one market sector went down, another market kept going. In this downturn, two things have changed. This is one of the few downturns in 50 years that has cut across almost every market. Maybe medical did not get hit as bad.

More importantly, in the last ten years, the demand of the consumer to get the most optimized performance out of every little widget that they buy has caused power management solutions to become more specialized to that end market. So the day I make that switching regulator and I sell it to satellites, cells phones and computers is gone. The analog market has to track the end market. They are coupled and can't break away from each other.

EE Times: Any exceptions to the rule?

Williams: Maybe LED backlight is a little more generic, but it also has been more commodized. People who did have LEDs had to start dropping the prices to keep out the people that didn't have LED. But suddenly, the number of suppliers in that market went from maybe five to 40 or 50. You know what that means? The price keeps hurdling on down. The products that are still generic are the industrial market. In maybe the future, that will become more specialized.

EE Times: Any signs of an IC recovery?

Williams: We've definitely had increased business. Our lowest bookings month was November. Every month since then has increased. Not necessarily linearly. We believe the top-tier guys are showing true demand increases. The third- and fourth-tier suppliers, especially in the handset space, are handled through distribution. There you can't tell what's replenishment and what's demand.

EE Times: Do you see a second half recovery?

Williams: It's hard to tell, but the bookings are picking up gradually. The handset is picking up. The smartphone is not picking up as fast as the ultra-low cost handset. The gadgets are not picking up as fast. GPS and hands-free car kits have not picked up. The notebook has not picked up in a big way.

EE Times: Will we ever see major mergers and consolidation in analog?

Williams: The lack of a common platform is why that doesn't happen in analog. If you find that special sauce that each analog company has, it's like a recipe. And it's different. It's like consolidating RC Cola, Pepsi and Coke. They just don't believe in the other guy's recipe. And there is no way to mix those product lines. So they will die fighting to preserve their secret recipes. That's why M&A has been challenging and unsuccessful in the broad-based analog market. For example, if you take a mid-tier company and merge it into On Semiconductor, all that happens is that the competitiveness of the mid-tier company goes away.

EE Times: You are a fabless analog player. Some say you need a fab to survive in analog. Some say you need to work with a generic foundry in analog. Which is the real case?

Williams: We started our company on the principal that both views are wrong. A general broad-based foundry commercializes every bit of intellectual property they have among all the analog customers. And that takes away some of the advantages that they have on their competition.

We know the fab model has inflexibilities. For example, I was at Siliconix for 18 years. We ended up with a German fab. Then, we stopped asking what the customer needs. We started asking: 'What can we use to fill that fab?' It was like operations had suddenly taken over the company. We were not asking about the market. We were only asking about the fab. That's the problem. When owning a fab, it's almost obsolete before you sign the paperwork.

EE Times: What is your manufacturing strategy?

Williams: Our vision is to be fabless without foundries. What we did is we partnered with basically DRAM fabs. DRAM fabs have the highest volume, lowest cost and most advanced technology on earth. They only know make DRAM. They have to keep moving the curve and when they do, they have to vacate their last fab. Then, they have to backfill it with something. Normally, they have to backfill it with LCD line drivers and camera image sensors.

So we went to those fabs and took our own process technology. We say: 'Make that process for us and we'll show you how. And then you'll have all of our business or half of it as long as you use our technology and don't sell it to anyone else.' We have now gotten to the point where we can bring up a process without anybody knowing exactly what it does.

EE Times: Who makes your product?

Williams: The two fabs we work most closely with are MagnaChip and Vanguard. We've also worked some with a China fab call CSMC. We have more fabs than we know what to do with it.

EE Times: South Korea's MagnaChip is an IDM and foundry provider. Taiwan's Vanguard was in the DRAM business years ago, but it now focuses on the specialty foundry business. Are you interested in partnering with TSMC?

Williams: Absolutely not. Whatever I teach them, they will use against me.

EE Times: In your fabs, you have developed a so-called ModularBCD process. Is that a generic BCD (bipolar, CMOS, DMOS) process?

Williams: It's a very unique way of doing BCD. BCD is preferred in high voltage and power. In analog, it gives you different combinations of transistors. Bipolar is good for precision analog. The CMOS is very good for digital, because you always have to have interface and CMOS is also good for low-power analog. The DMOS makes robust power devices.

The problem is that everybody's BCD process in the world evolved the B from the bipolar processes. And those old bipolar processes come from the 60s and 70s. And it's high-temperature with a layer of silicon called epitaxy grown on top of wafer. The thickness of the epi sets the voltage. If you change your mind and want to go from 40 volts to 60 volts, now you have to grow thicker epi. Once you do that, all the design rules change.

ST and TI have more BCD processes than anybody else. They also have like 25 flavors of different processes. If I had huge design teams, I can do that. Because we created a process that is modular, the 3-, 5-, 12-, 18-, 30-, or 60-volt sections don't even know they are on the same silicon. There is no epitaxy. There no isolation diffusion. There is no high temperature processing. It's all done in low temperature. Because of that, the 5-, 12- or 30-volt is just a module. That means the circuits are also modules.

EE Times: What node are you on?

Williams: Most of our competitors are at 0.5-micron. And some of them are using 3-micron. We're 0.35-micron on the mainstream effort we're putting in now. The next node we will bring up is 0.18-micron. We will skip the 0.25-micron node. We don't think 0.25-micron is worth it. 0.35-micron has a huge installed capacity. 0.18-micron has a huge installed capacity. The problem is 0.18-micron is more expensive. If you look at the cost of the mask sets at 0.35-micron, I don't have to consider them in my ROI. At 0.18-micron, the glass cost is enough that I have to consider will I make my money back?

EE Times: When will you roll out 0.18-micron devices?

Williams: No earlier than next year. We're actually looking at our first design concepts right now. If it were to be in production at all next year, it would be a limited list of modules.

EE Times: What is your company's strategy?

Williams: It's been the same from the very beginning. We started out with the linear regulator and power savings smart switches and port protection. Then, we added the charge pump for the LED battery. Then, we added the converter for the camera flash. Then we added some voltage regulators and DC-to-DC convertors. Then, we added the battery charger. Then, we had to add over voltage protection. And lastly, we added audio.

Maybe in the future we'll have some interface functions. Now all the pieces are there. Now, we're offering various degrees of integration. For example, a lighting management unit that mixes a camera flash, the backlight and the regulated voltages for the camera module. We also did a mini power management unit (PMU) that mixed one or two DC-to-DC converters with any number of linear regulators and it's all programmable. Many times, those mini PMUs are useful in personal media players, GPS solutions and hands-free car kits. Then, you have the full-blown power management IC, where the battery charger is inside the PMU.

EE Times: In your latest conference call, you talked about the company's diversification strategy. What does that mean?

Williams: Anything that's a one-cell lithium-ion hand-held product, we already have a total power management solution. Now, it's time for us to go after a couple of new vertical markets. One is large screen LCD. In that, we are putting together a total power management solution, including matrix LED backlighting.

The next one after that will be in the netbook space, where there is chaos in architecture. We like chaos. Because when there is chaos, it's bad for TI and Maxim and good for us. If Maxim is strong in notebook, they will automatically become the winner in netbook. But the netbook is not the same architecture as notebooks. So that's good for us. We see it as a spectrum of products and we see a lot of opportunities. We have all the building blocks and chaos is our friend.

EE Times: Are the company's revenues too tied to the handset?

Williams: Under normal orders, it's 60 percent handset, 20 percent computing and 20 percent consumer. We want to continue to grow handsets in absolute dollars, but reduce the percentage of contributions from handsets by getting more revenues in these new verticals, such the TV monitor, notebook and the netbook.

EE Times: Who are your closet competitors?

Williams: Maxim is number one. Maxim, using albeit older technology, still has the skills to make a product in a competitive way. They are very good in product definition and implementation. Occasionally, TI shows up. TI flirts with every market. No one can figure out what is TI's real strategy. They go out on a few dates and then they decide later if they want to make it a permanent relationship.

The very high gross margin companies have gradually been losing market share. Linear Tech was once throughout Asia. They are much less prevalent in Asia today, because they weren't willing to go down the price curve that a Samsung demands.

EE Times: You seem upbeat despite the downturn. Are you feeling doom and gloom now?

Williams: Not at all. This is the best time to do R&D. We see this as an opportunity.