Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > RF/Microwave

Achieving analog success through predictability

Posted: 01 Apr 2008 ?? ?Print Version ?Bookmark and Share

Keywords:Lothar Maier interview? analog industry?

Maier: We just have a business strategy that values predictability.

Linear Technology Corp. CEO Lothar Maier describes the analog chipmaker as "predictable" and even "boring." But Linear is far from dull. In this interview, Maier disclosed that the company considered a private-equity buyout, refocused its product efforts and moved to cut its capital spending. He also discussed Linear's new module strategy and Texas Instruments Inc.'s efforts to grab share in the PoE device market.

What is the overall outlook for analog in 2008?
Look at the World Semiconductor Trade Statistics forecast for analog. It's not always the most accurate forecast, but it's predicting that analog will grow about 10 percent in calendar 2008. The overall semiconductor market is growing at 7 percent. Last year was flat overall; 2008 won't be a banner year. It is not atypical for the analog market to grow 15 to 18 percent per year. But 10 percent is better than a flat year.

Where are we in the analog cycle?
I think analog is somewhat less cyclical than, say, DRAM and even digital electronics. The markets go up and down. Analog is not immune from that. The ups and downs in analog are softer, because it's so diverse. Analog is also very fractured.

What are Linear's growth goals, and how would you characterize the company?
Internally, we guide that the company is planning to try to grow 15 to 20 percent a year. Some years we're successful and some years, we're not. I hate to say it, but we're kind of the boring analog company. We tend to have just a business strategy that values predictability. That doesn't make for very good press. But over the last 25 years, we've gotten pretty good at it.

What do you think of private equity buyouts for the semiconductor industry?
That fad is looking as if it may have passed a little bit. It's certainly something we have thought about in the past. We did some financing to buy back our stock, and certainly, one of the internal discussions was should we buy it all back through some kind of private equity. I think companies that are candidates for that type of financing are companies that need to be fixed up and then sold again.

Looking at our company, we couldn't decide what needed fixing up. We just need to do more of what we're currently doing.

Is there any room for new, VC-backed analog startups?
The question is, is there an opportunity for a startup company to get into the high-performance analog space? It's very difficult, because those opportunities tend to involve relatively small amounts of customers and relatively low unit sales. Venture guys don't usually like to fund a lot of money and have to wait a decade for a return. It's a hard market to break into.

What changes have you seen in Linear's product focus?
Military and space products, in the early days of the company, were important. Later, industrial was important. Then, automotive was important and then communications and consumer. Now, we're coming back around full circle, and consumer is no longer as important as it was a few years ago. Now, automotive/industrial is important for us.

What happened to the consumer markets?
The growth of the company from the post-dotcom bust to a few years ago was driven by high-end consumer products. It was driven by cellphone products. But those markets became more and more commoditized.

Other growth markets for Linear?
The opportunities include industrial. Automotive is certainly a good market. A few years ago, automotive was only 3 or 4 percent of our sales. In the last reported quarter, it was 10 percent of our sales. It's a $100 million-plus business for us. Finally, believe or not, we're getting excited about military and space again. Linear had sales of 3 or 4 percent into the military space. Last quarter, it was close to 5 percent of our sales. That's driven by a lot of satellites, which are going through their life cycles.

What's going on in the PoE device market?
Linear has had a strong presence in PoE. Three or four years ago, we introduced our first generation of products. We're presently in the process of introducing our second generation of products. It's sampling. In the next three months, we'll announce that.

Is TI bombing the prices in the PoE device arena, as some have speculated?
The central theorem of their business is to bomb prices. That's not a new discovery. TI has parts. There are pluses and minuses with their parts. We'd like to think we have the better product.

Linear just rolled out a line of modules, which cover the entire signal chain. What's the strategy behind that?
We think it's a significant change in direction for selling products. Historically, we sold products on a discrete-component basis. We still sell discrete solutions. The modules have certain advantages, which are important to some people and less important to other people. The modules are a complete solution for the customer, meaning that all they have to do is take the module and solder it down to the board. Some customers could do that with discretes. However, the amount of engineering involved is great. What we're finding is more and more companies don't have the in-house skills to develop these analog solutions with discretes.

What goes into a module?
Most of the products we've sold are general-purpose products. The first six or eight products have been power products, such as a DC/DC power supply in a package.

What is Linear's capital expenditure budget, and have you cut it?
We'll probably do $30 million to $40 million this year in capex. We previously said we'd do $60 million, so we're a little bit lighter than we forecasted. We're not consciously pushing things out. Some projects that we had put into our forecast before are not going to be completed as quickly as we had planned to do them. We had some investments in process technology and some additional expansion of office space. However, for most semiconductor companies, this is a pretty modest amount of capital investment.

Linear has two 6-inch fabs. What are your design rules?
Our finest features that we do are 0.6?m. Our widest feature is 7?m. In analog, it's not transistor density that drives the performance of the product. It's volts.

Is Linear pushing the process technology down to lower feature sizes?
I can see us getting to 0.35?m in the future. But I don't see us in a mad race to 45nm.

Many digital IDMs are embracing foundries. How about analog companies?
Analog doesn't lend itself very well to a foundry model. The foundry model is to do everything the exact same way and make thousands and thousands of them. That's not true with analog. One of the tools in the engineer's toolbox is the process. If you go to the foundry, you lose that tool, because all of your competitors have that same tool. That's one of the strong arguments why a lot of analog companies have their own wafer fabs, even though there are fewer and fewer fabs going forward.

The analog space also consists of thousands of products. To get your products to customers with a reasonable lead time is a bit of a challenge with a foundry. Another reason we haven't outsourced is from a quality perspective. We think that we can make our products with higher degree of quality than if we have to go to a foundry.

Has Linear investigated the foundry model?
We do have some foundry work. We do a little bit less than 5 percent of our wafers that we buy through foundries. And we use a little bit of outside assembly. Taiwan Semiconductor Manufacturing Co. Ltd is one of our wafer foundries. We use NS Electronics Bangkok Ltd and Carsem Inc. as our assembly contractors. Typically, we do processes with the foundries that we don't have internally and in-house.

Isn't analog moving toward commoditization?
There always has been, and there always will be, commoditization in analog. The products of a few years ago, which were high-performance analog at the time, will tend to be more commoditized over time.

Is Linear looking to make any acquisitions?
There are not any mergers imminent specifically for Linear. We are always looking for opportunities. But in the past, we've never been able to find one that makes sense for the company. We've done small technology acquisitions before, but we've never taken on a significant acquisition. Besides, the success of an acquisition is not exactly 100 percent.

- Mark LaPedus
EE Times

Article Comments - Achieving analog success through pre...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top