Electronic Design
Analog Chipmakers’ Announcements Signal Trends

Analog Chipmakers’ Announcements Signal Trends

Examine a year’s worth of new products, and what do you see?

What’s happening to “standard” analog parts like amplifiers and data converters? How are analog semiconductor makers improving them? Or have they reached performance plateaus dictated by their inability to accommodate the shrinking geometries of Moore’s law? Is the application-specific analog device the wave of the future?

One way to answer those questions is to look at what new products the larger analog semiconductor companies have announced in the past year to see where they’re placing their bets. This analysis is largely about what Analog Devices, Texas Instruments, Linear Technology, Maxim Integrated Products, and Intersil announced in 2011, with the expectation that, whatever the answers, 2012 will be more of the same.


It’s tempting to look at recent introductions in analog semiconductors and conclude there has been a sea change in the kinds of ICs that companies are making and buying. The shift of manufacturing to China and the growing markets in the developing world have created an environment in which old-fashioned bench engineers are being eclipsed by chip designers (see “Competing ECG AFEs Reveal Chipmakers’ New Business Paradigms” at www.electronicdesign.com).

Offshore manufacturing (read: “China”) specifically and globalization in general have driven traditional analog chip companies to reduce development efforts for new standard parts like op amps and data converters and focus on application-specific analog/mixed-signal devices. Along the way, there have been fewer startups dedicated to small- and mid-volume end-user products in North America and Europe.

The analog designers who would at one time have worked for those companies have, for the most part, migrated into the semiconductor companies. There, they work either as chip designers or applications engineers.

After all, it has always been a challenge to design a front end that is perfectly suited for matching a sensor to an analog-to-digital converter (ADC). As the specs for end-products and standard analog parts have become tighter with each generation, the challenge has only grown more difficult. That’s why there used to be module companies. (Okay, Datel still exists as a subsidiary of Murata. At the extreme end of the ruggedness and performance scale, there will always be a need for custom design.)

Outside of filling low-volume, space, high-rel, and research custom-module needs, however, why not have the semi companies do the really tricky integration on a single die?

Also, now that so many of the performance requirements for new end-user products are coming out of the Far East—while they’re being written by engineers who are primarily interested in manufacturing and manufacturing economics—why shouldn’t the analog chip companies help influence the specs? That’s what a lot of semiconductor program managers have been asking (see “The Method Behind This Analog Product Announcement Analysis”).


Texas Instruments made 22 announcements related to analog products in 2011 (Table 1). Thirteen products or product families fell into the standard category, and nine new products were novel and highly integrated. The standard part announcements comprised four about op amps, three about digital-to-analog converters (DACs), five about ADCs, and a digital-output temperature sensor.

The more sophisticated parts comprised a Class-G headset amp with digital input; a Class-D amp for big-screen TVs and soundbars; an audio codec that integrates a DAC, ADC, and phase-locked loop (PLL); a development kit for a 2.4-GHz wireless headset chipset; a family of sub-1-GHz devices consisting of a transceiver, a receiver, and a transmitter; another sub-1-GHz mesh network kit; a sub-1-GHz RF booster; a near-field communications (NFC) receiver for controlling wireless battery charging; and a psychoacoustic speaker-driver for enhancing the apparent spatial separation of tablet and big-screen TV speakers.


With 35 announcements from Analog Devices, 17 deal with standard analog parts, while 18 are more complex and/or customized for specific applications (Table 2). There are eight instrumentation or sensor-type amplifiers, two DACs, four ADCs, a family of microelectromechanical systems (MEMS) gyros, an RMS-to-dc converter, and two RF transceivers for industrial control.

The complex, custom side features an HDMI transmitter and a dual-port receiver for portable devices, plus an HDMI audio/video front end, a MEMS microphone, multiband mixers with a bandwidth from 700 to 2800 MHz, a MEMS dual-axis accelerometer ruggedized for down-hole use, IF diversity receivers, ADCs optimized for cellular basestations, and an ADC front end with a 5-GHz bandwidth

Also, there’s an inertial-navigation MEMS device that combines three accelerometers, three gyros, three magnetometers, and a barometer; a MEMS inclinometer that’s accurate to a tenth of a degree; a MEMS gyro ruggedized to meet automotive standards; a multi-axis vibration sensor that includes a spectrum analyzer; a “tactical grade” MEMS gyro that can replace fiber-optic gyros for less money; and a 16-channel audio DAC with a 118-db signal-to-noise ratio.

Analog Devices is unique in the depth of its commitment to MEMS. In its case, counting the MEMS products as “analog” makes a great deal of sense. Essentially, the challenge is an extreme case of industrial-control analog design. In the case of a MEMS gyro, for instance, the displacement of the vibrating ring by Coriolis forces is detected using interleaved capacitor structures whose plates are displaced by increments on the order of the diameter of an atom. Sensing the changes in capacitance and extracting a signal in the presence of all kinds of noise requires extremely sensitive electronics.


Linear’s June-through-November “analog” announcements comprise 17 “standard” products, including 14 ADCs, an op amp, bus buffers, and ideal diodes (Table 3). Its more highly integrated products include an integer-N synthesizer and current-sense amplifiers with built-in reference and comparator. Yet Linear also offered innovative products on the “power” side, such as the LTC6803 EV battery monitoring chip (see “Automotive Applications Benefit From Advanced ICs” at www.electronicdesign.com).

The company’s “power” devices range from that battery-stack balancer to energy harvesters and dc-dc converter modules. Its specialized power products support the notion that a big part of the analog business is moving from “standard” to “advanced.” Essentially, Linear, with fewer resources than Analog Devices and TI, chooses to win sockets with ICs that command high margins and long product life-cycles.

“There is price pressure in every market, but price is not number one with the industrial market. The automotive market has had a surge of electronics integration. They have everything from navigation systems to stop-start systems to safety systems and then the special set of automotive hybrid and electric cars requiring some very sophisticated battery stack monitoring chips,” said Linear founder Robert Swanson.

“So we said, okay, we’re going to do that. And we’re also going to continue to be interested in communications, but not handsets. We’ll focus on the build-out of the basestations for 3G, 4G, LTE (Long-Term Evolution) or concentrate on the Ciscos of the world,” he said.


Maxim challenges its engineering teams to create new products so quickly that the company turns out a new product every week—but you don’t hear about them all. You don’t hear about the ones that provide secure communications for point-of-sales networks and Smart Grid metering. Not everything gets a press release.

Maxim’s six classic analog products from 2011 include a current-sense amplifier, three op amps, a low-noise amplifier (LNA) for GPS, and an IO-Link physical-layer (PHY) transceiver (Table 4). Its five more advanced products consist of a stereo audio codec; a power-line communications analog front end/modem that conforms to the P1901.2 standard, which the company helped define; ambient light sensors; a serializer-deserializer (SERDES) chipset for sending video and audio over twisted pair; and an energy-measurement system-on-a-chip (SoC) for industrial control.

Some of these products target the generation, transmission, and distribution aspects of the Smart Grid. Maxim is going after this market just as Linear is pursuing the automotive market.


Intersil provides some of the most surprising examples of innovation in the pursuit of new analog markets with highly customized chips in the whole industry. It’s not as if its managers are betting the company on these products. But when Intersil perceives a solid opportunity, it makes the commitment.

One example is security video. In 2011, Intersil produced a chip set designed for sending analog surveillance video over Ethernet twisted pair in addition to another for sending Internet protocol (IP) video over coaxial cable (see “Surveillance Camera Chips Help OEMs Envision New Markets” at www.electronicdesign.com).

Expect more unusual products for security-cam video in 2012. (There was an “official” announcement of the security-video chips in 2011, but the media coverage occurred in 2010.) In 2011, however, the Intersil pendulum swung back in the direction of standard analog, with a voltage reference, op amps, Class-D audio, and ADCs (Table 5).


Keeping score, 62 of these announcements were about standard analog products with attention-getting characteristics, and 36 were highly integrated parts customized for vertical market segments. Thus, there is plenty of life left in the “classic” analog product category.

Yet many of these new standard parts simply represent smaller footprints, lower power consumption, and lower pricing, compared to existing devices with roughly the same performance. The last blockbuster standard part to hit the industry was National Semiconductor’s 3.6-Gsamples/s, 12-bit ADC12D1800 (see “3.6-Gsample, 12-Bit ADC Revolutionizes SDR Designs” at www.electronicdesign.com), which TI is now happy to acknowledge as the jewel in its high-speed crown.

There may be a few more surprises like that in the lab, struggling to be born. But in general, Moore’s law works against improving analog performance, unless the chip designer adds more digital signal-processing blocks alongside the analog core. (Look for an upcoming announcement from Samplify.)

And isn’t that precisely the message that analog chip makers are sending us with more than half of the new products they announced in 2011?

The Method Behind This Analog Product Announcement Analysis

To test my hypothesis that chip design is eclipsing old-fashioned bench engineering, I looked the product announcements of five semiconductor manufacturers for January through November 2011: Analog Devices, Intersil, Linear Technology, Maxim Integrated Products, and Texas Instruments.

The companies archive those announcements on their Web sites, usually somewhere under the “Newsroom” heading, all sorted by announcement date. I’ve summarized those announcements here in five tables with columns for the announcement date, part number, descriptions, and key features. Also, devices with “traditional” analog functionality have blue backgrounds, and all the “new wave” devices have yellow.

By no means are these all the companies’ new products, though. TI, for instance, says it introduces 600 new products a year, 75 of which are data converters. So, I only surveyed the products that had press releases.

Also, Linear Technology is a special case. Its site only goes back to June, so you’ll have to mentally extrapolate to gauge what the whole year was like at Linear.

Finally, TI’s table omits the contribution from National Semiconductor, which TI bought in September. On the other hand, as TI’s high-speed products product marketing manager Chuck Sanna notes, while National’s existing product lines, skill set, and production capacity complement TI’s, in terms of “analog” market share, National effectively added only 3% to TI’s existing 22% share (according to Data Beans).Therefore, omitting National from this study does not skew the TI picture significantly.

Then there is the question of what exactly is an “analog” IC. The tables include all the parts I arbitrarily selected as falling into the classic “analog” category. I picked all the “industrial” (including automotive and medical) devices, plus the RF devices (excluding a few all-digital down-converters and anything that was all-baseband).

I omitted all the “power” products, even though one might consider a low-dropout regulator (LDO), for example, to perform an “analog” function. This choice also shortchanges companies like Linear that have a strong power-management product focus. On the other hand, it allows me to keep the sample size more or less manageable by omitting the many companies that focus exclusively on power management.

What the tables do show is what the five semi companies that are considered to be the big players in analog and mixed-signal considered the products that they wanted to boast about in 2011.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.