For most of electronics, 2011 came in like a lion and is going out like a lamb. Carryover from a strong bounce back year in 2010 set markets booming in the early part of the year, but dragging effects—chiefly macroeconomic concerns—have slowed business considerably as the year draws to a close.
But 2011 will be remembered for a number of things apart from the second half slowdown. Natural disasters, including the devastating March 11 earthquake and tsunami off the coast of Japan and the monsoon flooding in Thailand beginning over the summer, wreaked havoc on the electronics supply chain. Mobile products, including media tablets and smartphones, continued their ascent to the top of the consumer electronics heap.
But like any year, 2011 will also be remembered for its share of missteps in the electronics world. EE Times has compiled a list of 10 blunders from 2011 that we believe will be remembered for years to come. We proudly present them here:
10. CSR buys Zoran for $484 million, dumps its crown jewel
At this point, CSR’s acquisition of Zoran for $484 million earlier this year prompts only one sensible response: What were they thinking?
It didn’t take even a year for CSR to figure out it needs to dump Zoran’s digital TV and silicon tuner product lines, and lay off about 800 employees.
Why the connectivity technology specialist CSR wanted Zoran in the first place was a mystery to many in the industry in 2011.
Earlier this year, CSR said that it was looking to bring an “audio-visual context” to content that resides increasingly in the cloud. One of the CSR executives had also said that it wanted to add “Zoran’s strong heritage in imaging” to its own core competencies.
Such a message ("adding audio-visual context to content in the cloud") was an idea a little too conceptual for anyone’s taste; some investors simply dismissed it as "a difficult story to sell."
In the minds of those in the industry and investment community, the bigger questions for CSR were: If you’re really after image processing technology [used in digital cameras, etc.], weren’t there better options than Zoran? And did you really need to buy ALL of Zoran?
And another thing.
Zoran’s real worth might not have resided in the run-of-the mill digital imaging technologies used in cameras, but rather, in things like novel frame rate conversion IP.
Before getting acquired by CSR, Zoran had already gone off the deep end--in the home entertainment world. The company was hip deep in the consumer market, supplying chips for digital HDTVs, DVDs and Bluray recorders, segments where most global chip companies, not just Zoran, have struggled to make money.
Zoran had bought digital HDTV chip vendor TeraLogic through its acquisition of Oak Technology and separately acquired silicon tuner specialist Microtune. Through these moves, Zoran presumably had a number of advanced video technologies used in HDTVs, set¬tops and DVD systems. Zoran also owns novel frame-rate conversion IP, which has been licensed to Toshiba, Funai and Sylvania. One could argue that all this IP were really the crown jewels of Zoran’s crown jewel, which CSR’s brain trust promptly pawned off because they didn’t know what to do with them.
So, who’ll snatch up the jewels, and at what price? We await the dropping of the other shoe. -- Junko Yoshida
Awaiting Act II of CSR-Zoran Drama
9. The unfortunately named 'smart grid'
We reported that the builders of the smart grid know they're running a marathon, not a sprint.
"We should anticipate that new power technologies will give rise to innovative systems that may leapfrog our familiar lineup of power plants, wires and meters," Theodore F. Craver, Jr., chairman of the Electric Power Research Institute, stated in EPRI's State of the Technology 2011 report.
The buildout is a monumental task that will take decades to accomplish. And why is it called "smart" in the first place. Isn't building out the next-generation of power generation a natural evolutionary massive infrastructure problem that needs millions of dollars and plenty of ingenuity?
The 'smart grid' is a misnomer. It should just be labeled the 'next-generation power distribution infrastructure'. There is nothing intelligent about it, except for the fact that newer digital technologies are replacing older analog, or complementing them, and the entire power grid is tied together in a cohesive and efficient manner.
The to-do list items for the smart grid's architects include integrating communications across the grid; developing advanced control methods; tackling advanced sensing, metering and measurement issues; designing advanced grid components that incorporate superconductive materials, power electronics and microelectronics; and hammering out a support and human interface methodology.
There are enough challenges in this list to keep many people innovating and busy. But it is not to make the current power grid 'smart', just updated. As one of our online readers noted:
"The goal should not be to simply do a generational change. At this point it is not enough to do an upgrade. Instead they must create the capability to do future upgrades much more quickly. Take a lesson from the computer network security world, where the infrastructure had to be put in place to respond quickly to the evolution of threats as they are discovered. This is not merely a marathon, it is going to be an ongoing process for the forseeable future. Hopefully they are planning (and budgeting) appropriately for that." -- Larry M.
So the "smart" in the "smart grid" should refer to the intelligence of its builders. And making it secure is the smartest thing smart grid builders can do. — Nicolas Mokhoff
In Pursuit of a Smarter Grid
8. AMD's CEO search saga
Did AMD's board of directors commit a blunder when it forced CEO Dirk Meyer to resign in early January? That depends on your point of view.
But the blunder here from our perspective was the process of identifying Meyer's replacement—a seemingly perpetual CEO search that spanned more than seven months and reportedly included overtures to longtime Intel Corp. exec Pat Gelsinger, Apple Inc. COO Tim Cook, former Hewlett Packard Co. CEO Mark Hurd and Carlyle Group Managing Director Greg Summe, all of which reportedly turned AMD down.
AMD's board and acting-CEO Thomas Seifert said over and over that it was more important that the board identify the right candidate than fill the vacancy by a set date. It's hard to argue with that logic. Still, more than seven months without a permanent chief executive at the helm is no one's idea of a good strategy. AMD is very likely the only electronics firm of any significance that spend the majority of 2011 without a permanent CEO.
To AMD's credit (and Seifert's), the sky didn't fall with nobody sitting in the corner office. The company did brisk business through much of the year, with one Wall Street analyst—Craig Berger of FBR Capital Markets—predicting that AMD was poised to pick up a few points of market share at old buddy Intel's expense. It was only later in the year that AMD's sales started to slip, and that was largely blamed on manufacturing yields at AMD's foundry partner, Globalfoundries (more on that later).
But even as AMD continued to perform for much of 2011, important long-term strategy decisions presumably remained on hold, pending a new leader. Considering that one knock against AMD was that it hadn't moved quickly enough to offer products for the suddenly red hot tablet market (some believe that is why Meyer was shown the door), the extra delay couldn't have helped. Wall Street analysts complained that the lack of a permanent CEO was dragging on AMD's stock price and hurting the image of the company.
AMD finally put the matter to rest in late August, when it named Rory Read, a longtime IBM exec who had most recently been president and chief operating officer at Lenovo Group, the company's new president and CEO. To the relief of many, Read quickly got to work laying the groundwork for AMD's new strategic direction (although those many probably do not include the 1,400 workers AMD let go in November, in Read's first major move). -- Dylan McGrath
No new CEO as AMD's Q2 sales slip
AMD appoints former Lenovo exec CEO
7. RIM nixes Java
Research in Motion announced earlier this year it will unify its Blackberry, Playbook and QNX embedded operating systems sometime in 2012, but the resulting BBX environmentwill not support the custom Java APIs that are the basis for today's Blackberry smartphone applications.
Time will tell whether or not this is a bonehead move. At the very least it's a questionable gamble and one that RIM did not communicate clearly and broadly at its developer conference in San Francisco.
The vast majority of today's Blackberry apps are based on RIM's Java APIs. The company's big message to developers this year was to shift to either HTML5 or native code.
Sure, the future for mobile Java is iffy at best. Java's owner, Oracle, says it will shepherd mobile Java in to the future, but it makes its money in back-end systems for the cloud. Meanwhile it is suing Google over its implementation of Java in Android.
But it's a bad time for RIM to be giving developers a reason to go elsewhere. Android has become the most popular mobile platform by far, Apple still gets all the attention and Windows Phone is making a comeback.
Maybe someday BBX will be the next hot mobile environment. But for today, RIM's market share is slipping, its Playbook is getting panned and many of its developers are going back to the drawing board. -- Rick Merritt
6. DTV chip business exodus
The digital TV chip business developed a bad rap in 2011.
Broadcom, in September, shut down DTV operations in Toronto, China and Pennsylvania. In October, Intel acknowledged it was ditching its digital TV SoC business. Most recently, CSR announced Dec. 12 its decision to dump Zoran's DTV chip business. Trident, which previously acquired Micronas and absorbed NXP's consumer IC team, has been fading fast on the consumer market over the last 18 months.
So, who's left?
Of course, MediaTek and MStar in Taiwan, and a whole bunch of others emerging in China. No surprises there.
If history is any lesson, getting out of the DTV chip business—whether smart TV SoCs or a DTV demodulator chips—just as Taiwanese and Chinese take over the CE market and drive down prices—is only logical. After all, who wants to stay in a business where the margins are shrinking faster than Moore's Law?
And yet, not everyone believes that it's game over.
EE Times agrees.
CEVA announced Dec. 15 that it will enter the DTV chip business by offering IC vendors and OEMs a "universal DTV demodulation reference architecture" based on a CEVA-XC DSP core.
The facile argument is apples vs. oranges: CEVA, not a chip company but a semiconductor IP company, has a different business model from regular fabless chip companies or IDMs. But, well, maybe something like a universal DTV demodulator – addressing different geographies and products (terrestrial, cable, satellite and mobile)—could tempt SoC companies to rethink DTV strategies all over again in 2012.
Stay tuned. -- Junko Yoshida
5. Blue Origin's gaffe
Blue Origin, a secretive rocket developer started by Amazon's Jeff Bezos that also gets NASA funding, was forced to blow up a test rocket in September after a malfunction at 45,000 feet. Bezos waited a week to reveal the incident—after word leaked out. The result? A lot of bad press and the sowing of more distrust about the Bezos operation. -- George Leopold
4. Globalfoundries' yield woes
An overriding theme of 2011 was yield issues at advanced technology nodes by most if not all foundries. But one vendor, in particular, appeared to have been bitten by low yields at the 32- and 28-nm nodes.
On June 16, Globalfoundries Inc. abruptly announced a management shakeup. Doug Grose, the company's CEO since its inception, was replaced on an interim basis by Ajit Manocha (the appointment was eventually made permanent). At the same time, the company announced the pending resignation of Chia Song, chief operating officer. The company attributed the leadership change to demands from customers for more capacity, faster technology development and greater agility.
But a few days later, analysts at Nomura Equity Research said the management changes were related to problems with 32-nm chip yield and the slowness of creating a foundry-like operation at the firm's facility in Dresden, Germany.
In September, AMD warned that its third quarter sales would be less than expected, primarily due to 32-nm yield issues at foundry provider Globalfoundries. AMD eventually exceeded its revised targets, but acknowledged that a series of manufacturing issues limited chip supply and damaged margins during the quarter.
Things got worse last month, when it was widely reported that AMD decided to cancel APUs that Globalfoundries was set to build on its 28-nm process technology and instead use 28-nm gate-last high-k metal-gate manufacturing process technology from rival foundry TSMC. Globalfoundries declined to comment on the reports, but sources at AMD later said the company was making significant changes to its 28-nm roadmap.
"AMD has had so many problems with Globalfoundries this year, from low yields to low ramps and high cost," In-Stat analyst Jim McGregor told EE Times in November. "The relationship between the two does not seem to be providing benefit to either company at the moment."
AMD reportedly renegotiated its agreement with Globalfoundries after experiencing yield problems. Under the new terms, AMD pays Gloabalfoundries only for viable 32-nm dies. That agreement expires Jan. 1, after which AMD will presumably go back to paying for every wafer, viable or not.
Globalfoundries has made a lot of noise and spent a lot of money in an attempt to challenge TSMC for supremacy in the foundry business. One of several advantages Globalfoundries has had in this endeavor is its natural relationship with AMD. But if the events of the past year have damaged that relationship, it could be a significant problem for Globalfoundries going forward. Worse, if the company can't get its yield issues at advanced nodes straightened out, it could have serious problems landing and retaining customers. -- Dylan McGrath
3. Cisco does Flip-flop on consumer camera
Cisco System's zigzag moves into consumer electronics started to unravel in April, when the company killed the popular Flip video camera it acquired with Pure Digital Technologies for $590 million in 2009. The unraveling was complete by July, when Cisco laid off 6,500 people in a major shakeup. The Flip represented the latest and most high profile piece of a patchwork consumer strategy for the company.
The networking giant had spent nearly a decade assembling pieces of a consumer play, starting with the 2003 acquisition of home networking specialist Linksys. In March 2009 it bought startup Pure Digital for a whopping $590 million, arguing that Flip cameras would fuel growth of video on the Net, driving Cisco's core router and switch business.
John Chambers, Cisco's chief executive, warned the company had lost its way in April, in the middle of a year of disappointing results. Cisco was suffering market share losses in its key networking markets. Meanwhile it faced difficulty keeping pace with the fast rise of video capabilities on smartphones.
The July re-org that promised on-going process reviews assuaged Wall Street's concerns for the company. In an EE Times Confidential analysis, we argued it did not go far enough in getting Cisco to re-focus on its core back-end business and give up its consumer ambitions. -- Rick Merritt
2. Solyndra scandal sparks solar meltdown
The U.S. solar industry had a virtual meltdown in August, when three panel startups went belly up including Solyndra, sparking a political scandal and investigation.
Solyndra's bankruptcy came just 15 months after President Barack Obama visited the company's high-profile manufacturing plants along Silicon Valley's I-880 highway. Obama praised the company's innovation in using a novel cylindrical panel based on emerging copper indium gallium selenide (CIGS) technology.
The U.S. Department of Energy granted Solyndra a $535 million loan guarantee. But meanwhile China was pouring even larger subsidies into more mature and lower cost crystalline technology, driving prices as low as $1.20 per Watt, below Solyndra's manufacturing costs.
Analysts forecast China and Taiwan together could command as much as 60 percent of the global solar panel market when the books are closed in 2011. Anticipating the trend, some U.S. equipment suppliers and panel makers including Applied Materials had already shifted operations to China.
In a resulting Solar-gate, politicians drew links between Solyndra backers and Obama supporters leaking emails urging quick approval of the loan guarantee. Solyndra CEO Brian Harrison took much of the hit for failing to inform even committee Democrats of his company's "perilous condition" prior to its collapse.
Despite the hoopla, the Energy Department is continuing to invest in the sector, including a $150 million loan guarantee to 1366 Technologies which is developing a way to use bulk silicon to make wafers used in solar cells. It remains to be seen if Solyndra's problems will sour investors on still-promising CIGs and thin-film solar technologies. -- Rick Merritt
Solar cell maker files for bankruptcy
1. HP's TouchPad/PC/webOS mess
When it comes to the biggest blunder in electronics in 2011, there is no debate: Hewlett-Packard Co.'s Aug. 18 announcement that it would discontinue operations for its webOS devices—including the much-hyped TouchPad media tablet which had been available for a grand total of 49 days—and explore the spinout of its PC business was not only No. 1 with a bullet, it was one of the all-time greats.
Even if you happen to believe that HP did the right thing by throwing in the towel on TouchPad, you can consider the blunder to HP's bringing the product to market in the first place. There's really no room for middle ground here: either HP flubbed by introducing the product, or it flubbed by abandoning it a month and half later.
But the events of Aug. 18 were a multi-faceted blunder, one which resulted in then-CEO Léo Apotheker being shown the door after less than a year on the job.
First of all, the news of HP's announcement was leaked hours before the official announcement and reported by the Wall Street Journal, the Bloomberg news service and others. (Remember how much HP's board loves leaks). By the time HP's announcement hit the wires after market close, the world was already scratching its head about what was going on at HP.
At the time, Apotheker indicated that dropping the webOS products—which also included Palm Pre handsets—was a difficult decision, but that the products were not selling adequately.
Confounding the drama, the HP TouchPad briefly become the No. 1 selling media tablet after HP discontinued it, reportedly selling out in a matter of hours (albeit at a reduced price, $99, compared to the original $499).
On the PC business, many were taken aback by the proclamation that the world's No. 1 PC vendor would consider alternatives for selling or spinning out that part of its business. Several months later, after Meg Whitman replaced Apotheker and undertook a review of the action, Whitman reversed the decision, saying the PC business would stay put.
And, as a final, sad footnote to the events of Aug. 18, HP said earlier this month that, unable to find a buyer, it would open source webOS. That move, along with others, raises serious questions about which now must, in hindsight, be considered one of the biggest blunders of 2010—HP's $1.2 billion acquisition of Palm Inc.
A fine mess, all the way around. -- Dylan McGrath
HP to spin out PCs, drop webOS products