Friday, April 30, 2010

SEMI India benchmarks India's NSM on global FIT best practices

SEMI India organized a panel discussion in Bangalore, India, today to discuss public policy principles on feed-In tariffs (FITs) and their relevance to the Jawaharlal Nehru National Solar Mission (JN NSM).It also released a white paper on the policy principles and recommended global best practices for solar FITs, with comments on its relevance to the Indian context.

The paper was released jointly by Sathya Prasad, along with K. Subramanya, CEO, Tata BP Solar and Chair, SEMI India PV Advisory Committee, and Dr. J Gururaja, Honorary Executive President, REAF (Renewable Energy Advocacy Forum).

I believe the paper can be downloaded from the SEMI/PV Group's website.

A feed-in tariff (FIT) is said to be a “Government policy to encourage adoption of renewable energy by requiring regional or national electric utilities to buy electricity from renewable sources at above-market rates." If you look at the global FITs statistics, 37 countries adopted FITs in 2008, which now stands at 45 countries.

According to Sathya Prasad, President SEMI India, policy drives the solar PV market. For instance, over 80 percent of 2008 PV demand came from the FIT supported market.

Key findings from the white paper
* India’s JN-NSM framework compares favorably with long-established successful FIT regime (Germany, etc).
* The FIT must be continued until grid parity is achieved,
* Following FIT best practices ensures lower policy cost, faster ramp of manufacturing and growth of PV market.
* Efficient execution/implementation of JN-NSM is crucial.

More details later! ;) Keep watching!

Thursday, April 29, 2010

Manage your software efficiently for significant cost savings!

Ever wondered why you are unable to update your anti-virus software? Chances are that you are running a pirated version of the Windows OS! That itself opens your system and network up to cyber threats and other attacks! Yes, I know! Prices of software, at least the relevant ones, aren't that low! However, you don't have much choice, do you?

Now, if you were using original software in your computer, you won't really face this problem! It means, you are mananging your asset -- in this case, the software. To run your feature rich programs, you need robust software that is sometimes (or, nearly, all the times) expensive!

Okay! Imagine an enterprise using counterfeit or pirated software. Will it face problems? Surely, very serious ones, in the short and long terms. Now what if a public sector undertaking was using pirated software? Perhaps, that would directly impact the services, and in most cases, e-governamce services that it offers.

Keeping all of these in mind, the Business Software Alliance (BSA) has launched a SAM (software asset management) program for public sector undertakings (PSUs) based in the Indian state of Karnataka. This is the first SAM initiative for PSUs in partnership with Centre of e-Governance, Government of Karnataka.Dr. D.S. Ravindran, CEO, Centre of e-Governance, Government of Karnataka, said: “In the last year, the IT spend by the government was close to Rs. 300 crore. All PSUs are now also coming up with their own IT needs and it is important to adopt this standard of SAM with good IT governance practices in order to enable the state of Karnataka to be more productive and cost-efficient.”

The Centre of e-Governance, Government of Karnataka has two key mandates -- putting in place an entire core network infrastructure across the state for e-governance services, and, to expand the capacity and program of running the e-governance initiatives.

Dr. Ravindran added: "We have connected 30,000 offices across the state thru the wide area WAN. We have done over Rs. 22,000 crores worth of procurement over our e-governance platform."

As for capacity building, the Centre assists the State Government Departments to build e-governance initiatives. The PSUs in the state of Karnataka are among the very significant IT spenders. This makes it necessary to create an awareness on the best practices regarding software usage among the PSUs, as 'there is a persistent move to bring in more IT into governance.'

He said there is a need for the PSUs to adopt the good IT practices. In this regard, M.N. Vidyashankar, Principal Secretary, Department of e-Governance, Government of Karnataka, has been providing excellent leadership. A lot of PSUs within the state are participating in the initiative, while some others are on the path of doing so.

Dr. Ravindran said that the Centre is also working on a Karnataka government portal to offer all kinds of services. The need to bring IT into administration is to improve the internal efficiencies and deliver better citizen services.

Lizum Mishra, director, India, BSA, added that SAM has started as a pilot in Karnataka. Maharashtra has been the second state to get into this activity. Other states should follow thereafter, over a period of time. She said that SAM as a concept is still new to India. SAM is an ISO standard -- ISO/IEC 19770:1-2006, published by the BSI (Bureau of Indian Standards) in 2006.

Keshav S. Dhakad, Chair, India Committee 2010, BSA, highlighted that the BSA has programs running in over 80 countries. In 2009, Karnataka partnered with the BSA for the SAM leadership program. Maharashtra joined the program in 2009-10, becoming the second Indian state to do so.

Dhakad noted that SAM is about people, process and software. Effective management of software can lead to significant cost savings. Yet, many corporatons, of all sizes, simply do little or nothing to manage their software. Time for all that to change!

Wednesday, April 28, 2010

Semiconductor-IP directory for FPGAs indexes over 17,000 IP blocks and FPGA devices!

Today, I came across a very interesting story, which stated that Parallel Engines has launched the world’s largest semiconductor-IP directory for FPGAs. According to the company, the site --, indexes over 17,000 IP blocks and FPGA devices!

How does this help the global semiconductor industry? Most critically, customers can now search for semiconductor-IP and retrieve IP vendor datasheets, IP meta-information, and FPGA device configurations. Also, the meta-information includes IP interfaces, LUT, BRAM, I/O and embedded IP resources, costs and packages.

According to the release, Parallel Engines is the brainchild of its CEO, George Janac, Electronic Design Automation pioneer, founder of Chip Estimate; High Level Design Systems, and startup investor. “FPGA design has long been served by a disaggregated IP supply chain,” said Janac.

The next thing to do was to get in touch with George Janac and get his thoughts.

First, I quizzed Janac about the need for such a site. He said that today, most IP portals are really the outgrowth of IPs for ASICs and SoCs.

Janac added: “The FPGA IP market really has no central IP place of its own. Also there is a unique need in FPGA to combine both IP and devices. Much of what is ASIC and SoC hard-IP (I/O, PHY, memory, PLL, etc.) is really embedded in an FPGA device. Hence, the need for a specialized portal. Also, many ASIC and SoC suppliers do not sell in FPGA and vice versa.” Very interesting indeed!

If this is the case, why develop such a site now, and why not earlier?

Janac explained: “Timing is driven by the sizes of the new generation of FPGAs, especially the recent announcements of the upcoming 28nm FPGA devices from Altera and Xilinx, respectively. These device will put the FPGA devices two to three generations ahead in IC technology compared to ASIC. It means that more and more systems that were ASIC, could be placed in FPGA.

“Additionally, we are seeing more heterogeneous FPGA devices from companies like Actel. These have high embedded content for analog, and ARM cores. Finding this kind of IP and mapping to these devices needs a new approach.”

Giving FPGA designers visibility of semicon-IP choices
So, how will the FPGAIPDirectory website help the global semiconductor industry?

He added: “Our main goal is to give FPGA designers the visibility of what the semiconductor-IP choices and help them find the best devices for their applications. We also provide FPGA IP-vendors a place where they can congregate around and interact with users.

“There are approximately 17,000 parts in the directory from 300 vendors. We also list devices from Xilinx, Altera, Actel and Lattice Semiconductor.”

What's the roadmap ahead for this website? Janac noted: “Our roadmap is to provide a set of tools that can combine IP into a platform, and map it onto the most appropriate and cost effective FPGA device. This is our FpgaRFQ beta program that we have announced.”

George Janac and the entire team at FPGAIPDirectory need to be congratulated for this commendable effort.

LTE should benefit from WiMAX beachhead!

This is the concluding part of the Maravedis seminar on 4Ggear: Equipment market update and chipset trends, which I had the opportunity to participate late last month. Here is the perspective from Maravedis' LTE and WiMAX Equipment Analysis Service - Q1 2010, presented by Adlane Fellah, Research Director - Maravedis and Robert Syputa, Senior Market Analyst, 4Ggear - Maravedis.

4G equipment executive summary 2009
One, mobile WiMAX chipset shipments surged. Further, there has been a shift in devices. The year 2009 has also down as the year of transition between WiMAX and LTE. In fact, WIMAX has established a beachhead, while LTE has been gaining in momentum. It is apparent that the LTE chipsets landscape is already crowded.

Maravedis also presented some BWA/WIMAX statistics for 2009 -- 5.6 million chipsets, 4.8 million devices, and 3.5 million new subscribers (Source: 4Ggear Quarterly Report –March 2010).

When one looks at the WiMAX chipset breakdown by device during 2009, indoor fixed CPEs segment accounted for 49 percent, followed by USB dongle and PC cards at 42 percent. Embedded PC (netbook, notebook) contributed 4 percent, while outdor CPEs added up to 3 percent and indoor fixed CPEs accounted for 2 percent, respectively.

Now, let us look at some selected LTE and WiMAX key trends, as per the 4Ggear report.

Among WiMAX chipsets, the vendors have offered differentiated chipsets to address the emerged markets. The aggressive chipset prices have led to higher volume and optimized platforms. In case of LTE chipsets, as of now, the early solutions support LTE only. It may be pointed out here that the early suppliers may not be the long term winners.

Turning to devices, in the case of WiMAX devices, there have been diversified deployments of low cost CPEs, dual-mode USB dongles, and smartphones. In the case of LTE devices, the demonstrators have single-mode followed by dual-mode USB dongles.

As for the 4G equipment equipment itself, it is clear that WiMAX has already established a beachhead for technological progress. Definitely, LTE stands to and will benefit from all that.

WiMAX and LTE chipsets trends/landscape
Now, let's have a look at the WiMAX chipset trends as well as the LTE chipsets landscape and challenges. First, WiMAX!

According to Maravedis, shipments of mobile WiMAX chipsets reached 5.2 million units in 2009. There was a surge in Q4-09 accelerating device shipments in H1-10. Notably, the top five chipset vendors had a combined market share in excess of 95 percent. On the technology front, 802.16e owns the lion’s share with a share in excess of 95 percent as well.

There has also been a shift to very integrated and lower cost WiMAX chipset. It started with the nerging of functions -- WiMAX+WiFi, WiMAX+NPU, BB+RF, etc. Next, there was a migration to lower process geometry in order to save cost and power consumption, leading to 65nm boom. Thereafter, a pressure on price has led to fierce competition and market demand.

There is a distinct move toward LTE, leveraging OFDM technology and BWA ecosystem expertise. Mergers and acquisitions are expected in 2010, owing to scale and lack of funding, as well as perhaps, limited market share.

Now to the LTE chipsets landscape and challenges.

For starters, it has been a crowded LTE baseband market with as many as 12 players in the fray durung Q1-10. The incumbents are Qualcomm, Nokia, ST-Ericsson, and NEC/PMC/Fujitsu (and recently Icera). LG and Samsung are the new cellular entrants. The leading WiMAX players include Altair, Beceem, Comsys, Sequans and Wavesat, have also entered this segment.

Obviously, all of these different actors have different strengths, different strategies. The incumbents are focusing on dual-mode and the cellular ecosystem. The new entrants are paying attention to their time-to-market and broader patents portfolio. On the other hand, the WiMAX vendors are looking at OFDM and TD-LTE leadership.

So, what are the key challenges ahead for LTE? Possibly, one, software complexity, with issues such as interoperability, hand-over, and global coverage. Two, as of now, the size of the niche markets unknown -- whether TDD or LTE-only. Lastly, there could be fierce competition ahead for the existing players with potential late entrants such as Broadcom, Infineon, Marvell, and MediaTek, and possibly Intel!

Makes for a compelling year ahead!

Tuesday, April 27, 2010

Freescale in strategic partnerships with Mentor, Enea and Green Hills

Freescale Semiconductor signed strategic partnership agreements with Mentor Graphics, Enea and Green Hills Software to establish comprehensive enablement solutions for Freescale’s QorIQ, PowerQUICC and StarCore processors.

Freescale and its strategic software partners intend to share IP, invest jointly in product and technology roadmaps, and collaborate on go-to-market activities.

According to Raja Tabet, VP of Systems and Software, Networking Division, Freescale, the company's strategic alliance with Mentor focuses on enabling highly optimized Linux solutions for PowerQUICC and QorIQ processors. Mentor and Freescale will use a common Linux distribution for PowerQUICC and QorIQ processors.

He said: "Mentor is our strategic partner on Linux, We will be aligning with Mentor around a common Linux development distribution methodology. We will no longer have our own distribution methodolgy. The objective for us is to allow seamless migration for our customers from free Linux to commercially supported Linux.

"With Green Hills -- a premier provider of OS and tools --we are making sure that we have a broad support for our product portfolio. When our products are launched, people will have premium support from our partners such as Green Hills. These are for PowerQUICC and QorIQ processors.

"With Enea, the alliance is on enabling the highly optimized versions of OSE, OSEck, Optima Tools and related software for PowerQUICC and QorIQ processors as well as StarCore DSPs."

On the timing of these agreements, Tabet said: "We have been working on a model for quite some time. Ths announcrment, we've been planning it for a while. We wanted to put our silicon out there and add software system partners. The name of the game was pushing the clock speed. We see move from single to multi-core.

"We need to invest much more heavily among ourselves and our partners. We need different partnership strategies. We need to decide what investment to put into software and what investment into our partners."

He added: "We have been working on these strategic alliances for quite some time now and as the leader in providing processors into the networking market wanted to lead the way in setting a high bar for silicon-software vendor collaboration for multi-core devices.

"We believe that effective multicore support necessitates deep, collaborative and early investment between silicon vendors and software partners to truly provide an optimized silicon-software experience. Additionally, given the recent consolidation in the industry, we wanted to reaffirm our commitment to enabling a healthy independent ecosystem of partners and help customers maintain a choice of software vendors for our devices."

So, what do all the three partners -- Mentor, Green Hills and Enea -- stand to gain?

For our partners we hope that we will see an expanded market share. They recognize, and end customers are also recognizing the importance of embedded platforms. They know that Freescale is no. 1 in the overall networking market. Additionally, we sit down and have a discussion around the technologies between us and partners. The customers will also have a choice on the number of software solutions."

On Mentor, Tabet added that the EDA provider has had an embedded division for quite some time. "The growth they have had from Linux is more than many of the existing Linux distribution houses. We are alighing with them very strategically. Linux is becoming very important for networking. The same drivers and IP made available to Mentor, are also made available to other two partners."

He said: “Freescale is expected to benefit from these strategic alliances through enabling our partners (IP sharing and close technical collaboration) to provide Freescale device-specific optimizations and deliver software to customers close to silicon availability. Over time, we expect that such close partnerships will deliver unparalled time-to-market, cost reduction and application performance for our mutual end customers."

The key message here is: Freescale opted not to go and create an all inhouse software strategy. "We have enough inhouse capability and are self sufficient. We also felt that a companion strategy would benefit us. The partners woud have their own value add and differentiating technology to offer."

Monday, April 26, 2010

Analog Devices opens new development center in Bangalore

Jerald G. Fishman, president and CEO was recently on a trip to India, apparently, his first ever, to celebrate 15 years of Analog Devices’ presence in India, and to inaugurate its new development center in Bangalore.He said: “We have opened a new development center in Bangalore, India. We are focusing on signal processing. We are a leader in high performance signal processing solutions. We sell our products into virtually every application.”

Commenting on some trends prevailing today, he added: “Everything has to be greener and consume lower power. We also have products for security, video surveillance, etc. Another trend is healthcare. Our technology enables images that doctors need. For example, we have a solution that enables you to see medical imaging of the heart in four dimensions.”

Speaking on the company’s Bangalore center, he said: “For many years, the Bangalore center has been at the forefront of several designs. We have got the capabilities here as good as anywhere. The people here will become more important for us over the years.”

Dr. S. Karthik, engineering director, precision signal processing (ASC), India Product Development Center (IPDC), Analog Devices India Pvt Ltd, said that the India center opened in 1995 to focus on SHARC processors. The SHARC processor family dominates the floating-point DSP market with exceptional core and memory performance and outstanding I/O throughput. He added that the India PDC is home of the SHARC processor.

In 2002, Analog Devices India added analog and mixed-signal design group to the IPDC. In the same year, Analog Devices and IIT-Madras announced India’s first DSP learning center — the ADI-IITM DSP Learning Centre. The IIT-M faculty, in consultation with ADI engineers, designed the course material and the lab experiments. The centre is designed to train approximately 500 students every year, of which about 400 are from various engineering colleges in the region and about 100 are from industries in India, Asia-Pacific, and Europe.

In 2006, Analog Devices opened the IC test development facility. Karthik said that the company has also added the DSP application engineering capability. In the past 15 years, the IPDC has done products from concept to development in India.

Sunday, April 25, 2010

LTE will see larger deployments, higher volumes than WiMAX!

Late last month, I had the pleasure of attending a Maravedis seminar on 4Ggear: Equipment market update and chipset trends.

It also included a market perspective from Sequans Communications, presented by Craig Miller, VP, Marketing & Business Development. This post will highlight Craig’s presentation. Maravedis’ post will follow later.

4G trends: Device volumes and devices
During 2010, WiMAX device shipments are on pace to triple vs. 2009. The volume is well balanced in 2010. Key growth drivers include handset adoption, deployments in India as well as continued growth in US, Japan, SE Asia, and the MEA.

As for 4G devices, in the beginning (ca. 2006‐2008), the device shipments were dominated by fixed broadband CPE. Today, the device ‘mix’ is shifting toward mobile broadband devices, netbooks and mass-market multimode handsets. Tomorrow, we shall witness more mass-market handsets, plus mobile Internet devices (MIDs) and other CE devices, as well as the emergence of M2M applications.

According to Miller, mass-market prices are here now, enabled by low‐cost, high integration chipsets.

Technology trends
Coming to the technology trends, he said that there have been innovations in WiMAX chipset solutions. These include continued cost, size and power reductions. He added that Sequans has pioneered 65nm single‐die BB+RF integration.

Integrated applications processors have been another trend. Here, Sequans pioneered dual-CPU integration. Some other related trends include hostless VoIP, networking stack, CM, security and OMA‐DM.

Improved mobility is another trend, with enhancements such as improved handover performance, and mobile VoIP support. There have been PHY enhancements as well. In fact, Sequans was first to champion 2Tx for the mobile device. This is said to close the uplink vs. downlink budget imbalance. Uplink MIMO (Matrix A) and improved OTA performance: 64QAM UL, Cat 5/6 HARQ, etc., are also among the PHY enhancements.

There is room for improvement! Areas include: RF front‐end — with FEMs (front-end modules) that support 2Tx, improved PA efficiency, and reduced size and cost. Another area is multi‐radio co-existence and interworking. Here, in the WiMAX‐WiFi space, TDM co-existence solutions are required. Also, an improved handover performance is required for 3G + WiMAX.

There is also a need to focus on production test and calibration, with a definite requirement to improve the test and calibration times.

Complex LTE chipsets to come later!
Presenting some thoughts on LTE, Miller said that Sequans is already providing LTE chips to China Mobile for its TD‐LTE activities, as well as to operators in India, Japan, US and Europe. However, industry‐wide, LTE is still a work‐in‐progress.
Here’s why! One, chipsets are numerous, but aren’t optimized. Two, the IOT (interoperability test) is complex and incomplete. Also, devices are currently scarce, and networks are few and far between. Finally, the frequency plans are extremely complex.

Expect the early LTE devices to be data‐centric, he added. These could be LTE ‘thin modems’ or co‐processor designs. The complex 2G/3G/4G chipsets and products will surely come – but later!

Miller added that LTE will eventually see larger operator deployments and higher volumes than WiMAX — something, which is acknowledged. The volume crossover vs. WiMAX will take place in 2012 or 2013.

As of now, WiMAX is ‘mass market’. The volumes are ramping: in the region of well over 10 million units this year. Also, the product mix is shifting away from CPE. Compelling consumer devices are available. Costs too, have fallen significantly. Finally, chips are maturing, and innovations are in the market.

On the other hand, LTE is coming! While it will be a much bigger ecosystem, the volume crossover is not likely until 2012 or later.

Wednesday, April 21, 2010

Symantec's Internet threat security report on India has few surprises!

Actually, no surprise, really! India is definitely shooting up — in the wrong direction — as far as Internet threats are concerned! The India edition of Symantec’s Internet threat security report, presented by Vishal Dhupar, managing director, Symantec India, has several key findings that will make you sit up and think! Let’s start!!Here are just two among the many data points. One, India, Brazil and Poland — all witnessed growth in malicious activity. In 2009, India accounted for 15 percent of all malicious activity in the APJ region, an increase from 10 percent in 2008. Also, 19 percent of attacks targeting India, originated in India itself in 2009. So, India is rising — both as the country of origin and a target for attacks! Wonderful!

Another one: after the US, Brazil and India are prominent in countries where Web-based attacks originate. Okay, India was also one of the highest ranked countries for Zeus infections in 2009!

So, the key findiings of the threat landscape are as follows: The underground economy remains unaffected by the global economy. Hence, users are still plagued by Web-based attacks. Targeted attacks focus on enterprises — no surprise! Next, attack kits make it easier for novices to indulge in information theft. Finally, malicious activity takes place in emerging countries (read India, among them). I will deal with all of these a bit later.

Dhupar elaborated on some best practices as well that we all — enterprises and end users need to follow. These include:
* Defense-in-depth strategies
* Proactive policy based approach to security
* Test security, and update definitions and patches.
* Educate management on security.
* Emergency response procedures with backup and restore.

As for the way ahead, cybercriminals will continue to innovate to fuel the underground economy. New age Internet technologies and usage will encourage novel propagation vectors. The global scale and origin of attacks requires international co-operation.

The threat landscape
Let's start with the threat landscape first! Credit card information and bank accounts still top advertised items on the underground economy. Credit card dumps actually saw a marked increase in advertisements. Next, multi-tiered commercial model is exploiting the innocent and needy, and their PCs. The cyber mafia is also increasingly mashing up new and old criminal techniques.

There are targeted attacks focused on enterprises. We all know this, don't we? These attacks are frequently carried out by advance persistent threats (APTs). These threats remain undetected to penetrate deeply into the network.

Next, there are attack kits, which allow the novice, unskilled attackers to enter the market with sophisticated tools. The increase in kit activity is notably marked by Zeus.

Key fact and figures on India
First, the bot mania continues! India had an average of 788 bots per day, and 62,623 distinct bot-infected computers in 2009. However, Symantec observed an average of 10,440 active bots per day in the APJ region -- an 11 percent decline from 2008. The reason being -- many command and control centers were seized. Dhupar advised that we need to protect all end points, including mobile phones/smartphones, etc.

Still on bots, Mumbai accounts for 50 percent of total bots in India, up from 37 percent in 2008, followed by Delhi at 13 percent (7 percent, 2008), Hyderabad at 7 percent (4 percent, 2008), and Bangalore 6 percent (same in 2008). Some other cities in this list include Cochin, Chennai, Pune, Bhubaneshwar, Ahmedabad and Kolkata.

Next, malicious code trend types. India ranked no. 1 in APJ and 2nd globally (after the US), for malicious code! Wow!! What are we doing? We are toppers in this area! There has been an increase in worms and propagation through remotely exploitable vulnerabilities, primarily due to Conficker. Also, trojans made up 56 percent of the volume of the top 50 malicious code samples reported in 2009, thankfully, a decrease from 68 percent in 2008.

Still on malicious code trends, India had the highest number of potential worm infections in the APJ during 2009, unchanged from 2008. This, in a country which takes pride in its IT strength is really strange! Aren't the IT guys, especially, protecting themselves enough?

As for the propagation vectors among the malicious code trends, in 2009, Symantec identified over 240 million distinct new malicious programs -- a 100 percent rise over 2008! In 2009, propagation through file sharing executables accounted for 72 percent of malicious codes that propagate -- up from 66 percent in 2008. Also, there is a lack of information centric policy driven end point, data loss prevention and messaging security.

Coming to the phishing trends, brands are being phished by sectors. As per the APJ rank percentage 2009, the financial sector tops at 79 percent, followed by ISP at 8 percent, retail at 4 percent, and several others, such as insurance, Internet connectivity, telecom, consumer hardware, etc., at 2 percent each, respectively.

Of the phishing URLs identified in India during 2009, a whopping 91 percent targeted the financial sector. Oh, 1 percent of the world's phishing hosts and 7 percent of the regional phishing were in India. India continues to lead in these areas too!! Why? Some more India facts on phishing, One, cyber criminals are increasingly localizing phishing attacks. Two, recent examples include phishing attacks on the Indian tax department.

Now to spams! Spam accounted for 88 percent of all email messages observed by Symantec during 2009. No surprises again -- India was the third-highest spam originating country in the world and first in APJ!

Am sure you love forwarding emails to friends and colleagues. We all do it, don't we? How many of us realize that this is spamming? Even if we do, that still doesn't stop us from forwarding fun emails, and definitely, emails that have content dealing with religion, or even sex, jokes, latest buzz, scandals, cricket, IPL (Indian Premier League), politicians, and so on and so forth! See, how we all contribute 'lovingly' to this huge number!

Protect yourself!
Enterprises and end users, do try and protect yourselves better against Internet threats. Follow your company IT policies as strictly as possible! And, keep your Internet security program updated at all times -- especially, the home users.

Am sure, you love checking emails on your smartphones, or even updating your Facebook, MySpace or Twitter accounts, since it is all so convenient. No harm in doing that, but just ensure your smartphone is not open to hackers. Do remember, you are also opening up another end point for possible security breach or attack!

I do tell some of my friends -- do log out completely of the web sites you visit, especially, your personal email programs. Never save your username and password on the Internet -- simply because it is easy to log into a website/an email program. Oh, the next time you forward a 'fun' email, do think of how much Internet bandwidth is being wasted in that activity, besides you spamming your friends!

Finally, hello my dear fellow Indians -- what are we all doing? Why are we all so keen to be well known for "Internet notoriety"? Isn't there something better for us to do?

Tuesday, April 20, 2010

Altera intros 28nm Stratix V FPGA family

Altera Corp. has introduced its next-generation 28nm Stratix V FPGA family, said to be the industry’s highest bandwidth FPGA. Offering up to 1.6 Tbps of serial switching capability, Stratix V FPGAs leverage a myriad of new technologies and a leading-edge 28nm process to reduce the cost and power of high-bandwidth applications.Manufactured on TSMC’s 28nm high-performance (HP) process, the Stratix V FPGA family provides up to 1.1 million logic elements (LEs), 53-Mbits embedded memory, 3,680 18×18 multipliers and integrated transceivers operating up to an industry-leading 28 Gbps.

Gangatharan Gopal, country sales manager and office manager for Altera India, said that the FPGAs are suitable for devices used in next-generation high bandwidth systems. According to Altera, these offer 35 percent higher performance than alternative process options, as well as 30 percent lower total power versus other generations. These also enable the fastest and most power efficient transceivers.

He pointed out that Altera has been delivering innovations from the core to the I/O that provide higher system performance at lower cost and power.

Altera’s 28nm Stratix V FPGAs are said to have broken through the bandwidth barrier. The company is also said to be dramatically improving the density and I/O performance of the FPGAs, and further strengthening their competitive position versus ASICs and ASSPs.

Altera’s devices incorporate the industry’s highest level of application-targeted hard intellectual property (IP) for increased system integration and performance without the cost and power penalty.

The FPGA family itself includes four variants that address a broad range of applications in the wireless/wireline communications, military, broadcast, computer and storage, test and medical markets. These variants include:

* Stratix V GT FPGA – Industry’s only FPGA with integrated 28-Gbps transceivers targeting 100G systems and beyond.
* Stratix V GX FPGA – Supports a wide range of applications with 600-Mbps to 12.5-Gbps transceivers.
* Stratix V GS FPGA – Optimized for high-performance digital signal processing (DSP) applications with 600-Mbps to 12.5-Gbps transceivers.
* Stratix V E FPGA – Highest density FPGA ideal for ASIC prototyping, emulation or high-performance computing applications.

New innovations
According to Gopal, Altera’s new innovations are said to be taking the industry beyond the benefits of Moore’s Law. These include embedded HardCopy blocks, partial reconfiguration — a feature said to be announced for the first time, and embedded 28Gbps transceivers.

The 28Gbps transceivers are said to be an industry first. Other features that facilitate these FPGAs to break through the bandwidth barrier include 1.6Tbps serial switching capability, 1,600Mbps DDR3 interfaces and 1,840 GMACS or 1,000 GFLOPS.

The FPGAs are said to deliver higher system performance at lower power and cost with the aid of the industry’s first variable precision DSP block, register-enriched ALM, higher performance embedded RAM architecture, and highest level of hard IP integration on any FPGA. Besides, the Stratix V FPGAs lead the path to HardCopy ASIC.

Gopal also touched upon the increased efficiency and system performance brought about by the new ALM architecture and the new M20K block and MLAB. The ALM architecture allows adding up to 800K additional registers on the largest device. It is ideal for heavily pipelined and register-rich designs. The new M20K block and MLAB facilitate improved area efficiency and higher system performance, as well as 53Mbits of embedded RAM.

Another feature is the embedded HardCopy block, which has 700K logic elements and 14 million ASIC gates. It facilitates 65 percent reduction in power and 2X performance improvement vs. soft logic. Also, it allows three-to-six months turnaround time for new variants to address new target applications.

Some next-generation applications enabled by the Stratix V FPGAs include multi-100G Ethernet line card, multi-device OTN muxponder, replacing ASSPs with a single-chip solution, and replacing DSPs in radar systems.

Thursday, April 15, 2010

ST outlines future vision; to increase presence in MCUs and advanced analog

I had the pleasure of meeting Francois Guibert, executive VP and president Greater China & South Asia Region, STMicroelectronics, along with Vivek Sharma, regional VP, Greater China & South Asia region -- India Operations and Director, India Desgin Center, STMicroelectronics, during the recently held ISA Vision Summit in Bangalore. Naturally, it was great to meet up with them again, at ST's sprawling facility in Greater Noida last week, where the company outlined its future vision.According to Guibert, ST has been the world's no. 5 semicon supplier for some time now. Its revenue was $8.51 billion in 2009. The semicon giant currently has key collaborations with Nokia, HP, IBM and Ericsson, respectively. In February 2009, the company had formed a JV with Ericsson Mobile Platforms to establish ST-Ericsson.

In terms of the company's global set up, ST has 13 main production sites and advanced R&D centers in 10 countries.

Giving a breakup of ST's FY09 revenue by region, Guibert said Apac accounted for 30 percent, followed by EMEA at 28 percent, Greater China at 25 percent, America at 12 percent and Japan at 5 percent, respectively.

As per its various groups, the ACCI (automotive, consumer, computer and telecom infrastructure) accounted for 38 percent, industrial and multisegment sector (IMS) 31 percent, wireless 30 percent and others 1 percent, respectively. "We have a solid product position via industry verticals," Guibert added.

ST's product strategy for the future includes:
* Maintain world leadership -- especially in industrial, power conversion and MEMS.
* Consolidate leading position -- especially in wireless ICs, digital consumer ICs, automotive ICs and computer peripheral ICs.
* Increase presence in MCUs and advanced analog, with leadership in selected segments.
* Gain market share on selected standard product families.

ST's new products aim provide a better future for all. The focus is on:
a) Energy savings -- New products addressing the issue of volatile and soaring oil prices, and contributing to reduction of global warming.
* Microelectronic devices in home appliances.
* Electric/hybrid cars
* PV technologies, including control and conversion.

b) Healthcare -- New solutions improving the quality of life and addressing the problem of skyrocketing healthcare costs.
* Lab-on-chip for diagnosing influenza and other diseases.
* Insulin pumps, other microfluidics, etc.
* Remote monitoring of patients with chronic diseases.

Guibert also touched upon some of ST's Q4-09 solutions and design wins. These include:

Digital consumer
* Adoption of ARM Cortex-A9 processor for upcoming STB and digital TV SoC ICs.
* Launch of new Freeman SoCs.

* Major design win for a 32-bit MCU based automatic transmission platform from a leading US car manufaturer.
* In powertrain, design wins from major tier 1 OEMs in Europe and Japan.

Co-operation with Mayo Clinic for remotely monitoring of patients.

* Portfolio expanded with next generation acoustic MEMS microphones.
* Design wins for motion sensors in various applications.

Smart cards and MCUs
* STM32 family achieved an industry first for MCUs based on ARM Cortex-M processor cores with devices featuring 90nm embedded flash.
* World's first security certification at level EAL6+ for secure MCU that targets security-demanding government applications, such as ID cards and passports.

Power and advanced analog
* New series of highly accurate LED drivers with automatic power saving.
* Design wins for controller and high-voltage converter chips in LED backlit LCD TV power supplies.

Greater China and India perimeter
Guibert pointed out that ST's Singapore fab represents 56 percent of the company's manufacturing capability. According to him, India, China and Taiwan are the strategic markets for ST in Asia.

Traditional markets include Thailand, Malaysia, Singapore, Hong Kong, Australia and New Zealand. The emerging markets in Asia include Vietnam, the Phillippines and Indonesia.

Giving some more information, he added that the 2009 revenue from this region was $3.4 billion (including revenue from ST-Ericsson). This region also has a 40 percent of ST's total global workforce of 21,000. ST has three IC design centers in this part of the world -- India, China and Singapore. The region also accounted for 56 percent of all of ST's wafer production volume, 81 percent of assembly and test production volume, etc.

Guibert added: "India should focus on developing the local electronics industry. We also help start-ups in India. Further, we speak and discuss with governments on how to go about doing things." He noted that a very strong percentage of ST's libraries are developed in India.

What's happening with the global semiconductor industry?

This is turning out to be quite a week in the global semiconductor industry! First, Intel reported a record Q1, with first-quarter revenue of $10.3 billion. It reported operating income of $3.4 billion and net income of $2.4 billion. Great stuff!

Next, at the Intel Developer Forum in Beijing, China, it outlined plans for a new Atom processor-based SoC. It is codenamed as the Tunnel Creek SoC for IP phones, printers and in-vehicle-infotainment systems for cars. Excellent!

Then, at its 2010 Technology Symposium, TSMC announced that it will skip the 22nm manufacturing process node and move directly to a 20nm technology. In fact, it proposes to enter 20nm risk production in the second half of 2012. Brilliant!

Now, I have a release from Future Horizons that outlines the state of the global semiconductor industry. On the one hand, Future Horizons indicates that semiconductor sales have continued to be very strong. This looks set to continue for the rest of the year, resulting in a 2010 that is massively improved on 2009.

On the other hand, several companies still remain unjustifiably pessimistic and confused about the state of the market. Companies now have an opportunity to dominate the market. Instead they continue to be cautious, undermining their own prospects of making some serious money. Isn't that confusing for the industry? Or, is it confusing itself?

I also have a report from Accenture titled 'Flying blind in the semiconductor industry', which you can read on the PC Semicon Blog.

According to Scott Grant, managing director with Accenture’s Semiconductor Business, the fallout from the global recession, massive fragmenting of the value chain, the rise of a more diverse world economy, and new sales and distribution models have created tough challenges for semiconductor companies when it comes to understanding and predicting demand for their chips and managing their supply chains. This lack of understanding is a dangerous liability in a world characterized by unprecedented volatility and competition.

Oh my, these are clearly mixed signals all over again! Or, is the global semiconductor industry having problems with its 'place-and-route' strategies? Just a fiigure of speech!

For one, is the global semicon industry truly flying blind? Scott Grant has given suggestions as to how the challenges can be tackled. He advises semiconductor companies to focus on three priorities: sales force effectiveness, supply chain integration and optimizing their collaborative planning and fulfillment capabilities.

Look, I'm not an expert! There are several questions that need to be asked, and I hope some knowledgeable folks can answer those.

For one, should the global semicon industry continue to revel in 'inappropriate pessimism',' how will it affect its fortunes in the short and long terms? Or, are those strategies, as advised by Accenture, enough to help the industry? Next, why this need to skip process nodes? What happens to those betting on 22nm? Okay, will all of that have some impact on the semiconductor equipment industry in the long term? What's really happening with the global semiconductor industry?

There is a need to swing back to optimism, folks, as Malcolm Penn of Future Horizons says in his monthly update.

As I'm about to call it a day (or evening or night), comes the news that EDA industry organizations, Accellera and The SPIRIT Consortium, have completed their merger!

Didn't I tell you at the very beginning that this is turning out to be quite a week in the global semiconductor industry?

The next week promises to be fun, especially in the Indian semiconductor/VLSI/electronics industry! Well, it has to do with microelectronics! You'll find out soon. :) Keep reading this blog, friends.

Monday, April 12, 2010

Latest global semicon sales forecast estimates: Cowan’s LRA model

This is a continuation of my coverage of the fortunes of the global semiconductor industry. I would like to acknowledge and thank Mike Cowan, an independent semiconductor analyst and developer of the Cowan LRA model, who has provided me the latest numbers.

Here are the latest forecast results for 2010 global semiconductor sales estimates associated with the forecasting model — the Cowan LRA model for predicting worldwide semiconductor sales.

The table provided below summarizes the latest updated global S/C sales forecast estimates derived from the Cowan LRA Model and is based upon the just published (by the WSTS) February 2010 actual sales results.Source: Cowan LRA model

The updated sales forecast estimate for 2010 (of $298.88 billion) shows a large drop from last month’s forecast estimate (of $316.20 billion).

This corresponds to a decrease in the year-over-year sales growth estimate of 7.6 percentage points, namely from 39.7 percent to 32.1 percent.

It should be noted — the latest WSTS actual monthly sales numbers for Feb. reveal a (strong) downward revision to last month’s Jan. sales (down $0.530 bn) as summarized here.

It should be emphasized that each month’s actual global sales number published by the WSTS is a “lagging indicator” since it is released a full month after the fact.

The Cowan LRA Model, however, “turns” this lagging monthly sales number into a “leading indicator” by virtue of its near-term forecasting capability looking out over the next five quarters.

This is the “beauty” of the model and, therefore, makes it dynamic in the sense that it can be run each month utilizing the most recent actual global S/C sales number published by the WSTS. Thus it allows “rigorous tracking” of the near-term sales forecast outlook for the global semiconductor industry on an “almost” real-time basis.

Consequently, the model’s monthly sales forecasts do not “sit still” but “evolve” with each month’s latest sales number. Since conditions change rapidly and unexpectedly in the semiconductor industry, market forecasters are hard pressed to keep up with these changes.

Saturday, April 3, 2010

Some thoughts on VLSI manufacturing in India

This particular post is an extension of the previous one. Here, Rajat Gupta, MD, Beceem Communications, shares some thoughts on what can be done in India for VLSI manufacturing. So, rather than these comments getting lost in the blog, it deserves an individual post.

VLSI manufacturing: The government of India has to give appropriate incentives to this industry and suitably phase these to allow an ecosystem to build up. The fabless model is fine for companies to transact their business. But, India needs to have this technology or some semblance of it to start with or else it would never be able to count herself amongst the leading nations of the world. How can we achieve that?

a) One of the quickest ways to get started is to create an organization (with equity wholly or partly owned by the government of India) with the charter that that company invest in/fractionally own some of the leading semiconductor manufacturing facilities in Taiwan. TSMC may be very difficult to own even modest percentages, but simultaneously buying reasonable stakes in UMC, Chartered Semiconductor, the IBM multi-platform consortium may be possible with (my guess) at under $1 billion that gives adequate leverage. This organization then makes available fab capacity (that is available to them now due to their fractional ownership) to fabless product companies with intended product sales in India.

b) Repeat the above for packaging, assembly, and VLSI test operations. (The challenges are less here though, but these are just as important elements of the ecosystem).

c) With this, India will “own” leading edge VLSI technology although in a circuitous manner, but given that we missed the boat by over 20 years I think this is a reasonably low cost but effective way to get a starting foothold.

d) Next, and in parallel, the government ought to give incentives to companies to create semiconductor manufacturing capacities on Indian soil but buying re-furbished and lower cost equipment at previous generation technology nodes.

Some policies are required here that allow easy transport back-and-forth of sub-assemblies for repair and replacement.

The business model here would be that the government provides land (to drive the concentration of such technology around three to four zones distributed across the country) but companies should be able to do this on their own should they so choose to. These companies in return bring in (a) captive manufacturing load (whether for sales in India or export, both are ok) and additional spare capacity (or, equivalently, a commitment to scale up with an incentive/penalty structure) to stimulate manufacturing by Indian fabless companies.

e) Eventually, the organization(s) in (d) above will “catch” up with those in (a) above. There will need to be phased modification of the incentives, reducing some, changing some such that over a six-seven year period all incentives can go away and the eco-system will be self sustaining.

f) Well, not quite, we will also need to address (i) the semiconductor manufacturing, assembly, test equipment manufacturing and (ii) the semiconductor manufacturing raw materials (silicon wafers, silicon grade pure chemicals, etc etc). The focus to these technologies have to be phased to the decade after the steps outlined in (a) to (d) have yielded some measurable results.

Friends, please feel free to share your thoughts, comments, etc., and add value.

Friday, April 2, 2010

Did you know that the Indian semicon policy had expired and now requires an extension?

Interesting, isn’t it! I wonder how many folks in the Indiian semiconductor industry, and for that matter, in India and in the Indian technology industry, are actually aware of this! The Indian semiconductor policy was announced back in 2007. It has now supposedly expired on March 31, 2010!

Let me refresh your memory!

Back in September 2007, the Department of Information Technology, Ministry of Communication and IT, Government of India, came up with the Special Incentive Package Scheme (SIPS) to encourage investments for setting up semicon fabs, and other micro and nanotechnology manufacturing industries in India!

The "ecosystem units" were clearly defined as units, other than a fab unit, for manufacture of semiconductors, displays including LCDs, OLEDs, PDPs, any other emerging displays; storage devices; solar cells; photovoltaics; other advanced micro and nanotechnology products; and assembly and test of all the above products.

What has happened since? Lots of initiatives announced in solar PV, including the National Solar Mission. However, as of now, no semicon fabs! SemIndia and HSMC had planned to start one, but somewhere down the line, one doesn't get to hear much about those. In between, the global recession happened, starting Q3-08 and went on to cover most of 2009. Consequently, the global semiconductor industry took a solid beating, and perhaps, the Indian semicon policy got buried in the midst of all of these.

Now, there are published reports of De Core Nanosemiconductors setting up an LED fab in Gandhinagar, Gujarat. It also has a plant for LED lamps in Noida, UP. The Karnataka government recently announced its own semiconductor policy at the India Semiconductor Association's (ISA) Vision Summit 2010.

As per published reports again, the Karnataka government recently approved investment projects worth Rs.943 billion. The proposals include a 50 MW solar PV power project of Moserbaer, electrical switchgears by Easun Reyrole, aerospace components by Swiss firm Starragheckert, and CFL bulbs, energy efficient tube light and LED lamps by Surya Roshini. That's great!

Are these enough? Perhaps, not! Should other states also propose similar semicon policies? Perhaps, yes!

Most importantly, should the Indian semicon policy get an extension? Why not? Why should a policy have an end date in the first place? Do you, for a moment, think that in the year 2050, there won't be the need for any electronic device? Or, even in the year 2020? If yes, then more chips would need to be manufactured. So, the question is: should the Indian semicon policy be extended or enhanced? Perhaps, the latter!

Am certain that the concerned stakeholders will be taking corrective measures regarding India's semiconductor policy.

I also happened to speak with a few industry figures to get their views on India's semiconductor policy. Here are some of their views.

Manjunatha Hebbar: Vice President & Head - Strategic Services (ERS), HCL Technologies Ltd

While I agree that we need an extension of this policy, I feel it also needs a validation from a country competitive standpoint. For example, what other countries are doing to favor industrialization in this space and how they are benefiting from the same? This will help strengthen the policy, while making it relevant for every stakeholder/beneficiary.

Also, consideration for these policies should be from a long term perspective – covering at least 10 years minimum with some periodic validation, say, every two years, to bring more relevance vs. progress made and considering other environmental factors.

Raju Pudota, Managing Director, Denali Design Systems Pvt Ltd

I believe you should highlight that the extension is needed till 2015. Now that the Indian economy has opened up, the proposals will start within the companies who are interested. Since the level of investment is high, they would need time internally to get approval for investment in India. Once this is done, then they would need to approach the Indian government for permits, etc. I believe we need this policy to be in effect till 2015.

Also, note that initially the focus would be to increase the electronic manufacturing ability – and following that, the real semiconductor consumption drive would start. I believe that the policy in 2007 was ahead of its time. We needed an ESDM policy at that time which was followed by a Semi policy right about now.

If the electronics system design and manufacturing ESDM) ecosystem focus is going to start now, we need to give it a couple of years to bear fruit, and then drive the demand for local semi content.

S. Uma Mahesh, co-founder CEO, Indrion Technologies

More electronics from India can still be achieved without a fab, although a fab will surely be of NO harm. Requirements of state policies is well said though specifics about the drawbacks in Karnataka policy and how to overcome them (example, funding amounts) will be good. We have already suggested some changes sometime back, for example, something like including 'rebate', 'discount' to the tune of Indian manufactured amount in the system solution, for one.

I believe manufacturing capability (not necessarily 'self sufficiency') is a sine qua non for any country's long term success. And, it need not be semiconductor (read chips) alone. As we said in our fabcity report -- manufacturing can be any -- of LCD, LED, solar, and of course (as possible) semicon. Interestingly, it appears now that we have missed the manufacturing cycle point for semicon fab to start in India.. and may have to wait for the next such cycle to begin.

Another point of concern is to correct the usage of the original Fabcity.

The simple point is: all we have to do at the minimum is to have a policy and the business will decide whether it will utilize it and if they do decide to, the policy will have influence on it (our Fabcity approach). Not having the right policy will mean a definite no for manufacturing, which is like depriving the basic vitamins for a patient/person.

Appreciate your efforts.

Rajat Gupta, Managing Director, Beceem Communications Pvt Ltd

The Policy certainly needs an extension, both in time and in content. My take:

1. Photovoltaic and solid state lighting (white LED) manufacturing: The government incentives ought to be all linked to sales post-manufacturing not for getting started on manufacturing. In any case, unless there is business viability this will not take-off and if there is, there will be investors willing to put money – the government does not have to become an investor (except in providing real estate subsidy to drive the selection of location where all such manufacturing can be concentrated). How can we achieve that?

a) India has 600K+ villages and all of these have borewells the power of which is (supposed to be) provided free by the government. Assuming about 50 per village (not sure) and ea 10 KVA approx: this is about 500 KVA of “committed” power that the government is actually unable to supply beyond 5-15 percent during peak demand months in summer.

The total such capacity demanded is thus about 300,000 MVA ~ 250,000 MW. These should be completely switched over to solar and the government subsidy worked out such that the savings to the exchequer for say three years is passed on to the manufacturers. I believe this is completely workable and “creates” so much of additional power capacity from the existing power plants. It also gives farmers reliable power based on solar. This ecosystem of solar cells à solar power driven pumps / motor / genset à Govt subsidy that makes this possible and financially viable for both manufacturing and the government.

b) Lighting: Existing incandescent lamps are only about 5 percent efficient, Fluorescent and compact fluorescent are marginally better at about 7-10 percent. White LEDs are probably at about 20 percent, but potentially for significant improvement through R&D over the next decade or so. This needs to be stimulated.

Of the total power being generated and consumed in India, about 20 percent is due to lighting. If this is replaced with white LED, the improvement will be about 10 percent add-on of virtual capacity increase without any additional investment. Again calculating this virtual increase as due to investment, the government can divert that as some form of incentive to make it attractive to the growth of this business/ecosystem.

The above alone may meet India’s CO2 norms for the next several years and the additional carbon credits once these are monetized can provide additional incentive to the government to think along these lines. Their investment could actually become virtually funded by other countries!

2. VLSI manufacturing: The government has to give appropriate incentives to this industry and suitably phase these to allow an ecosystem to build up. The fabless model is fine for companies to transact their business. But, India needs to have this technology or some semblance of it to start with or else never be able to count herself amongst the leading nations of the world. How can we achieve that?

a) One of the quickest ways to get started is to create an organization (with equity wholly or partly owned by the government of India) with the charter that that company invest in/fractionally own some of the leading semiconductor manufacturing facilities in Taiwan. TSMC may be very difficult to own even modest percentages, but simultaneously buying reasonable stakes in UMC, Chartered Semiconductor, the IBM multi-platform consortium may be possible with (my guess) at under $1 billion that gives adequate leverage. This organization then makes available fab capacity (that is available to them now due to their fractional ownership) to fabless product companies with intended product sales in India.

b) Repeat the above for packaging, assembly, and VLSI Test operations. (The challenges are less here though, but these are just as important elements of the eco-system)

c) With this, India will “own” leading edge VLSI technology although in a circuitous manner, but given that we missed the boat by over 20 years I think this is a reasonably low cost but effective way to get a starting foothold.

d) Next, and in parallel, the government ought to give incentives to companies to create semiconductor manufacturing capacities on Indian soil but buying re-furbished and lower cost equipment at previous generation technology nodes. Some policies are required here that allow easy transport back-and-forth of sub-assemblies for repair and replacement.

The business model here would be that the government provides land (to drive the concentration of such technology around three to four zones distributed across the country) but companies should be able to do this on their own should they so choose to. These companies in return bring in (a) captive manufacturing load (whether for sales in India or export, both are ok) and additional spare capacity (or, equivalently, a commitment to scale up with an incentive / penalty structure) to stimulate manufacturing by Indian fabless companies.

e) Eventually, the organization(s) in (d) above will “catch” up with those in (a) above. There will need to be phased modification of the incentives, reducing some, changing some such that over a six-seven year period all incentives can go away and the eco-system will be self sustaining.

f) Well, not quite, we will also need to address (i) the semiconductor manufacturing, assembly, test equipment manufacturing and (ii) the semiconductor manufacturing raw materials (silicon wafers, silicon grade pure chemicals, etc etc). The focus to these technologies have to be phased to the decade after the steps outlined in (a) to (d) have yielded some measurable results.

I will add more views, as and when I get those.

Thursday, April 1, 2010

Global semicon trends and spotlight: UWB and PV developments

Here are the excerpts from the Global Semiconductor Monthly Report, March 2010, provided by Malcolm Penn, chairman, founder and CEO of Future Horizons. This post covers market trends and semicon developments. Those interested to know more may contact Future Horizons.

Market trends – UWB
Ultra-wideband (UWB) is a low power, short-range radio technology that uses a large portion of the radio spectrum to carry high bandwidth data. UWB uses pulse-coded information with sharp carrier pulses using a number of centre frequencies. It has traditional applications in radar imaging.

As well as transmitting data, it has been used for sensor data collection as well as precision locating and tracking applications. UWB communications transmit in a way that should not interfere with other more traditional 'narrow band' and continuous carrier wave used in the same frequency band.

However studies have shown that a number of UWB transmitters increases the noise level and can make traditional communications services more difficult in proximity to the UWB transmitters. This may affect the stability and reliability of existing systems.

Ultra-wideband (UWB) is an unlicensed radio technology that can provide audio/video data streaming over short distances. It is expected to deliver the bandwidth and Quality of Service (QoS) for multi-channel consumer video equipment in the home, which is more difficult for other data transmission technologies like WiFi.

UWB transceivers can communicate at a data rate of up to 480Mbits/sec and operate at 3.1 to 10.6 GHz. ISO/IEC and also the European Commission have approved the standard. As well as streaming high-definition video, UWB can also be used for transferring digital data between domestic entertainment and computer equipment.

Agreeing standards is problematical but UWB is a good concept. It uses a wide slice of the radio spectrum, is efficient, and sends data at almost minimal power over a short distance. Technically it is quieter than background radio noise, yet it can theoretically transmit data signals of 480MBits/s. Realising this potential has been more difficult in practise.

Lack of agreement in UWB technology choice was beginning to hinder the market potential for UWB as established wireless LAN (IEEE803.11) has developed higher bandwidth solutions in the new ‘n’ specification, which has the possibility to steal some applications in higher speed media streaming.

The lack of agreement caused the IEEE Standards Association to disband the IEEE 802.15.3a Task Group. Nevertheless, some of the world's top chip firms still consider the UWB market important, especially for the high volume and potentially lucrative home consumer market. The industry was helped, during 2006, by some more concrete applications for wideband wireless Bluetooth version 3.0 and wideband wireless USB links.

On 28 March 2006, the Bluetooth Special Interest Group announced its selection of the WiMedia Alliance Multi-Band Orthogonal Frequency Division Multiplexing (MB-OFDM) version of UWB for integration with current Bluetooth wireless technology (although it does also see the Bluetooth protocol stack being used with WiFi as well). However, in 2009, the Bluetooth SIG made an announcement concerning Bluetooth 3.0 High Speed, which (notably) did not mention UWB and only 802.11 as the physical layer, which must come as a warning sign for UWB technology.

Unfortunately, UWB has seen a number of technical and standards issues and although the technology shows some promise, the full potential has yet to be realised. Despite the initial optimism over the use of UWB a number of companies have either been taken over or ceased operations. Tzero Technologies joined a shakeout of UWB manufacturers in 2009 that also claimed Focus Enhancements. WiQuest and Artimi also merged with Staccato to pool resources. Intel also announced that it was stopping development in November 2008.

Bluetooth technology using the UWB physical radio layer has the capability to meet the high-speed demands of synchronising and transferring large amounts of data but at the moment appears to be beset by technical problems with a very much lower than expected data rate in current silicon. However, these problems are likely to be temporary and it could still be an ideal solution to enable high-quality video and audio applications for portable devices, multi-media projectors and television sets.

Home video networking applications cannot easily be met using existing wirebased technologies (for installation and aesthetic reasons), and modified existing wireless technology is struggling to meet the latest video networking requirements. UWB could resolve most of these issues at least in a single room -- at short range. It does, however, need the broad agreement of the consumer electronics industry on standards for this to happen. If it does, then home video networking applications could drive UWB with connections likely to be seen on a broad range of consumer products.

In summary, UWB has lost some traction because of standards and technical problems, but Future Horizons believes the technology is delayed rather than dead and our forecast for unit sales shows steady growth from 2011 onward.

Semiconductor spotlight – Photovoltaic developments
Photovoltaic (PV) cells are arrays of cells that convert radiation from the sun into (direct current) electrical energy. This conversion happens without intermediate steps although the efficiency of the conversion can vary. Semiconductor materials used for photovoltaic devices include various types of silicon and other semiconductors with dopants including boron and phosphorus.

The increasing interest in green energy sources including photovoltaic modules has spurred research and development in PV especially in the last four years. Production has been increasing, especially, as government grants and incentives have become available for smaller installations.

Although the total power produced by PV modules installed worldwide has increased significantly in the last two years it is still a very small percentage of the electricity produced by fossil fuels and is still far behind nuclear power stations. The worldwide production from photovoltaic sources is approximately 21 Gigawatts and for nuclear is 370 Gigawatts.

The PV effect is caused by photons of light stimulating electrons into a higher state of energy. When a photon is absorbed, the energy of the photon is transferred to an electron in an atom of the semiconductor cell photodiode. The higher energy electron is able to escape from its ‘normal’ position associated with the atoms in the material to become part of the current in an electrical circuit. The absence of the electron in its usual position causes a hole to form and the current flows through the PN junction with enough voltage and current to drive a load viz.charge a battery or light a light bulb.

Polysilicon (c-Si) is the primary material of wafers used to fabricate crystalline silicon solar cells but cheaper alternatives are being developed such as thin film (CdTe) casting wafers instead of sawing, thin film copper indium gallium and selenium (CIGS), as well amorphous and microcrystalline silicon.

Almost all photovoltaic devices are an adapted photodiode with a large light collecting area. The semiconductor interest is primarily in the PV cell itself, but there is also the need for control and management of the system using microprocessors and also the components necessary for the inverter.

The inverter is needed to bring the low voltage direct current battery or from the solar cell to AC mains voltages for use in the premises or for onward transmission to the grid. Advances in technology and increases in manufacturing competence have resulted in price reductions for PV solar cells and modules, which is typical of other semiconductor devices.

Financial incentives, including attractive feed-in tariffs for solar-generated electricity, have also led to growth of solar PV installations in many countries where these incentives exist. Australia, China, Germany, Greece, Israel, Japan, Spain and the United States are examples of countries where incentives are offered and others are set to follow.

Solar PV installations can either be stand-alone or connected to the grid depending on the location. Stand-alone applications include cellular base stations, telemetry, electrical power for remote buildings, rural communities, parking meters and emergency telephones. Grid connected systems are used in houses and in industrial buildings as a supplementary source of power. Some more extensive arrays for commercial energy production are also grid connected.

Many governments are pushing green energy and solar cells and arrays are important components in the mix of renewable energy options, especially for smaller installations and also on a larger scale in more extensive arrays in suitable locations. Because of this, the demand for PV cells has almost doubled every two years for the last seven years, despite the relatively high cost of installation and a long time for payback (tens of years).

Further reductions in the cost of PV installations will reduce this payback and encourage future market growth. The main suppliers of PV modules include Suntech, Sharp, JV Solar, Q-Cells, BP Solar and SunPower.

The economic downturn did have an effect on the PV market but it has shown some resilience and despite a build up of inventory in the early part of 2009 the inventory has mostly been consumed during the upturn in the second half of the year. The forecast in the figure shows the growth in the generation capacity of PV modules to 2014.

Thin film based technologies will grow its share of total production from 15 percent in 2008 to over 35 percent by 2014. The technology is advancing and prices are falling which will encourage uptake. On the other hand, this growth will be negatively affected by the gradual reduction of government subsidies either as direct grants or the benefit of generous feed-in-tariffs.

To summarise, the PV market is still growing and is forecast to grow at a CAGR of 54 percent between 2008 and 2014. The effective price of PV silicon and modules will reduce by approximately 40 percent during the same period, which will encourage this growth. The forecast does not include the essential peripherals such as inverters, which will also add to the semiconductor total for these installations.