Wednesday, June 30, 2010

Upgrade to WiMAX 2 uncertain as TD-LTE gains in momentum!

Here's the concluding part of the Maravedis seminar on Mobile WiMAX Deployment and Migration/Upgrade Strategies.

Robert Syputa of Maravedis focused on the evolving technology and market landscape. He said, the landscape features skyrocketing wireless broadband demand, such as the iPhone. That's also translating into fixed broadband networks. 'Hot'web devices would drive the industry growth. Also, the WiMAX IC vendors were the first to market with LTE.

Some other features of this landscape include multiple-mode WiMAX/LTE: extend or transition? Also, the WiMAX operators have been indicating a shift to LTE. Finally, the upgrade to WiMAX 2 was uncertain as TD-LTE was gaining in momentum. It is still early to tell what the commitments to WiMAX 2.0 would be.

Definitely, the early 4G adoption trend is exceeding many operator’s expectations due to broadband traffic demand.

Commenting on the deployment considerations, he said the fit of wireless technologies would be according to the market needs. This would take into consideration factors such as degree of mobility and roaming, build or migration to flat IP, and response to competition with 3G/LTE.

Some points to consider for device and network upgrades and migration include: how to migrate devices from WiMAX to LTE, changes in core networks, and impact of SONs, femtocell. The degree selected market is driven by mobile web devices. Maravedis also listed various candidate bands identified for IMT.

Who will follow WiMAX?
So, who will likely follow the WiMAX path? According to Syputa, these would be many of the greenfield operators, operators with local industry requirements, vertical applications, and semi-licensed deployments.

The 3GPP(2) operators will likely follow the LTE path! There will be a lower probability to adopt WiMAX 2.0, but there may be a need for some WiMAX collaboration. Also, there could be an availability of heterogenous networks, and some symbiosis with LTE. the caveat: the market is in a state of flux as LTE looms!

Avait Networks happens to be the no. 1 as the WiMAX infrastructure market share leader in India ((>42 percent), besides being a global leader in wireless backhaul network evolution. It has installed over 500,000 systems globally.

Jonathan Jaeger, WiMAX Solutions Marketing, at Aviat Networks discussed the mobile WiMAX deployment and migration/upgrade strategies and highlighted the critical elements of migration/upgrade. Note that ALL elements are necessary for success. These are: Spectrum/technology -- frequency, channel size, bandwidth, and WiMAX release 2 (802.16m) or TD-LTE; mobile/subscriber stations/base stations -- baseband and radio/antenna, and ASN-GW/network elements.

Migration/upgrade decisions
According to him, 16m/TD-LTE migrations/upgrades are not viable options for all operators. All elements listed by him are said to be necessary for success. Operators also need to focus on the frequency, amount of spectrum, services, installed base of devices, system architectures. There is a need to also focus on 16e 'Enhanced.'

Saturday, June 26, 2010

Mobile WiMAX deployment and migration/upgrade strategies

This week, I had the pleasure of attending a Maravedis seminar on Mobile WiMAX Deployment and Migration/Upgrade Strategies, sponsored by Aviat Networks. Thanks a lot to Maravedis for providing me this opportunity.

Adlane Fellah of Maravedis provided a general overview of the mobile WiMax scenario, while Robert Syputa of Maravedis touched upon the evolving technology and market landscape. Later, Jonathan Jaeger, WiMAX Solutions Marketing, at Aviat Networks discussed the mobile WiMAX deployment and migration/upgrade strategies.

Giving a general overview, Adlane Fellah said that the global WiMAX industry in 2009 was as follows: the total WiMAX market size $1.36 billion. About 5 million WiMAX chipsets were shipped -- +332 percent growth Y-o-Y. Also, 3.5 million new WiMAX CPEs were deployed. The WiMAX base station sectors –10.9 percent. However, the end-users want broadband wireless now, while the operators are confused by future evolution.

Key trends in LTE and WiMAX
Looking at the LTE and WiMAX key trends from 4Ggear report, specifically, for WiMAX chipsets, vendors have been offering differentiated chipsets to address the emerged markets. Also, there has been aggressive chipset pricing: higher volume and optimized platforms.

As for LTE chipsets, the leading chipset vendors include Qualcomm, ST-Ericsson and Nokia. The early solutions support LTE only. However, it is believed that the early suppliers may not be the long term winners.

Coming to WiMAX devices, we have seen diversified deployments of low cost CPEs, dual-mode USB dongles, and smartphones, etc. As for LTE devices, demonstrators = single-mode followed by dual-mode USB dongles.

With regard to 4G equipment, WiMAX has established a beachhead for technological progress, and LTE will surely benefit from it.

WiMAX chipset trends
Next, we take a look at the WiMAX chipsets trends. Shipments of mobile WiMAX chipsets reached 5.2 million units in 2009. There was a surge in Q409, thereby accelerating device shipments in 1H10. It must be noted that the top five chipset vendors market share is over 95 percent. Also, 802.16e owns the lion’s share with over 95 percent.

There has been a shift to very integrated and lower cost WiMAX chipset. Functions, such as WiMAX+WiFi, WiMAX+NPU, BB+RF, have merged. We have also seen a migration to lower process geometry to save cost and power consumption – leading to 65nm boom. There is a pressure on prices leading to fierce competition and market demand. Operators definitely need more devices.

We have had a move towards LTE, leveraging OFDM technology and BWA ecosystem expertise. Some have already sampled chipsets.

Some mergers or acquisitions are also likely in 2010, because of the scale and lack of funding, and limited market share.

The concluding of this report — in the next blog post — will look at the impressions of Robert Syputa of Maravedis on the evolving technology and market landscape, as well as Jonathan Jaeger of Aviat Networks on mobile WiMAX deployment and migration/upgrade strategies.

Thursday, June 24, 2010

Students, show off your technical talent @ Mentor’s University Design Contest 2010!

I really envy the students of today! They live in such lucky times!! There are so many opportunities for them to learn, grow and flourish, and so many avenues to venture!!!

Here’s one more opportunity for those brilliant students from various engineering colleges across India. Friends, you can participate and show off your techical talent by participating in Mentor Graphics’ University Design Contest 2010!

I was elated when Veeresh Shetty of Mentor Graphics informed me today about this great contest! Thanks a lot, Veeresh, and Mentor Graphics India, especially, my friend Raghu Panicker, for taking time to think about spotting talent among students.

Time to participate and win, dear students!
First, what do the students get out of this contest? Well, there’s prize money involved! The top three winners will earn a Certificate of Recognition. There will be one Winner team followed by two runner-up teams. The winning team will be awarded a cash prize of $3,000, 1st Runner-up – $2,000 and 2nd Runner-up – $1,000, respectively.

So, what’s the purpose behind this contest? Well, this is the inaugural Mentor Graphics University Design contest 2010 in India.

The objective/focus of this contest is to provide an opportunity for engineering students to showcase their technical talent and competency using Mentor Graphics tools. The company expects to grow this contest with increased representation and participation, and draw maximum number of contestants.

What’s the aim?
The contest aims to provide an opportunity for engineering students from different technology streams to participate in the contest and demonstrate their technical capabilities, problem solving, project management and design skills in a contest against their peers.

It will also provide all participants with new perspectives on how design teams from different disciplines and different regions of the country approach a common design problem. The engineering students will also get a chance to meet experts to showcase and present their ideas, and network with fellow participants to discuss newer ideas of working on a project.

And here’s the contest!
What are the contestants supposed to do? The DUT will primarily consists of five blocks:
1. Fetch (Must be designed in System Verilog)
2. Decode (Must be designed in Verilog)
3. Execute (Must be designed in VHDL)
4. Write Back (Must be designed in Verilog)
5. Testbench (Must be designed in System Verilog)

What happens once students have submitted their projects? The contest involves the following three stages:

Stage 1: Preparation and Submission of Paper/Project
Stage Two: Poster Presentation
Stage Three: Paper/Project Presentation

Technical committee and judges
The technical committee for this contest will be techno-experts from Mentor Graphics, headed by Rajeev Sehgal, Development Engineering Director. Each team’s paper submission will be reviewed by the technical committee based on:
* System Functionality : The hence designed MCU will be required to correctly execute sample instructions file.
* In case the complete system is not functional unit level functionality will be considered.
* RTL Code (Apart from TestBench) must be synthesizable.
* Timing of unit blocks must be honored.
* Coding style and Block partitioning will be considered.
* Synthesize the DUT and show it working on a FPGA board. Use of ethnography and contextual research.

In addition the next two stages will be reviewed based on:
* Clarity and organization of the project/solution.
* Relevance and clarity of presentation material (slides, etc).
* Quality of argument used to justify why the solution is worthy of consideration.

Early this year, I was also part of an initiative, again involving Mentor Graphics, this time with the RV-VLSI Design Center, where a similar design contest was rolled out for students. Look forward to seeing more such initiatives! Best of luck to the students participating in this contest!

Wednesday, June 23, 2010

Xilinx intros 7 Series FPGAs

Xilinx has introduced its 7 Series FPGAs that claim to slash power consumption by 50 percent. These also reach 2 million logic cells on the industry’s first scalable architecture.The company has introduced the Virtex-7, Kintex-7, and Artix-7 FPGA families, which promise:
– 2X more usable performance.
– 2X better price-performance with new class of FPGA.
– 2X better system-performance with Ultra High End FPGA.
– Innovative Virtex-based high volume offering.

Power is said to be the key limiting factor. At Xilinx, it is the top priority at 28nm. The Xilinx 7 Series pushes the boundaries for FPGAs. The three new families are based on a unified architecture.

The Artix 7, when compared to Spartan-6, gives 30 percent more performance, 35 percent lower cost, 50 percent less power and 50 percent smaller footprint. The Kintex 7, when compared to Virtex-6, gives comparable performance, 50 percent lower cost and 50 percent less power. Likewise, the Virtex 7, when compared to Virtex-6 is 2.5x larger, and provides up to 2M logic cells, 1.9Tbps serial bandwidth, up to 28Gbps line rate, and EasyPathcost reduction.

Users can easily scale applications with Series 7 unified architecture and design portability. Enhanced productivity is possible with Targeted Design Platforms. The foundation for Xilinx’s next generation Targeted Design Platforms are:
* Optimized ISE Design Suite.
* Expanded eco-system enabled by Plug and Play AXI based IP.
* Targeted reference designs accelerate development.
* Scalable boards using FMC Base.

Victor Peng, senior vice president, Programmable Platforms Development, Xilinx, also touched upon Programmable Foundation for Extensible Processing Platforms (EPP). It has unrivaled levels of system performance and integration, and ARM based software centric design environment.

This 7 Series will drive the next wave of FPGA adoption, thanks to:
* Increasing market demand for bandwidth coupled with flexibility.
* 2x improvement in power/performance/ capacity/price.
* Major productivity boost with Targeted Design Platforms.
* Fewer ASSPs to choose from/ASICs too risky.

Synopsys’ Dr. Aart de Geus at SNUG 2010 India!

"Moore's Law is absolutely alive and well," stated Dr. Aart de Geus - CEO & Chairman of the Board, Synopsys Inc., while delivering the keynote '20/20 Vision for 2010' at the Synopsys Users Group (SNUG) 2010 event in Bangalore today. He added that systemic collaboration is getting more and more important. "We need to have a joint vision of where our field is going."To make the 'machine' called SoC work, one needs to look simultaneously at economics and technology, and hence the word, techonomic.

Commenting on the global economy, he said, the industry had just come out of a very severe recession. Last year, Dr. Geus had introduced the recession compiler.

Today, there's a clear sense of turn, and a huge shift in the global economy during the recession. According to him, China will pass Japan and become the second largest economy. China has continued to evolve quite a bit, and so has India. Dr. Geus also introduced the recovery compiler six months ago.

He added that people on one side are looking at how to minimize costs and risks. How does this impact semicon? Most semicon companies are now reporting good results. Today, there has been about 5.8-6 percent of growth, indicating a steady state. Semicon is in the center to drive growth.

In the foundry world, there has been some consolidation, and you now find some really large ones. So far, semicon has rebounded somewhat much faster. The memory folks are also feeling pretty good. It must be noted that during the last three years, they invested really nothing in capex, and some players also disappeared.

If one were to look at cool killer applications today, there's certainly a theme around video more and more on mobile apps, HD, 3D, etc. All of this is leading to the fact that bandwidth and storgae will grow even more. Smart grids are also clearly becoming more important In future. The word 'smart' will be critical. "Eveything around us will commmunicate in some form or another, in future," Dr. Geus added.

Touching on active advanced designs and tape-outs, during Q1-Q2 there was some slowdown. Also, the 32nm/28nm nodes have been coming up pretty rapidly.

The race for advanced nodes is still on -- you can see it among foundries. Also, some node skipping has been happening as well, essentially taking advantage of some platform. Finally, the older nodes are getting more and more squeezed for costs.

Cost is key! Hence, it is interesting to see that Moore's Law is still evolving. It is far from over, noted Dr. Geus. The RoI on Moore's Law is also decreasing. As we go on to smaller nodes, it is becoming a value of integration. More functionality is also moving toward software.

To a query, he responded that consolidation of EDA is massive and it will continue. "We also need to be able to be more sensitive to costs, even as more design centers take place here."

Dr. Geus added: "We are by far the no. 1 in our field. We spend over 30 percent of our revenue in R&D." Synopsys uses about 10-15 percent of the revenue for acquiring new technologies.

It was really great to see a jam-packed room, besides the fact that the keynote was held first thing in the morning!

Monday, June 21, 2010

Growing Indian power electronics market provides host of opportunities

In case you are an enterprenuer, or an aspiring one, in electronics and wish to invest or manufacture in the sector, but don’t know where to go, here’s a welcome relief — in form of a report on the power electronics sector in India.

The India Power Electronics Market Report – 2010 has been developed by Dhaval Dalal and Ram Kumar, on behalf of Innovatech Switching Power India Pvt Ltd in Bangalore, India. I was delighted on being contacted by Ram Kumar, MD, who was kind enough to share some bits of this report.

The unprecedented growth in the Indian electronics demand (estimated at $50 billion for 2009), has spawned a corresponding spurt in the domestic power electronics industry. While this growth has been acknowledged in industry circles, no specific data exist to understand this phenomenon – this report aims to fulfill this gap.

The report highlights the peculiarities of Indian industry by identifying unique areas of growth which require special attention from industry participants. It also highlights the gap between the domestic demand and supply which is currently fulfilled by imports. Conversely, areas where the Indian industry contributes to the global demand by exporting products/services are also highlighted.

Coming from technology/strategic marketing background and with an unmatched access to the decision makers and trendsetters in Indian electronics industry, the authors are able to provide a highly credible and comprehensive account of the market that goes well beyond the surface data and helps identify actionable agenda for the reader.

So, here’s an opportunity for folks to enter the Indian power electronics segment, which offers a host of opportunities.

Some excerpts from the report are reproduced here.

Energy segment in India
Power backup (including batteries) is a $1.1 billion revenue industry in India. This industry has several facets uniquely shaped by Indian market realities of customer need and supply-demand dynamics. Two types of power backup devices commonly deployed in India are (a) the UPS; and (b) the inverter.

In the Indian context, the UPS systems refers to a power backup source for limited times, up to one hour or less, and a true-sine wave output is generated.

These are generally on-line systems where power is processed continuously and no „transition‟ to backup mode is seen by the load. Desktop computers, data centers, hospitals, telecom exchanges, and mission critical applications use UPS systems. Battery ratings are consequently lower and form less than 30 percent of the system cost. One of the reasons for shorter back-up time is that there is an alternative power generation source (such as diesel generator), that comes on-line soon after utility power fails.

The inverter system refers to a power backup source for extended time of power outages, typically from one hour to eight hours or more, and a pure sine-wave output (older products produce quasi-sine wave output) is produced by the system. Battery ratings (and the battery chargers also) are typically of high capacity, consequently the battery costs can form 60 percent or more of the system cost. These are popularly deployed in homes and offices to drive ceiling fans, TV sets, fluorescent and incandescent lights, or even elevators in housing complexes during power outages.

One salient characteristic of these applications is that they are essentially power electronics end equipment, and hence, power electronics is the driver and forms a large portion of the cost of the system. We estimate the power electronics content in the UPS and inverter markets combined at $494 million per year.

Demand and growth rates for power backup systems
The figures provide unit volume and PE content for power backup segment in 2009.Source: Innovatech Switching Power India Pvt Ltd.

UPS systems
We estimate the UPS market in India has a power electronics content of $290 million. The industry has seen declining revenues the past two years of 8-9 percent, but prior to 2008 the industry grew at a CAGR of 18 percent+ for five-six years.

The sub-5 kVA systems make up 35 percent (by unit volume) of the UPS and Inverters sold in India‟s market. Within this, more than 70 percent by unit volumes shipped are rated 600 VA and below, which cater to the desk-top computer and linked to the growth of this market. Recent growth in the penetration of laptops and net books and the erosion in desktop PC sales are perceived as the reason for declining sales in 2009 in the 600 VA and below segment. The 10-100 kVA systems earn the highest revenues in the online and line-interactive UPS systems market.

The outlook for 2010 remains depressed to flat, and beyond that recovery to a CAGR of 18-20 percent is our forecast given the power shortage in the country. Several projects for generating power, largely coal-fired plants, are in the pipeline and a recent initiative to tap into solar power has begun, however, this power hungry nation will still see power outages due to the supply-demand gap in power generation for the near future.

Inverters
We estimate 1.8 Mu inverters are sold annually in India, with more than 60 percent rated below 1.5 kVA. We estimate the power electronics content at $204 million. This market is expected to remain steady in 2010 and accelerate to a growth of 10-15 percent as the power shortage in the country is expected to ease only after 2012. This industry is currently much more fragmented than the UPS industry.

The unbranded inverter industry in India is estimated to have a power electronics content of $60 million, and more than 80 percent is sourced from China/Taiwan presently. Cost pressures are at the root of this trend and manufacturers tend to become traders in this high volume business segment. Unless significant policy changes happen to encourage access to low cost components for the local manufacturing industry, the trading opportunity will remain attractive.

Interested folks can buy the report from Innovatech for $1,495; or Rs 65,000 plus taxes.

India will surely benefit from more such industry reports as these would definitely help grow the Indian electronics ecosystem.

Thursday, June 17, 2010

Indian semicon market grows 15.6 percent in 2009, but don’t rejoice yet!

The India Semiconductor Association (ISA) today released the ISA-Frost & Sullivan India Semiconductor Market 2009-2011 update. It was released in New Delhi by Dr. M.M. Pallam Raju, Minister of State for Defense, Government of India.

Executive summary:
• India semiconductor market grew 15.6 percent in 2009 in contrast to the global market that shrunk by 11 percent from 2008.
• Wireless handsets, 3G networks, WiMax, notebooks, set-top-boxes and smart cards to primarily drive the semiconductor market in India.
• Telecom infrastructure development related domestic semiconductor consumption to grow by a massive 132.5 percent from 2009 to 2011.

Key findings:
• The Total Semiconductor Market (TM) revenues poised to grow from $5.39 billion in 2009 to $8.04 billion in 2011. The market is estimated to grow at a CAGR of 22.1 percent.
• The corresponding period is expected to witness a phenomenal CAGR of 34.8 percent in the Total Semiconductor Available Market (TAM). TAM revenue is anticipated to climb to $4.84 billion in 2011 from $2.66 in 2009.
• The rapid growth of manufacturing index of electronic products in the country is seen in the following segments- wireless handsets, telecommunications and IT & Office Automation (OA).

Poornima Shenoy, president, ISA said: “The semiconductor industry in India is growing at a CAGR of nearly 22 percent. This points to the rapidly growing domestic market which necessitates the need to have electronics as a national agenda. It can contribute substantially to the GDP in the years to come.”

While this growth looks phenomenal when compared to what has happened in the global semiconductor market during 2009, it is still too early to rejoice! It is well known that the size of the Indian semiconductor industry is not too large at the moment. So, any growth will be positive!

As per the report, the TM revenues for semiconductors in India during 2009 were $5.39 billion. End user segments of wireless handsets, communications and IT segments have played a key role in the expansion of overall semiconductor revenues in 2009. Mobile handsets, communication infrastructure and drive to extend IT are likely to be the growth drivers till 2011. Registering a CAGR of 22.2 percent, TM revenues for the Indian semiconductor industry are likely to touch $8.04 billion during this forecast period.

Electronics hardware manufacturing holds the key
The ISA-F&S report focuses on one key area — electronics hardware manufacturing – which holds the key to the Indian semiconductor industry’s growth. Just a couple of days ago, nearly the entire Indian semiconductor industry was hotly debating on how to kick-start an electronics ecosystem! Undoubtedly, the time for action is now!

That brings me to ESDM!

There has been a lot of talk about Electronics Systems Design and Manufacturing (ESDM) over the past six months or so. Instead of “easier to say, but difficult to manage”, the ESDM should reflect the message — “Electrifying start to dreams of many (in India)” with suitable action! The real test for ESDM starts now! A clear roadmap has to be in place to achieve the growth ahead.

Friends, I will try and review the entire report, time permitting!

Monday, June 14, 2010

What needs to be done to build an Indian electronics ecosystem!

A raging debate has been going on for long within the Indian semiconductor industry — what really needs to be done here to build a robust electronics ecosystem. Well, a highly engaging panel discussion, titled: “The electronics ecosystem: Plans and strategies for innovation and growth, organized by the India Semiconductor Association (ISA) was held today.Moderated by Rahul Arya, director, marketing and technology sales, Cadence Design Systems (I) Pvt Ltd, the panelists included:

* Lou Hutter, senior VP and GM of Analog Foundry Business unit, Dongbu HiTek, Korea;
* N. Ramakrishnan, president and CEO, Mandate Chips & Circuits Pvt Ltd;
* D. Rajakumar, GM, Third Millennium Test Solutions (3MTS);
* Ganapathy Subramaniam, president and CEO, Cosmic Circuits Pvt Ltd; and
* Arnob Roy, president-engineering, Tejas Networks.

Indian electronics product industry provides biggest opportunity for semicon
According to Tejas Networks’ Arnob Roy, product companies addressing a large local market have the potential to trigger an ecosystem. He added that the Indian electronics products industry is the biggest opportunity for Indian semiconductor industry.

Presenting an overview of the Indian electronics product market, which stands at $45 billion today, inclusive of $18-20 billion in telecom, $20 billion in consumer, industrial, automotive, etc., actually presented a great opportunity for the semiconductor industry. The Indian electronics product market represents close to 10 percent of the global electronics products market, which has been growing at approximately 22 percent year-on-year compared to 3-4 percent globally.

The Indian electronics product market is likely to grow to $120 billion by 2015. Roy added that domestic products for this market can be the biggest trigger for the semiconductor ecosystem. India consumes semiconductors worth $8 billion today, which would grow to $30 billion by 2015.

India has all the key ingredients — such as a large, growing domestic industry, a large talent pool of technical and managerial resources, and tremendous leverage of “cost of innovation”. However, the Indian market has its own innovation needs. Domestic systems companies need to drive the innovation requirements for the semiconductor market.

The Indian electronics industry can contribute significantly to the GDP as well and reduce the trade deficit. Its contribution today is negligible, as compared to the USA – 40 percent, Israel – 22 percent, and Korea and China – 15 percent, respectively.

Roy said that there is a need to nurture the Indian industry to meet domestic and strategic needs. There is also a need to create a ‘market pull’ for “made in India” products. In this regard, it would be wise to adopt a similar approach as China, while it was creating companies such as Huawei, ZTE, Haeir, etc. There is also the need to make available risk capital, and develop skills in product conceptualization, product management, marketing, branding, etc.

Dongbu HiTek’s Lou Hutter stated, “We believe in the Indian industry and we would like to be part of and help drive growth.” He added that Dongbu HiTek has been visiting India with the goal to establish something here! This is the first clear sign of a global foundry having shown keen interest to be a major player in India, which is excellent news!!

He pointed out that the global analog/power market has been growing at a CAGR of 2x times the semiconductor market While India has advantages such as a world-class education system, vibrant emerging market and proven engineering strengths, it also needs to address major challenges such as trying to build up market presence, lack of or presence of few ‘Indian’ global brands, developing marketing and business knowledge, and have access to manufacturing.

On the subject of how Dongbu HiTek can help the progression of the Indian semiconductor industry, he said that the company works with IDMs, engineering services companies, fabless and systems companies, and it could do the same in India. “Collaboration is really the essence in an ecosystem,” he added.

D. Rajakumar from Third Millennium Test Solutions noted that while the test market has been moving up, the Indian market has been coming up as well, although growth has been slow. Touching upon the ecosystem, he said that there was a need for design and test groups within companies to work side by side. In this regard, 3MTS can help develop or evolve a more complete collaborative ecosystem.

India needs more OEMs/ODMs
Cosmic Circuits’ Ganapathy Subramaniam pointed out that India lacked the presence of companies having revenues in excess of $10 million. “We need to have several such companies. We also need more OEMs/ODMs.”

He projected the electronics industry to touch $36 billion by 2015, but added that there were few or no Indian companies that could take the benefit of this market as yet! Even the lack of IP companies or the lack of fab/foundry were not India’s current pain points.

He highlighted certain gaps in the Indian ecosystem. These are:
* OEMs and ODMs developing products for India and working with fabless/IDM companies in India.
* Serious efforts needed at the level of the ISA to put a fabless policy in place.
* Need for a centralized lab for reliability, FIB and for FMEA purposes.

He advised the Indian industry to take advantage of the Karnataka state government’s semiconductor policy. He concluded that there was a need to view design as a value add, contrary to the current situation.

N. Ramakrishnan of Mandate Chips & Circuits Pvt Ltd stressed on the need to collaborate for semiconductor product development. “There is a need to understand what customers want, and then give them what they want,” he added. “You also need to look at the complete product lifecycle. Unless you have the customer in mind, people do not pay attention to the aspect of the product lifecycle.” He highlighted the need to maintain the sanctity of the value chain.

Sunday, June 13, 2010

Evolution of various semicon analysts growth forecasts for 2010

Earlier this month, Mike Cowan, an independent semiconductor analyst and developer of the Cowan LRA model, had provided me with the latest update on forecast results as gleamed from running the Cowan LRA forecasting model incorporating the "actual" April sales.

As a follow-on to his April 2010 global semiconductor forecast numbers, Mike Cowan constructed (and updated) a table (sourced from the GSA website in order to compare the latest 2010 sales growth forecasts from a large number of leading market researchers to his latest sales growth forecast estimate of 33.4 percent.

Notice that for the thirteen (13) yellow-highlighted market researchers shown in the attached table (including mine), 12 of the market watchers have increased their most recent forecast year-over-year sales growths to a range of 22.6 percent to 33.4 percent with a mean sales growth forecast of 28.7 percent (28.4 percent without Cowan's forecast number).

As revealed in the table, Cowan's most recent 2010 sales growth forecast estimate is the most bullish of the bunch (at least for this month; stay tuned for my monthly forecast numbers as the year plays out!).

Also note that the just published (last week - June 8 and 10, respectively), WSTS and SIA Spring 2010 forecast sales growth results for 2010 are included in the table.Source: Cowan's LRA model.

Wednesday, June 9, 2010

Indian industry proposes to extend deadline of India's semicon policy up to March 2015!

If you recall some time ago, I'd mentioned that the Indian semiconductor policy, which was announced back in 2007, had supposedly expired on March 31, 2010!

Now, the Indian industry has come up with recommendations, which include extending the Indian semicon policy up to March 2015!

For those who have come in late, 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!

I am very grateful to the India Semiconductor Association (ISA) for sharing the details of the summary of inputs and amendments to India's semiconductor policy.

As I mentioned, the recommendations include extending the Indian semicon policy up to March 2015!

This is fair enough, although I am sure it can extended for an even longer duration. Some other key recommendations include lowering of threshold limit for ATMPs and other ecosystem units, development of ecosystem, faster project appraisal time, etc. -- and all of these are significantly necessary.

Should India have a fab? Why not?
There may be debates over whether India should have a fab or not! However, it is pleasing to see that there is still hope. Actually, why not have a fab or a foundry? Use it to serve the global market! The Indian semiconductor industry needs a serious rethink in terms of strategy. Maybe, it cannot survive on chip design services alone!

About 15 months ago (see Feb. 2009 archive on this blog), I had written "Can the Indian semicon industry dream big? (And even buy Qimonda?)! Well, time, I repeated that story! Here's why!

If you have noted, early June 2010, ATREG, a division of Colliers International, has been appointed as advisor to market the sale of the advanced 300mm manufacturing campus of Qimonda in Dresden, Germany. Now, here's a great chance for India or some Indian investor to grab this fab!

The highly accessible campus, located in the State of Saxony, features a state-of-the-art 300mm semiconductor fab including 281 advanced front-end semiconductor manufacturing tools, an advanced 300mm R&D fab and 360,000sq. ft. of administrative space. The campus also includes excess land, which could be used to construct a mirror-image of the existing 300mm fab, potentially doubling production capacity.

ATREG will focus on finding an operational purchaser for the facility. As originally constructed, the facility was capable of producing DRAM chips with a maximum volume of approximately 10,000 wafers per week.

The fully automated, world-class Qimonda Dresden 300mm wafer fab was built in 2001 and was the world's first 300mm production facility. The state-of-the-art, 300mm R&D facility was built in 2005. The world-class 300mm equipment includes hundreds of advanced front-end manufacturing tools, many of which were used in volume production at 65nm. Though production has ceased, the cleanrooms remain in a production-ready state so that a potential purchaser could restart operations quickly.

There, I've said it!

There's an opportunity waiting out there to be grabbed!! If India or someone in India does not move fast and attempt to buy, someone else, from somewhere else in world will surely buy this fab!

Let's go back to February 2010, when the Karnataka State Government had announced its semicon policy. Perhaps, here's an opportunity for Karnataka to take a lead!

Industry recommendations for semicon policy
Now, back to the recommendations from the ISA and the Indian semicon industry, which has been reproduced below:

Government of India Semiconductor Policy 2007: Summary of Inputs and amendments: proposed by committee members and industry.

1. Extension of the Policy:
The World Bank has confirmed that global economies are currently facing a recession. The global economic slowdown has severely impacted the semiconductor industry leading to piling up of inventories and reduced capital expenditure. While it is difficult to project the year of revival of the industry, the experts opine that the present situation for the semiconductor industry would last for at least two plus years. It is, therefore, proposed that the deadline of March 2010 may be extended till March 2015.

The extension will provide time:
* To market the Policy and improve the prospects of India to attract investments.
* To highlight the importance of the domestic market to the potential investors. It may be mentioned that the domestic semiconductor market (TM, i.e., the Total Market), as per the ISA-Frost & Sullivan India semiconductor market update report ( 2008-2010) is projected at $ 7.6 billion in the year 2010.

2. Lowering of threshold limit for ATMPs and other eco-system units:
It is suggested that the threshold limit for certain categories of eco-system units like ATMPs, optical LEDs, storage devices, LCD, FPD, Photovoltaics, fuel cells, micro and nano technology products (as defined in the SIPS) needs to be re-visited, as these units may not require large investment of the order of Rs. 1,000 crores.

Lower threshold limits (to qualify for the incentives available under the policy) is expected to generate interest for such categories of eco-system units, which has not been seen so far. This lowered threshold limit needs to be defined in consultation with industry experts.

The above areas, though low in technology vis-a-vis the technology required for wafer fab, form an important part in the value chain of chip manufacturing. Location of several of the manufacturing infrastructure in the above ecosystem areas in the country can also serve as a pull factor to set up wafer fabs.

It may be mentioned that countries like Hong Kong, Singapore, Malaysia, etc., took this approach initially and went on to have wafer fab facilities.

3. Financial closure commitment:
The present Policy requires the investors to produce legally binding commitment of equity holders and debt financiers to provide or mobilize funding for achieving financial closure (at least 90 percent of the project cost).

Projects envisaging large scale investments are implemented in phases. As the major markets for the products from the fab units and the ecosystem units are presently targeted at exports, the investors find it difficult to produce legally binding commitment of achieving financial closure of this large magnitude from the Financial Institutions (FIs) at the time of submission of the proposal.

It is, therefore, suggested that the achievement of financial closure (as proportion of the project cost) may be lowered and the investors may be asked to submit bank guarantee for the differential amount (amount required to achieve financial closure less own equity contribution plus the commitment received from the FIs), as a collateral.

4. Extension of Policy clauses:
The policy has been formed on the need to have a (semiconductor) wafer fab unit in India. Besides acting as a strong catalyst to promote the growth of domestic electronic equipment manufacturing market and product development, the wafer fab would also meet the requirements of the strategic applications.

In case the deadline for the policy is extended till March 2015 and the proposal for a chip fab unit is not received, say, by March 2013, it is proposed that higher incentive structure could be proposed to make it more lucrative. As the ceiling on the number of such units as per the present Policy is kept at 2-3, we could limit this number to 1-2, in case the higher incentive structure is proposed to seek investments for the wafer fab.

The following points may be considered while making amendments to the policy w.r.t this main point:-

i) The proportion of incentive structure (higher than the present structure) should be arrived at on the basis that the earlier budgeted incentive to be extended to 2-3 fab units is now available for 1-2 units. This will not entail any excess outgo of incentive from the Government of India against the amount of incentive budgeted earlier.

ii) The window of time period to provide for this higher incentive structure (April 2013 – March 2015) is only a suggestion and the same can be decided amongst the various stakeholders.

5. Development of the ecosystem:
It is in the overall interest of the country to have an ecosystem covering all kinds of units. A recommendation is that in the event that proposals from different categories of ecosystem units are not received, multiple proposals from the same category of ecosystem unit may also be considered under the policy, without putting any limit on number of such units to be covered (within the ceiling on the total number of such ecosystem units at 10 as per the present Policy) e.g., Government of India has received several proposals to set up solar PV manufacturing facilities.

6. Appraisal process:
It is also suggested that the Government of India specifies the time period and dates to complete appraisal of the proposals received to date and makes an early announcement of the approvals of the proposals which have been submitted to it. This step would send positive signals to the other investors.

7. Clarity on certain points in the Semiconductor Policy:
i) A number of investors are not fully conversant with the steps they need to take to qualify for the Government of India subsidies. It is felt that the investors will be reluctant to make investments in their projects till they receive approval of their proposals and the Government incentives.

ii) The Policy is not clear about inclusion of costs on utilities, interest charges, etc., as part of project cost. It is important that total project cost of running the plant including the items mentioned above is considered for the award of the incentives.

iii) The present Policy provides for the release of the incentive being linked to the achievement of the threshold investment. It is proposed that the payment of the incentive to the investors may be advanced, though the threshold investment may not have been reached as yet. The suggested measure would provide the investor with funds at an early stage, which is crucial for timely implementation of the project. In this case, as also proposed at Sl. No. 3 above, the investors may be asked to submit bank guarantee for the incentive released as collateral.

Monday, June 7, 2010

Context-aware traffic mediation can help telcos manage data tsunami: Openwave

Given the rising (and no end in sight) surge in demand for mobile content and data services, mobile network service providers are facing the challenge of effectively managing exponential growth in data traffic. Service providers must also find ways to maximize bandwidth without sacrificing end user experience.

In conjunction with the Mobile Marketing Association Forum (MMA Forum) APAC event held on April 13-15, I interacted with Anand Chandrasekaran, director of Product Management, Openwave Systems Inc., which also did a global launch of it product -- the Analytics Express at the event.

Managing data traffic challenges
Despite claims of vendors to have solved growing data traffic challenges, those still remain. How can Openwave really help manage this?

According to Chandrasekaran, a fundamental shift has occurred in the industry. He said: "The demand for mobile data that we planned for years ago is finally here - only it’s bigger than everyone predicted. The proliferation of new devices like the iPhone and HTC Incredible, along with vastly improved user experiences and unlimited data plans (to date), has caused a tremendous and unprecedented surge in mobile data demand – AT&T disclosed this year that 3 percent of its users consume 40 percent of its bandwidth resources. This increase in traffic and the competitive pressure to keep data plans flat are squeezing service providers’ margins."

Now let us look at how service providers can tackle the bandwith issue. As per Chandrasekharan, until now, one approach has been to add network capacity through additional equipment CAPEX. Unfortunately, this strategy is expensive and provides only a short-term solution.

Not all service providers have the financial strength to simply throw money at the problem, nor does that guarantee a sustainable solution. Service providers need to take a more holistic approach in developing solutions that will maximize available bandwidth while being able to monetize this surge of mobile data traffic.

An effective way for mobile service providers to handle the approaching data tsunami is to deploy context-aware traffic mediation software that sits in the data path, empowering them with a full view of their network, their subscribers’ profiles and the mobile devices in use. Context-aware traffic mediation enables service providers to monitor, manage and monetize traffic by creating and delivering smart policy-driven services.

According to him, Openwave’s Traffic Mediation solution runs on an open, IP-access platform that acts as a single control point for traffic management and provides services such as content adaptation, web and media optimization, network security, smart policy control and dynamic charging and campaigning.

Is user experience getting compromised?
However, it is claimed that user experience is still getting sacrificed and also sidetracked, as not all networks are as robust or have good coverage. What can be done here?

Chandrasekaran said that among the various forms of mobile data, video is quickly becoming the dominant data type in carrier networks. Given the popularity of video sites such as YouTube, Hulu, and Metacafe on wireline broadband networks, it should come as no surprise that consumers tend to exhibit the same behaviour on wireless networks.

"As video traffic quickly consumes more of the available bandwidth, mobile service providers are experiencing tremendous strain on their networks. Implementing media optimization technologies and best practices can help service providers cope with the exponential growth of mobile video data without resorting to expensive CAPEX/OPEX solutions, while also providing an enhanced user experience and increased network efficiency.

"To effectively manage this explosion of mobile data traffic, mobile service providers need a solution that can reduce data traffic volume by using different optimization techniques such as compression to deliver optimal visual quality. They
can improve the user experience through intelligent caching and delivery mechanisms.

"Monitoring consumption intelligently will help facilitate introduction of appropriate policies for services and tariffs with differing rates. Improving network resource usage between different users and traffic types can ensure an overall optimal response to the user," he added.

Puzzle of dynamic profiling data cracked?
Everyone claims to have solved the dilemma of 'storing and serving up of dynamic profiling data,' which is far from true. It would be interesting to find out what Openwave has to say.

Chandrasekaran added that mobile service providers must be creative in how they define data service plans that appeal to the subscriber, while covering their rising network costs. They should closely monitor what is happening on their network and they must have “total contextual awareness.”

"Deploying solutions that allow for new policies in real time, based on time of day traffic patterns, special events and user profiles, can be a way forward. Openwave Smart Policy combines an IP mediation framework residing directly in the data path; a context-aware policy engine that aggregates and analyzes session data relating to each user’s activity; operational analytics to influence policy decisions in real-time; inline charging and notification services to bill the user for service when appropriate.

"It benefits the operators as the IP Framework sits in the data path monitoring all traffic while enforcing appropriate policy rules in real time. The Policy Manager aggregates IP data into session data relating to each subscriber’s activity, over potentially multiple access networks. The Integrated Subscriber Profile Repository (SPR) chooses the appropriate policy rules to apply based on total context, whereas, the Operational Analytics guides policy changes in real time.

"Inline Charging and Notification helps operators notify subscribers nearing volume limits, offer them extensions or upgrades, and bill them in real time when appropriate – for instance a North American operator recently reported revenues of $1M/month using their ability to bill users transparently for 24-hour and other short term unlimited data usage (particularly when roaming).

"On the other hand, tailored service plans can be developed to match specific subscriber needs based on their online behavioural profiles. For example, the “Movie Buff” is provided with a plan with incentives to download larger volumes of data when the network is likely to be less congested. “The Mobile Breadwinner” can go online at any time while travelling abroad without worrying about costly roaming fees. “The Talk of the Town” is provided with high-speed access to their favourite social networking sites and offered premium messaging services," he elaborated.

Aggregating data close to source
Are telcos really able to aggregate data as close to the source as possible?

The Openwave analytics offering allows for integration with various types of data close to the source, including from other vendors. “Big data” is one of the fastest growing technology areas, and was recently covered in an Economist report. However, the fact is that raw data is just that and what is required is insights that are actionable, either for the operator themselves and in a way that can be exposed to various ecosystem end points.

The other key factor is the ability to present these insights in a way that the industry can understand – for example, Openwave recently partnered with Nielsen Co. to build an “insights” engine that combines the mobile footprint within the operator network with segmentation data from Nielsen’s Prizm product to help advertisers and marketing folks better understand their users.

Expect to hear more from Openwave in this area in the coming quarters.

Openwave's predictive solutions
What are these with predictive solutions of Openwave mobilizes the Internet fueled by real-time analytics? How are they helping operators?

The Openwave Analytics Express is a cloud-based solution, which features analytics and a set of reporting tools designed to perform essential monitoring of mobile internet usage and behaviour, enabling operators to improve operational efficiency without significant infrastructure investments.

Analytics Express enables operators to gain key network and usage related insights across a wide array of operator data sources and includes reporting functionality that can help mobile operators build profiles and segments, and leverage insights into new revenue streams. As a hosted solution, it is uniquely positioned to provide these insights at a low-cost, low hardware footprint and rapid time to market. The offering is also highly scalable and can address specific customer requirements with rapid customization.

Trends in Apac/India
Going forward, let's look at some of the trends in Apac/India and how they compare globally?

Chandrasekaran said: "We recently did a workshop at the MMA APAC Forum where we co-presented with AdMob, InMobi and BuzzCity and also met with operator customers. The maturing of the voice market is a global trend and we heard similar observations by Malaysian and Singaporean operators. The same applies to the coming “data tsunami” trend as well.

"Globally, the advertising market is forecasted (Source: Chetan Sharma) to grow at 100 percent YoY to $3.1 billion in 2013, and this market can leverage some of the data assets currently locked away within operator networks. We’re observing that inclination to be a global trend as well.

"APAC is a growth market for our industry, and operators here are very willing to monetize this demand for mobile data from a growing range of smarter devices."

Openwave in India
Besides customers, Openwave has an offshore development center in New Delhi, India.

Recently there has been a great increase in use of data services in India for specific applications such as Facebook as well as the Indian “ABC” of Astrology, Bollywood and Cricket. The recent 3G spectrum auction also provides new opportunity for services and revenues. However, the operators there are in one of the most competitive environments in the world with voice rates at less than a cent a minute and 12-15 mobile operators grappling for market share.

Openwave is said to be approaching these operators and offering opportunity for clear differentiation based not just on price, but on services, quality and user experience.

"We have solutions targeted at 2.5 and 3G data usage which enable a real internet experience on mobile whilst enabling operators to conserve bandwidth, get a full 360 degree understanding of who is using the network and customise services for each tier or niche group of users. Some of them have already been mentioned earlier.

"We have two channel partners in India, HCL and Tech Mahindra. We will be taking these solutions proactively on the road initially with Tech Mahindra later this month (June) focussing on approx six operators in the Delhi/Mumbai areas," he concluded.

Friday, June 4, 2010

Ten commandments of effective standards!

Do you know how to distinguish between a ‘good’ standard and a ‘bad’ standard? Or, how would you go about trying to develop a standard in the first place? What makes a standard ‘effective’?

Are you wondering why am I asking such questions? Here’s why: Today, Synopsys sent out a release stating that under the imprint of Synopsys Press, it has published The Ten Commandments for Effective Standards.

I was absolutely thrilled when I discovered the author’s name — Karen Bartleson, Senior Director, Community Marketing, Synopsys and IEEE-SA Corporate Advisory Group member, who I met just about three months ago, during a seminar in Bangalore on IEEE Standards for Design Automation: Their Impact on an industry.

Back then, I enjoyed discussing with her how Synopsys was making use of the social media to connect with its customers. You can find the blog post in the March 2010 archive.

The Ten Commandments for Effective Standards is available for a retail price of $29.95 hardcover, $19.95 softcover and $14.95 eBook through bookstores and online, including through Happy About and Amazon.com. So, I hopped on to Amazon.com to have a look. Here are the chapters! Quite interesting!

1. Why standards?
2. Why effective standards?
3. The 1st commandment: Co-operate on standards, compete on products.
4. The 2nd commandment: Use caution when mixing patents and standards.
5. The 3rd commandment: Know when to stop.
6. The 4th commandment: Be truly open.
7. The 5th commandment: Realize there is no neutral party.
8. The 6th commandment: Leverage existing organizations and proven processes.
9. The 7th commandment: Think relevance.
10. The 8th commandment: There is more than one way to create a standard.
11. The 9th commandment: Start with contributions, not from scratch.
12. The 10th commandment: Know that standards have technical and business aspects.
13. Go Forth and Standardize.

Now, I’m not so lucky to lay my hands on Karen’s latest book. However, I’ve requested her to send me a copy, if possible.

But guess what, I managed to speak with Karen tonight, after all! I started by asking what compelled her to write this book?

She said: “As Synopsys created its publishing imprint, Synopsys Press, I volunteered to write the first book in the Business Series. Because standards are one of the areas where I have the most experience, and because I’d written short blog posts about good practices for standardization, it seemed like a natural topic for a book. I’d also searched for a similar book and finding none, decided that there could be a demand for it.”

I also quizzed her about ‘good and ‘bad’ standards. She added, “While I list several characteristics of both “good” and “bad” standards, the overall distinguishing factor of a “good” standard is that it’s widely adopted.”

And, how can the overall standardization process be improved? Her reply, “Leveraging proven processes and organizations, understanding business concerns, using alternate ways to create standards when appropriate, and learning to cooperate on interfaces while competing on products.”

Since it’s too late here in India, I only had three more questions on standards. First — the use of when mixing patents and standards. She said: “Patents that must necessarily be infringed upon in order to implement a standard pose a real challenge to standardization. These patents, called “essential patents”, can stymie a standards effort. Imagine if you implemented a standard, then were sued by an essential patent holder for doing so. There are several ways to deal with essential patents, some of which I describe in the book.”

There’s a chapter ‘Realize there is no neutral party’ — what’s that supposed to mean? Karen said, “It’s important to recognize that everyone who participates in a standards project has a reason for doing so. It puts people’s behaviors in perspective.”

Finally, ‘There is more than one way to create a standard’! What’s the secret? Karen concluded: “The traditional formal standards committee is what most people think about when they think about how standards are created. However, there are other ways to create them. One example is the open source model for creating standards that Synopsys pioneered in the EDA industry.”

Many congratulations and best wishes to Karen! Thanks a lot for taking the time to speak with me tonight!

Thursday, June 3, 2010

Global semicon sales forecast estimates based on 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 semicon sales estimates associated with the forecasting model — the Cowan LRA model for predicting worldwide semicon sales.

The World Semiconductor Trade Statistics (WSTS) yesterday (6-02-10) posted the April 2010 actual sales numbers on its website. Consequently, this is the latest update to Mike Cowan’s forecast results as gleamed from running the Cowan LRA forecasting model incorporating the “actual” April sales.

The actual April global semiconductor sales as released by the WSTS came in at $23.385 billion, which is up 43 percent from last year’s April sales of $16.354 billion and down 11.9 percent from last month’s sales of $26.553 billion, (which was revised upward slightly from last month’s published sales of $26.533 billion).

Therefore, the latest updated 2010 sales number predicted by the Cowan LRA model is $301.865 billion corresponding to a year-over-year sales growth of 33.4 percent for 2010.

These latest sales and sales growth are up from last month’s reported forecast estimates of $294.981 and 30.3 percent, respectively. These upward results reflect the relatively strong April sales as reported by the WSTS.

The full complement of the latest sales and sales growths forecast estimates for 2Q, 3Q, 4Q and 2010 are summarized in the table below along with 1Q’s actual numbers.Additionally, the May 2010 global semiconductor sales forecast estimate is projected to be $22.743 billion, which would yield a May 3MMA sales forecast estimate of $24.227 billion, which is normally published by the Semiconductor Industry Association (SIA) each month in order to characterize the semiconductor industry’s growth posture.

Wednesday, June 2, 2010

May 2010 global semicon update: Four quarters of sequential growth, yet still no one believes! Wake up, says Future Horizons

Here are the excerpts from the Global Semiconductor Monthly Report, May 2010, provided by Malcolm Penn, chairman, founder and CEO of Future Horizons. There are a lot of charts associated with this report. The report also covers market trends. Those interested to know more may contact Future Horizons.

March’s total semiconductor sales came in at $26,533 billion, slightly above our February expectation, closing the quarter at $69,181 billion. This was up 2.8 percent over Q4-2009 and one of the strongest first quarter performances ever in what is normally a negative growth quarter. We have now had four straight quarters of industry growth, yet still no one believes in the strength of the recovery.

Of course, something unexpected can always go wrong but the industry fundamentals have never been better aligned. Just as 2001 ushered in the conditions for the so-called the perfect (semiconductor) storm, 2010 is now wallowing in the inverse effect. Surprisingly, few firms are tough. Most are too timid, too cautious or too scared. Welcome to the brave new world of semiconductor company ambivalence and life-threatening risk aversion. “Hello”.

Future Horizons presented its review and forecast for the global semiconductor market on the first day of their ongoing 19th International Electronics Forum (IEF) 2010 in Dresden, Germany, May 6-8. Our overall prediction was that the 2010 chip market would have a barnstorming year; only a disaster of the Lehmann Brothers scale could now derail the market.

The overall five-year forecast presented was:
* 2010: +31 percent, with still some scope for upwards revision.
* 2011: +28 percent; based on: peak of the structural cyclical boom (could stretch into 2012).
* 2012: +18 percent; based on: normal cyclical trash cycle starting 2H-2012 (1H-2013?).
* 2013: +3 percent based on: market correction in full flow (could be negative, cap ex overspend and inventory build depending).
* 2014: +12 percent; based on: start of the next cyclical recovery (single digit, if 2013 is negative).

This would take the industry to around $300 billion in 2010 with a CAGR of 11.8 percent between 2010-14. It would also signal a 180-degree reversal in the industry’s fortunes following its ‘zero growth’ 2000-09 lost decade of growth. Moreover, despite the apparent bullishness of these numbers, given the now unavoidable 2010-11 fab shortage, the growth upside for 2010-12 is still huge.

The real tragedy however of what ought to have been good news for the industry was: (a) still, no one believes in the numbers; and (b) it was entirely predictable.

We first presented this forecast in January 2009, at the high point of the industry’s economic and business uncertainty. The only change we have made in the last 17 months was to increased 2010’s growth number from 15 to 31 percent number. Whilst all other industry analysts, business leaders, trade associations and economists alike wrestled with what was happening, we alone never lost faith in the industry or what the underlying fundamentals were saying.

This cycle’s forecast was the easiest we have ever had to make. All we had to do for the IEF meeting was to adjust for the fact that the 2009 recovery was faster and steeper than even we had dared to predict. The bottom line? The industry fundamentals may often get distorted by events but they never lie, ignore them at your peril.

We were ridiculed for our optimism in January 2009 and throughout the year when we stuck to our guns. We never stopped believing in the numbers however and never subscribe to industry fashion, trend or sentiment, despite this sometime being out on a limb with industry consensus. We are proud of the fact only we got this analysis right, but equally sad that no one had the courage to listen. This was not forecast luck either; this was simply doing what we do best, making a considered analysis and then believing in what the forecast tells us.

Just to recap then on these industry fundamentals, before the 2008-09 collapse the chip market was in remarkably good shape with negligible inventory, increasing ASPs, low levels of Cap Ex and tight fab capacity.

With the global economy – albeit fragile and vulnerable – now recovering, the industry has started its recovery with shortages and all of the other aspects still positively aligned. It simply does not get much better than this, but despite what the numbers say, “Ah but” is (still) driving the industry consensus … still no one looks too far beyond the next quarter!

Granted economic uncertainty remains the biggest short-term risk, not helped by the fact that the global financial system remains fundamentally flawed. One cannot however run a business based on what might go wrong with the economy. You need a plan for growth with ‘Plan B’ if the worst-case scenario then happens. Top of the industry’s paranoia and fear, uncertainty and doubt (FUD) factors, largely due to a pandemic the loss of collective confidence following ‘five bad years’ of industry growth and pandering to Wall Street’s greed, fickleness and quarterly reporting bullying, are:

* Is unit demand overheating or sustainable?
* Is inventory starting to get out of control?
* Will the economy slip back into recession?
* Is the capacity crunch a blip or more fundamental?
* What will happen to the economy when the stimulus funding dtops?
* Will the end market demand hold up or slip back?

Yet, such risk aversion is not risk management. As recessions draw to a close, few are ready to believe that it is over, exaggerating the catch-up reactions, amplifying the cycle peaks and troughs. By treading water in fear of a 2H-2010 dip, the industry will simply exacerbate the next trash cycle. By the time the industry waits for 2010 clarity (September?), it will be too late to rescue 2011.

As we have constantly mentored, the industry fundamentals - economy, unit demand, fab capacity and ASPs - don’t lie; believe in them or die! The current capacity famine was instigated over two years ago, well before the crash, meaning today’s shortages were inevitable and accurately predictable, given the investment lead times. The next trash dynamic has not yet been triggered and, if overinvestment-driven, is unlikely to happen before 2011 at the earliest, meaning, 2012 impact, as reflected in our forecast. Each fundamental however always carries a health warning.

Economy: Economic disruption will always derail the chip market. The only questions are: “by how much and for how long?”

IC unit shipments: Managing this dynamic always takes judgement. The monthly run rates vary dramatically from the trend line making it impossible to balance medium term supply with demand (demand changes in days, supply changes takes months).

It is this mismatch that makes it ‘feel’ capacity expansion is out of control. That said, you always need to keep a paranoid eye on inventory and double ordering, especially in tight supply and demand cycles. Unfortunately reliable visibility into the demand chain is the weakest link in the whole chip market equation.

Fab capacity: Long-term wafer supply security is an industry prerequisite, yet increasingly trivialised and often overlooked. It is the fundamental fabless (Fablite) achilles heel, and the reason that FSA (now GSA) was formed in 1994. New capacity takes a year to come on stream meaning the next four quarter’s capacity is already cast in stone. Even if we splurged cap ex today, there will be no new sales impact until May 2011.

ASPs: ASPs are diagnostic and complex; systemic (Moore’s Law); structural (capacity/wafer Size); and sabotage (price wars). After a long period where all of these worked against industry pricing trends, we can now look forward to an extended period of structural ASP recovery potential.

As we mentioned in our previous reports, our 22 percent January 2010 forecast for 2010 was based on the relatively benign quarterly growth pattern of -1.0, +1.0, +6.2, +2.0 percent; in essence a very weak year. Q1 came in at +2.8 percent and no one we speak with is seeing lower than 3 percent positive growth for Q2. That alone would bring the year on year growth up to 28 percent.

It is now impossible for 2010 to be single digit growth; even low double-digit growth is incomprehensible to comprehend!

Why then are the trade associations and the industry leaders still talking down the industry? This is not caution, but complete irresponsibility. Barring an epic 9/11, Act Of God or immoral banker style disaster, growth of anything less than around 30 percent in 2010 is now all but impossible!

Tuesday, June 1, 2010

Indian Microelectronics Academy (IMA) formed to build, nurture and grow start-ups!

Friends, it takes great pleasure to announce the formation of the Indian Microelectronics Academy (IMA), a brainchild of Cre8 Ventures, which was launched by Mentor Graphics UK Ltd in 2005 as an independent network. Cre8Ventures aims to help start-up electronic design companies achieve business success.

The idea of the IMA was conceived and developed by Carson Bradbury, the mastermind behind the European Microelectronics Academy (EMA), who has helped this India chapter get off the ground. You can see him raising the salient points of the IMA to a select audience from the Indian microelectronics industry.Carson also happens to be the director of Cre8ventures, the initiative Mentor Graphics took back in 2003 to help the startups in Europe. During this period, Cre8ventures has been able to help over 80 companies and create a healthy ecosystem for the startups.

On the IMA's objective, Carson says: "It is to help enable a new breed of microelectronic companies that have the promise to become global by owning value chains. The strategy would be to establish new international communities whose tantalizing messages on the world stage is exploitation (US), execution (Asia) and innovation (Europe)."

As for the areas of focus, he adds that it will be "wherever there is a big unmet business need and the opportunity to own a value chain through microelectronics."

What will the IMA do for start-ups in India? Carson says: "The IMA will uncover unmet business needs from big companies from Coca Cola to Tata and will solve the innovation gaps by creating/connecting start-ups and leveraging open innovation strategies from multinational semiconductor companies.

"The IMA will help start-ups get to momentum through the experience, trust and influence of the high profile individuals which represent the IMA’s executive team and the contribution from industrialists (e.g. EDA, Foundry, IP, Assembly and Test, Software and Design Services) who’s ‘take’ is first mover advantage and who’s ‘give’ is risk sharing business models (e.g. spin-out, spin-in, IP donation/extraction).

"The IMA will then aim to take start-ups further down stream so when they get to Series A they will have triple A board rooms, markets defined, early adopter customers an first product/silicon in place."

Now, I sincerely hope this is the story that the Indian semiconductor and electronics industry has, perhaps, been looking forward to! There is a need to build, nurture and help start-ups to develop and grow. The IMA has the goal to make that happen! In that respect, this is your -- the Indian microelectronics industry's -- story!

The IMA's initial steps would be to actually start getting connected with the Indian startups who require immediate help and attention, and come out with its first success story, and more! I sincerely hope that the IMA can interest the other leading industry bodies in India to join hands and make this a very robust ecosystem that will serve the Indian industry in the years to come.

Why need an IMA?
So, why need an IMA in the first place? Well, the IMA has been formed in India with the objective of creating an ecosystem of bringing together all of the necessary pillars required to create successful startups and bridge the innovation gaps among the leading companies.

The idea is to also grow the GDP of the region based on a knowledge based economy. This is a proven concept, which worked for seven years in the form of Cre8 Ventures in Europe. In 2010, Cre8 Ventures first set up the EMA in Europe and now, the IMA in India!

Who's involved in IMA?
Naturally, Carson Bradbury, the brain behind all of this, is heading this initiative.

The others in this team include Raghu Panicker, sales director, Mentor Graphics India, who plays the role of a non executive director in the Academy and brings the world class EDA tools required for the startups to the table.

Himanshu Rawal, account manager, Mentor Graphics India, will play the role of director, Cre8 Ventures India. His charter in the Academy is to discuss and uncover the interests of the various startups, understand the innovation gaps of the leading companies and bring these in front of the Academy.

On a personal note, many thanks to Himanshu for also providing valuable inputs for this post.

Yours truly has also been asked to join the IMA. I hope I can play a decent role in connecting the right dots together!

Now, a note on the Academy’s structure. The IMA will have a chairman, a set of non executive directors and a set of advisory board members.

It will have platinum members who should be willing to invest dollars of investment in the form of services, EDA, foundry services, IPs, etc.

The Academy will also have financial partners – the VC community, as well as knowledge based partners, such as colleges/universities offering incubation to startups; serial entrepreneurs to guide and give the success mantras to the innovators; bluechip partners – big companies who see big markets in the horizon and who need the Academy to fill in their innovation gaps; and a media partner who can transmit the right things that the Academy does at the right time.

The Academy itself will be closely working with the innovators who might come out with solutions such as spin-in, spin-outs, joint ventures etc.

So far, the Academy has identified a couple of VCs who are ready to invest if the Academy approves an innovator’s idea; a couple of successful serial entrepreneurs in the microelectronic Industry; an incubator to help startups getting access to EDA tools; big microelectronics companies who see big markets coming; a platinum partner who is willing to put its money in the right places.

The IMA has now started functioning and a couple of uncovering sessions have been done with startups, which will be put in front of the Academy members during its next meeting scheduled in July 2010. There is an uncovering session planned with a major healthcare company to put their idea in front of the Academy as a project.

With the formation of the IMA, a platform is now ready for the startups in India in microelectronics – semiconductor, VLSI, solar, electronics, telecom, etc., and get benefitted by the core members of the Academy.

Calling all interested start-ups: to start a dialogue, please feel free to contact me -- Pradeep Chakraborty -- through this blog. You can also contact Himanshu Rawal at himanshu_rawal[at]mentor.com.