Sunday, January 31, 2010

TSMC leads 2009 foundry rankings; GlobalFoundries top challenger!

Recently, IC Insights released the rankings for the world’s top pure-play and IDM foundries. No surprises, as TSMC continues to lead! The surprise entrant is of course GlobalFoundries, which ranked fourth, after having started operations in March 2009.However, all of the foundries, barring Tower Semiconductor, registered negative growth during 2009. Tower, which acquired Jazz Semiconductor in 2008, was the only foundry to post positive growth during 2009.

IC Insights further stated that if the revenues of Chartered Semiconductors, which was purchased by GlobalFoundries recently, were combined with GlobalFoundries, their combined sales would have amounted to over $2.6 billion in 2009. That’s not very far from UMC, which is ranked no. 2!

Now, I am not an expert to comment on which foundry has done really well, despite the recession, or whether GlobalFoundries can really challenge and overtake TSMC in the future. However, I followed with great interest a discussion on one of my groups on LinkedIIn on this topic.

Daniel Nenni, a critically acclaimed blogger, and an industry colleague, had recently blogged on this subject. There were some interesting comments following that post. I sought the permission of Malcolm Penn, chairman and CEO, Future Horizons, to use some of his remarks for my blog post.

Penn said that there is a reason why TSMC is the no. 1 foundry in the world. However, as competition breeds innovation, hence, the foundry business will be much more interesting to watch with GlobalFoundries challenging TSMC.

He added: “There was also no reason why Intel is #1 in PC MPUs or Microsoft #1 in PC OS except for one key factor. Once a competitor gets to be a certain size, no amount of innovation will help swing the balance of power.” The real question therefore is: “Has TSMC now passed this tipping point?” I fear it already has.”

Fact of the matter is in the foundry business capacity is king and TSMC already outguns all of the other competitors combined. Given the minimum two to three year lead time to build and ramp new capacity, TSMC has incredibly clear visibility to see any competitive threat coming.

Very interesting! Some other observations later!

Saturday, January 30, 2010

Indian semicon industry starting to make a noise and be heard!

Several years ago, I was extremely fortunate enough to get to see Cirque du Soleil perform Alegría! I remember one of lines going something like — when you can’t shout, scream! Although I don’t quite remember the context, this line really stood out!

Another anecdote! “Did I not instruct, that box 5 was to be kept empty?” boomed Gerard Butler as the angry Phantom of the Opera, in the movie featuring him and Emmy Rossum. The Phantom really made his statement. He made noise!! Deafening noise, that was hard to close your ears to!

Now, what’s the relationship between Cirque du Soleil’s Alegría and The Phantom of the Opera? Well, nothing really! However, both are fantastic, distinct examples of how to create a noise — and be heard!

Why am I saying all of this? Well, here’s a wonderful comment left by a reader, Sanjay Agarwal, president and CEO, Zelig Semiconductors, to one of my earlier blog posts titled Indian semicon industry: Time for paradigm shift!

He says: “I have worked with India-based semiconductor design houses since 2001. The quality of work has steadily improved. I am sure that India will have more of semiconductor design development including whole product development in few years.”

That’s great to hear! The Indian semiconductor industry is really starting to make a lot of noise and be heard, globally, irrespective of having or not having a fab! I sincerely hope that this momentum gained by the Indian semiconductor industry continues in the years to come!

The forthcoming India Semiconductor Association (ISA) Vision Summiit 2010 — on Feb. 1-2 in Bangalore — will provide even opportunities for the Indian semiconductor industry to really make a lot of noise, and be heard, globally.

In the meantime, here’s a photograph from Oct. 28th, 2004 — the day, the ISA was born in Bangalore.Notice the stalwarts? From left: Dr. Ananda, Dr. Madhu Atre, S. Uma Mahesh, Rajendra Khare, Dr. Sridhar Mitta, Dr. Anand Anandkumar, V. Veerappan, S. Janakiraman and Dr. Satya Gupta.

The ISA, and indeed, the Indian semiconductor industry have come a long way since! Both have made a lot of noise during this journey. Back in those days, it was a bit difficult for me to explain to folks in India that I wrote about semiconductors! Or, for that matter, EDA. Never mind. That was the ‘age of innocence’ for the Indian semiconductor industry!

Nevertheless, I was there!

The journey of the ISA and the Indian semiconductor industry, post the formation of the ISA has been pleasant, so far. One, it got all leading companies under one umbrella. Two, we started seeing some India specific reports on the industry. Does not matter if the statistics presented were/are agreeable to all. Three, it led to the semicon policy. Again, it does not really matter, how much of it has succeeded as far as setting up a wafer IC fab is concerned.

What it has not achieved so far is that these events have not really led to the creation of a knowledge economy one would have liked to see. However, that will change for sure in the times ahead! Do understand: semiconductors is not an easy subject, nor easily understood by all. I’m still a student and have much more to learn about this industry!

Continuing with the journey of the ISA and the Indian semiconductor industry, starting from 2005, when the Indian semiconductor industry and the ISA were discussing design hub challenges before India and whether the country should go fabless vs. fabs, it has since led to the formation of the Indian semicon policy in 2007. That sparked off a huge debate about having fabs in India in 2008, to the falling off of fab plans by early 2009! The strength of India in embedded electronics and design services also came to the fore during this period. Last year, it led to the emergence of solar photovoltaics (PV) and solar fabs in India. How this market segment plays out in the coming years would be interesting to watch.

As of now, the Indian wafer IC fab story lies dead and buried! Will it get revived? No, for the present. Perhaps yes, in the future.

The new mantra in 2009 became Made in India, Made for India. Now, the emerging mantra in 2010 seems to be product development and product development companies, and India as a market with growth momentum.

All of these really make for very compelling reading, especially for followers of the Indian semiconductor industry. However, everyone associated with the Indian industry will agree that while a lot of noise has already been made and it has been heard globally, a lot more really needs to be done to make India a truly semicon powerhouse.

Hope to see you at the ISA Vision Summit 2010. More, later!

Monday, January 25, 2010

Electronics for energy efficient powertrain

Early this month, an Auto Expo was held in New Delhi, the capital of India, organized by CII, SIAM and ACMA with the support of the Government of India. The ISA ExCITE was part of the Auto Expo, a seven-day exhibition with a one-day conference – where, the India Semiconductor Association (ISA) organized a one-day conference on smart and green automobiles.

The Auto Expo is the premium event for auto sector in India with global participants.

Session 1 of the conference focused on electronics for energy efficient powertrain.

Environmental concerns and driver comforts are major factors for deciding on next generation automotive power-train electronics. Most countries are championing cleaner and greener automobiles. Today's high-end ECUs (Electronic Control Units) are combining these demands to create efficient engine management systems that would be driving tomorrow's people's cars.

In his presentation, Praveen Acharya, vice president, Semicon Solutions, KPIT Cummins Infosystems Ltd, said that powertrain was both complex and critical. Highlighting the automotive/ECU market, he added that while this market had matured in Japan, USA and Europe, big growth was likely in BRICs.

He highlighted some industry challenges for powertrain and electronics, which include:
* Meeting stringent emission controls globally.
* Fuel efficiencies across fuel variety (gasoline, diesel, EV, hybrids).
* ECU reductions in bill of materials and cost.
* MCU features, and performance and reliability.
* Development cycle time for new generation MCU/ECUs.
* Reduction in number of MCUs for powertrain and platform based approach.

Dr. A. Zahir, vice president, Bosch, discussed sustainable individual mobility. He focused on the reduction of CO2 emissions as well as technologies to support CO2-reductive comportment.

He highlighted some measures to manage vehicle energy. These include: combustion engine optimization, demand-responsive energy management, stop/start, hybrid, electric vehicles, components optimization, and waste heat recovery.

Dr. Zahir said global warming and population were rising. Hence, there was a strong increase of urbanization and energy hunger in the emerging countries. All of these challenge sustainable individual mobility.

Sustainable individual mobility can be achieved by reducing CO2 emissions and by increasing fuel economy. With a system approach based on enabling technologies from the Bosch portfolio, there is an enormous potential to reduce the CO2 emissions and increase fuel economy.

All of the required functionalities require cost effective electronics running complex control algorithms and diagnostics. The proliferation of smart and highly integrated.semiconductors will continue to accelerate in the automotive domain.

Suraj Mukundrajan, director Automotive Development, Infineon Technologies India Pvt Ltd, touched upon how semiconductors can enable fuel efficiency.

He highlighted certain global CO2 targets, wherein, the European Union (EU) proposes steep fines to cut car CO2 from 2012. For cars – 120gCO2/km by 2012 +10 g coming from biofuels. Now, the law is: 35mpg CAFE (Corporate Average Fuel Economy) by 2020 for cars.

Mukundrajan advised that emission reduction could be achieved by optimizing different areas in a car. The full spectrum of semiconductor technologies would be required to achieve energy efficiency in automotive electronics Less fuel consumption and cleaner engines will happen due to better performing MCUs.

He also highlighted the advantages of automotive hybridization. For instance, plug-in hybrid saves 40 percent to 60 percent energy compared to conventional combustion engines.

Already, the 2013 annual production will likely rise to over 3 million vehicles, which is about ten times the production of 2005 with 335.000 hybrid cars. Further, the semiconductor content in hybrid cars is much higher -- $525-$900 as compared to conventional combustion engine cars -- $225-$300.

Mukundrajan added that the CO2 target must be achieved. Although, the customer did not clearly realize the value today, the required technologies do exist. However, they are not included in the drive cycle in a huge manner, at least not yet!

Sandip Sarkar, head - Electrical Systems, Controls & Software Engineering, General Motors - Technical Center India, highlighted the huge opportunity for smart and green automobiles. While there was 70 million sales in 2007, in 2016, it is likely to touch 95 million in sales.

According to him, the automotive DNA of the future would include: electrically driven automobiles, energized by electricity and hydrogen, powered by electric motors, controlled electronically and 'connected.'

Sarkar highlighted the sustainability benefits as well. These include:
* Lower mass and smaller size vehicles.
* Reduced battery, hydrogen storage and power costs.
* Predictable routing, network-wide traffic management, reduced travel times, real-time congestion pricing.

The vision is to have electrically driven vehicles in the future – equipped with sensors, V2V, GPS, digital maps, electronic controls and actuators, etc. This will enhance roadway safety, besides providing real-time traffic management. Smart intersections would be part of the roadway of the future. All of these together, will enhance energy efficiency and reduced emissions.

Wednesday, January 20, 2010

Qualcomm, AMD head top 25 fabless IC suppliers for 2009; Taiwan firms finish strong!

So, IC Insights has revealed the top 25 list of fabless IC suppliers for 2009! No surprises, Qualcomm still leads!However, AMD is the surprise runner-up, for now. The reason being: AMD became a fabless company by including its Dresden, Germany fabs as part of GlobalFoundries spin-off. IC Insights included all of AMD’s sales for 2009 in its study.

Some other interesting points
First, as many as nine fabless IC companies — all of the top nine companies — had sales of $1 billion or more in 2009. These are: Qualcomm, AMD, Broadcom, MediaTek, nVidia, Marvell, Xilinx, LSI Corp., Altera and Avago! And you still believe there was a recession in H1-09?

Movers and shakers
So, who are the leading top movers and shakers?

* nVidia dropped down two places from 3rd in 2008 to 5th in 2009.
* Marvell also dropped down two places from 4th in 2008 to 6th in 2009.
* With AMD coming in 2nd place, Xilinx, LSI, Altera and Avago — all dropped down one place each.
* CSR dropped down three places from 12th in 2008 to 15th in 2009.
* MegaChips dropped down four places from 15th in 2008 to 19th in 2009.
* Conexant had the steepest drop — dropping 11 places down, from 14th in 2008 to 25th in 2009.
* MediaTek moved up one place from 5th in 2008 to 4th in 2009.
* Realtek moved up three places from 16th in 2008 to 13th in 2009.
* Mstar moved up five places from 19th in 2008 to 14th in 2009.
* Richtek had the steepest climb — moving up by 11 places, from 35th in 2008 to 24th in 2009.

The list comprises fabless IC suppliers from the USA — which has 17 representations, including nine suppliers in the top 10! Taiwan has six representations, with one — MediaTek — figuring among the top 10, well, top five actually! Europe and Japan have one representation each — in CSR and MegaChips.

True story in stats!
If you look at the percentage change, the true story of 2009 reveals itself! By the way, AMD is excluded here.

First, Taiwan’s MediaTek was by far the most impressive, registering a very strong 22 percent increase in sales to $3.5 billion.

The other impressive performers — all from Taiwan — have been Realtek with 15 percent growth, Mstar with 33 percent growth, and Richtek with 12 percent growth.

From the USA, Atheros and Silicon Labs have registered 6 percent growth during 2009, respectively. Besides these two, Qualcomm registered 2 percent growth.

So, Taiwan firms have indeed finished strongly in 2009, as compared to the others, with four of the six representative firms registering positive growth during 2009.

The rest of the companies in the top 25 list registered negative growth!

So, all of this means that only seven of the top 25 fabless IC suppliers (excluding AMD), had a good 2009 — or a year of growth, with the remaining top 18 IC suppliers having a poor year.

Will we see more of Chinese companies making this list? Probably yes, in the years ahead!

Will we see any Indian company enter this list? Not so soon, perhaps!

According to the GSA Dec. 2008 figures, the distribution of fabless IC companies is: Canada 29, USA 606, Europe 151, Israel 61 and Asia 510 — China 222, Taiwan 196, Korea 47, Japan 16, Taiwan 16, Singapore 7, Malaysia 4, and India 2-28 (two fabless companies and 28 design services/IC providers).

Business continues to grow in India and it has spilled into third party design and IP houses, and some into a few fabless companies, and assembly and test facilities. This means: there’s lots of room for growth! India has all of the enablers to move up higher!

Tuesday, January 19, 2010

Indian semicon industry: Time for paradigm shift!

Early next month, the India Semiconductor Association (ISA) will be hosting the ISA Vision Summit 2010. On the backdrop of this, I did a short study on the strengths and weaknesses of the Indian semiconductor industry, and whether a paradigm shift is the need of the hour to further enhance the industry’s image!

First, the strengths — what really makes the Indian semiconductor industry tick and appear attractive to the overseas players!

Key strengths…
Dr. Pradip K. Dutta, corporate vice president and managing director, Synopsys India, and treasurer, ISA, said, “India’s strengths include its huge market potential, availability of good engineering talent with the ability to scale up as required; and a pretty well established IP/judicial regime.”

As per Raju Pudota, managing director, Denali Software, India defintely has a great talent pool and a lot of work has been done over the last 20-25 years. “Over a period of time, India has developed a variety of skills and capabilites. We have forayed into multiple domains. There is a variety of domain knowledge that has been developed. As captive design centers, companies are having more ownership as before,” he added.

Rahul Arya, director, Marketing & Technology Sales, Cadence Design Systems India Pvt Ltd, listed strengths from business and technology perspectives. From a business perspective, it would include an expertise built over years, scalability of operations, etc. “If you look at any other emerging country, expertise could be there, but can you scale that?” Also, cost arbitrage is still valid.

From a technology perspective, India’s strengths include embedded design, verification and an increasing ownership of end-to-end SoC designs, which are well known globally!

…and weaknesses!
Now, on to the weaknesses, which I’ve actually, written about a lot in the past!

Synopsys’ Dr. Dutta said: “Although we have lot of talent, it is raw talent. The industry is more services oriented and not product oriented, with very little of systems manufacturing. Maybe, the services companies became more successful, so there has been risk averseness. However, it can change overnight with some mavericks coming in. As for systems manufacturing, the govenment of India should put together a national agenda to promote electronics manufacturing.”

Denali’s Pudota agreed that the Indian semiconductor industry is still in the services center mode. “We have not yet managed to translate the talent base into product startups. Somehow, the mindset and value proposition is: how do we get margin on the work we do. Its been like that in the semiconductor industry. You don’t see many home grown startups,” he added.

Cadence’s Arya also advocated that India needs to move away from service to product mindset. According to him, some other weaknesses include — not enough basic research and not many PhDs in the fraternity, as well as a gap of new hires — the time new hires take to become productive, and third — a gap in analog and RF engineers.

Tackling weaknesses
Now, how can the Indian semiconductor industry go about tackling these so called weaknesses?

Particularly on the subject of having more product companies in India, Arya advised: “It is a momentum we have to build, and it is more of a mindset issue. We need to keep re-inforcing and showcasing successes. We should leverage our reach to local market, understand customer pain points and serve that need. We should not box ourselves that if we don’t have a fab, and focus on applications instead.”

As for the lack of PhDs, etc., the government is already working on this area. Arya added: “There should be some good stipend to pursue PhD and MTech. We also need to scale the availability of good faculty. That is where the government can play a big role in incentivizing them. There should be better industry-academia partnerships as well.” However, he added that the univesity-academia-VC model is currently missing in India. This aspect, it is hoped, will get plugged in soon.

According to Pudota, specific to the semiconductor industry, overall, while there’s some risk taking potential, there may be some semi-product-semi-services companies. However, he believes that the investing capital available in the country has so far been more into services startups. Hence, the need for more product companies.

He noted: “Product companies definitely require two to three years gestation. Some few million dollars need to go into these companies. Somehow, the investing indusrty in India does not seem to like such kind of models. The semiconductor industry can only follow the software industry. I feel that one of the primary weaknesses is the ecosystem, which is not conducive enough to promote product type companies.”

Many system companies are moving away from chip design and are doing system integration. For the semiconductor companies., the only way to move up the value chain is to become a systems company. “We can’t move into VAS, so for us, the only thing we can do is by developing products,” he said.

Need for paradigm shift in Indian industry!
So, are all of these really the weaknesses of the Indian semiconductor industry or is there a serious need for a paradigm shift or change within the industry? Perhaps, the latter!

Every single conference I’ve attended in the past six months and this year, you get to hear speakers highlight the need for having more product development companies. I’m all for it!

Let me also mention about an event! Recently [actually, in the middle of December 2009], I was invited to a session hosted by Carson Bradbury from Cre8Ventures! Launched by Mentor Graphics UK Ltd in 2005 as an independent network, Cre8Ventures aims to help start-up electronic design companies achieve business success. Interesting!

The audience at this session, which largely comprised Indian product start-ups, got to learn more about the European Microelectronics Academy (EMA), as did I, and helped Cre8Ventures learn more about India’s interests.

Will there be an Indian chapter or an Indian Microelectronics Academy (IMA)? Too early for me to say! However, should that happen, I would surely love to be a part of such an initiative!

Wednesday, January 13, 2010

NSM formally launched, but why so much focus on thermal solar?

So, the Jawaharlal Nehru National Solar Mission (NSM) has been launched! The NSM was launched by the Prime Minister of India, Dr. Manmohan Singh, at the Solar Energy Conclave organized by the MNRE at Vigyan Bhavan, New Delhi this week in the presence of Dr. Farooq Abdullah, Union Minister for New & Renewable Energy, Sharad Pawar, Minister of Agriculture and Minister of Consumer Affairs, Food and Public Distribution, Jairam Ramesh, Minister of State (Independent Charge) for Environment and Forests, and Montek Singh Ahluwalia, Deputy Chairman of the Planning Commission, Government of India.

In his address, the Prime Minister of India, Dr. Manmohan Singh, said: “Increased use of solar energy is a central component of our strategy to bring about a strategic shift from our current reliance on fossil fuels to a pattern of sustainable growth based on renewable and clean sources of energy. I sincerely hope that this solar Mission will also establish India as a global leader in solar energy, not just in terms of solar power generation but also in solar manufacturing and generation of this technology.

“The importance of this Mission is not just limited to providing large-scale grid connected power. It has the potential to provide significant multipliers in our efforts for transformation of India’s rural economy. Already, in its decentralized and distributed applications, solar energy is beginning to light the lives of tens of millions of India’s energy-poor citizens.

“The rapid spread of solar lighting systems, solar water pumps and other solar power-based rural applications can change the face of India’s rural economy. We intend to significantly expand such applications through this Mission. As a result, the movement for decentralized and disbursed industrialization will acquire an added momentum, a momentum which has not been seen before.

“I am happy that the Federation of Indian Chambers of Commerce and Industry (FICCI) has been associated with this event. The role of industry in this Mission’s success will be critical. Eventually, if the ambitious roll out of the Mission is to become a living reality, we will have to create many ‘Solar Valleys’ on the lines of the Silicon Valleys that are spurring our IT industry across the four corners of our country.

Dr. Farooq Abdullah, Union Minister, stated in his keynote: “This Mission is named after India’s first and visionary Prime Minister, Jawaharlal Nehru. For him, India’s development needed to be anchored in its mastery over cutting-edge technologies. The Solar Mission is very much in line with his vision, which has made India today, a leading nuclear and space power. He would have been equally keen and proud to see India attaining the same level of advancement in solar energy. I am confident, that under the leadership of our Hon’ble Prime Minister, Dr. Manmohan Singh, we shall make India a global solar power as well.”

The focal point, for the next three years, will be the NTPC Vidyut Vyapar Nigam (NVVN), which is the power trading arm of the NTPC. NVVN will purchase solar power at rates fixed by the Central Regulatory Electricity Commission (CERC) and for a period specified by the latter.

Now, I wasn’t able to attend this conference, but was able to download the presentations from the MNRE website. I’ve seen the first two presentations so far. These are: Jawaharlal Nehru National Solar Mission – Dr. B. Bhargava, direcor, MNRE, and Solar Grid Power Projects Guidelines – I.J. Kapoor, Director Commercial, NTPC Ltd.

First, the targets are quite clear:

Targets set for development of solar power under NSM
Installed Capacity (MW)
Phase – I (up to 2013) 1100 MW
Phase – II (up to 2017) 4000 MW
Phase – III (up to 2022) 20000 MW

I’ve seen some points in these two presentations, which I don’t quite understand, especially in the second one. Those industry colleagues who have attended this solar conference at Vigyan Bhavan, may I request you to share your opinions!

For instance, in the selection of solar power developers (SPDs), the ratio of solar PV to solar thermal is proposed to be in the ratio of 40:60! I am tempted to ask: why so much focus on thermal solar? Yes, an NSM target is to create favorable conditions for solar manufacturing capability, particularly, solar thermal for indigenous production and market leadership. But, why maintain the 60 percent for thermal solar all the way up to 2022? Can’t it be toned down in the later years and phases to accomodate newer technologies? Won’t PV prices have dropped sufficiently by then?

Second, I wonder how the net worth of the SPD for the past three years will be calculated, as well as the SPD’s turnover of last three years. How many have been really around for three years, unless they are large industrial house? What about newcomers? Since I wasn’t present at the conference, please share your opinion, friends.

Third, the readiness of PPAs and PSAs by 31st March 2010, which means, preparing the implementation mechanism will take another two months. Fourth, the readiness for signing of PPAs and PSAs is by 31st October 2010. Isn’t eight months a very long time? Nearly the entire 2010 will be gone by then — the first year of the NSM period.

Fifth, a committee under former Secretary-IPP will be formed to suggest measures to establish solar thermal power industry in the country. By when? Was a date given during the conference? I don’t know, as I wasn’t present there, and request those who are aware, to share the information.

Am sure a lot of thought has gone into the NSM. I am just raising these points out of my ignorance. In some ways, the NSM is also a triumph for the Indian semicon policy.

I quite like the off-grid solar applications, which have a target of 200 MW capacity in the first phase (up to 2013). The focus would be on:
- solar lights
- rural power supply
- replacement of diesel
- telecom towers
- solar water heaters
- solar cooking for institutional use

There is a link on MNRE’s web site (, which says: ” Your suggestions about Jawaharlal Nehru National Solar Mission.” Those who have any suggestions or queries, should post those to the MNRE.

Tuesday, January 12, 2010

Global semicon industry on rapid recovery curve: Dr. Wally Rhines

Thanks to Mentor’s Raghu Panicker and Veeresh Shetty, I had the pleasure of an exclusive meeting with Dr. Walden (Wally) Rhines, chairman and CEO, Mentor Graphics, post his technical keynote at the recently held VLSID 2010.

We discussed a range of issues, such as the global semiconductor and EDA industries, as well as the Indian industry.

Global semicon industry
According to Dr. Rhines, the global semiconductor industry is currently on a rapid recovery curve. However, that the semiconductor total available market (TAM) would decline in 2009 due to the weakness of H1-09.

He added that the global semiconductor industry’s growth could even be as much as 22 percent during 2010, as also advised by Malcolm Penn of Future Horizons. This number can definitely change, rather than remain so optimistically high, as the year goes by. (Yes, Malcolm’s an optimist!)

Dr. Rhines said, “If you look at the major semiconductor companies, most of them will have an opportunity to grow — after the Q4 results are in.”

On global EDA industry
So how will all of this contribute to the well being of the global EDA industry?

According to Dr. Rhines, the EDA industry has been a little different. Mentor and Synopsys have probably had more growth and stability. “Mentor could be the only company to grow. Bookings in Q3-09 have grown by 15 percent and revenue by 3 percent. We are perhaps the only major EDA company that has grown,” he said. In the long term, EDA tends to trend with semiconductor R&D. Semiconductor R&D was flat in 2008, down in 2009, and will probably grow in 2010. So, EDA will probably lag in 2010.”

Advise for Indian semicon industry
I also requested Dr. Rhines to advise the Indian semiconductor industry.

He said: “The Indian semiconductor industry needs to look at more of systems and IC design, as well as embedded software development. India should also have more product start-ups. What makes India unique is its talent, education, etc. India is now producing its own electronic design architecture and embedded software, and it also has the systems infrastructure. India can greatly advance what it already does so well.”

Commenting on the industry’s weaknesses, he added that the Indian semiconductor industry needs to increase the infrastructure for its local start-ups. That infrastructure would require things such as VCs and especially, a cultural acceptance of failure. Also, if India does not have the intrinsic semiconductor manufacturing capability, then it needs to stay current on the evolving technologies.

Does India need a fab?
On being asked this question, Dr. Rhines’ reply was immediate: “I don’t think India ever needed a fab! You can easily have closer relationships with other fabs. The highest paying jobs are design and innovation, etc.. The recession has been driving down costs and innovation has been happening. There are big growing markets — in India and Asia.”

Sunday, January 10, 2010

Global semiconductor sales forecast: Cowan’s LRA model (based on Nov. 2009 sales data)

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 2009 and 2010 global semiconductor sales estimates associated with the forecasting model — the Cowan LRA model for predicting worldwide semiconductor sales.

The presently updated global semiconductor sales forecast estimates are based upon the recently published November 2009 actual sales numbers released by the WSTS (posted on its website on Jan 4, 2010).

The table below details the latest, updated forecast numbers covering the next five quarters, that is, from 4Q09 through 4Q10, respectively, as well as for the full years of 2009 (including 4Q09) and 2010.

NOTE – This is the last forecast for 2009 since next month’s sales data from WSTS, namely for Dec 2009, will “wrap up” the year (2009 will become history!) and the model will “shift focus” to 2010 forecast expectations and beyond!

As the table below shows, the latest updates for years 2009 and 2010 chip sales forecast estimates increased by +1.9 percent and +0.5 percent, respectively, to $222.1 billion and $236.5 billion compared to last month’s sales forecast estimates of $217.9 billion and $234.0 billion, respectively.

The updated 2009 and 2010 chip sales forecast estimates correspond to yr-o-yr sales growth forecast estimates of -10.6 percent and +6.5 percent, respectively, which represent a continuing improvement compared to last month’s sales growth prediction of -12.6 percent for 2009 but, however, no change from last month’s same sales growth forecast estimate of +6.5 percent for 2010.

Table Summarizing Latest Cowan LRA Model’s Sales & Sales Growth Forecast Estimates For The Next Five QuartersCowan's LRA Model: Sources: WSTS / SIA (Actuals); Cowan LRA Forecasting Model (Forecast Estimates).

Remember that the model is dynamic, that is, is recalculated each month as the year plays out; therefore today’s latest, updated full year sales growth prediction will not sit still but will evolve over the coming months.

Additionally, the model also projects a sales forecast estimate for next month, namely for December 2009. Thus December’s (actual) sales forecast estimate is projected to be $20.023 billion, which corresponds to a 3MMA (three Month Moving Average) sales forecast estimate of $21.28 billion as is normally published by the SIA.

Thursday, January 7, 2010

Future research directions in EDA: Dr. Prith Banerjee @ VLSID 2010

Dr. Prith Banerjee, senior VP of research at HP and director of HP Labs, discussed some promising areas for research while delivering his keynote on future research directions in EDA at the ongoing VLSID 2010.

He highlighted eight areas that HP Labs is currently working on. These are:

– Digital commercial print
– Content transformation
– Immersive interaction
– Information management
– Analytics
– Cloud
– Intelligent infrastructure
– Sustainability

EDA challenges
So, what are the EDA challenges? According to Dr. Banerjee:
* Today, EDA develops automated tools for designing ICs. However, there is a need to address automation for electronic systems at higher levels.
* There is a proliferation of new modes for communication and collaboration has resulted in the explosion of digital information.
* An intelligent IT infrastructure, which can deliver extremely high performance, adaptability and security — will be the backbone of these developments.
* In future, you need to look at design automation for entire systems.
– networks and data centers
– electronics and photonics
– performance and sustainability

Growth of new modes of communication and collaboration has led to an explosion of digital information. The IT industry would need to develop novel ways to acquire, store, process, and deliver information to customers. An intelligent IT infrastructure, which can deliver extremely high performance, adaptability and security, will be the backbone of these developments.

Intelligent infrastructure
This is required to capture more value via dramatic computing performance and cost improvement. HP Labs’ contribution has been on three big bet projects, namely, next generation data centers, networking and next generation scalable storage. HP Labs’ research contribution has been:

* Exascale
An Exascale data center that will provide 1000X performance while enhancing availability, manageability and reliability and reducing the power and cooling costs. HP Labs is working on the design of a sustainable data center that reduces total cost of operation (TCO) and carbon footprint, while meeting the current quality of service goals.
— Designed across components, interconnects, power and cooling, virtualization, management and software delivery.

* Photonics
– Replace copper with light to transmit data

* 1000X gain in performance

Dr. Banerjee said that there is a need to create brand new optical technologies that can work in exascale. Photonics interconnects make use of light for data communications. The transmit or receive optical bus is a simple modular system in four elements — transmitter media, optical tap, optical source, and optical receiver.

Non volatile memory and storage is another area that HP Labs is working on. Dr. Banerjee highlighted the memristor or a resistor with memory. In future, it has the potential to replace DRAM, hard drives and Flash memory.

“The memristor has the potential to revolutionize electronics,” he added. It is structurally simple and easy to fabricate. Also, it switches in nanoseconds and has many year lifetimes. HP Labs is said to be in discussions with leading memory makers for further.developing and licensing this technology.

Next-generation displays is yet another key research area at HP Labs. These are said to be unbreakable, conformable, ultra-thin and lightweight displays. Such displays have paper like qualities + video capabilities.

Technologies such as memristors, photonic interconnects, and sensors will likely revolutionize the way data is collected, stored and transmitted.

Sustainable data centers is another key research area. Among HP Labs’ research contribution, it is aiming to reduce TCO by 50 percent and carbon footprint by 75 percent.

Highlighting another industry challenge, Dr. Banerjee pointed out that the IT industry is only responsible for 2 percent of total carbon emissions. The global economny contributes the remaining 98 percent of total carbon emissions from other industries such as aviation, transportation, retail, etc. IT has a significant role to play here.

Naturally, all of the research areas would require sophisticated system-level design automation tools. Dr. Banerjee said: “In the past, the EDA research focused on chips. In the future, we need to look at entire systems.”

Wednesday, January 6, 2010

Delivering 10X design improvements: Dr. Walden C. Rhines, Mentor Graphics @ VLSID 2010

Friends, here is an overview of the remarks made by Dr. Walden C. Rhines, CEO, Mentor Graphics, during his technical keynote titled: Delivering 10X design improvements, at the ongoing VLSID 2010 in Bangalore, India.

Today, the exponential rise in complexity has quickened its pace as the industry moves toward adoption of 28 nm and below. Dr. Rhines discussed how over the next five years, 10X improvements in design methodologies are required in four areas: high-level system design, verification, embedded software development, and back-end physical design and test. According to Dr. Rhines:

* Reduced cost per function will continue on a predictable learning curve – long after Moore’s Law is obsolete.
~ 40 billion transistor (5-10 billion gate) ASIC designs in 2018.

* EDA tools for 40 billion transistor designs are available today:
— Design abstraction.
— Advanced verification methodology changes.
— Merging of physical verification with new router architectures.
— Parallelization.

* Embedded software automation (ESA) will automate software development and verification
— Just as EDA did for hardware design.

So, why has Moore’s Law been a useful approximation for 40 years? It is based upon a basic ‘Law’ of nature – the learning curve. Moore’s Law is a special case of the learning curve when two things are true:

1. Cumulative transistors produced increase exponentially with time (e.g. 2x cum volume => fixed percentage cost decrease).
2. Almost all cost reduction has come from shrinking feature sizes (and growing wafer diameter).

Learning curves
* Cost per unit decreases by a fixed percent every time total cumulative volume doubles.
* Applies to all free market products and services (over centuries) when measured in constant currency.
* Used to predict future costs.
— Aircraft industry
— Semiconductor industry
* Also true for subsystem or component costs and improvements in reliability, quality, yield, etc.

A non-linearity of growth in transistor consumption led to Moore’s Law adjustments.

Can Moore’s Law support 10X complexity?
Going forward, Moore’s Law doesn’t matter! The innovative assembly and packaging of multiple chips will drive lowest cost. The cost per function will continue to decline long after Moore’s Law is obsolete.

Growth in unit volume distinguishes semiconductors from other industries (see graph). The question is: Will transistor unit volume continue to grow at 49 percent annually for the next 10 years?

There has been constant or increasing IC unit growth. There have been only three years of negative growth in history. There has also been constant transistor unit growth. An order of magnitude increase in complexity Is likely by 2018. Therefore, what are the applications that will require 10X more transistors (per package) by 2018?

Now, computers and cell phones account for over 70 percent of semiconductor revenue. Also, semiconductor recessions are typically followed by the rapid growth of new semiconductor applications.

The economic cycle provides a boost to new DRAM intensive applications. Further, new NAND Flash applications will be accelerated by price decreases beyond Moore’s Law. The affordability of memory enables new applications and re-ignites the old appetites. The reduced memory cost drives new architectures and applications.

Dr. Rhines gave examples of how digital cameras achieved high volumes as falling prices ignited volume sales. Low prices generally tend to open up whole new markets. ASICs have even enabled sub $20 mobile handsets. The integration of new functionality also ignites demand in both new and old applications — an example being products and applications incorporating GPS.

As another example, the desktop PC market growth is slowing after 30 years. However, notebook/netbook/smartbook market appears to be headed for many years of growth ahead! Similarly, while the Internet penetration as well as mobile phone adoption are saturated in mature economies, both of them are in a high-growth mode in the emerging economies.

10X design complexity to increase by 2018
According to Dr. Rhines, the 10X design complexity will increase by 2018. The learning curve will continue to provide predictable cost reduction after Moore’s Law is obsolete. The transistor unit volume growth of nearly 50 percent per year will continue to drive the 35 percent per year cost reduction. Also, new applications will emerge to utilize the 10X growth of transistors per package, accelerated by recession-driven price reductions.

We will see a ~10X increase in transistors over the next eight years, leading up to 40 billion transistors by 2018, he added.

Delivering 10X and beyond design improvements
Four principal areas will also require 10X improvements in design methodologies. These are — high-level system design, verification, embedded software development, and back-end physical design and test.

In system level design creation, there is a need to raise the level of design abstraction from RTL to transaction-based.
* Transaction-based model generation
* System-level design and modeling standards
— C++, SystemC, UML, SysML, MatLAB

A new level of abstraction will meet 10X complexity requirements. The industry is due for next stage in abstraction. Design tools for 40 billion-transistor ICs are probably already available today, noted Dr. Rhines.

In functional verification, verification is falling further behind. We need exponential growth in verification capabilities to keep pace. A 10X increase in the number of transistors will require 1000X increase in verification. How can that be achieved? You can do the same thing, but do it faster. Or, you can eliminate redundant verification. You could also have transactional test benches, as well as mixed dynamic and formal verification. For instance, the intelligent testbench automation eliminates 90+ percent redundant verification.

Emerging Verification Techniques Near Term Productivity Gain
Emulation 100X-1000x
Intelligent Testbench 10X-100X
Transactional Testbench 500X-1,000X
Dynamic Formal 100-106X

In physical design and verification, there has traditionally been a new routing architecture every 2 ½ -3 technology nodes. You can do parallel optimization—MCMM, timing, etc. You can also merge place and route with full physical verification to reduce, or eliminate, ECO routing. Futther, you could opt for full parallelization of routing.

Lastly, the SoC design costs are forecasted to exceed $100 million within three years. Most of the SoC escalating design costs are attributable to the software development. All of these lead to the case for embedded software automation (ESA) – development and verification — in terms of re-use, automation and open standards. For instance, the Android platform has accelerated the embedded development beyond mobile phones.

Tuesday, January 5, 2010

India needs to develop the right products: Dr. Bobby Mitra @ VLSID 2010

The 23rd International Conference on VLSI Design and the 9th International Conference on Embedded Systems (VLSID 2010) kicked off this week at the NIMHANS Convention Center in Bangalore.

Inaugurating the conference today, wirelessly, along with the other distinguished guests, Dr. Biswadip (Bobby) Mitra, President & Managing Director, Texas Instruments India, said that the technology behind the conference has already started. This year, the conference is being taken to greater heights — VLSID 2010 is being webcasted live for the first time! Dr. Mitra added, “Taking the conference to the people is absolutely wonderful.”

Delivering the opening keynote, Dr. Mitra presented his views on how he foresees the change in India from a VLSI design community to a semiconductor community.

He said: “So far, the industry and the academia have been focused on developing products right. As we enter the new decade, another new vector is likely to become a guiding point — that is, developing the right products. We will be developing better chips for our customers the sooner we can better understand their system aspects.”

He urged the Indian semiconductor/VLSI community to continue developing products right, and also to develop products that really benefit the customer. “Understanding the end application is going to be very vital.”

According to him, India’s growing importance in semiconductors would be critical during this decade. However, India, as a market for semiconductors, will help everyone in learning more — when the customer is at your doorstep. The amount of consumption in semiconductors in India has been amazing so far, and will only grow in future.

He added that companies based in India — both MNCs and local — have been really innovating. India is a not-to-be-missed market! He concluded, “The time to invest in the Indian semiconductor market is now, not later!”

Later, welcoming the delegates, Dr. Mahesh Mehendale, General Co-chair VLSID 2010 and Texas Instruments Fellow & Director, Center of Excellence for Digital Video, Texas Instruments India, said that Bangalore had really emerged as hub for semiconductors and IT. In fact, 2010 indicates 25 years since this revolution started.

Commenting on this year’s conference, he said: “Our aim was to push this conference nationally and globally, by taking the conference to the desktops.” Dr. Ruben A. Parikhji, Program Co-chair, apprised delegates of the technical sessions.

This was followed by two technical keynotes:

* Nanoelectronics challenges for the 21st century, by Prof. Dimitri A. Antoniadis, MIT, and
* Deep submicron CMOS technology – the challenges for semiconductor IDM, Prof. Dr. Hermann Eul, Member of Management Board, Infineon.

Monday, January 4, 2010

Want to enter solar off-grid business? Build your own solar LED lanterns and emergency lights!

I’ve been looking at the search results on my blog. It doesn’t surprise me that solar is a top search topic with searches such as ‘emerging opportunity in solar cell manufacturing’ ranking very high.

Some other interesting solar-related searches — ‘list of new solar manufacturing companies’, ‘vertically integrated solar/PV companies’, ‘making home solar power in India’, ‘FDI in India for solar projects’, ’solar PV production in India’, ’solar products manufacturing opportunities’, ’solar cells manufacturers’, ‘India solar opportunities’, ‘top rated solar products’, etc.

Wow, I am overwhelmed! Friends, I have touched on each one of these aspects over the past year and will continue to do so in future.

There has been a lot of interest in off-grid solar applications in India and elsewhere for quite a while now. I’ve been discussing how one can easily enter the solar business with my associate Usha. Friends such as Ravi and Shilpa have got into discussions with me regarding the solar business in India as well. There have been a few requests on LinkedIn too, to comment on NSM and also, how to get into the solar LED lantern business.

That’s very interesting and heartening to learn that folks are getting serious about solar and LEDs. Many thanks to those who think I am capable enough to write about such topics. It is very humbling!

Now, if you search the Internet, you can find sites that offer tips on how to build you own DIY solar LED lantern — which, I consider a very easy way to enter the off-grid market in India.

Build your own solar LED lantern
Let me attempt this with the solar LED lanterns. First: how do you enter the off-grid solar applications business in India or anywhere? There are three ways of doing so.

One, either you build/manufacture a product yourself. Two, become a franchisee for some local manufacturer. Three, simply import products from manufacturers in China, Taiwan or anywhere else, and sell it in India, or wherever you are!

What would you actually need to build a solar LED lantern or even solar emergency lights?

* Solar cell/panel of say, 2W
* Two or four LEDs — you can go for high-brightness LEDs too!
* Small transparent plastic containers, preferably with holders to hang, if needed.
* Terminal block
* Wires, toggle/selector switch and other basic circuitry to connect to the LED
* 2A or 3A batteries

If you want to build a rechargeable unit, add a NiCd/NiMH rechargeable battery and connect it to the solar cell/panel.

Now, am sure you can assemble this! If not, find some electrical engineering student to do it for you. Or, some electrical mechanic! Or, search the Internet for sites who tell you how to do this, if you want to save on the assembly cost too!

Where can you source these materials? From electrical suppliers or components distributors! Or, from the open market itself.

What will be the development cost for building a basic solar LED lantern? I reckon, not even Rs. 200 (US $5), or even less!

Where can you sell these? Anywhere! Hope you have seen the Hindi movie — Mumbai Se Aaya Mera Dost, starring Abhishek Bachchan, where he brings a TV to his village and how it changes the village’s lifestyle. Or, I hope you’ve seen the Hindi movie, Swades, starring Shah Rukh Khan, where he literally ‘brings’ electricity to a remote Indian village!

Try visiting your own village (if you belong to one), or your home town. Am sure there are power cuts galore, if you happen to live in India. Just go there and ask around if the people would be willing to try out your product. Once they can see it in use and understand its benefit, who knows, you could be in for major success. Most importantly, think, how you can help improve the quality of life of the rural folks!

Oh, think about our farmers too. If you can develop good quality solar irrigation pumps, that would be great! It won’t be a bad idea either to visit the Auroville and The Matri Mandir in Pondicherry. They use solar cookers!

If making this solar LED lantern looks easy, then try and building your solar decorative lights, solar lamps, solar camping lanterns, solar flashlights, solar emergency lights, etc. The choice is yours! Share your success stories with me!

Maybe, there are too many makers in this field, some would say. I feel, the only differentiator would be in the quality of the LEDs used and how long your product would operate in a day — if 10-12 hours or more, brilliant!

The National Solar Mission (NSM) has ambitious plans. Maybe, here’s a way of participating and contributing to its success.

Finally, I am a layman, who doesn’t understand engineering. What I’ve said here is in simple, layman’s language. Am sure there are many knowledgeable folks out there, who can share their experiences.

Saturday, January 2, 2010

Top semiconductor and EDA trends to watch out for in 2010!

This is a continuation from my previous blog post, where I was in conversation with Jaswinder Ahuja, corporate vice president and managing director, Cadence Design Systems (I) Pvt. Ltd. I also took this opportunity to quiz him on the top semiconductor and EDA trends for 2010.

Semiconductor and EDA industry trends forecast for 2010
According to Jaswinder Ahuja, as we look ahead to 2010 and beyond, the challenges are great for the semiconductor and EDA industries – but so are the opportunities.

He said: “We must learn to thrive in an increasingly globalized world and in a collaborative yet competitive ecosystem. EDA providers and semiconductor companies must collaborate to reduce the escalating cost of high-quality design and remove productivity bottlenecks, and work together to bring the semiconductor industry out of the doldrums.

“In terms of the worldwide semiconductor industry, collaboration will continue to accelerate. The Common Platform, which brought together competitors including IBM, Chartered, and Samsung to share the costs of next-generation process technology, is one example.”

Deep, pro-active collaborations between EDA vendors and semiconductor providers have become crucial. Cadence in 2009 announced collaborations on advanced process node design with ARM, TSMC, SMIC, Global Foundries, UMC, IBM, Common Platform, and other partners.

Ahuja added: “In terms of EDA, providers must not only help customers reduce design and verification costs and solve technology problems, but must also help them differentiate themselves from competitors, and collaborate with partners on a worldwide basis. To do this, the EDA industry must focus on the integrated solutions and close partnerships, rather than simply selling point tools.

“Time-to-market pressures and design complexity are critical challenges that design teams face today. Design teams want to “get it right” the first time, and improve the predictability of design process. EDA will have to sharpen the focus on making the verification process cost effective and focus not only on design but on IP integration so that design predictability can be ensured for customers at optimum costs. Design for manufacturing (DFM) technologies including extremely accurate modeling solutions for both mask-making and manufacturing to avoid expensive silicon re-spins will be another focus area.”

End consumers are driving our customers to incorporate more functionality into the products. This demand for new functionality is driving both the importance and complexity increase in mixed signal and RF (radio frequency) design.

“As our customers face significant cost pressure, outsourcing certain elements of a production-grade design environment can address certain customer needs for reduced cost and risk. The need for business optimization gave a fillip to models like Software as a Service for electronic design, whose flexible engagement models to access state-of-the art design environments helped design teams reduce risk and cost while increasing productivity, in 2009. This will continue to gain traction in 2010.

“As designs migrate to sub-90nm process nodes, power management will be an imperative across the entire design and manufacturing chain,” noted Ahuja.

Strengths and weaknesses of Indian semiconductor industry
In terms of strengths, India’s talent base is a clear advantage. The availability and scalability of talent has attracted IDCs (India development centers) here for many years now.

Second, India has a wealth of technical expertise and proven delivery record of designs from spec to GDSII by design services companies and by IDCs, and in the embedded software area as well.

Third, the Indian semiconductor ecosystem is maturing and the IP regime in India is considered just and favorable for business.

In terms of weaknesses, we are yet to develop full product definition and system design capability from India. This is a challenge for the Indian semiconductor industry. Second, we are yet to develop a competitive manufacturing capability and capacity, although we have taken some steps in this area through ATMPs and solar fabs.

Yes indeed! The Indian semicon industry has many, many miles to go and several promises to keep!

Friday, January 1, 2010

Need for Indian semicon industry to think beyond embedded and design services

Howdy, folks! Warm greetings for a very happy and prosperous new year and wish you all the success in 2010.

A friend recently called me a ‘maverick’ blogger! I wonder why! Nevertheless, let’s start off the new year with a topic that has been coming up for discussion time and again.

It is said: Indian industry should think beyond embedded and design services. What should those be? What is that extra or more that the Indian industry should do to grab the attention of overseas players? I had the pleasure of discussing this aspect with Jaswinder Ahuja, corporate vice president and managing director, Cadence Design Systems (I) Pvt. Ltd.

India should become next innovation hub
According to Jaswinder Ahuja, the Indian semiconductor industry has come a long way from the early eighties when a few companies set up centers to offshore non-critical design work to India.

“Today, India is a preferred destination for chip, board, embedded software design and development. Captive design centers and design services companies now own end-to-end design. Fabless companies and IP providers have also emerged, resulting in the gradual, steady rise of India’s semiconductor ecosystem. India now needs to move to the next level i.e., become the next innovation hub for experimenting and launching new products,” he said.

The mobile phone revolution in India has taught us that India is a very large, viable and profitable market if all elements of the ecosystem align and there is appropriate innovation at the technology and business level.

The Indian market has some unique characteristics and provides Indian entrepreneurs a once-in-a-lifetime opportunity to innovate, create and capture value by building products and solutions for the local market, which can then be extended to global emerging and developed markets.

He added that Indian companies also need to get comfortable with collaborating across the value chain within India and globally. “Sectors like energy, healthcare, communications and education hold immense potential and we are already starting to see early signs of Indian companies innovating for the Indian market.”

All of this does make a lot of sense!

Yes, like you, I too am looking forward to seeing lot of product development and innovation. It is definitely not going to serve anyone by making negative statements as to what hasn’t happened in India so far! We all know what’s happened and what hasn’t! It would serve everyone’s purpose if much more attention was paid in India to actually developing and building products and solutions for India, which would solve local problems.

The second part of this discussion will look at the top semicon/EDA trends to watch out for in 2010.