We have become aware that our Apps Processor market share numbers have been challenged and that prompts us to explain our methodology and to give more details on this specific question regarding Marvell versus MediaTek.

Firstly, Strategy Analytics has a rigorous top-down and bottom-up methodology to measure the mobile processor markets. With our extensive apps processor methodology, we can account for every single smartphone apps processor ever shipped and every single dollar ever earned by mobile processor companies. We source the smartphone device shipment data from our “best-in-class” Wireless Smartphone Strategies (WSS) service. Also, we look at every single smartphone’s apps processor supplier using our popular SpecTRAX database service. Using these comprehensive inputs from our device research service teams, we built an extensive quarterly mobile processor tracker models, which are well received by our clients for their granularity and attention to details. In addition, we regularly check our numbers with leading mobile processor vendors in the industry.

The commentator asserted that Strategy Analytics “totally missed” Marvell which is unfounded. We are one of the first research firms to report Marvell’s TD-SCDMA smartphone growth (see our Q1 2012 PR and tracker report). We wrote the following at that time. “Marvell’s strong TD-SCDMA smartphone applications processor shipments at China Mobile were not sufficient to offset its declining shipments at Research in Motion. As a result, Marvell dropped out of the top-five.” Marvell's TD-SCDMA customers include RIM, Sony Ericsson, Samsung, Motorola, LG, ZTE, Huawei and Hisense. We are well aware of Marvell’s momentum in the Chinese TD-SCDMA smartphone market.

We estimate that Marvell’s smartphone apps processor shipment declined 26 percent Y/Y in 1H 2012. Marvell’s shipments to RIM declined 55.2 percent Y/Y in 1H 2012 while its shipments to non-RIM customers increased 44.9 percent in the same period. Please also note that Marvell is not the sole apps processor supplier to RIM.

The commentator also disputed our apps processor estimates for MediaTek. With our extensive OEM and chipset vendor relationship matrix we can account for the significant number of shipments for MediaTek in 1H 2012. Please note that MediaTek also participates in non-China smartphone markets such as Africa, Eastern Europe, India and Latin America. We are also aware that not all MediaTek’s apps processor shipments are 3G (UMTS / TD-SCDMA). MediaTek’s EDGE smartphone processor MT6516 is also popular in sub-$100 Android smartphones in emerging markets.

We are as keen as anybody when it comes to observing and tracking mobile processor markets. We always keep our ears to the ground and capture every single signal coming from primary sources. We are confident in our methodologies and numbers.

Sravan Kundojjala

Broadcom announced today that it will acquire privately held 4G chipset vendor Beceem Communications for $316 million. Beceem, founded in 2003, is the leading provider of WiMAX basebands and RF chips for portable and handheld WiMAX devices. The company also supplies WiMAX radio chipsets for gateways, routers and other fixed WiMAX customer premise equipment. Recently Beceem started to shift its emphasis to LTE with the announcement of its BCSM500 baseband, which will support both LTE and WiMAX. Broadcom so far hasn't announced its LTE plans publicly, and this acquisition signals that the company is looking to the future in the wake of growing competition and consolidation in the cellular baseband supplier base. Broadcom acquired GSM/GPRS/EDGE and W-CDMA technology through acquisitions, and these are now paying off, so it appears that Broadcom is repeating this strategy for LTE.

Beceem supplies basebands for all of the WiMAX USB dongles and data cards sold by Clearwire, the largest WiMAX operators in terms of subscribers. Beceem generated $43.7 million in revenue in 2009 with net loss of $17 million, but with shipments surpassing 1 million chips per quarter in Q3 '09 and sales having tripled since 2008, the company appeared headed for profitability by late 2010. Buoyed by this success, Beceem filed for an IPO initially thought to be worth more than $100 million in April '10. This would have made a nice down-payment to investors, which include Intel Capital, DoCoMo Capital, Samsung Venture Investment and NEC. Investment by all VCs in Beceem totals close to $200 million to date.

By August 2010, announcements of support for LTE by operators around the world made it clear that Beceem faced a weakening, or at best slower growth WiMAX market, cutting the value of its proposed IPO. With its new LTE-WiMAX baseband, Beceem could benefit from the launch of LTE by Verizon Wireless, AT&T Mobility, and MetroPCS in North America, however, Beceem will not ship samples of its WiMAX-LTE baseband before Q1 ’11. After shipping its first samples and getting them designed into products, Beceem would still have to go through interoperability testing (IOT) with infrastructure vendors and demonstrate support for FDD and TDD in multiple channel bandwidths. The IOT process could take 18 months or longer, pushing LTE chip production out to mid-2012 or later for Beceem. In light of this, investors evidently saw risks in continuing to support Beceem, and decided to take Broadcom’s offer.

From Broadcom's perspective, the acquisition will allow Broadcom to offer home gateway / router platforms that incorporate 4G, Wi-Fi, cable, DSL and / or fiber optics, which would allow multiple users in a household to share a single wired or wireless broadband connection, cutting the monthly price per user. Of even more potential significance, the acquisition will allow Broadcom to eventually expand into LTE mobile devices. Until then, Broadcom can draw on Beceem’s success in WiMAX USB dongles, PC data cards and handsets to help fund development of LTE chipsets.

It seems likely that merchant baseband suppliers Qualcomm, ST-Ericsson, Altair Semi, Icera, Renesas and Intel / Infineon will beat Broadcom to market in LTE, particularly in the rapidly growing “non-handset” (our term) segment of the mobile market consisting of USB dongles, PC data cards, M2M modules and related data-centric wireless modem devices. By our estimates, basebands for the non-handset mobile market accounted for about 12 percent of the total baseband market in revenue terms in 1H 2010. Currently Qualcomm, Icera, Infineon and ST-Ericsson have the leading market share in basebands for non-handsets, a market today dominated by 3G. An increasing proportion of future non-handset WWAN data modems will support LTE plus 3G, making LTE capability essential for firms to continue supplying basebands into this important segment. Note that compared to LTE + 3G, it appears that LTE + WiMAX mobile devices will at best represent a small niche market.

When Broadcom enters the LTE baseband market, we expect the company to offer chipsets with comprehensive support for legacy GSM air interfaces through W-CDMA / HSPA+, multi-mode / multi-band RF transceivers, plus optional Wi-Fi / Bluetooth / GPS, and optional ancillary chips for mobile TV and advanced video processing. Broadcom had only about 3.3 percent share of the cellular baseband market in unit terms in 1H 2010, but the company's market share in basebands is growing rapidly with design wins at top-tier OEMs.

Broadcom's acquisition of Beceem leaves other baseband start-ups Icera, Altair Semiconductor, Sequans and Wavesat as potential acquisition targets for established baseband / chipset vendors looking to expand into LTE. Of these start-ups, Altair Semiconductor, which sampled its first LTE baseband in September ’09, appears to be in the lead with LTE basebands now reaching early production status. Another firm to watch in LTE is Intel, which recently acquired WiMAX / LTE chipset start-up Comsys and announced its intention to acquire the cellular chipset product lines of Infineon. Intel Capital has a 20 percent stake in Beceem, so we can only speculate that Intel decided that acquiring Comsys and Infineon would offer better synergies and returns on investment given that Intel already ships WiMAX chipsets into the embedded laptop market.

Aside from LTE chipsets, Broadcom also has to solve is its smartphone applications processor challenges; by our estimates, Broadcom had no share in smartphone applications processors in 1H 2010. Broadcom scored design-wins in Nokia's Symbian^3 smartphones recently with its BCM2727 co-processor, but we believe that Broadcom has to integrate BCM2727 functionality into its basebands to score significant smartphone design-wins in the future.

Last week we published our cellular baseband market share estimates for 1H 2010. By our estimates, cellular baseband unit shipments grew 23 percent and revenues grew 15.5 percent from 1H 2009 to 1H 2010. The non-handset segment, which includes PC data cards, PNDs, netbooks, tablets wireless USB dongles and M2M, accounted for 12 percent of total baseband revenues in 1H 2010 and grew faster than the handset baseband market.

Here are some highlights of our cellular baseband market share estimates:

• Qualcomm continued its revenue share lead with 39.6 percent share in 1H 2010. Qualcomm ranked number one in CDMA and W-CDMA baseband markets. We believe much of the Qualcomm's growth came from smartphones and non-handsets in 1H 2010.

• MediaTek ranked number two in revenue share behind Qualcomm with 16.3 percent share in 1H 2010. Contrary to popular perception, MediaTek currently doesn't compete head-to-head with Qualcomm and only recently launched its first W-CDMA baseband. By our estimates MediaTek had less than 1 percent W-CDMA share in 1H 2010 but the company ranked number one in GSM/GPRS/EDGE (GGE) and TD-SCDMA baseband markets in revenue terms. MediaTek’s baseband shipments to grey handset OEMs showed signs of decline in 1H 2010 in the wake of government-led crackdowns in India and China.

• ST-Ericsson finally showed signs of recovery in 1H 2010 after a difficult 2009 with its improved and rationalized product line up for GGE, TD-SCDMA, W-CDMA and LTE markets. However, the company has to shift its focus to its stagnant W-CDMA market share to compete against Qualcomm and Infineon.

• Infineon continued to gain share in GGE and W-CDMA markets with the help of its top-tier handset OEM relationships. By our estimates Infineon is emerging as a strong competitor to Qualcomm in the non-handset market also with 9 percent unit shipment share in the non-handset segment of the W-CDMA market. Infineon is expected to improve its non-handset baseband share further once Intel completes its acquisition in 2011. Texas Instruments' exit from the baseband market creating opportunities for Infineon and Broadcom at Nokia.

• Broadcom's baseband revenue grew 245 percent in 1H 2010 compared to same period the year before thanks to its design-wins at Nokia and Samsung. By our estimates W-CDMA revenues accounted for 17 percent of Broadcom's total baseband revenues in 1H 2010 and the company continued to improve its product-mix.

• Spreadtrum, Marvell, Icera, and Via Telecom also made progress in 1H 2010. Renesas' baseband revenues were down slightly but the company is expected to emerge as strong player by 2012 with LTE chipset design-wins at Nokia.

Looking forward, we believe that the non-handset market and new radio technologies (HSPA+, LTE and TD-SCDMA) to create growth opportunities for baseband vendors.

Hear insights from Stuart Robinson, Director of Strategy Analytics’ SpecTRAX service, who will talk about the important trends in recent global cellular handset model announcements. Join us on Thursday 9th September for a complimentary live online webinar.

Topics covered by the webinar include:

• OS penetration; how many phones feature Android compared to LiMo, Windows and Symbian?
• Analysis of current spec trends; Bluetooth, touchscreen, QWERTY, form factor, screen & camera resolution….
• Shelf share analysis by region; which carriers offer the most Android phones?
• Typical specs of Android phone vs handsets running other operating systems

The webinar is targeted at existing clients of the SpecTRAX service as well as product planners, product marketers and portfolio managers across the industry.

Attendance to the webinar is free however pre-registration is required. REGISTER –>>>           

Date:                 Thursday 9th September 2010           

Duration:           30 minutes, followed by 15 minutes Q&A           

Time:                  2pm BST UK / 9am EDT Boston, repeated at 5pm PDT LA

Once we have processed your registration, you will receive a confirmation email message with instructions on how to join the event. Feel free to forward this message to your colleagues who may be interested in this webinar.

Nokia announced today that Renesas will acquire its wireless modem technologies business for $200 million and the transaction is expected to close before the end of 2010. Renesas has offered GSM PAs and transceivers for the mainstream GSM market for more than a decade but lacked basebands. More recently, Renesas entered the baseband market with W-CDMA basebands for Japan and plans to provide complete chipsets and PAs for the global market. Renesas merged with NEC Electronics in April 2010 and most recently joined the Symbian Foundation.

Previously Nokia transferred its 3G IC design operations to ST-Ericsson, licensed its W-CDMA/HSPA modem technology to Intel and collaborated with Infineon for LTE RF transceivers development. This announcement raises a question whether Nokia will continue to own the IP and earn royalties for legacy basebands, GSM through W-CDMA / HSPA+?

This is a significant development considering Renesas' lack of baseband design-wins outside Japan. Strategy Analytics estimates that Renesas and NEC Electronics cumulatively held just 1.3 percent of the baseband market (in revenue terms) in 2009. Renesas originally licensed baseband technology from Japanese network operator NTT docomo to produce the SH-Mobile G series baseband processors, which also integrate Renesas' application processing technology. Most recently Renesas sampled a fourth generation of SH-Mobile G baseband processors, the SH-G4, which supports the HSPA air interface.

This partnership with Nokia provides Renesas with three different sources for fundamental 3G modem IP, NTT docomo, NEC Electronics and Nokia. This situation is similar to ST-Ericsson's, which also has three different 3G modem IP sources, EMP, NXP and Nokia. It appears that Renesas will use Nokia's 3G modem technology to compete for design-wins outside Japan. The combined entity of Renesas and NEC Electronics now boasts Fujitsu, Sharp, NEC, Sony Ericsson and Panasonic as its baseband customers . Renesas' expansion into the global baseband market would increase competition in the marketplace.

This move comes at a time when the 3G market is growing, accounting for close to one third of global cellular baseband revenues in 2009. Nokia's existing 3G chipset suppliers include Texas Instruments, Qualcomm, Broadcom, ST-Ericsson and Icera. It is worth noting that the first Nokia 3G phones based on Qualcomm, Broadcom and ST-Ericsson's basebands are expected to debut in late 2010 or early 2011. Infineon is noticeably absent in this list although Infineon is a key GSM/GPRS/EDGE baseband supplier to Nokia.

We don't expect the first products based on this partnership to debut until late 2012, the time when Texas Instruments completes its baseband exit. In 2013 Nokia's 3G chipset suppliers list would include Qualcomm, ST-Ericsson, Broadcom, Icera, Renesas and potentially Infineon and Intel. While Nokia can afford to have multiple suppliers considering its scale we could see some further acquisitions and mergers among these baseband suppliers.

Today’s announcement is likely to affect several companies...

• ST-Ericsson: ST-E will probably be worst hit by this announcement as it brings into doubt the strong relationship ST-E has had with Nokia over the past 3 years. ST-E has spent the last couple of years consolidating the 3G technologies it acquired from EMP and NXP with Nokia's 3G IP into its flagship processor the U8500 which is expected to appear in handsets at the end of 2010. Nokia will obviously be working with Renesas in future HSPA/LTE projects, thus potentially diluting the relationship with ST-E.

• Qualcomm: Qualcomm has enjoyed a much more positive relationship with Nokia during 2009/2010 since the companies resolved their long-standing IPR legal battles. Qualcomm is expected to supply chipsets to Nokia for its upcoming W-CDMA/LTE products. Nokia's stronger relationship with Renesas may now affect that.

• Infineon: Why did Nokia not choose Infineon? Nokia and Infineon already have a close relationship around LTE RF transceivers, but Nokia may have considered Infineon's expertise in HSPA and LTE modems to be too weak.

• NTT docomo: Renesas currently licenses 3G modem IP from NTT docomo for use in its SH-Mobile G processors which are desgined into handsets from Sharp and Fujitsu. Renesas' strong relationship with Nokia now makes it more likely that Renesas will use NTT docomo's IP for products aimed at the Japanese market, and will employ Nokia's HSPA/LTE IP in new chipsets for us in Nokia handsets. It will also allow Renesas to expand its client base beyond Japanese waters.

The number of mobile operating systems just keeps getting bigger and to our count close to ten platforms are now competing in this space. They include Symbian, Android, iPhone OS, BlackBerry OS, WebOS, MeeGo, bada, Windows Mobile, LiMo Foundation-compliant platforms, Brew Mobile Platform and Linux variants. The question that obviously comes to everyone's mind is who will win in this mobile software platform market.

There are some interesting angles to think about this.

• Can the smartphone market size accommodate this many platforms?

• Do consumers care about operating systems?

• Is it a zero-sum game or can there be a more than one winner?

• Everyone is talking about carriers becoming dumb pipes but what about handset OEMs? Will Google and Microsoft make handset OEMs as dumb pipes or add value?

• Should every handset vendor build their own OS? Can they cope with incremental mobile OS R&D costs?

• Can horizontal and vertical business models co-exist in the mobile industry?

• Is there enough developer support to address every OS?

• How many platforms operators can support?

We think R&D capabilities, scale, brand, and strong partner and developer ecosystem are necessary ingredients to sustain a long-term smartphone OS and only few of the above mentioned platforms check most of the boxes. Strategy Analytics estimates that smartphone OS vendors currently spend close to $100-$200 million on R&D on average and this is expected to grow incrementally.

Nokia (Symbian and MeeGo): 2010 will be remembered as a software transition year for Nokia and 2011 will see a settled software platform strategy with both Symbian and MeeGo (Maemo) getting much waited Qt UI and app frameworks. Nokia has a unique software strategy and is the only vertical vendor who is open sourcing its proprietary platforms. Nokia already has scale, brand, and R&D capabilities and if the company can excite the developer community and consumers with Qt then it should keep its smartphone lead for a long time to come. Nokia is well positioned to democratize the iPhone and BlackBerry experience to the masses. Nokia's software transition will be watched closely as competitors becoming stronger.

Microsoft: We believe Microsoft can not afford to lose in the mobile platform market given its synergistic importance to its other businesses and Windows Phone 7 Series (WP7 Series) announcement just reiterates that. It remains to be seen how WP7 Series platform will fare in terms of price points, hardware requirements, OEM support, developer and consumer reaction which can decide whether Microsoft can get the scale that is required to sustain a long-term smartphone OS. Microsoft will face a strong challenge from Google as both companies are targeting the same OEMs. Microsoft may position older Windows Mobile versions for low-end smartphone market but it remains to be seen how OEMs will react given more appealing nature of WP7 Series platform.

Android: Google and its partners executed very well so far and 2010 could be the Year of Android. Unlike Microsoft's WP7 Series platform, Android can also run on low-end chipsets which can increase its addressable market. Android and Symbian are the only mobile OSs that are targeting mass-market smartphone segment currently. This gives edge to Android to gain scale as handset OEMs push Android to feature phones. However some key risks like fragmentation and future Google moves that may cause friction to OEMs and operators will be watched closely. We also believe that the expanding size of the smartphone market should create enough opportunity for horizontal software platform players as Google and Microsoft currently lack high volume handset customers.

BlackBerry OS: RIM continued to gain with its market leading "Push" technology and security features. However, its web browsing and touch screen experience needs to be addressed. RIM already trying to fix the web browsing problem with new WebKit-based browser but touchscreens remain a challenge.

iPhone OS: Apple continues to win mindshare from developers and consumers alike and is well positioned to be a long-term player in the smartphone OS market. Apple's web browser is equally capable even if the trend shifts to web applications in the future. Apple may come under pressure from Android and resurgent Microsoft in the future but the growing size of the smartphone market should give enough room for iPhone OS to grow.

LiMo Foundation: Being a middleware platform and because of its governance structure, the LiMo Foundation currently lacks a direct relation with the developer community and currently depends on its operator partners for ecosystem. 52 LiMo-compliant handsets have been announced so far and 24 major global carriers are members of the Foundation. We believe that the LiMo Foundation's middleware platform is technically closer to Nokia's MeeGo and it remains to be seen how the Foundation will pursue its strategy given global OEMs are currently choosing Android.

Palm's WebOS: Despite seeming to have one of the best mobile platforms, Palm's limited geographical reach, and limited R&D budget could prevent the company from scaling as much as its competitors Nokia and Apple. Palm's lack of brand-awareness outside North America is not helping the matters. We believe Palm's lack of scale could hurt it in the future as its competitors become even stronger.

Samsung bada: Samsung bada's definition and market positioning remains still subjective given its kernel-configurable nature. It remains to be seen how Samsung's bada will keep pace with Android, iPhone OS, Symbian and others. Like Android and Symbian Samsung's bada also expected to target mass-market smartphones.

Qualcomm BREW Mobile Platform (BMP): We believe Qualcomm is positioning BREW as a low-end smartphone platform but it will face a stiff challenge from Android and Symbian as they are also trying to address the feature phone market. Qualcomm's BMP is also not supported by other chipset vendors but Qualcomm is expected to leverage its existing relations with operators and handset OEMs.

Despite the desire of OEMs to control their platforms, we continue to believe that the growing R&D costs to develop a smartphone OS and ecosystem around it could potentially force them to consolidate their efforts around licensable operating systems while also building OS-agnostic differentiating layers on top of that. Examples include Nokia (Qt / OVI), Motorola (Moto Blur), Sony Ericsson (UX), HTC (Sense UI), Samsung (TouchWiz) and LG (S-Class).

I was asked recently how many phones were HD-capable, and realised that the question is actually quite complex.

I should start with a definition for what I believe constitutes HD, as many handsets already claim some sort of High Defintion credentials, but very few actually deliver it...

- True HD, 1080p, is a 16:9 ratio screen with a resolution of 1920x1080 pixels and a frame rate of 60 frames per second (or 24 fps for Blu-ray players);

- 1080i is the same resolution, but the image is interlaced, effectively halving the frame rate;

- 720p is also 16:9, but with a resolution of 1280x720 pixels.

The highest resolution display on a handset at present is WVGA (800x480 pixels) and there are just under 200 handsets with that spec, the biggest being the HTC HD2 which has a very nice 4.3 inch screen, but WVGA is nowhere near big enough to be classed as HD.My take on HD is that we will see it on phones in certain functions: a)       Capture? Yes: The Samsung i8910 HD (also known (incorrectly) as the Omnia HD) was the first phone capable of capturing 720p HD video and the quality is immense. I think we will see many more phones in the near future being capable of capturing 720p or even 1080p HD;

b)      Playback on screen? No: The Samsung i8910 HD has a very impressive 3.7 inch, 640x360 pixel OLED screen which gives a fantastic representation of high quality (if not HD) video, but it’s not true HD playback on the screen. In fact it’s only nHD (ninth full 1080 HD) at 640x360 pixels. Converting the Samsung i8910’s specs into dots per inch gives a figure of 198dpi.  I would argue that the highest definition screen we will ever see on a phone is QHD (960 x 540 pixels) or at a real push 720HD (1280x720 pixels) which, on a 4-inch display, would result in 275dpi or 367dpi respectively. Putting a 1080p screen on a phone is a complete waste of time (and money) because the human eye cannot distinguish anything greater than about 300dpi;

c)       Playback off screen? Yes: We may see more usage of playing content from the phone onto a larger screen using the TV Out or mini HDMI capabilities of the phone, or maybe even using a built-in pico projector. Nokia has been a strong supporter of TV Out and we will certainly see many more phones in future with a mini HDMI port. As for pico projector phones, I’m not sure that they are going to appeal to the network operators because power consumption will be so poor that the phone will be dead after an hour or two, then there’s no revenue stream for the operators;

d)      Storage: This is where I think there is potential for HD. Using Moores Law, I estimate that in 10 years time the average phone will have 50GB internal memory and some high end phones will have around 1TB of NAND Flash memory, and probably for about the same price as about 32GB of NAND Flash today. What we will do with that memory is another matter. With 1TB you could store about 375 hours of full HD (about 180 movies) so perhaps we will all have our collection of HD movies on our phones ready for viewing when we’re on a business trip for example;

Battery life will certainly be affected by HD though, so we will probably see much bigger batteries (maybe 2000mAh) in handsets with HD capture or playback.

Finally, it's worth noting that several chipset companies have chips in, or close to, production capable of supporting HD capture including TI OMAP 3440 (12MP, 720p); NVIDIA Tegra APX (12MP, 720p); Broadcom 2727 (12MP, 720p); Qualcomm MSM8260 (16MP, 1080p); TI OMAP4 (20MP, 1080p), so it looks like HD capture and storage on phones will be a booming segment.

Stuart Robinson

At the London “official” launch of Samsung’s new mobile operating system, bada (which means ocean in Korean), many questions remain unanswered regarding what bada is all about.

bada’s goal is “smartphones for everyone”, everywhere in the world, starting with 50 countries when the first handset is launched sometime during the first half of 2010, but extending to global coverage eventually.

Samsung’s first handset based on the bada platform will also be the first phone to run Samsung’s TouchWiz 3.0 user interface. The company has a desire to deliver a rich user experience to everyone. Although Samsung declined to comment on its strategy for other OSs including Symbian, Android, LiMo and Windows Mobile, we believe that bada has come about as a result of Samsung’s frustration with the user interface capabilities (or lack of) of Symbian over the last couple of years, forcing the handset OEM to develop its own solution.

Samsung declined to estimate the number of shipments it expects to achieve with bada during 2010, but said it is currently working with mobile network operators to estimate the likely demand for its first handset.

Full details of the hardware requirements for bada were not discussed during the announcement, but Thomas Richter, Director of Portfolio Management at Samsung, confirmed that the platform will support 3D graphics, multi-touch capacitive touchscreens, Flash, C++, HTML5, multiple sensors and many other hardware and software functions. Application developers will have access to some of the core phone functions such as the proximity sensor, weather sensor and call functions. Richter added that there will be an extensive developer site and developer support centres around the world. In the same way that Google drummed up support for its Android platform, Samsung is launching a developer competition for bada, with total prize money of US$2.7 million and a first prize of $300K for the winner of the best application. Developer days will be held in Seoul, London and San Francisco to begin with during 2010.

Samsung said it expects to ship a total of 40 million touchscreen phones during 2009 and “many more” during 2010. Estimates from other sources suggest that around 50% of Samsung’s portfolio in 2010 will consist of touchscreen phones and that the company may ship over 100 million touchscreen handsets during 2010.

Stuart Robinson

Follow me on Twitter: @Stuart_Robinson

Strategy Analytics feels Marvell’s lack of exposure to high volume handset makers, lack of a visible long-term baseband product roadmap and missing technology pieces like RF transceivers will force the company to reconsider its position as a cellular baseband processor supplier. Marvell is a prominent player in application processor and connectivity market but a minor player in the cellular baseband market. Since acquiring Intel’s cellular business in 2006, Marvell has made limited progress in the cellular baseband processor market where as competitors Infineon, Qualcomm, and ST-Ericsson built scale during the same period. The company's total cellular revenues (including basebands, application processors, connectivity) are still less than one billion dollar.

Research in Motion (RIM) is Marvell’s biggest cellular baseband customer but Marvell is expected to lose to share to Qualcomm in the long-term. Qualcomm already supplies basebands to over 25 percent of the RIM’s handsets (mostly CDMA). Freescale also recently scored some baseband design-wins at RIM at the expense of Marvell although Freescale recently gave up its baseband efforts. Marvell’s other cellular baseband customers include Palm, Asustek, and some Asian ODMs. Most recently the company scored Ophone TD-SCDMA Android smartphone design-wins at China Mobile and we believe that most of the TD-SCDMA baseband IP belongs to China Mobile and affiliates. Marvell's competitors MediaTek, ST-Ericsson are also investing in TD-SCDMA and most recently Qualcomm also expressed its intention to compete for TD-SCDMA design-wins in China next year. We don't see how a niche market strategy can be successful for this R&D intensive technology.

The company's lack of RF transceiver products is likely to have an impact on its penetration in entry level and mid market handset segments where transceiver-baseband approach is appropriate. The lack of visible LTE roadmap could put Marvell behind its competitors Qualcomm, ST-Ericsson and Infineon. We believe that growing R&D costs to develop advanced communication IP could put pressure on Marvell to reconsider its long-term position as a baseband vendor.

We feel that Marvell could perhaps partner with companies like MediaTek or Infineon who lack high-end application processor technology. Marvell's application processors are already popular in the Chinese smartphone market and could be a good fit for MediaTek. However it is not clear whether these companies are interested in Marvell's XScale technology.

Sravan Kundojjala

Client reading: Baseband Processor Profile: Time for Marvell to Reconsider Its Baseband Position

Qualcomm, Gold sponsor of the 2009 Symbian Exchange and Exposition, today announced a new subsidiary called QuIC (Qualcomm Innovation Center, Inc) to focus on optimizing open source software such as Android, Symbian and Chrome to its wireless chipsets. Qualcomm settled its legal disputes with Nokia last year and earlier this year Nokia awarded a 3G / 3.5G design-win to Qualcomm. Qualcomm agreed to port Symbian operating system to its MSM7000 and 8000 series chips and Nokia's handsets based on this partnership are expected to debut in 2010.

We believe that Qualcomm will play an important role in Nokia's high-end smartphone portfolio in the coming years as Nokia continues with its multi-vendor strategy for baseband suppliers.  Nokia's previous custom ASIC approach, primarily using Texas Instruments (TI), is gradually reducing and Broadcom, ST-Ericsson and Qualcomm will ramp up at Nokia to compensate for this. We expect Nokia to use Qualcomm and ST-Ericsson in the high-end portfolio while Infineon and Broadcom are expected to play a role in the low-end to mid-range portfolio. It is worth noting that with the exception of Infineon, the remaining three are all fabless semiconductor companies. This is in line with Nokia's strategy to adopt standard off-the-shelf ASSP solutions instead of custom ASICs. Earlier Texas Instruments didn't have a 3G baseband design of its own and acted like a foundry for Nokia's in-house designs.

Currently Qualcomm has a relatively weak position in the smartphone market primarily because it did not support the Symbian OS. However, the company's recent settlement with Nokia, its SnapDragon processors and multiple Android and Windows Mobile design-wins could potentially make it competitive in the smartphone market. We believe that initiatives such as QuIC could contribute positive results in improving Qualcomm's smartphone position.

Sravan Kundojjala