Replies to post #17386 on Qualcomm Inc (QCOM)

[stricklybiz]

sbono, I'd venture a guess that--regardless of the average price of handset offered--poor people who buy cheap phones don't pay even $38, much less $38 more.

[Eric]

The Single-Chip in Handsets for BRIC

<< But does payoff explain why 3 out of 4 Indian customers choose GSM over CDMA, despite the fact that GSM handsets cost $38 more on average and GSM service costs about 2X as much on a per minute basis (although ARPU is not appreciably higher)? For some reason or another, customers see value in GSM, and I can't imagine that someone is bribing them one by one... >>

Interestingly, people who fall into the demographic for low-cost platforms don't always head straight to the bargain bin. Despite being a little more expensive, Nokia phones are still quite popular in India and have picked up steam in China. This isn't surprising to people who have experience in developing markets, like Ian Drew, a vice president of segment marketing for ARM Ltd., which is introducing its smaller, more efficient Cortex-M3 processor to replace the ARM7 in low-cost phones. ... "People often save up [for] one or two or three months to go and buy a phone, and they don't want the cheapest, no-brand one. They want to have something they can use and continue using," Drew said. "So there will be demand for the ultralow-cost handset, but I think there will be equally if not bigger demand for the one above it."

>> Low-Cost Handsets Drive Single-Chip Designs

The developing world is the target for slimmed-down phones, but designers must balance integration and cost with utility.

Mike Clendenin
EE Times
June 12, 2006

http://www.eetimes.com/news/design/showArticle.jhtml?articleID=188702141

A year ago, the notion of an ultralow-cost handset for the masses was largely conceptual. There was talk of using single-chip designs to weed out cost, and slimmed-down software stacks to strip mobile phones to the bare necessities--voice and maybe SMS, just enough to satisfy the basic needs and budgets of the next billion users. Now it's starting to happen. In developing economies like Brazil, Russia, India and China--the so-called BRIC countries--phones that sell for less than $50 are popping up on store shelves. By the end of the year, the single-chip platforms that were much talked about will start to appear in those phones, driving the cost down even more.

The goal is to hit $25 or less for materials, and some more-aggressive predictions say $10 or less is possible. The lower the better for people like Lourd Augustine, a 29-year-old Bangalore, India, resident who uses a Nokia 1100 that cost $49. The phone is voice-centric, with a black-and-white screen and Short Message Service capability. Yet even this no-frills model cost Augustine a fortnight's wages, a sacrifice he was willing to make in order to work as a driver for a local family.

"These days people expect a chauffeur to already have a mobile phone if he is to have a better chance of landing a job," Augustine said.

The market for ultralow-cost handsets is expected to grow quickly over the next three to four years, as more than two-thirds of new subscribers come from developing countries. The GSM Association expects several hundred million ultralow-cost handsets will ship by 2010. Market watcher iSuppli Corp. estimates that ultralow-cost handsets, lumped in with entry-level devices, will account for 35 percent of the market in 2006, or about 308 million units. By value, the segment will generate $14.6 billion, or about 13 percent of the overall revenue.

However, selling phones to the masses will get increasingly tougher. By 2010, as ultralow-cost handsets expand their share of the low end to 9 percent of units shipped, revenue will tail off quickly to $5.4 billion, or about 4.5 percent of the market. Naturally, chip prices will drop, too, as handset prices decline: iSuppli pegged the average selling price of an ultralow-cost handset at $50 in 2005. As single-chip solutions replace discrete chip sets, that will drop to $42 this year and $32 by 2010.

"The people who are chasing that [single-chip market] are probably the ones who are going to be more involved in the bloodbath, and driving the prices down to the point where it is questionable whether they will make any money at it," said Doug Grant, director of business development for RF and wireless systems at Analog Devices Inc.

There's disagreement over the right silicon approach for the market--a system-on-chip (SoC), including the RF, vs. a system-in-package or even a tightly integrated module or discrete silicon. ADI's Mercury reference design is a multichip module. Philips also opted for a module approach in its Nexperia 5130, released late last year, but said it will ultimately move to a single-chip Edge solution.

Others are moving straight into SoCs. Infineon Technologies is already on its second-generation SoC, called the E-GOLDvoice (ULC2), which uses fewer than 50 other electronic components. Unlike its predecessor, the chip integrates power management and SRAM, and by using 130-nanometer process technology is able to pack it all into just 4 cm2. Samples are expected in July.

RF CMOS expert Silicon Labs recently developed baseband technology so that it could offer an SoC, the Si4905, which includes power management and uses TTPCom's AJAR protocol stack for ultralow-cost handsets. The company claims a footprint of 3 cm2.

Texas Instruments Inc. will be a strong contender, too, and is working with Nokia to get its LoCostco SoC, which uses separate power management, into the market in the second half. And even though the market for ultralow-cost handsets is considered GSM turf, CDMA leader Qualcomm Inc. is making strong moves with its single-chip QSC6010.

"You have to basically have a [single-chip] solution this year to be able to compete in this market. If you haven't shown some sort of RF capability that can be integrated with the baseband, like an RF CMOS transceiver, then you're probably not going to get there in time," said Scott Smyser, an analyst with iSuppli who issued a report on the low-cost handset market earlier this year.

The GSM Association's target is to give 80 percent of the world's population access to mobile communications by 2010, even if every person doesn't actually own a handset. To accomplish this, the role of network operators will be an important factor, although probably secondary to the "cash barrier" presented by the handset, said Ben Soppit, director of strategic initiatives for emerging markets at the GSMA.

"Cost of service is really about usage, not penetration," he said. "The lower the service cost, research has shown, the more people use it. So we have two different things--cash barrier being the primary determinant of penetration and service cost determining usage."

In India, one of the fastest-growing markets for cellular services outside of Africa, Bangalore driver Augustine pays $10 a month for access. A pittance to some, or one-tenth of a monthly salary when viewed through his eyes. The concern over cellular bills in developing markets has led to some interesting ways for people to use the network without paying for it. In India, "missed calls" are used as a signal that someone has arrived at a meeting place or returned home safely. In China, where both callers and receivers have to pay for calls, a similar tactic is used, in addition to a preference for cheaper SMS services.

Various operators are testing out ways to lower service costs and expand coverage. India's telcos are experimenting with shared networks to spread the cost of infrastructure in rural areas. In Africa, lower-cost prepaid cards are used. In other markets, micro credits are issued wirelessly, or middlemen disburse small blocks of minutes from prepaid cards.

While all of these methods have the goal of pairing ultralow-cost services to ultralow-cost handsets, it is the latter that will always be the biggest barrier. Fortunately, progress is being made quickly. Already, handset prices in India are dropping fast. Motorola's C118, which won a tender for ultralow-cost handsets from the GSMA, is $45. And CDMA handsets from LG are proving competitive, too. Telecom operator Reliance Infocomm offers one with a service package for $36, while another service provider, Tata, offers an "Ace" brand CDMA phone with service for $33.

"We have five OEMs working on platforms for the single-chip solution, so you can expect those later this year or early next year," said Terry Yen, senior director of marketing for Qualcomm CDMA Technologies.

Interestingly, people who fall into the demographic for low-cost platforms don't always head straight to the bargain bin. Despite being a little more expensive, Nokia phones are still quite popular in India and have picked up steam in China. This isn't surprising to people who have experience in developing markets, like Ian Drew, a vice president of segment marketing for ARM Ltd., which is introducing its smaller, more efficient Cortex-M3 processor to replace the ARM7 in low-cost phones. Drew spent several years working in or covering developing markets like China, Vietnam and India. In the latter, he worked on the Simputer, one of the earliest low-cost, mobile computing platforms.

"People often save up [for] one or two or three months to go and buy a phone, and they don't want the cheapest, no-brand one. They want to have something they can use and continue using," Drew said. "So there will be demand for the ultralow-cost handset, but I think there will be equally if not bigger demand for the one above it."

Already, chip makers are looking to add on features to their low-cost platforms, blurring the line between ultralow cost and entry level. "The more you penetrate emerging markets, the more the population will get aware of new features and want to have more than just the bare phone," said Renzo Pellandini, marketing director of the entry-level/midend phone segment at Philips. He calls it "ultralow cost plus one."

Initially, that "plus one" feature will likely be an FM or AM radio, but it will eventually migrate to MP3 playback or Bluetooth. Motorola, which will ship 18 million ultralow-cost handsets this year, has been working on more advanced mass-market handsets in anticipation of users demanding features, as they do in mainstream markets.

Also similar to mainstream markets, the company is looking for any special features that will help set it apart, such as batteries that provide two weeks of standby time--important for many consumers in emerging economies that don't have ready access to electricity, said David Taylor, director of strategic operations in Motorola's division for high-growth markets. "Lack of electricity also means a lack of light, so our phones have a 'lantern' facility that adds to their usefulness at night or in dark environments," he said. "And for many consumers, radio is their primary source of news, so our C168 handset incorporates a radio."

In China, where ultralow-cost handsets aren't as popular yet, local system designers are still interested in the single-chip and module solutions, said Cary Wong, manager of segment marketing at SST, which is finding a new market for its low-density 16-Mbit to 32-Mbit flash in the low-cost segment.

"They are using chip sets that were originally targeted at the ultralow-cost platform, and then they are packaging it along with other subsystems to create not just a simple ultralow-cost handset but a low-cost feature phone," he said. It's called xiao ma twui da che, or using a "small horse to pull a big cart."

It's no surprise that the next billion users will probably do what the last billion have: start with basics, and then crave more. The difference may be that they will want those extras, but will be unwilling or unable to pay substantially more for them. So even though ultralow-cost handsets are a new market, the challenge will be familiar to chip designers --
do more for less. <<

- Eric -

[budro]

Well, of course I do not REALLY know what is going down over there. I was just thinking that something underhanded almost has to be taking place, and that it could easily happen in a country like India. Do we actually know what they are charging for each customer?....the total bill....the one that may reflect such BS as line charges, surcharges, items other than actual handset prices, etc etc. Are there other incentives? Maybe they get a coupon for a free TV and it doesn't show up in any of the reports we see. I just feel that there has to be another angle not seen. Of course they would not bribe each and every potential GSM sub, and maybe there really isn't anything totally underhanded. But have we missed something? Even in this country, people tell me sometimes about some deal they got, and when I look into it, I find all these hidden incentives not mentioned in the beginning.
Just trying, as you are, to make sense of this India thing.
thanks
B