What’s A Cell Phone?

By Ed Sperling
Just because a smart phone is sold by Verizon or AT&T mobile no longer means that it will be used primarily as a phone.

That distinction may sound trivial, but it has deep implications for the components that are used inside of these devices, how they’re used, and who wins the designs. Shifts such as this can also lead to broad changes in who buys the tools to develop the components, which tools they buy, and what sorts of flows they create with those tools.

There are several fundamental reasons why this shift is occurring, and all of them intersect and support the others.

Generation, geography and culture
First, there is a huge generational and geographical gap between what’s important in phones. For older users, voice conversations are the most important feature. For younger users, texting and games are key. And for business people on the go, the most useful features are a combination of voice and e-mail.

“This trend began in China, where a phone is not considered a voice device,” said Charlie Cheng, CEO of Kilopass. “It’s very textual and graphical. You use it for text, Facebook and browsing. Young people all use it that way, too. My kids think it’s a novelty when I call them on the phone.”

The tablet has blurred the lines even further. While most of the comparisons have been between tablets and personal computers, the real volume market overlap will be between smart phones and tablets. Both are capable of texting, videoconferencing and e-mail, and each can go places and do things that the other cannot. A tablet has huge possibilities in the business world and in places such as hospitals, where a touchscreen is preferable to a keyboard because it can be wiped clean. It’s also better for making presentations. But while it fits in a briefcase, it doesn’t fit in a pocket—something that may change as flexible screens begin production.

“A phone is no longer a phone,” said Vishal Kapoor, vice president of product management for SoC realization at Cadence. “The three most important issues are security, management of data—including how much of that is local information—and the video or graphics. Even bandwidth is no longer a problem technologically, although not all of the phones can take advantage of 4G yet.”

Power and performance
The second major change is in performance and power. While the two typically are tradeoffs on the same SoC, they’re not necessarily tradeoffs in the same package.

For the past couple of generations, smart phones have been able to hold their own as full-fledged number-crunching and computing devices. They can be used to surf the Web, do e-mail, download documents and photos, and even update those documents. While the form factor is limiting, the tradeoff in portability may suffice for executives or salespeople on the road.

But the real opportunity is less in conventional desktop or notebook computing than a raft of new applications. Apple reportedly is working with Visa, for example, on “swipe and go” technology, where a smart phone is used as a checkout device that can replace credit cards using near-field communication technology. Phones already are being used as boarding passes on many airlines, particularly in Europe.

These features are a sign of just how far performance has increased on these devices. Apple, MIPS, Freescale, ARM and Synopsys (through its ARC acquisition) all have developed very powerful multicore processors that draw very low power when used in conjunction with such approaches as power islands and power gating.

“What’s changing is that you’re starting to put a lot of personal information into these devices,” said Cheng. “There’s a lot of money involved in this and there’s a lot at stake.”

The Android effect
A third factor that is contributing to this shift is Android. The operating system developed by Google is spawning a revolution in how devices such as smart phones are used—and who wins the designs.

MIPS, which was one of the first adherents of the Android platform, is experiencing huge growth—much of it because of Android. The company’s revenues grew 44% in Q2 of 2010 vs.Q2 of 2009.

“The playing field is wide open,” said Art Swift, vice president of marketing and business development at MIPS. “What’s changed is those companies that were the leaders in the past in mobile will not necessarily be the leaders in the future. That’s especially true with tablets. Part of the market driver here was Android, and it’s wide open. There’s a whole cast of new players.”

Not all of it is happening in the usual places, either. John Koeter, vice president of marketing for IP and systems at Synopsys, says Android is opening up other markets that didn’t exist in the past.

“We’re not going to play in the shootout between MIPS and ARM,” said Koeter. “But we are seeing new markets for things like picture frames that can be interconnected and run applications. There are all sorts of new and interesting applications.”

Samsung's new Android-based Galaxy tablet.

Design challenges and opportunities
The challenge for chip companies working in these transitional technology markets is figuring out where the volume adoption will be, how to best utilize the technology to serve multiple markets, and how to add in enough flexibility so that certain features can be given priority where necessary.

In some cases this may utilize a system-in-package approach with an interposer technology or a network-on-chip architecture for improving signal traffic flow. In some cases it may be multiple cores within a complex SoC that serve the same purpose. And in the future, it will likely be multiple chips on a 3D stack, where different functions can be developed and then manufactured as needed for different markets.

What will have to change, however, will be the pace of tool adoption and development. One of the big complaints among system-level tools vendors is that not everything can be integrated into a flow because high-level tools don’t necessarily work perfectly with older tools that utilize lower levels of abstraction. That has stymied the growth of high-level synthesis and software prototyping, for example.

But these kind of changes may bring new players into the market, raising the competitive stakes to develop chips more quickly, more efficiently and with more flexibility. That means newer tools will be required, and it can quickly force a competitive upgrade among existing companies and spur growth in areas that have been slow to develop, particularly in the ESL space.

Conclusions
So what will be the most important features on a phone in the future? That will depend to a large extent on the applications and what’s important to users and companies that buy these devices. A phone will still have to be able to make phone calls, but that may be just a lesser feature on these increasingly complex devices.

“A phone will have to be reliable and clear,” said Koeter. “But once it meets that standard, then it’s all about a whole new experience.”