Apple

The Apple rumor mill kicked into high gear again Monday regarding the much discussed, yet still mythical Apple tablet. But a significant new piece of information emerged.

Tech blog Venture Beat is reporting that PA Semi, the chip company Apple bought last year whose specific function within Apple has thus far been unclear, will be designing the chips for an Apple tablet in house. Dean Takahashi wrote Monday, "PA Semi's team was split into two parts, one designing portable ARM-based processors for iPhones and iPods, and another designing a processor for the tablet device."

The PA Semi team has some renowned chip designers, and there have been recent Wall Street Journal reports that Steve Jobs has been personally overseeing the development of a tablet-like device, and that Apple also has a large chip design project in the works. Taken together, all three reports seem to make sense.

But the round of rumors regarding the supposed tablet from earlier in the day point to an October launch time frame for the device. There's no actual evidence for that. Apple did do a Mac notebook event in October last year, and it's conceivable it could do another one, bringing out a touch-screen tablet in time for the holidays.

But October is very soon. Reports from Apple analyst Gene Munster, beginning in May, have repeatedly pinpointed a tablet launch in early 2010.

Apple seems almost ready to bridge the gap between the iPhone and the Macbook with a new type of mobile computer.

(Credit: Tom Krazit/CNET)

Is the world finally ready for the mobile minitablet?

It's become quite clear over the last several months that Apple is ready to bridge the mobile computing gap, with plans to develop a device that fits somewhere in between the iPhone and the MacBook. A recent Wall Street Journal article proclaimed that during his medical leave, CEO Steve Jobs has been working on that midsized mobile device, bigger than an iPhone but smaller than a MacBook.

And just this week, BusinessWeek reported that Apple is developing a "media pad" that would let users watch videos on a larger screen than an iPod Touch or Amazon Kindle, but on a device that's more portable than notebooks and lacks a keyboard.

The personal computer industry has long tried to make such a device a reality, but apart from some early success for the Kindle, no one has managed to convince the public that the attempts released to date--such as the Ultra Mobile PC--are worth buying. Instead, PC companies looking for increased mobility are finding ways to shrink the notebook PC as opposed to a finding a new way to use computers.

As far back as 2000, Microsoft founder Bill Gates was evangelizing Tablet PCs, but a combination of price and uninspired software doomed that category to niche status. Intel and Microsoft then turned the hype machine to the UMPC (later rebranded MID, or Mobile Internet Device), which several years later aren't exactly flying off store shelves.

More recently, PC companies have embraced Netbooks, small inexpensive mininotebook computers that are designed for basic Web surfing and e-mail. Netbooks, however, are further depressing the PC industry's gross margins and attempt to cram a full-fledged notebook user interface into a small package, and it doesn't seem that Apple is all that crazy about this category.

But Apple has developed a few unique ideas for mobile computing over the past two years that have resonated with designers, developers and users; namely, the iPhone OS and the App Store. So, is the timing finally right for the tweener computer?

BusinessWeek reported that the iPad (name stolen from Silicon Alley Insider for its brevity) would be about the size of the Amazon Kindle, but with a screen that covers a greater portion of the surface. The Kindle is 7.5-inches long, but the screen is just 6 inches; by comparison, the iPhone sports a 3.5-inch display, while the smallest MacBook uses a 13.3-inch display.

Previous attempts at making keyboard-less devices with 7-inch or 8-inch screens--such as UMPCs and MIDs--haven't captured the public's imagination. Microsoft and Intel had high hopes for the concept in 2006, which was also known as Project Origami inside Microsoft. Samsung made perhaps the best-received UMPC, but that wasn't saying much, and interest in the category quickly faded after the launch of the iPhone.

The main issue with UMPCs was a lack of compelling software. They were designed to run Windows XP, which itself wasn't designed to run on a device with such a small screen and limited methods of input. And at launch, Windows Vista was actually a step backward in terms of its suitability for mobile devices.

Samsung's Q1 was perhaps one of the best UMPCs/MIDs, but it never amounted to much in the market.

(Credit: CNET)

Intel tried to shift MIDs to Linux to get around the resource problems of Windows Vista, but its partners have yet to gain any traction. And neither attempt was able to galvanize third-party developers into creating applications designed specifically for a mobile platform.

Apple's iPhone OS, however, was designed for a small-screen mobile environment. Installing the iPhone OS 3.0 on the iPad would allow Apple to preserve the user interface from the iPhone and iPod Touch and keep the device simple: a more complicated (and power-hungry) operating system isn't needed for a computer like this.

This would also allow Apple to take advantage of the App Store, giving the iPad thousands of applications at its disposal right from launch. One potential problem with that approach is that developers will have to rewrite their iPhone applications to adapt to the larger-size screen on this new device, said Craig Hockenberry of Iconfactory, creators of Twitterific.

Hockenberry, who is very confident that Apple has such a device in the works, doesn't think this will be a huge obstacle, but developers will have to gauge whether the extra development effort is worth their time. One thing Apple could do is set aside a separate section of the App Store for iPad-optimized applications, while finding a way to run older iPhone applications in some sort of compatibility mode.

"It wouldn't be hard for Apple to have a "Classic" environment on a tablet that provided a 320x480 window for running one or more iPhone applications," Hockenberry said in an e-mail. "It would be a smart thing for them to do: there are instantly tens of thousands of apps and users are presented with a familiar interface (something that looks a lot like Dashboard in Mac OS X.)"

The iPad could also be the first Apple product to surface with a chip designed by P.A. Semi, which Apple didn't buy on a lark. A custom chip could solve two problems for Apple--the need to keep software compatibility with the ARM-based chips used to run the iPhone while delivering more performance for HD video playback or more robust games that competitors might not be able to immediately match.

What might such a device cost? There are two schools of thought on price.

UMPCs, at around $700, were considered too expensive but because they didn't offer any value, not because of the sticker price itself. It would not be hard for Apple to argue that an iPad with an HD screen, thousands of applications, and a superior mobile browser is worth just slightly more than what people were willing to pay for the original iPhone.

A $699 iPad would slot nicely between the iPod Touch and MacBook in Apple's product lineup and preserve Apple's profit margins, while allowing the company to reduce the price over time if needed similar to the original iPhone.

But Apple could also hook up with a wireless carrier--we'll save the AT&T versus Verizon debate for another day--to subsidize the iPad. The company has reportedly been in talks about distributing MacBooks through wireless carrier friends like AT&T, which already sells 3G-equipped Netbooks with a data plan subscription.

A device such as this would be infinitely more attractive with wide-area wireless networking, as opposed to just Wi-Fi--especially if carrier subsidies bring the price down to around $499, just above the largest iPod Touch. It's hard to see something this big replacing a mobile phone--you're not going to hold one of these up to the side of your head--but there are certainly plenty of headsets available in the world.

This is the last frontier of the promised convergence between computers and communication devices: the midsized device. That shift has already happened to the smartphone, but it seems very reasonable that for many people, smartphone screens are too small for serious computing.

If Apple is indeed working on such a product, it will have to get the implementation right to avoid duplicating the failures of so many other mobile computing aspirants. But by having awakened the public to the promise of basic mobile computing, Apple could be best positioned to capitalize on the need for something more.

Axiotron has a Mac tablet already, could Apple be planning to join them?

(Credit: Axiotron)

Familiar Apple rumors are making the rounds again this week as Macworld looms in the distance, one month away.

Trip Chowdhry of Global Equities Research put out a research note Monday morning spotted by D: All Things Digital saying that Apple plans to release something from "a completely new device category" next year. He fails to explain exactly what such a product might encompass, but speculates that it will be based around a processor designed by the former P.A. Semi engineering team.

As a result, the usual rumors (Mac tablet, iPhone Nano, iKindle) are under discussion within the Mac universe. Chowdhry believes that Apple patent filings will be released early in 2009 that will make everything clear, and also notes that P.A. Semi should have an iPhone chip out around that time that improves performance and battery life.

Earlier this year CEO Steve Jobs implied Apple was watching small-device categories like tablets and Netbooks to see if they actually take off as a mega-trend, but in the meantime the company had other priorities. Mac tablet rumors stretch back for years, and in the past Jobs has quickly shot down talk of an Apple-produced competitor to Amazon's Kindle, which has been interpreted as a sign Apple was doing just that.

Mark Papermaster was chosen to lead a team developing future iPhones, even though the company worried about his lack of direct consumer electronics experience.

(Credit: Apple)

Apple knew full well that Mark Papermaster would have to learn his new role as iPod and iPhone chief "on the job" when it hired him.

Papermaster's former employer, IBM, filed a lawsuit against him last week, claiming that he broke the terms of a noncompete contract with IBM in accepting a high-profile job with Apple. InformationWeek spotted Papermaster's formal response Friday morning, in which he declares that the two companies are not competitors and that his experience at IBM is not the primary reason why Apple sought his services.

The court filings reveal the interesting process Apple used to hire Papermaster to replace Tony Fadell, a longtime executive in charge of Apple's iPod group. And contrary to speculation, it appears that Papermaster--a well-respected chip executive--will have nothing to do with chip design at Apple on Day 1.

Apple began searching within the consumer electronics industry in October 2007 for a lieutenant and eventual successor to Fadell, but it couldn't find anyone it liked, according to the court filing. Instead, the company decided to search for an executive with strong overall technology skills who would be a good fit inside Apple, modeling the search on the process used to find current Mac hardware leader Bob Mansfield.

Mansfield was the one who suggested Papermaster as a candidate, though he didn't exactly roll out the welcome mat for his college buddy. On a list of potential candidates sent to Vice President of Human Resources Danielle Lambert (who is married to Fadell), Mansfield described Papermaster: "Mark fits the bill wrt (with respect to) systems and semiconductor understanding, but in every other way is a long shot."

Nonetheless, he was brought into Cupertino in February 2008 to interview with CEO Steve Jobs and Fadell. Apple liked Papermaster in many ways, but it wasn't sure that his experience in server development was the proper background for the role--especially in light of the fact that in February, Apple was working hard on getting the iPhone 3G out the door, and it wanted someone who could hit the ground running.

However, as Lambert said in a statement attached to Papermaster's response, "nobody questioned Mark's ability to lead a development team." The company offered him a role designing laptops, and while Papermaster was intrigued by the possibility of working at Apple, he wasn't all that crazy about that particular role.

But after Apple got this year's crop of iPods out the door in September, the search for Fadell's replacement intensified. Papermaster was offered that role, and he jumped at what he called "the opportunity of a lifetime."

As Papermaster sees it, his role is fairly narrow: he's tasked with overseeing the development of iPod and iPhone hardware. He won't be working on personal computers, he won't be working on servers, and perhaps most surprisingly, he won't be working on microprocesors.

Papermaster's court filing says the P.A. Semi team acquired by Apple earlier this year is part of the group managed by Mansfield, not part of the iPod and iPhone group. That's a surprising organizational decision, given that Jobs has said publicly that P.A. Semi was brought into the fold to work on chips for the iPhone and iPod Touch.

Despite a career spent developing chips at IBM, Papermaster supposedly won't have anything to do with Apple's budding chip team.

(Credit: CNET)

The filing notes that Apple currently acquires iPhone microprocessors from an outside vendor, widely believed to be Samsung. Unsurprisingly, it stops short of mentioning whether or not Apple plans to eventually design and develop its own microprocessors for that product, as seems evident.

In a declaration accompanying the formal court response, Papermaster notes that "it is also my understanding that I will not be responsible for developing the microprocessors that are used in the iPod and iPhone products, but rather those will be procured from sources outside my group." Whether those sources are outside Papermaster's group but inside Apple is left unstated, but Papermaster also says, "I will be acting solely as a product manager--I am not being hired to develop technology across product lines."

An Apple representative declined to comment on Apple's organizational structure or the court filings, only to say, "We think IBM will see that the iPod and iPhone are not competitive with their business." IBM likewise declined to comment on Papermaster's response beyond the statements they have already made regarding their intention to "vigorously" pursue the case against Papermaster.

Papermaster's argument against the lawsuit is that since Apple and IBM aren't true competitors, and since he isn't working on the small slice of Apple's business--servers--that does overlap with IBM's business, the noncompete should not apply. Likewise, he believes that he's not in a position to divulge any IBM trade secrets because "Mr. Papermaster's position at Apple will involve a completely different product using different technology that Mr. Papermaster will have to learn on the job."

It's quite possible that Papermaster's lawyers are deliberately downplaying his connections to Apple's budding chip design team in order to make this lawsuit go away, since the chip angle is IBM's only real argument. Even if Papermaster isn't directly involved on a day-to-day basis with the P.A. Semi team, he will be in charge of specifying the hardware requirements for the iPhone, and part of that includes the chips that go into that system.

Apple appears to be making a bit of a gamble with this hire, entrusting the care of what has become its most important product to an executive who, though well-regarded, has no experience working inside the fast-paced consumer electronics industry. And the most relevant part of his IBM experience doesn't appear to be part of his marching orders at Apple.

But the company believes that his leadership skills will serve him well at Apple, according to one of the court filings: "Apple has hired Mark Papermaster because he has strong general engineering skills, is an outstanding leader, and because we believe he will be a good cultural match at Apple."

Future iPhone processors may be designed exclusively by Apple's new P.A. Semi team of engineers.

(Credit: CNET)

The magic of social networking has confirmed that Apple plans to make its own ARM-based processors for future versions of the iPhone.

The New York Times spotted the LinkedIn profile of Wei-han Lien, Apple's senior manager of CPU development, over the weekend. Lien's job description, according to his profile, involves managing the ARM CPU design team for Apple, an extension of his previous work at P.A. Semi.

It's been pretty clear for a while that Apple bought P.A. Semi earlier this year to work on developing its own chips for the iPhone, but as the Times points out, Apple has never specifically said that it planned to make CPUs: many different chips inside a modern smartphone use a processor based on one of ARM's cores.

However, CEO Steve Jobs did say that Apple was planning to have the engineers make "system-on-chips" for the iPhone, which implies they were tackling the whole package, CPU included. Another solid hint was the disappointment expressed by Intel's Pat Gelsinger over missing out on the design win for the next-generation iPhone, as well as Apple's apparent decision to secure an architectural license for the ARM core.

A new architectural licensee revealed by mobile-chip designer ARM might just be an old friend.

ARM Chief Executive Warren East revealed on an earnings conference call on Wednesday that "a leading handset OEM," or original equipment manufacturer, has signed an architectural license with the company, forming ARM's most far-reaching license for its processor cores.

East declined to elaborate on ARM's new partner, but EETimes' Peter Clarke could think of only one smartphone maker who would be that interested in shaping and controlling the direction of the silicon inside its phones: Apple.

And at first blush, the theory makes a lot of sense. Apple acquired P.A. Semi earlier this year to put CEO Don Dobberpuhl and his team of designers to work on developing new chips for the iPhone and iPod Touch, which currently use chips manufactured by Samsung, Marvell, and NXP based on the ARM core. Dobberpuhl has a long history with the ARM architecture as leader of the StrongARM processor design team eventually acquired by Intel.

ARM's architectural license gives holders the freedom to tinker with the ARM architecture and processor cores, whereas most of its customers sign licenses for specific core designs to incorporate into their chips.

The company is thought to have very few architectural licensees. Although they are believed to include companies like Qualcomm, Marvell, Texas Instruments, Freescale, and Samsung, an ARM representative did not immediately respond to a request for the full list of architectural licensees.

Note the common ground between those companies: they're all chipmakers. No other major handset vendor--with the exception of Samsung--has taken a similar degree of interest in designing processors, which makes Apple a likely candidate for the new architectural license, given its plans for the P.A. Semi designers.

Ties between ARM and Apple go back to the early 1990s, when ARM was actually created out of a partnership between Apple and Acorn, ARM's forerunner. ARM got the design win for the ill-fated Newton back in those days, but it made a strong comeback at Apple, inside the iPod line in the early part of this decade.

Apple may have taken a look at the future of mobile chip development and decided to forge its own path.

Future successors to the iPhone 3G might use a chip completely designed by Apple.

(Credit: Apple)

The New York Times scored an interview with Apple CEO Steve Jobs following Monday's Worldwide Developers Conference keynote, and buried inside a rambling exchange about parallel processing and Mac OS X Snow Leopard was this little nugget about PA Semi, the chip company Apple acquired in April. "PA Semi is going to do system-on-chips for iPhones and iPods," Jobs told the Times.

System-on-chips, or SOCs, are pretty much what they sound like: complete computer systems on a single chip, including the processor, memory, graphics, networking, and all the regulator chips needed to manage things like power consumption. ARM's licensees, such as Texas Instruments, Samsung, and Nvidia build SOCs around ARM's processor cores for smartphones such as the iPhone, and Intel wants to head down this path with its Atom processor family.

It's well known that Apple has played an active role in the design of chips that go into its system for years, but the acquisition of Dan Dobberpuhl's PA Semi team means it will apparently play an even more active role in the future. Jobs has previously said that Apple acquired PA Semi for its talent and patents--not its products--but had not shared many details about its plans for that talent.

In an interview with Nvidia CEO Jen-Hsun Huang last week, we got to talking about mobile processors and the evolution of that market, and he insisted that Samsung, widely thought to be the processor supplier inside the iPhone, merely "fabbed" the chip. In his view, Apple was the chief designer of the ARM-based processor that's used to run the iPhone--and presumably the iPhone 3G unveiled Monday--with Samsung just providing the factory. The PA Semi engineers would allow Apple to draw up a complete design in-house and take it to a chip foundry without having to let any other mobile processor companies in on its plans, Huang said.

The companies that license ARM's instruction set are increasingly butting heads with Intel as the ARM community tries to move up from smartphones into more powerful mobile computers, and Intel tries to shoehorn its PC processing know-how into a mobile environment. There has been much speculation over the past year or so that Apple will one day add processors for mobile devices to the invoices orders it sends Intel every quarter for Mac processors, but the PA Semi acquisition apparently means Apple is prepared to go it alone.

Apple will be a chipmaker--of sorts--for a while after its acquisition of PA Semi as it satisfies demand for that company's current product.

The Register is reporting that Apple will support PA Semi's current Pwrficient processors after it finalizes the purchase of the company. Apple CEO Steve Jobs indicated after his company bought PA Semi that its primary interest in the chipmaker was for the company's talent and patents, not the actual chips themselves.

But PA Semi counts some pretty influential organizations among its user base, such as the U.S. Department of Defense, which uses military systems built around Pwrficient chips, and apparently likes them. EETimes had reported that Raytheon and Lockheed Martin, two prominent defense contractors, planned to complain to the DoD after hearing that PA Semi couldn't guarantee the supply of those chips after it was acquired.

Apparently the parties have managed to work out a deal that satisfies all concerned. The DoD could have caused problems for Apple by holding up the acquisition, but Apple could fairly easily cut a deal with a third-party foundry to manufacture the Pwrficient chips and dedicate a few PA Semi employees to managing that business, as El Reg suggests. Apple will not, however, continue to produce new versions of the Pwrficient chip, meaning the defense industry will have to look elsewhere at some point.

PA Semi, acquired by Apple earlier this week, has been telling its former customers that Apple has no interest in its products, according to a report.

EETimes is reporting that PA Semi started telling its customers in the military hardware industry that it was about to be acquired by a company that wanted only its intellectual property and employees.

On Tuesday night, we learned that Apple was that company via a Forbes report that said the Mac maker wanted to put PA Semi's PWRficient processors in its iPhone.

The PWRficient processors are used in military systems built by defense contractors such as Raytheon and Lockheed Martin, companies reportedly annoyed by the possibility that the PWRficient processor might be coming to a quick end. According to EETimes, PA Semi told those contractors that after the acquisition, supply of the chips could not be guaranteed.

And that, apparently, could cause problems for Apple. The report says those contractors are going to complain about the acquisition to the U.S. Department of Defense.

I can't imagine that the deal would be overturned just because of a few complaints to the military, but Apple might be forced to figure out a way to transfer that portion of PA Semi's assets to another chipmaker that could satisfy Uncle Sam's need for the chips.