Nanotech - The Circuits Blog

If the Apple tablet emerges as expected, this will be another big device market, following media players and smartphones, that the PC industry cedes to Apple.

Tablet: Is this the best WinTel-HP can do?

(Credit: Hewlett-Packard)

The writing is already on the wall already for Microsoft and smartphones, as spelled out in a previous post and as documented in shrinking market share numbers.

That's not to say that Microsoft, Compaq (later Hewlett-Packard), and Intel didn't have a chance. Remember the Compaq iPAQ PDA that debuted way back in 2000, powered by an Intel StrongARM chip running an early version of Windows Mobile?

That device had a lot of potential. The operative word being "potential." An iPAQ could have been an iPhone. Or at the very least an iPod. And everybody could be drooling over iPAQs today instead of iPhones. Or using iPAQs instead of BlackBerrys. But of course things didn't turn out that way.

Fast forward to 2010 (January?). Apple announces a tablet and suddenly everyone wants a tablet. (Or iSlate, if you will.)

Whatever happened to this Intel-powered Asus MID?

(Credit: Asus)

And what have Microsoft, Intel, HP, and others been offering in the interim years when they had every opportunity to come out with a blockbuster tablet? Unattractive, bulky, half-baked convertible laptops that, let's put it this way, have not taken the PC market by storm.

So, here's the $64,000 question, uh, make that the $64 billion question. Why can't the combined R&D smarts, market clout, and overall technological resources of Microsoft-Intel-HP-Dell come up with a thin, sexy compelling tablet and/or media pad that will turn heads and convince the unbelievers (the average why-would-I-need-something-like-that consumer) that a tablet is a must-have product?

Answer: Because Apple will.

Here's a not unlikely scenario. Apple brings out the tablet/media pad, wows U.S. (and world?) consumers, sells a ton of units, Microsoft-Intel-HP-Dell follow suit with slavishly copied devices that don't sell very well comparatively.

iPAQ PDAs: Missed opportunity?

(Credit: Hewlett-Packard)

That's how the market for successful newfangled devices works these days. Apple creates the market and everyone else follows in a panic.

Then there's the Intel factor. Intel also wants to be a player in this space. But Intel and its coterie of PC makers can't get off the traditional-design laptop gravy train. Plus, as formidable a chipmaker as Intel is, it is still behind the Qualcomms and Texas Instruments of the world in building the power-efficient system-on-a-chip silicon that goes into smartphones and will likely go into tablets.

So, here's my question for Intel et al: How many people will be buying Netbooks or Intel-based MIDs (mobile Internet devices) in 2011 if Apple has a more compelling alternative? Answer: a lot less if the Apple tablet exists.

OLPC tablet concept: Can't a PC maker do this?

(Credit: OLPC)

And add Asia-based device makers offering tablets using an Nvidia Tegra 2 chip to that. A number of these tablets are expected too in 2010. In fact, Nvidia is already doing what Intel should have finished doing a long time ago: make a competitive system-on-a-chip that powers small devices. Intel had the chance to make XScale (what StrongARM eventually became) into something big for small devices six years ago. But it didn't. And now Intel is trying to reinvent the wheel by squeezing the upcoming "Moorestown" Atom chip into smartphones.

Intel, I'm sure you think Moorestown is a great idea, but it's a little late. Apple beat you to it by about three years.

Intel is about to roll out some pretty impressive mobile processors, finally bringing its "Nehalem" Core i architecture to the mainstream mobile space. So, what can we expect from Apple?

First, a little recent history. Apple, so far, has elected not to use the Core i7 quad-core mobile processors announced back in September in its MacBook Pro line. Which isn't that surprising. The first crop of laptop i7s barely qualify as mobile processors: they have a desktop-like TDP (thermal design power) of 45 watts that wreaks havoc on battery life.

That said, as an example of what an Apple rival has chosen to do, the Hewlett-Packard Envy 15 now offers both the i7-720QM (1.6GHz, 6MB cache) and i7-820QM (1.73GHz, 8MB Cache)--both quad-core Core i7 processors.

A big imponderable is whether Apple will adopt a future 32-nanometer version of the quad-core mobile i series for the 17-inch MacBook Pro. Presumably, this would have a lower TDP and be more amenable to Apple.

Now, for some armchair analysis. Arrandale: One of the burning questions (at least among some in the tech media) is whether new MacBooks will use Intel's "Arrandale" mobile Core i series of processors. Arrandale is significant for two reasons: it is part of Intel's 32-nanometer chip roll-out and is the first instance of Intel combining the graphics function with the main "CPU" processor. This results in better overall power efficiency and integrated graphics performance that "doesn't suck" anymore, as some observers have put it.

One school of thought is that Apple will not use the processor. If there is any truth to that rumor, that makes for a head-scratching scenario since Arrandale will be the pillar of the mobile Core i3 and i5 lineups. A likely scenario is that Apple--one way or another--chooses to attach Nvidia or ATI discrete graphics processors to Arrandale, or a facsimile thereof.

Nvidia or ATI: And speaking of Nvidia and ATI, instead of trying to second-guess Apple on all of the possible graphics chip permutations, the easier question to ask is: which graphics chip supplier will prevail this time around? Nvidia--despite defects in some of its past offerings--has been dominant over the last year or so across the MacBook lineup. Will this continue? Or will Apple strike more of a balance between Nvidia and ATI?

Remember, that Apple is touting the general compute function of the graphics processing unit, or GPU, in OS X Snow Leopard. "Now a new technology in Mac OS X Snow Leopard called OpenCL takes the power of graphics processors and makes it available for general-purpose computing," according to Apple ad copy. (Translation: using the GPU more like general-purpose CPU.) Are Nvidia and ATI OpenCL equals?

MacBook Air: And what, pray tell, will happen to the MacBook Air? Which is coming up on its second anniversary in January. I won't venture a guess (not yet at least), though I have a personal interest in this subject since I have been using an Air since February 2008.

Blu-ray: And finally, next-generation optical drives. Will MacBook Pros finally get Blu-ray? Maybe. There seems to be some pessimism about Apple adopting Blu-ray, as reader comments (and threads on other forums) suggest.

Apple, in all of its wisdom, will provide the answers to these questions soon enough.

Updated at 6:05 p.m. PST: adding Blu-ray discussion.

Intel's checkered past on some large projects means the chipmaker must prove that Larrabee isn't a development flub that will simply be kept on life support for the next few years.

An Intel session on Larrabee technology in 2008 when hopes were high.

(Credit: Brooke Crothers)

As reported last week, Intel's Larrabee graphics chip was killed as an initial product offering after protracted delays, demonstrating that as successful as Intel is, it's not immune to major product missteps.

If certain product histories are any indication, the challenge could be daunting. Intel's XScale processor for small devices--which was used in Compaq handhelds back in 2000--was sold off to Marvell in 2006 after an unsuccessful run. And its Itanium processor has been the object of perennial ridicule as a product hanging on for dear life after getting off to a very rocky start back in 1998. Sun Microsystem's former CEO Scott McNealy eventually dubbed the chip the "Itanic" as a play on the word Titanic.

"If you go back in history when they started down the Itanium path in the mid-90s, they said they were going to have a really whiz-bang product, but by the time they finally got it out, it was decidedly ho-hum or even worse," said Nathan Brookwood, principal analyst at Insight 64.

"They've learned that you don't ship a product the first time around so that when it finally does appear people go 'What was all the fuss about?'" he said.

Larrabee, as we know now, was not ready for prime time. "It was for all intents [and] purposes an Intel project--a test bed, some might say a paper tiger," said Jon Peddie, president of Jon Peddie Research, writing in a blog.

For better or worse, Intel is expected to forge ahead with Larrabee, with a real product not appearing until ... Read more

It's too tempting not to pose that question as the monthly Apple tablet rumors fly.

The Apple 'media pad' could look like this--just bigger with ad copy that says '4x faster.'

Conjecture about future Apple products is always an interesting exercise because it requires a lot of imagination to make up for the copious lack of hard data. This is especially the case for the rumored Apple tablet, despite analyst claims about product specifications, such as the oft-repeated 10.1-inch screen.

But there is one theme that keeps popping up that is highly plausible: it will be a device to view media and book content (rumor: 30/70 revenue split between Apple/publisher) in a "better" way. Hmm...let me think for a minute: the same way many people now use their iPhone? (Certainly, the media part.) So--though not a phone per se--basically an iPhone (or iPod) with a bigger screen, faster graphics, and better interface (and/or subscription model) to view content.

Does that about cover it? That doesn't sound like a tablet to me, with all the negative connotations that the word "tablet" has in the PC industry: fat, heavy, kludgey, boring.

Which is why "media pad" sounds a lot better. That, to me at least, connotes thin, light, modern, exciting. But I'll leave the branding to Apple. So far, they have a pretty good track record.

The trend-setting MacBook Pro and Air both now face tough competition from Hewlett-Packard, which has the resources to match, and in some cases exceed, Apple laptop designs.

HP Envy 13

(Credit: Hewlett-Packard)

I will expand very briefly on a previous post where I compared, on technological merits, the 13-inch Apple MacBook Pro and Air laptops with an HP Envy 13 in response to some of the comments attached to the post.

I had stated, as an opinion, that the aluminum-clad HP Envy 13 had eclipsed Apple MacBooks technologically in some crucial areas. Namely, processors offered, screen resolution, graphics, and battery life.

The assertion that the HP Envy 13 has surpassed, in some important respects, the MacBook Air and 13-inch MacBook Pro in technology shouldn't be that surprising considering the financial and technological resources that HP has.

Companies like HP and Dell bifurcate their lineups into inexpensive (typically retail consumer models) and more expensive (often business models). Some models are of decidedly lower quality than Apple--as many comments quickly point out--but some are equal to or better than a roughly equivalent Apple laptop both in quality and technology.

The Envy 13--which is HP's entry into the luxury laptop category--falls into the better-than-Apple-laptop-technology category, in my opinion. The luxury Adamo line from Dell is also making a play to, at the very least, achieve parity with Apple's MacBook line.

Again, this is an opinion, not a be-all, end-all verdict on the fate of Apple. And not a review per se that gets into benchmarks. I'm just looking at the raw technology.

Opinion pieces invariably elicit strong counter arguments--not to mention strong opinions (or invective). Especially when Apple is involved.

Firefox has a CPU usage issue and, consequently, can cause overheating problems in some laptops, particularly ultraportables. That's what I've found over the last couple of years.

But don't take my word for it. This is documented on a Mozilla support page entitled "Firefox consumes a lot of CPU resources." The page states: "At times, Firefox may require significant CPU [central processing unit] resources in order to download, process, and display Web content." And forum postings like this one about a Dell Netbook are not uncommon: "Mini9 would get way too hot."

The Mozilla support page goes on to say that "you can review and monitor CPU usage through specific tools" and describes ways to limit CPU usage, such as: "A Firefox add-on, called Flashblock, allows you to selectively enable and disable Flash content on Web sites."

Let me describe my experience. I find that tab for tab, Firefox uses decidedly more resources than other browsers--Safari, for example. And in the past (when I was actively using a Windows Vista-based machine) Firefox also compared unfavorably with Microsoft's Internet Explorer for CPU usage.

More specifically, here's the behavior as I see it. When I'm accessing sites with multimedia content such as the CNET front door, Firefox CPU usage will bounce around between 30 and 60 percent, and sometimes spike higher (80 percent and above), as indicated by the Mac OS 10.6.2 Activity Monitor.

On the other hand, the Safari CPU usage with the same pages open is much lower--typically between 2 percent and 10 percent.

My theory is that most users don't notice this because in mainstream laptops, this isn't an issue. But it can become an issue in ultraportables--typically under an inch thick--which are more sensitive to heat because of the design constraints. The ultrathin Apple MacBook Air, which I use as my main machine, is a good example.

The fan is usually an audible indicator of CPU usage issues. When I'm using Firefox and I have tabs open on multimedia-rich sites (which is par for the course these days), the Air's fan will almost invariably kick on and stay on until I close the tabs. As I write this, the fan has finally shut down after I closed the Firefox tabs (e.g, CNET front door). Those same tabs in Safari are still open and not causing any significant spike in CPU usage or fan activity.

When I contacted Mozilla, a technical support person guessed that Safari is possibly better at optimizing Flash-based sites compared to Firefox. And that may be true. However, I had similar issues before when I was using a Hewlett-Packard business ultraportable (also very thin like the Air) that were not necessarily tied to Flash usage. In short, Firefox was less efficient with CPU usage compared to Microsoft's IE 8. And the behavior was similar. The HP laptop would quickly heat up and the fan would kick on.

Finally, let me reemphasize that I'm guessing that most users don't notice this because heat dissipation is not a big issue for mainstream laptops that are not necessarily thermally-challenged when accessing multimedia-rich Web pages. That said, this has been a steady problem for me because I use ultraportables almost exclusively and has forced me to limit my use of Firefox.

If the iPhone didn't finish off Windows Mobile in the smartphone market, the Motorola Droid may.

Windows Mobile is losing the last vestiges of its mojo--if it really had any to begin with--as the Droid and other phones based on the Android 2.0 operating system push the buzz meter needle into the red zone. Many in the media--which can play a big role in steering users to one technology platform or another--sense that Windows Mobile has now been relegated resolutely to has-been status.

The Motorola Droid's high-resolution screen.

(Credit: Verizon)

Let's do a quick canvas of what some in the press are saying now that we're at the start of the Droid era. A post on SFGate.com (the Web site of the San Francisco Chronicle) is, like other commentary out there, clearly dismissive of Windows Mobile. "Curiously, Microsoft is nowhere to be seen in this battle royal," the author states, referring to the iPhone and Android.

And there's this more damning comment from a blog at SeattlePI.com. "Rarely mentioned, however, is another player in the mobile OS market--Microsoft. Why not? Because not many people in the smartphone world seem to really give a hoot about Windows Mobile anymore."

The litany of like articles is long. This post on PC World asks: "Has Microsoft Placed Its Last Mobile Bet?" The article cites research from Canalys showing Windows Mobile slipping from 13.9 percent of the worldwide smartphone market in 2002 to 9 percent in the second quarter of 2009.

The numbers are even less favorable in an accounting by ad service Admob, which compiles data on which operating systems are in use on mobile devices that access online ads. In August, according to AdMob, Windows Mobile had only a 4 percent share of the mobile OS market worldwide, down from 7 percent in February.

But getting back to my original premise of no mobile mojo for Windows. The fact is that consumers don't care about Windows on smartphones. In other words, while Windows seems to be a prerequisite for many consumers when buying a PC, it just doesn't come into play in a big way in a smartphone purchase.

This will have ramifications beyond Microsoft of course. Companies like Toshiba (and its attractive TG01 smartphone) will probably not be as successful on Windows Mobile as they would (will) be on Android 2.0. Or, at the very least, will not get the necessary buzz.

Then there's the Intel factor. Intel also wants to be a player, eventually, in the smartphone space. If it is indeed able to beat back Texas Instruments (whose chip is used in the Droid), Samsung (iPhone), Qualcomm (BlackBerry), and Marvell, it probably won't do it by sticking to the tried-and-true "WinTel" combination that's been so outrageously successful in the PC space.

And Intel is chasing a fast-moving target. TI, and all the other ARM-based chip suppliers cited above, are slated to bring out dual-core designs that can hit speeds as high as 2GHz (think next-generation tablets and media pads). In other words, they'll also be able claim the coveted speed mantle on phones, such as the Droid, where Windows Mobile is no where in sight.

So the Droid may not be the iPhone killer but rather the Windows Mobile slayer. Microsoft, of course, will always have the unassailable PC franchise. But, wait, isn't Android coming to Netbooks next year? Maybe the real battle royal for Microsoft is yet to come.

Intel on Monday said it is investigating a glitch that prevents an Apple iPhone from synching with certain PCs that use a new Intel chipset.

"Our leading theory is a BIOS or system configuration issue, but we are still investigating," Intel said Monday. The BIOS, or basic input/output system, is the initial code that runs when a PC is powered on. The BIOS identifies and initializes system devices such as the chipset, graphics card, and hard disk drive. Makers of PC circuit boards, aka motherboards, typically offer their own BIOS.

The problem is thought to be tied to the Intel P55 chipset and desktop motherboards from certain manufacturers, according to a report in The Register.

The Intel P55 Express chipset supports the Intel Core i7-800 and Core i5-700 series processors, according to Intel. The chipset is new and has only been available since summer.

As depicted in an "Apple Discussions" thread, on certain PCs or motherboards with the P55 chipset, Apple iTunes 9 for Windows will recognize the iPhone, but when a sync is attempted, an "error 0xE8000065" is generated, indicating an iPhone connection failure. Windows 7 is also cited as a common problem, particularly the 64-bit version of Windows 7.

"Microsoft has not seen this particular question posed in the Microsoft Answers for Windows 7 community forum, nor in any of our call centers," Microsoft told the The Register on Friday. "If we determine this to be a problem specific to Windows 7, we will post an update on the Microsoft Answers site."

The Apple Discussions thread proposes a number of possible solutions.

Apple could not immediately be reached for comment.

Though the Motorola Droid and Apple iPhone have different chassis, their high-octane engines are similar.

The internal similarities begin with performance: both devices are fast. The iPhone 3GS is already distinguished for its speed. And the Droid is quickly garnering similar accolades.

The Motorola Droid has a radically different exterior compared with the iPhone but uses a speedy Cortex-A8 ARM chip like the Apple phone.

(Credit: CNET Reviews)

"The Droid makes a big leap in internal performance. Compared with its rather sluggish Android predecessors," CNET Reviews said, citing the speed at which the Droid opens applications and menus and scrolls through lists and switches display screens.

"We're really pumped to see all the industry excitement it's created," said Jeff Dougan, the OMAP 3 product marketing manager at Texas Instruments, which supplies the OMAP 3430 processor that powers the Droid. "This is the first handset that truly realizes the full potential of Android," he said, referring to Google's Android 2.0 operating system that runs on the Droid phone.

The TI processor, like the one in the iPhone, is based on an a new architecture called Cortex-A8 from U.K.-based chip design house ARM, whose wide variety of chips populate most of the world's cell phones. Dougan says most smartphones currently on the market use an older, lower-performance ARM architecture than the Cortex-A8--with the exception of the Palm Pre, which opted for the newer TI chip. The Cortex-A8 provides a "two to three times performance boost" over older architectures, according to Dougan.

Max Baron, an analyst at Microprocessor Report, says the chips in the Droid and the iPhone (see not below) are so alike that differences are more dependent on the operating systems the two chips use and how successfully each phone maker optimizes the OS. "With chips that have near-similar specs, the optimum OS and the look-and-feel of the user interface may make or break the product," Baron said.

The core of TI's OMAP3 processor.

(Credit: Texas Instruments)

"The caveat, however, is that even small differences in chips will surface and become important differentiators as soon as the market forces you to increase the screen size or add more pixels per screen, or execute more power-consuming applications," he added.

The raw MHz ratings on the chips are slightly different. The processor in the iPhone 3GS--which is believed to be based on the Samsung S5PC100 processor--runs at 600MHz, according to most accounts. The Motorola Droid's TI chip is rated at 550MHz though theoretically it can be run as fast as 600MHz, according to TI's Dougan.

Both phones also use PowerVR graphics from Imagination Technologies--a company that both Apple and Intel have invested in, testifying to how hot its ultramobile graphics technology is. The PowerVR SGX is renowned for its ability to process several million triangles-per-second--a key indicator of graphics chip performance--blowing away other phones and the previous version of the iPhone.

Other internal specifications are similar between the two phones, including memory capacity (either 16GB or 32GB) and communications chips that offer 3G, Wi-Fi, and Bluetooth connections.

So, internally the Droid is every bit the iPhone's equal. And future versions of TI OMAP 3 chips that may appear in upcoming Droids will be backed by formidable ecosystems, according to Baron. "Investments in application software may lean more toward the TI components," said Baron, given TI's strong support of the entire chip ecosystem, including auxiliary chips and software development tools.

Note:: Apple's and Samsung's reluctance to release information about the processor used in the iPhone 3GS has made it difficult to determine if the chip is based on the Samsung S5PC100, according to the Microprocessor Report's Baron. Many iPhone 3GS reviews and teardowns, however, state explicitly that the iPhone's processor is essentially the Samsung S5PC100 processor.