Don't get me wrong-- I think the Intel-TSMC alliance announced earlier this week is a good thing for both companies.
But the official explanation, that Intel wants TSMC's help to make Atom processor cores more widely available to the industry, just doesn't strike me as a sufficient reason for the deal.
Intel hardly needs TSMC's help to make SoCs (systems on a chip). Intel has been making highly integrated devices for the embedded market, as well as PC chipsets for a long time. It already has enough of the building blocks and enough experienced engineers to make Atom-based SoC products.
And it isn't as if Intel needs better process technology, or more fabrication capacity. Intel already has more of the best fabs in the world than any other company.
What's the one thing TSMC can do that Intel can't? Operate with low gross margins. In its most recent quarter, TSMC's gross margin was only 31.3 percent, while Intel's gross margin is still an industry benchmark at 53 percent. The difference is more than Intel's net profit--that is, if Intel had TSMC's gross margins, it would be losing money.
Low-margin component suppliers are a critical element of the embedded-systems market, which Intel identified as one of its target markets for this deal. Cost is king in consumer electronics, so high-margin suppliers like Intel rarely get a chance to participate.
Similarly, as average PC-selling prices decline, a growing share of the demand for processors and chipsets drops into price ranges in which Intel just can't afford to play.
The TSMC deal is Intel's way of taking a piece of these businesses without spending much money or taking much risk. For example, TSMC is already accustomed to helping its customers make SoCs for embedded systems. Intel could build such a business itself, but not at the margins it's used to.
Intel said in its press release that it will be porting its Atom cores to TSMC's technology. This is the sort of work that can get expensive in engineering time, but it's possible that the work will be made easier by a convergence between TSMC's processes and Intel's.
Last May, Intel agreed to cooperate with TSMC and Samsung in the transition to larger 450-millimeter silicon wafers (a little less than 18 inches across, up from the 12-inch wafers used today).
This doesn't necessarily mean that the three companies will co-develop fully compatible manufacturing processes, but with the 450mm transition being slated for 2012, there's still plenty of time left to drop that other shoe.
Anyway, this new TSMC deal is merely at the earliest official stage. The companies have signed a memorandum of understanding, but they have yet to work out the details. That could take a year, and it could be another year or two before Atom-based chips are ready to start rolling through the TSMC factory.
All in all, Atom SoCs might not become available from TSMC until 2012, at which point, they could, in principle, be made on a common Intel-TSMC process.
Not that Intel would provide its really good process technology to TSMC. In chips, as in other things, quality is expensive. Intel's best process technology, which it uses primarily for microprocessors, is at the leading edge of semiconductor manufacturing, with features such as a metal electrode acting as the transistor's gate, a hafnium-based insulation between the gate and the channel, and strained silicon in the transistor channel itself (where the current flows when the transistor is on). (See this Intel presentation for more details. Incidentally, did Intel ever announce which metal it's using? If so, I can't find it.)
TSMC may not need or want any of these features, and it would make sense for Intel to keep its best process technology to itself, anyway, if only to protect its high profit margins.
Even without a leading-edge process, TSMC can still make good money from Atom-based SoCs in the embedded market. That's enough to justify TSMC's participation in the deal.
But I'm not sure that explains Intel's motivation. Sure, Intel will make money it wouldn't have made otherwise, but it will also have costs it wouldn't have had otherwise. Intel may make a few bucks per chip in intellectual-property licensing fees, and perhaps this could amount to hundreds of millions of dollars a year, but that isn't a whole lot of money to a company like Intel, which makes tens of billions of dollars a year in gross revenue.
Why else would Intel be doing this deal?
Well, I think that the chipmaker could be setting itself up to kill off three of its biggest rivals.
There's already an x86 processor company using TSMC to make (some of) its chips: Via Technologies. Via isn't a big player, but it's been a thorn in Intel's side ever since it purchased the x86 processor operations of IDT (WinChip) and National Semiconductor (Cyrix) in 1999.
Via specializes in exactly the kind of processors that Intel can't afford to sell: low-cost, highly efficient designs aimed at low-cost PCs and embedded systems. Today's Atom is better than Via's best chips, but it's also more expensive. A cheaper TSMC-sourced alternative will hurt Via badly.
Most of the same reasoning applies to ARM, which licenses its processor cores to be used in SoCs made at TSMC, among other fabs. That's almost the same business model Intel is adopting with its own TSMC deal.
ARM dominates the market for microprocessors in cell phones. Intel's current Atom processors are too expensive and too power-hungry for that market. But remember, it'll be a couple of years at least before Atom-based chips start shipping from TSMC. The Atom cores of 2011 or 2012 will be more directly competitive with ARM's cores.
So put ARM on the endangered-species list too.
There's one other company that ought to be worried by this deal, and it probably isn't one you'd expect: Nvidia.
Nvidia is generally thought to be TSMC's biggest customer. It doesn't make x86 processors (though there are persistent rumors that the company is developing one), but it does make the ARM-based Tegra family, which would run up against these future Atom chips.
It's Nvidia's graphics chips that I'm worried about, however.
Intel is developing graphics chips of its own under the Larrabee code name. I wrote about Larrabee last August, and it seemed like a bad idea to me at the time. One of my key objections, however, was that graphics chips are inherently a low-margin business due to the strong competition between AMD and Nvidia, and I didn't think that Intel could afford to drag down its margins just to compete in that market.
The TSMC deal changes all that.
Larrabee's cores aren't Atom cores, per se, but they're similar enough that Intel might consider them to be covered by the language in the TSMC partnership announcement. Or if not, agreements can always be expanded later.
Making Larrabee chips at TSMC would solve the margin problem, putting Intel's graphics chips on a level playing field with Nvidia's. Larrabee would still be at a significant disadvantage because its x86-based design isn't as well-suited to graphics acceleration as Nvidia's chips, but Intel has a special ability to sell inferior products along with other chips its customers need--especially processors. That's reportedly how Intel's slow integrated-graphics chipsets ended up in so many systems during the Windows Vista transition, leading to many disappointed customers.
Or it's possible that Intel will not allow the TSMC deal to harm these companies, if only because Intel may still be in court defending itself against AMD's antitrust lawsuit.
But I wouldn't make that assumption, and I bet that ARM, Nvidia, and Via won't either. Intel isn't the only paranoid company in this industry.
I bought my 2.33GHz MacBook Pro about two years ago, shortly after it was introduced. It came with a 160GB hard disk, but that wasn't really enough for all my stuff, particularly when I wanted to add a Boot Camp partition for Microsoft's Windows Vista.
So last July, I upgraded to a 250GB drive, a process I described here ("A new hard disk for my MacBook Pro").
Samsung's Spinpoint M6 500GB mobile hard disk
(Credit: Samsung)That drive started feeling a little tight within just a few months, chiefly due to videos downloaded from the iTunes Store. Although I rarely buy videos from iTunes, there's a lot of free stuff there. I have a particular weakness for video podcasts about automobiles, such as VOD Cars and BMW's own video magazine, BMW-web.tv. Oh, and I've also lost some potential productivity to the Onion News Network video feed and the original Onion Radio News, which are also available through iTunes.
I hung tight through the 320GB generation of laptop hard disks, figuring that wasn't enough of a capacity improvement to justify the cost.
But shortly after Samsung started shipping the Spinpoint M6 model HM500LI, Montalvo Systems shut down, and I had other things to think about than upgrading my hard disk. I decided to wait for Hitachi or Western Digital to introduce a competing model, so I could make sure I was getting the best product when the time came.
Hitachi has a 500GB drive, but at 12.5mm thick, it won't fit in the MacBook Pro. Then Western Digital introduced the new Scorpio Blue, a 9.5mm drive with specifications pretty much identical to those of the Samsung drive. I was able to get a pretty good deal on the Samsung drive, so that's what I decided to go with.
I went through the same upgrade process I used last time, which I recommend to anyone upgrading a hard disk: back up the old disk to the new disk in an external enclosure before swapping in the new drive. With a Mac, it's easiest to do the backup by connecting both drives to another machine using the special feature called FireWire Target Disk Mode.
In this case, I only backed up the Mac partition this way, since Macs can't natively write to NTFS partitions; I used Windows to back itself up separately to a different drive.
After going through the usual grief involved in upgrading a MacBook Pro hard disk-- which I don't recommend to anyone who isn't very familiar with safe maintenance procedures for modern laptops-- everything just worked. The new drive is fast, silent, and huge, everything I love in a hard disk.
Well, all but one thing. The Boot Camp partition isn't so easy to migrate over. After booting from the new drive, I let the Boot Camp Assistant program create a new Boot Camp partition with an NTFS filesystem, then used Mike Bombich's NetRestore application to copy the old data to the new partition.
But although the copy proceeded normally and the new partition received all the files from the old one, it also received the old partition's size-- 20GB instead of the 32GB I had allocated for it. And it didn't come out bootable, nor would Parallels Workstation work with it, in spite of being configured to use the Boot Camp partition on the old drive.
I can't find anything online about migrating a Boot Camp partition when upgrading a hard disk. So let me ask all of you folks: does anyone know how to do this?
I'll post an update here when I get it figured out. In any event, I can always just wipe out the new partition and reinstall Windows...
Update: now solved! See my followup post: "Migrating and resizing a Boot Camp partition". Thanks to everyone who commented.
Samsung Electronics, an arm of the giant Korean company (second only to General Electric in annual revenue among conglomerates), held a press event in San Francisco last week to show off its products for the coming holiday season.
Samsung's new Series 9 LCD HDTVs use LED backlights to improve contrast and color quality
(Credit: Samsung Electronics)I'd been looking for an excuse to go up to the city, so off I went-- taking Caltrain rather than driving. Conveniently, the Samsung event was just a few blocks from the train station in San Francisco.
... Read more
Wednesday morning I visited the new Menlo Park, Calif., headquarters of Zonbu, makers of the low-cost, service-supported Linux computer I mentioned last month (here). I met with Zonbu CEO Grégoire Gentil, who gave me an overview of the company's business plan and a demo of the system. He also offered ... Read more
A big fuel cell from UTC Power was in the news here in Silicon Valley this week when Fujitsu installed it as a backup power source for its local campus. CNET's Michael Kanellos wrote a good story about the event here, and took pictures ("Photos: Fujitsu unveils king-size fuel cell").
Although the fuel cell itself runs on hydrogen, there's no convenient source of pure hydrogen in Silicon Valley, so UTC Power also provided a steam methane reformer that yields hydrogen from natural gas.
One place where pure hydrogen is readily available is NASA's Cape Canaveral facility, and BMW recently completed an eight-week test of its Hydrogen 7 prototype vehicle there. (See the Edmunds review of the car, and a story of the test, with a great photo of the car posed in front of the Endeavour before the recent launch.)
BMW has now handed the keys to one of these cars to actor Will Ferrell (BMW press release), though I suspect BMW will learn less from Ferrell than it did from the NASA testers.
Interestingly, however, the Hydrogen 7 is not a fuel cell car, in spite of stories like this one. BMW just uses its big 6-liter V12 engine with minor modifications allowing it to run on hydrogen as well as gasoline. This is may be the most practical way to run a car on hydrogen, but it's not the wave of the future.
Ford has made a true fuel-cell car, the Fusion 999, and it's considerably faster than the Hydrogen 7. In fact, Ford's unique vehicle, derived from its production Fusion sedan, recently set a speed record for fuel-cell vehicles, reaching 207.297 mph on the Bonneville salt flats. This isn't a car you'll be seeing on the road anytime soon; it has a 770-horsepower electric motor and several huge pressure tanks, it weighs 6,700 pounds, and its range is only just good enough for the high-speed runs on the salt. Richard S. Chang blogged about the event for The New York Times and there's also an interesting video on the Popular Mechanics site.
Ford worked with Ohio State University on the Fusion 999 and on OSU's scratch-built Buckeye Bullet 2, a fuel-cell streamliner that may be able to exceed 350 mph. There's a blog for that project, and it's fascinating reading if you like cars and high technology.
But when can we regular folks have fuel cells of our own? Other than expensive and clumsy solutions like the Trulite and Medis products I blogged about last month (somewhat disparagingly), it won't be soon.
EE Times recently reported that Toshiba, for example, expects it will take several years to bring practical fuel cells to market. "Practical," in this case, means fuel cells based on DMFC (direct methanol fuel cell) technology, which can be powered by inexpensive methanol (also known as wood alcohol).
Samsung has demonstrated a version of its Q35 ultraportable notebook running on a DMFC power supply, achieving 240 hours of operation over the course of a month, but don't get too excited-- the supply is fairly bulky (see some photos and a video on AVING.net) and I'm not entirely convinced that the full month's worth of fuel is stored internally.
It's no coincidence that Samsung chose the Q35 for the demonstration; even the best DMFC fuel cells have much lower power density (watts of output power per cubic inch) than lithium batteries, so they'll have to be very large to support high-performance notebooks.
I'm sure DMFC technology will reach the consumer market soon enough, and then we'll see how it compares with batteries. I suspect lithium batteries will remain the most popular solution for laptops, and I'm sure handheld electronics will stick with batteries unless there's some breakthrough in fuel cells. But it'll be good to have another choice in portable power supplies.
Samsung announced this week that it has begun producing a 64GB flash drive for notebook computers.
Although 32GB flash drives have been on the market for several months, most users need more storage, especially in Vista-based notebooks. I think the new 64GB drives will find a much larger market.
Samsung didn't announce a price for the new drive, but 32GB drives have been selling for around $500 as an upgrade for a few notebook models from Dell and other OEMs. The new drives will probably decline to that price over the next several months.
Meanwhile, of course, conventional hard disks continue to expand in capacity. I just ordered a 250GB hard disk for my MacBook Pro; it was only $279 including an external drive enclosure I can use with my old HD. (I'll report on the results when I install it.) An 80GB notebook hard disk is now a sub-$60 item.
So why go with a flash drive if it's seven to ten times as expensive for the same capacity?
Two reasons may make up the difference: power and reliability. A flash drive will give significantly better battery life; the smaller and more sophisticated the notebook, the greater the improvement. A flash drive will also survive accidents that would kill a hard disk.
But I reject the conventional notion that a customer should have to choose between a flash drive and a hard disk. I'd like to see notebook computers with room for one of each. Not just the hybrid hard disks now being offered by Samsung (which Microsoft refers to as ReadyDrive technology) or Microsoft's related ReadyBoost technology, where a flash drive acts as a cache for a hard disk.
Both approaches are good ideas, getting significant benefits from small amounts of flash memory-- a gigabyte or less-- but I'd like to see notebooks where a large flash drive is made available as an independent logical drive on the system. This way, files can be distributed according to their access characteristics. Files that see a lot of random reads-- the OS, applications, email data files, the virtual-memory swap file, etc.-- can be put on the flash drive. The hard disk would only need to spin up to transfer large media files, make backups of modified files from the flash drive, and other accesses that involve long sequential reads and writes.
This approach isn't practical with just a gig or two of flash, but with 32GB or 64GB, it starts making good sense. Within a few years, that much flash will be fairly affordable, especially for high-end notebooks. It'll also be smaller, allowing OEMs to tuck it into any open area inside the machine.
The combination will provide the best of both flash and rotating storage-- high capacity, low power, and high reliability. There's a cost factor to consider, but premium notebooks sell very well; there are many customers who will pay for such tangible benefits.
I know I would!
- prev
- 1
- next
