With Eyes on China, Intel Invests Billions In Mobile Ambitions

itwbennett writes The allure of mobile devices has led Intel to take some uncharacteristic moves, partnering with Chinese companies to build some smartphone and tablet chips, and relying on third parties to manufacture those chips. Intel is betting the partnerships will accelerate its business in China, where smartphone shipments are booming. But the company wants to regain complete control over manufacturing, and on Thursday said it was investing $1.6 billion over 15 years in a China plant for mobile chip development and manufacturing.

Big Mistake

[TheEyes]

And thus does Intel make the same mistake that hundreds of companies around the world have made before it: putting intellectual property in physical reach of the Chinese government. Fast forward five years and I'm almost certain we'll see Foxconn or some other Chinese company with ties to the Chinese government have a series of "research breakthroughs" that mysteriously parallel the exact technologies that Intel brought to its own plant, which is once again down for "inspection".

It's not like this sort of thing hasn't happened already.

Re:Big Mistake

[Mostly a lurker]

Perhaps. That is certainly a valid concern. However, the state of the art in this area is continually advancing very quickly. Just having an advanced fab in China does not mean that Chinese engineers are able to create the next generation chips and fabs. I think Intel's move is quite logical, and the danger of intellectual property theft not too serious in their case.

Re:

[cheesybagel]

You would be wrong. Lithography has not been advancing as fast as it used to be. For example the light sources have not improved at all for like a decade. Today the advances are made by fiddling with more complex masks and immersion fluids and crap like that.

China has already been trying to crack into this market with SMIC and the like. This is like adding fuel to the fire.

Re:

[Mostly a lurker]

If you believe advances in lithography are no longer occurring, you are clearly unfamiliar with the huge investments attempting to bring EUVL (extreme ultra violet lithography) to production fabs. However, in another sense, you are correct that other technologies (such as plasmonics) are going to be more important in developing the future 3D chips. You have not convinced me that Intel's move is going to make it much easier for China to become leaders in these areas than Micron's R&D centers in Shanghai

Re:

[lsatenstein]

Perhaps. That is certainly a valid concern. However, the state of the art in this area is continually advancing very quickly. Just having an advanced fab in China does not mean that Chinese engineers are able to create the next generation chips and fabs. I think Intel's move is quite logical, and the danger of intellectual property theft not too serious in their case.

Intel, like GM will gradually leave the USA for permenancy in China. Population 1.2 billion vs 350 million, Single party government vs democracy, better labour cost controls, lower overheads and government (universal) medicare.

Re:Big Mistake

[Trepidity]

It's definitely a risk, but I would be surprised if they're going to be using their latest-gen technologies there, partly for that reason. So Foxconn will be able to pilfer some older tech which is by that point less secret to begin with.

Intel already has one fab in China, in Dalian, but it's on a 65nm process, several generations behind the 14, 22, 32, and 45nm processes that they use [wikipedia.org] in their American fabs.

Re:Big Mistake

[Kjella]

It's definitely a risk, but I would be surprised if they're going to be using their latest-gen technologies there, partly for that reason.

You know, you could RTFA:

The chips made in the Chengdu factories won't be based on the latest process manufacturing technology, McGregor said, adding that Intel wants to protect its intellectual property and won't transfer its latest manufacturing process to China.

Re:

[PaddyM]

RTFA? Our ewe knew hear?

Re:

[cheesybagel]

There are trade sanctions against selling latest generation chip manufacturing tools to certain states including China. I do not think even Intel would try to challenge that.

Re:

[drolli]

I dont think that they will set up the EUV Production lines there.

Re:

[cheesybagel]

At this rate there will be EUV production lines nowhere.

Re:

[drolli]

i think there will be. Too big to fail....

Re:

[cheesybagel]

I have been hearing that for over a decade now. Plus they used to call this soft x-rays in the 1960s and it went nowhere back then as well.

Re:

[citizenr]

It was either that, or getting thrown out of China like Qualcomm. Intel was screwed no matter what, this way at least they get to slap a lot of intel inside stickers on cheap phones and getting piece of the market.

Re:

[Dorianny]

The US intelligence community has long warned that China's government has been waging a campaign of industrial and military espionage through hacking and even the White House asked them to "Stop hacking, pretty please". It really doesn't matter whether you have those secrets in China or somewhere else, the Chinese will get to them. I for one say hallelujah. For us the consumers, more competition is a good thing.

Re:

[tlhIngan]

The thing is, Intel pretty much buys the same equipment that everyone else has access to for fab technology.

The problem is the fab equipment is REALLY expensive and sourced from Japan, and it has to be that way in order to produce usable chips. With each wafer costing $1-3K each, sub-standard equipment brings the cost up VERY quickly.

In fact, this technology is pretty much open - even Intel has opened FinFETs to everyone, not because they're not cutting edge, but because if you can do it, you're already qui

What now?

[Opportunist]

Regain control of your IP and manufacturing OR invest in a Chinese manufacturing plant. Pick one!

Re:

[timeOday]

Guess it depends which you think is the bigger threat to Intel - technological inferiority to a peer competitor (first priority: protect IP), or technically inferior but "good enough" low-cost competition (first priority: ramp up low-cost production).

Both seem like serious threats. If you lose on volume long enough, you then also lose your technical edge. Just like Intel did to DEC, Sun, etc.

Hard times for a former monopolist

[JoeyRox]

Intel is learning that success doesn't come as easy in business segments where your tried-and-true monopolist practices don't apply.

Re:

[Kjella]

Meh, Intel wanted to buy enough market share that developers will make x86 Android builds and I think they succeeded. ASUS MeMO Pad, Acer Iconia, Dell Venue, Nokia N1 and a bunch of lesser brands now make x86 tablets which means you probably want to compile any native code for x86 not just ARM. That's their foot in the door, now that the contra revenue ends we'll see what they do with it.

Avoiding the ire of chinese government

[edxwelch]

Investing in Chinese companies is probably a tactic to get on the good side of the Chinese government.
Consider what Intel is doing in China: giving away subsidised SoC's that compete directly with local Chinese companies. That sort of anti-competive behavour normally attacts tariffs (a similiar sort of thing has happen before US Department of Commerce Ignores WTO, Imposes Preliminary Anti-Dumping Tariffs of 26-165% [cleantechnica.com]

Good Luck

[sdguero]

The high rate of employee turnover, politics, and finger pointing that goes on in the China office I work with (aprox 300 engineers in Beijing) is shocking. I don't know how a company can expect to be sucessful at doing real engineering and development work over there.

meaningless tripe

[Anonymous Coward]

Intel spends upwards of $7 billion to make a fab. What they are doing is not a fab.
Intel already has big assembly test manufacturing sites in China and has had it since last century. CD1 and CD6 are huge.
http://download.intel.com/newsroom/kits/22nm/pdfs/Global-Intel-Manufacturing_FactSheet.pdf

$2 billion in 15 years is $130 M/yr. The CEO, Krzanich, makes $9M/ year. This is only about 15x his salary. The Intel board take-home pay is larger than this "investment".

slashdot hype

[swell]

TFA article headline:
"Intel plunks down billions to expand in mobile market"
Slashdot article headline:
"With Eyes on China, Intel Invests Billions In Mobile Ambitions"

What's the difference? Every word in /. is in Caps!

What's the point of this? Is it a deliberate attempt to confuse and corrupt? Look at headlines in any respectable publication and you will find sensible, understandable headlines. Major words in caps, Trademarked words in caps ... Slashdot's insistence on every word starting in a capital letter