Intel believes its process technology could eventually give it the lead in low power processing.
LONDON – Though ARM may currently have an advantage in low power processing, Intel believes it could eventually take the lead if it maintains its current pace of advancement in process technology.
Speaking to the firm’s executive vice president and head of architecture David (Dadi) Perlmutter recently, EE Times learned that Intel is pushing ahead with its sub 20-nm and 14-nm plans, while the ARM ecosystem struggles to find a business model beyond 22 nm.
[Get a 10% discount on ARM TechCon 2012 conference passes by using promo code EDIT. Click here to learn about the show and register.]
Intel is biding its time. The semi giant appears quite prepared to wait a couple of generations until sub 20-nm nodes are breached before bringing the battle of low power processing to ARM and its mobile partners.
Indeed, Intel claims to have line of sight all the way to a 7-nm process, with all the firm’s CPUs becoming SoCs as of next year. Haswell will be Intel’s 22-nm SoC for tablets and Ultrabooks, while Merrifield will be the company’s SoC for tablets and smartphones.
After that, 14 nm is expected to be Intel's killer node with the most dramatic power improvements and integrated baseband--although the firm has yet to officially announce it.
Meanwhile, the ARM ecosystem has a rather more challenging road ahead of it, with low cost continuing to present a challenge for foundry supply, yield and materials science. ARM is seeing Moore's Law slowing down due to a lack of foundry spending and a deficit in R&D capabilities.
Even at the 28-nm, the ARM ecosystem is feeling the squeeze. Until very recently only one foundry (TSMC) could yield 28-nm chips, with Globalfoundries just starting to produce 28-nm in volume, and the entire industry is under-supplied.
With Intel pushing ahead to introduce FinFETs at the 22-nm and hoping to use EUV lithography at 14 nm and below, ARM faces an even bigger crisis of competition.
Meanwhile, Intel’s design teams work hand in hand with the the company's fabs, in one unified effort. “We always work very closely with the fab,” said Perlmutter. “Each time we put a new big challenge in front of the giant team,” he said, adding that every time the focus centered on new and different aspects of the technology.
“If you develop technology, every day you bounce into a problem which may seem to be a showstopper,” he explained, discussing the various challenges the teams had overcome hitting the targets for Haswell’s design.
“What looked to be hugely complicated 10 years ago looks like a piece of cake now. We have to overcome new things every time around, so we have an 'aha' moment every day, every minute in every single Intel product,” he said.
Is the paragraph below what Dadi said, or is this E.E. Times' conclusion about the non-Intel camp? It does seem like investments are fairly robust at TSMC, GlobalFoundries, and Samsung.
"Meanwhile, the ARM ecosystem has a rather more challenging road ahead of it, with low cost continuing to present a challenge for foundry supply, yield and materials science. ARM is seeing Moore's Law slowing down in its foundries, due to a lack of investment and also to a deficit in R&D capabilities."
It is neither. Analysts have been predicting difficulties in sub 20-nm for ARM for a long time, and the problems at 28-nm prove that the road ahead will only be more difficult. Yes, there is a significant investment going into foundries like TSMC and GlobalFoundries, but even representatives from those fabs don't like to argue about the business model sub 20-nm. It's simply going to get much more difficult and costly. Hardly a conclusion, but it is a deduction. Everything that Dadi said is either attributed to him or is in quotations.
least advanced intel chip being sold is still 2x the estimated cost reported on this site for A6 chip in iPhone. Almost every one of intel chip is more like 10x or greater cost of A6. Current Price list is at intel.com for all to see.
So intel will make some magic architecture at most advanced 14nm node, in brand new D1X and fab42...and will cut chip price in half?? I don't think so. Intel has enjoyed and abused monopoly for years, I do not see how intel can compete in volume against entrenched mobile foundry like TSMC, UMC, Samsung, etc. Anyway, current ARM ecosystem is much more consumer friendly than this one you suggest where intel "crush" ARM
Intel has always been ahead by 3-4 years of all those foundries you mentioned, and, if anything the distance will grow. As far as your "do not see"...just watch for the next couple of years and you will.
I'm not sure where you're looking but costs of Atom chips to Intel are 10x cheaper than their PC chips because they're so much smaller. 22nm chips are cheaper than 32nm and 14nm are cheaper than 22nm versions. Of course, the performances go up and the power down each new generation too. 14nm will be a killer chip for Intel that ARM won't be able to match
Intel is betting on EUV never happening in the future. It's a reasonable bet. Granted that Intel tends to introduce nodes earlier than foundries, it stands to lose if EUV suddenly becomes available, say between its 14 nm node and some lagging foundries' 14 nm node. But it will suffer the same inevitable costs of two generations of double patterning and two generations of quadruple patterning.
The recent announcement of a 100x higher energy EUV source by a Washington professor and ASML announcing a firmer EUV timeline ( and implied might be purchase of the higher power Zpinch plasma EUV source ) hints that all of a sudden EUV might gain viability it has been missing for years. With the higher power plasma source there might be all kinds of unexpected good things on a nearing horizon. And it will likely invigorate major players in the industry for faster node milestones
This part seems weird:
"With Intel pushing ahead to introduce FinFET to foundries and EUV at 14nm and under, ARM faces an even bigger crisis of competition.
Meanwhile, Intel’s design teams work hand in hand with the foundries, in one unified effort."
Why would Intel work with foundries? Foundries don't have x86-based customers.
Dadi must be having a laugh! It's Intel that will hit a wall with its obtuse pursuit of performance and power efficiency with x86 on top, banking solely on advances in the physical layer. These are at best linear in nature, whereas higher level optimisations can deliver much faster improvement rates. Moreover, are we to believe that Intel's investment in advanced technology nodes cannot be matched by all the rest of the semiconductor industry?? That would be arrogant and foolish. The reality is that the world has been heading towards open platforms and networks rather than oligopolies, for sometime now. It is this trend that has delivered unprecedented levels of higher performance and lower power AT LOW COST. Consumers realise that and they are not going back to the Wintel duopoly.
PS. I am surprised some people are still buying Intel's propaganda!
Well, you seem to have a strong opinion about the world moving toward open systems, etc. However, the facts do not seem to support your opinion. Just look at Apple and your theory about open systems. No one gives a damn about whether a new product comes from an "open system" or from a close done, the only question is value (more precisely perceived value) delivered to the customer.
Analysts say that Intel now has about a 2 year lead in process technology. This is a huge advantage. And as the lithography & equipment complexity & cost goes up with each new generation it's more difficult to keep up and even afford to build new fabs. Intels new 14nm fabs are about $7B each and the list of those who can afford it has gotten much smaller. Today, only Intel, Samsung & TSMC can afford them. This is one reason why Intel is getting into the foundry business too.
Gullible, unqualified "reporters" seem to be swallowing every bit of propaganda. 22 nm Tri junction was supposed to be the break through node to end ARM domination of SoCs. Why is the power savings for the latest 22 nm CPU not up to expectation ?
Come on EE Times hire some sincere reporters or you will soon lose credibility !
Yes. What happen to 22nm FinFET being Intel's savior?
Now Dadi is saying
"Intel is biding its time. The semi giant appears quite prepared to wait a couple of generations until sub 20-nm nodes are breached before bringing the battle of low power processing to ARM and its mobile partners."
This means 22nm is not competitive on cost, power, or timing !....
and maybe a solution now 2 years.
sylviebarak, can you follow-up on what happen to 22nm finfet propaganda ?
Intel cannot afford to implement commercially beyond 20nm.
Likely U.S. Government & Intel are concealing Corp. is looted & Bankrupt in future terms.
Sources are 5; 1) infiltration by cartel organized network crime; 2) banked cost constructing surplus barrier limiting competitors beyond 32 nm; 3rd, theft from stockholder’s to administer cost of channel price fix tie; 4th, theft from end customers charged price fix in invoice plus monopoly overcharges. Likely PC end buyers will see some recovery. 5th industrial theft from processors dumped at price less than cost. 5 categories record $178,000,000,000 unreported cost burden on Intel.
Where are present cost burdens? 1st, inventories of Xeon Westmere EX, Aarondale, Sandy Bridge Desktop and Mobile. Issues are 3; first, surplus processors banked in channels on deferred revenue recognition; 2nd, completed systems stalled in channels showing now what is occurring inside Intel; 3rd, surplus banked goods; processors & systems, repurposed for apps that will compete against Intel future product while placing burden on industry in total.
Finally, one must ask why Intel is sustaining price on prior runs instead of flushing at cost to recover investment burden?
For Intel to dump inventories means twin tower effect on supply chain. As processor margins eliminated AMD becomes system house to capture remaining downstream producer values. Same for other processor design producers impacted by surplus raining down.
Collapse flattens Intel’s long time channel & contract manufacturers finally take over for certain.
Final question was price hold a hidden condition in Docket 9341 consent agreement? It’s time for regulatory mechanisms in this country that are supposed to police monopolization, cartel and investment fraud, including at Intel, to do that job of be replaced with administrations that will do that job.
I am not an Intel fan at all, but considering this as a propaganda is a stretch. It still owns the best fab technology in the world. Laptops and servers are not going away or switching to ARMs any time soon. Mobile attempts definitely were a failure for Intel, but if they can learn from failures and fix their lack of understanding of the mobile market needs, the fab technology and capacities can give them an advantage. Besides performance and power there other aspects of mobile AP which can put an end to its life if overlooked at the architecture phase.
Which Segment will intel plan to crush ARM beyond 20nm? PC? It would be low digit or no growth this and next year. Tablet? Too crowded and Intel is lack of infrastructure advantage. I bet it would be in server market. ARM plans to jump into for a while.
Intel is a manufacturer, and ARM is merely a license, so this is comparing apples and oranges.
The difference is not the cores - end customers do not care about cores.
Look at Apple, they managed to CHANGE cores entirely, and barely missed a beat, and support multiple cores now.
The difference is in Manufacturing Capability, and there, intel really is the 600lb gorilla.
A smarter question is could Apple team with intel, on a new design ?
Apple adopts Intel CPU in iMac already. It would be interesting to see intel replacing Samsung to take Apple's foundry order in 22nm SOC. I remembered TSMC licensed Intel's Atom core for mobile chip manufacturing about 2-3 years ago. Is there any progress update till now?
what's not addressed in this article is that mfring node is not the only competitive dimension. The world does not start and stop solely with transistor current consumption. Architecture optimized for use cases can be a huge differentiator and more than make up for the gap in transistor vs transistor efficiency.
Kresearch: Intel attempted to get TSMC to fabricate ATOM at RISK production phase. However the two fabricator's process cost structures are similar. So at the time TSMC could not afford to fabricate Atom eiteher.
I believe Atom costs have come down into the
$0.25 per sq mm dice area but that is on Intel financial's check not economic assessment of the ATOM production long run.
Kresearch: More to consider . . . what is certain on Intel price sheet ATOM average price is up from RISK production phase on diriviative offerings. The question is where is average weighed price.
Just seen the launch of two very competitive medfield based intel phones that outperform similar pricec arm devices and have aversge mobile battery performance. Here in Europe. And they were just proof of concept.
Now chipzilla has the making of a true mobile device proc., 2+ yeafrs lead in process technology, gazillions cash on the balance sheet and more fab capability than tsmc and a few others put together.
We believe you! :-)
PS. This is not a matter of Intel vs. ARM folks. This is about long-term lower cost higher performance and open platforms vs. a return to the oligopolies of the past. The choice is clear...
For the time being, 28 nm foundry has closer density to Intel 22 nm than 32 nm. But the addition of the fins could offset any cost benefit initially. Foundry 20 nm vs. Intel 14 nm would be where process efficiency is really important.
Mike Bruzzone is fm IBM PC Board add on’s; Turbo, Memory, Network Card, 1st 2.5 D PC AT graphics card, 33 PC introductions including fastest 386 33 class workstation on memory write back; co host 1990 SIGGRAPH, first PC Standard MP C&T Corallary Chip set, first not Intel 387 math coprocessor, first not Intel 486SLC/486DLC, 486S Intel substitute microprocessors, first Nx586 replacement platform, first 16 bit bus 32 bit core Mini RISC, first 32 bit performance Mini RISC SOC, first I frame editor, first fast486120, some Alpha&NT, first low power C6 desktop & first less than $1,000 PC. Current research tracks include economics of fabrication; member Silicon Valley Round Table, intersection x86 & ARM, HPC/GPU/APU, resistive RAMs & advanced non volatile memory structures.
Mike Bruzzone is an Intel competitive strategist specializing in not Intel product commercialization who is invited discovery technical assistant by Federal Trade Commission Bureau of Competition for FTC v Intel Docket 9288, lettered to work report Assistant AG Antitrust Division of the State of California Department of Justice for defining Sherman Act Section One violation, a EUCC domestic U.S point of contact Monti/Kroes/Martinez Commissions, SEC recognized Relator by letter, Federal False Claims Act Original Source by letter U.S. Attorney Northern California District, recognized Relator Intel monopoly procurement theft in fifty States, four territories and the District of Columbia, invited discovery technical assistant FTC Chairman Referral back into Bureau of Competition Docket 9341, FBI original source of Intel Dealing Cartel 1996.
Criminal sub groups operating in Intel have in past positioned detractors as mentally ill, including their own employees up high in the enterprise who question Intel’s invented reality in pursuit of error correction, remedies, maintenance of democratic capitalism & industrial social responsibilities. For consultancy inquire email@example.com
Before, when buying a PC, the CPU (Intel or AMD or whatever) was the key thing. Now, when buying a tablet or smartphone, most will not care if it's an ARM or Atom processor inside. They would only care about the final product design by Apple, Samsung, or whomever.
One hope ARM has is that IBM's design for yield tools and SEM/TEM capabilities are helping GF get their Malta fab up and running faster. And ARM-GF have new long term partnership lately. IBM may have learned to be more agile and sneaky that in distant past when they let intel empower the cloning of their PC's by using the easy-to-rev-eng BIOS ROM chip. But its all about execution, so we will see over next 3 years. Maybe Warren Buffet realized this was a game he did not understand? Risk is very high next 3 years.
Human civilization continues to advance due to the species' unique ability to build on knowledge. Today's engineers are smarter (read Scientific American Sept 2012, p44). Dadi Perlmutter is as brillian an engineer and manager as you will ever meet, which explains his current responsibilities. Dadi "gets it" and has for more than 15 years. He also gets people, has the humilty and wisdom to attract and utilize the best among us.
- Kurt Robinson