How long can intel maintain such a large process lead over rivals?

[Arachnotronic]

Oh no, that one again. Yes, for people who are used to paying $200, $300+ for a CPU the 2500K is a good deal. But to repeat myself again, there are a lot of people who have never paid more than $100 for a CPU. I know hardware forums are full of people who buy every new things, but...

For instance since the days of the Celeron 300A my budget has and thinking has been constant: I spend around roughly £50 for a CPU and the same for a mobo and always overclock 30-50%, but I only upgrade if I can get twice my previous performance.

Thing is atm, I've had my 45nm Pentium E5300 @ 3.5GHz for a good while because there is nothing new which offers me that kind of value. About the closest would have been the Thuban's but even they were above that budget and I'd really like better idle power.

I can afford to pay $300 every year for a new CPU. Especially given the INSANE resale value of older Intel chips, it's really just a new mobo + a small delta for a new CPU. But hey, I guess I just don't think $300 is all that much for the best client CPU performance that money can buy.

 

[Arzachel]

I can afford to pay $300 every year for a new CPU. Especially given the INSANE resale value of older Intel chips, it's really just a new mobo + a small delta for a new CPU. But hey, I guess I just don't think $300 is all that much for the best client CPU performance that money can buy.

"Can afford" and "find it a reasonable use of money" are two hugely different things. Would you spend $200 to upgrade from SB to IB given that the performance delta is <10%?

 

[Idontcare]

Right. Qualcomm, NVIDIA, TI, and Samsung shareholders would be happy with break-even operation

It'd also be stupid for Intel to hit a wall with their process development ; saving gross margins now at the expense of keeping a technological lead will only result in Intel becoming an irrelevant player in all of their segments.

Since you mentioned TI, I can give you a real-world example.

In 2004/2005 Intel announced they were going into the HDTV market, direct competition with TI's DLP for HDTV. Only thing was we knew the margin story and knew there was no way Intel had the ability to compete if GM's fell below 40%.

Fast forward a year later, GM's had fallen below 40% and Intel announced they were scrapping all further development to get into HDTV.

Most businesses have a cost structure that enables them to operate at break-even so long as their gross margins are ~30%. No business is happy doing it, but shareholders (institutional, not talking the float traded by day-traders) are more forgiving of non-Intel companies having fluctuating GM's versus Intel.

Intel is simply held to a higher standard, and it is to their own detriment when it comes to limiting the markets they can diversify into. It was a limitation when it came to HDTV, and to their original plans for breaking into the handset market, as well as their plans for discreet GPU products.

 

[AtenRa]

I agree that everything has pros and cons, but some things have more pros than cons and thus become industry standard, and others are more balanced and become trade-offs or design choices.

SOI falls into the latter category. You can't really use that as an example of where Intel lags in process technology, because they had the ability to implement it and chose not to do so.

I was being sarcastic when i said Intel is behind because they didnt used SOI.

Can you give me any info to support that SOI is no longer viable?? It will be even more difficult for GloFo and TSMC at 22/20nm without SOI than 28nm bulk.

I believe that SOI (Especially FD) is cheaper to research and faster to implement for GloFo and TSMC than FinFETs. At 22/20nm they both will have lots of leakage problems as processes are shrinking bellow the 32/28nm and only with SOI FD or FinFETs will be able to continue. I believe both GloFo and TSMC will use SOI FD for 22/20nm and perhaps FinFETS for 16/14nm and below.

 

[ViRGE]

as well as their plans for discreet GPU products.

Discrete, IDC, discrete. Discreet is how you handle your mistress.

 

[Obsoleet]

To answer the OP- they'll always maintain the lead in x86.
The problem is that this market is increasingly irrelevant. The world is moving to ARM and GPU/APUs. Even businesses (high end and low end).
No one can stop Apple's march and Windows is always a day late and dollar short.

Intel is working on process leadership because they know x86 won't likely stop ARM at this point. That said, it's a good idea- it's the way to keep Intel relevant in the future.

 

[blastingcap]

From Dow Jones last week:

By Shara Tibken

Of DOW JONES NEWSWIRES

NEW YORK (Dow Jones)--The chief executive of ARM Holdings PLC (ARMH, ARM.LN) expects companies making chips based on its technology to capture a bigger percentage of the notebook PC market than rival Intel Corp. (INTC) will take in smartphones.

The two companies have been trying to expand in each other's markets, with ARM licensees making a push into computers and servers and Intel focusing on smartphones and tablets. ARM processors, used in virtually all mobile devices, have typically had an energy consumption advantage over Intel chips, while Intel's semiconductors, used in the majority of PCs, have traditionally been more powerful.

Warren East, in an interview Thursday with Dow Jones, said he expects companies making ARM-based chips to capture about 10% to 20% of the notebook PC market by 2014 or 2015. Conversely, he expects Intel to control about 5% to 10% of the smartphone market within a few years.

East called Intel's first smartphone on the market, the XOLO X900, a "perfectly adequate smartphone," but said there are about 20 other chip suppliers also targeting the highly competitive market. That will make it difficult for any one company to dominate the industry like what has happened in traditional computing.

"It's going to be quite hard for Intel to be much more than just one of several players," East said. "But they'll be a perfectly credible player."

Intel hasn't provided its estimates for phone market share, but Chief Executive Paul Otellini said in a recent interview that the company expects to have a significant presence in the smartphone market and be a major player in mobile in a few years. He also said he expects the industry to consolidate, with only about two to three mainstream mobile chip providers ultimately existing.

Intel spokesman Jon Carvill on Thursday declined to comment further about market share projections.

ARM doesn't build chips itself but designs processor technology that it licenses to companies such as Qualcomm Inc. (QCOM) and Nvidia Corp. (NVDA). Such power-efficient processors are used in nearly all mobile devices, including those from Apple Inc. (AAPL). But Intel believes it has just about reached power-consumption parity with ARM-based chips, and the first smartphone using its Atom processor recently was released in India.

Intel competes with Advanced Micro Devices Inc. (AMD) in supplying processors to computers and servers, but the two will face competition from ARM-based suppliers later this year when Microsoft Corp. (MSFT) releases its newest version of Windows. One flavor of the operating system, dubbed Windows RT, is the first to be compatible with ARM chips.

East said Windows RT won't have all the functionality of Windows 8, which runs on x86 chips from Intel and AMD, because older applications won't work on the ARM-compatible operating system. He noted ARM-based chip makers won't be addressing or targeting all PC use cases.

"If you look at a lot of consumer PCs, people just want to run an Internet browser, an email package, some Office applications and Adobe [Systems Inc.'s (ADBE)] Photoshop or something like that, and not much else," he said. "Therefore, we can put ARM processors into the heart of PCs to target a lot of the use requirements."

East said the ARM-based chips used in PCs will be significantly cheaper than PC chips from Intel, something that will make them attractive to the computer manufacturers. Typically, smartphone chips cost about $20, he said, compared to the approximate $80 to $200 pricing for Intel's Core line of PC processors. The ARM-based chips in PCs will likely command a slight premium to the smartphone processors, East said.

"It's not as if anyone will pay a lot more money to put an ARM processor in a PC versus in a smartphone," he said. "But selling a chip for $25 instead of $20 is a massive, massive improvement in profitability for the smartphone chip provider."

While Intel's PC chips are more expensive, its Atom line--targeted at mobile devices and low-end PCs--are priced more in line with ARM chips.

Intel's Carvill said Intel expects to be "very price competitive with the ARM licensees" with its Atom chip in consumer devices such as smartphones, tablets and PCs that convert between a tablet and a notebook.

He added that for Windows, Intel chips also bring "tremendous capabilities and expertise in performance, compatibility and power efficiency that will differentiate platforms running Intel architecture from new entrants to the market."

ARM shares trading in New York recently slid 3% to $22.34, down 19% in the year-to-date. Intel, meanwhile, slipped a fraction to $26.39. Its shares are up 8.9% in 2012.

 

[Arkaign]

To answer the OP- they'll always maintain the lead in x86.
The problem is that this market is increasingly irrelevant. The world is moving to ARM and GPU/APUs. Even businesses (high end and low end).
No one can stop Apple's march and Windows is always a day late and dollar short.

Intel is working on process leadership because they know x86 won't likely stop ARM at this point. That said, it's a good idea- it's the way to keep Intel relevant in the future.

Wot?

ARM =! Apple

Windows is the de facto standard for business software, and even if Microsoft utterly fails at everything over the next decade, they will likely continue to dominate the business computing sphere.

On the consumer end, Microsoft will become less relevant imo, but for that matter so will Apple as the products get better and cheaper from the competition.

ARM is also utterly terrible for certain tasks while being very efficient and effective at others. It's just a side effect of RISC architechture.

I also wouldn't put too many beans in the GPU/APU basket, Intel is headed there pretty fast, but in the very areas you're talking about, it's ironically ARM-powered devices that are replacing the units (think of how many folks pick up iPads or Galaxy tablets and use them to run dumbshit games and such on them rather than get a real laptop with a real GPU or Fusion/Llano/etc).

 

[BiG K]

I strongly disagree.

First of all Intel havent had _any_ competition since july 2006.

Secondly, Intel is forced to innovation not by AMD, but by consumers....................

You have missed my point entirely. Without a competitor, when a consumer (whether Server, Gamer, Mainstream or Low Power) needs or wants to buy a new computer, they are forced to buy Intel. So Intel can have the same processor on sale for years, because the consumer can not go anywhere else for their processors. If the customer doesn't like it that the same processor is on sale for years, tough, where else can they go?

Intel hasn't had STRONG competition since ~2007 but they have had competition of sorts.

It's because of AMD that Intel went with the "Tick Tock" model SINCE 2007 when Intel were experiencing STRONG competition from AMD. Each tick is a die shrink and each tock is a new microartitechture. There is expected to be one tick or tock.

Intel has stepped up the game in response to AMD. Hence, if AMD did not exist, Intel would have no incentive to further improve products on a massive scale as the consumer would buy the same processor regardless. It's the same reason why Windows XP stayed around for so long. OSX was non-existant. As soon as OSX started becoming a threat for Microsoft, Windows Vista was rushed (and botched) so Microsoft could retain their monopoly.

 

[ShintaiDK]

Intel doesnt make money selling a few CPUs. They need to sell 100s of millions. Thats also the only way to pay for fabs.

Its simple business 101.

Intels biggest competitor is itself to satisfy customers so they actually buy the product.

So no, Intel cant keep the same CPU on sale for years.

Vista started a major overhaul of Windows. So no, the part about OSX aint true either. Not to mention that at the time of Vista. OSX marketshare was pretty stable, including a year after.

 

[frozentundra123456]

You have missed my point entirely. Without a competitor, when a consumer (whether Server, Gamer, Mainstream or Low Power) needs or wants to buy a new computer, they are forced to buy Intel. So Intel can have the same processor on sale for years, because the consumer can not go anywhere else for their processors. If the customer doesn't like it that the same processor is on sale for years, tough, where else can they go?

Intel hasn't had STRONG competition since ~2007 but they have had competition of sorts.

It's because of AMD that Intel went with the "Tick Tock" model SINCE 2007 when Intel were experiencing STRONG competition from AMD. Each tick is a die shrink and each tock is a new microartitechture. There is expected to be one tick or tock.

Intel has stepped up the game in response to AMD. Hence, if AMD did not exist, Intel would have no incentive to further improve products on a massive scale as the consumer would buy the same processor regardless. It's the same reason why Windows XP stayed around for so long. OSX was non-existant. As soon as OSX started becoming a threat for Microsoft, Windows Vista was rushed (and botched) so Microsoft could retain their monopoly.

don't want to start an off topic debate, but I still use vista at home and win 7 at work. I really can't say that I have any issues with either.

 

[2is]

Intel doesnt make money selling a few CPUs. They need to sell 100s of millions. Thats also the only way to pay for fabs.

Its simple business 101.

Intels biggest competitor is itself to satisfy customers so they actually buy the product.

So no, Intel cant keep the same CPU on sale for years.

Vista started a major overhaul of Windows. So no, the part about OSX aint true either.

Simple business 101 also dictates competition brings better products at lower price points. Intel may not keep selling the same CPU for years but they can make incremental upgrades and charge a large premium compared to what they are doing today.

 

[Denithor]

Intel has stepped up the game in response to AMD. Hence, if AMD did not exist, Intel would have no incentive to further improve products on a massive scale as the consumer would buy the same processor regardless.

Except that you forgot one key fact: it costs Intel less money to make a 22nm CPU than it does to make a 32nm CPU. At each process node shrink, their cost per part has gone down.

So it's obviously in their own best interest to continue shrinking those nodes.

Now, new architecture development, that's a bit trickier story. For the last couple generations Intel has simply been competing against their own prior generation rather than anything from AMD, VIA, etc. So that's why we've probably seen some stagnation on the CPU side as they turned their focus to improving the GPU instead (where AMD is still a real threat).

 

[ShintaiDK]

Simple business 101 also dictates competition brings better products at lower price points. Intel may not keep selling the same CPU for years but they can make incremental upgrades and charge a large premium compared to what they are doing today.

Intel is already in the sweetspot right now between margins, volume and innovation. That wont change. Change any of those and their revenue and profit will go down.

Problem is competition doesnt bring better products in all segments. That philosophy often only works on very basic concepts without much change. Or in segments thats relatively new and dynamic with low entry level costs.

Competition quickly turns to a problem when companies want to protect their situation/marketshare. Or cant afford to take a innovative risk, in fear of the competition will kill them off or take marketshare by not suffering the same risk.

Intel used 10billion$ on IA64 to frogleap performance for the future. AMD used what, 50mio$ on a slap on x64 to keep us locked for another 30 years. Who won? Do you call it innovation? I call it stagnation.

The goal of competition and capitalism is monopolies.

When you reach the endgame like AMD/Intel or nVidia/AMD you need a monopoly for a certain time to change the field. Either via the sole company taking on those risk that couldnt be taken in a competitive environment. Or by innovative startups or outside change to the actual segment.

A duopoly is more damaging than a monopoly. If you wanted competition we would need much more companies competiting. But the ROI aint possible for that.

 

[Ferzerp]

Simple business 101 also dictates competition brings better products at lower price points. Intel may not keep selling the same CPU for years but they can make incremental upgrades and charge a large premium compared to what they are doing today.

You ignore one key point.

The lifespan of a CPU is such that if this is all Intel did, since only new purchases would be from replacing old systems, their revenue would sharply decline. They profit much more by innovating and providing compelling upgrades.

 

[2is]

Intel is already in the sweetspot right now between margins, volume and innovation. That wont change. Change any of those and their revenue and profit will go down.

Problem is competition doesnt bring better products in all segments. That philosophy often only works on very basic concepts without much change. Or in segments thats relatively new and dynamic with low entry level costs.

Competition quickly turns to a problem when companies want to protect their situation/marketshare. Or cant afford to take a innovative risk, in fear of the competition will kill them off or take marketshare by not suffering the same risk.

Intel used 10billion$ on IA64 to frogleap performance for the future. AMD used what, 50mio$ on a slap on x64 to keep us locked for another 30 years. Who won? Do you call it innovation? I call it stagnation.

The goal of competition and capitalism is monopolies.

When you reach the endgame like AMD/Intel or nVidia/AMD you need a monopoly for a certain time to change the field. Either via the sole company taking on those risk that couldnt be taken in a competitive environment. Or by innovative startups or outside change to the actual segment.

A duopoly is more damaging than a monopoly. If you wanted competition we would need much more companies competiting. But the ROI aint possible for that.

How do you know intel has reached a perfect equilibrium? All I see is a whole lot of speculation not based on much of anything. AMD took a risk with BD arch, Intel with tri-gate. The problems you suggest could happen with competition aren't actually happening.

 

[ShintaiDK]

How do you know intel has reached a perfect equilibrium? All I see is a whole lot of speculation not based on much of anything. AMD took a risk with BD arch, Intel with tri-gate. The problems you suggest could happen with competition aren't actually happening.

I dont see Trigates as a risk. Its what everyone wants to do. And roughly just a linaer progression.

And BD was basicly all AMD had left. I dont actually consider it a risk when its everything the R&D budget allows for. Plus it wasnt a gamechanger.

So I dont see the point there. Intel and AMD is still just giving us more of the same. No change, no risk.

The last risk was taken by Intel/HP in 1989. With first product in 2001. Countered by 2002. So in my view the last risk was taken 23 years ago.

 

[Charles Kozierok]

I dont see Trigates as a risk.

I don't see why not.

It was a completely new technology and could have cost them billions of dollars and been a complete failure.

 

[ShintaiDK]

I don't see why not.

It was a completely new technology and could have cost them billions of dollars and been a complete failure.

In my view I dont see it that way. For me a risk needs to be a game changing event. Something to set a true difference. HKMG, trigates etc is just a natural step on the process manufactoring development. If Intel had gambled on graphene, photonics etc. Then yes.

Trigates is soemthing TSMC, Glofo, Samsung etc also wants to do. And been planned so for ages. Its not a unique new path.

 

[2is]

But intel actually did it. Theres a big difference between wanting to do something and being the first one to actually take the chance and do it. It was a risk, one that Intel took even with competition.

Essentially what I'm hearing is that if an argument goes against your competition/economic model, you'll just shun it as a natural progression or a last resort.

Real life examples > hypothetical possibilities.

 

[Charles Kozierok]

In my view I dont see it that way. For me a risk needs to be a game changing event. Something to set a true difference. HKMG, trigates etc is just a natural step on the process manufactoring development. If Intel had gambled on graphene, photonics etc. Then yes.

I think you're conflating innovation with risk. Something doesn't have to be completely innovative to be risky to undertake.

And I disagree with you even on the technical side. How can you view using a different dielectric material for transistor gates as being equivalent innovation-wise to completely changing their structure?

When is the last time the basic design of the transistor was changed successfully, going from drawing board to shipping competitive product in just a few years?

PS As 2is said, who cares about what others "want to do"? I want to flap my arms and fly to Paris, and so does everyone else, but the guy who figures out how to actually do it will be rich and famous -- and for good reason.

 

[ShintaiDK]

Doing something first isnt a risk in itself. Doing something the others didnt consider or didnt want to=risk.

 

[pelov]

Intel used 10billion$ on IA64 to frogleap performance for the future. AMD used what, 50mio$ on a slap on x64 to keep us locked for another 30 years. Who won? Do you call it innovation? I call it stagnation.

I'm not sure that's a good point. The IA64 would have suffered from the same issues with x86 thus I don't think it was ever a matter of us the consumer winning but rather whoever joined the bandwagon. It's always been Intel's red-headed stepchild that gets brought up on occasion by companies other than Intel who's been trying to their hardest to forget it ever existed in the first place

 

[Insomniator]

Intel just poured 18 Billion into R&D last year, I think they'll have the lead for a while.

 

[2is]

Doing something first isnt a risk in itself. Doing something the others didnt consider or didnt want to=risk.

And making up your own definition of what risk is doesn't change the actual definition of the word.