How will AMD answer the challenge posed by Haswell?

[LogOver]

And can you get any of them in the US yet?

The reason you can't get it in US is the lack of LTE support. Intel is going to release new LTE chipset before the end of 2012. BTW, the lack of LTE support is the reason why US HTC One X & Samsung SGS3 models got Qualcomm S4 instead of Tegra3 & Exynos.

 

[NTMBK]

I dont know what you can get in the US. You can get Atom phones in Europe and Asia. And Asia is bigger than Americas+Europe combined.

The US is either the biggest or second biggest market (depending on whose estimates you go by of how fast China is growing), and not having a single Intel phone in that market is clearly a major failing. And that's still only three models of smartphone. Hardly stellar.

 

[cliftonite]

What is the ETA for haswell desktop chips?

 

[ShintaiDK]

The US is either the biggest or second biggest market (depending on whose estimates you go by of how fast China is growing), and not having a single Intel phone in that market is clearly a major failing. And that's still only three models of smartphone. Hardly stellar.

The US for example accounts for what of Intels revenue? 10%? Sub 10%?

Not to mention US telco infrastructure is...wayyyy behind. And that you have to make special models for it.

 

[ShintaiDK]

What is the ETA for haswell desktop chips?

Quadcores in march.

 

[pelov]

The US for example accounts for what of Intels revenue? 10%? Sub 10%?

Not to mention US telco infrastructure is...wayyyy behind.

So Europe is behind the US in Intel's revenue by geography, yet that's why we have no Intel phones? Intel made a great portion of their cash in China and India, both nations known around the world for their telco infrastructure?

Uh... no.

 

[NTMBK]

The US for example accounts for what of Intels revenue? 10%? Sub 10%?

21%, apparently:

Given that the "Asia-Pacific" band seems to cover China, Australia, Indonesia, Hong Kong, Taiwan, and Korea, it's not that big. But the US does seem to be a smaller portion of their revenue than I first guessed.

 

[ShintaiDK]

So Europe is behind the US in Intel's revenue by geography, yet that's why we have no Intel phones? Intel made a great portion of their cash in China and India, both nations known around the world for their telco infrastructure?

Uh... no.

Europe is 13%.

There are Atoms phones in europe.

China, Japan, South Korea etc got great networks.

And again, US needs their own version...just like with HTC X, Galaxy S3 etc.

 

[ShintaiDK]

21%, apparently:

Given that the "Asia-Pacific" band seems to cover China, Australia, Indonesia, Hong Kong, Taiwan, and Korea, it's not that big. But the US does seem to be a smaller portion of their revenue than I first guessed.

Americas=North+South America.

Little extra help for you:
http://en.wikipedia.org/wiki/Americas

Over 925million population.

 

[pelov]

Americas=North+South America.

In the same way you considered Japan to be part of their Pacific Asian region.

 

[NTMBK]

Americas=North+South America.

Little extra help for you:
http://en.wikipedia.org/wiki/Americas

Over 925million population.

D'oh. Good catch.

It's still only three phones though.

 

[ShintaiDK]

In the same way you considered Japan to be part of their Pacific Asian region.

I think you should reread the post again. I never said Japan was part of Intels Asia-Pacific breakdown. I just said Asia was bigger than Americas+Europe.

Help for you:
http://forums.anandtech.com/showpost.php?p=34067051&postcount=100

 

[Nemesis 1]

Motorola Razr I, Intel's reference phone (rebranded by several different countries), and ZTE Grand X IN; 3 hardly counts as "more and more". And can you get any of them in the US yet?

WHY does it hardly count? Who the heck are you? It makes no differance if you can get one in the USA at this time . The point is Intel forced the door open and they did it with a good product. Last year how many smartphones did intel have? ZERO! So 1/2/ or3 counts as more unless you don't understand the zero concept.

 

[Nemesis 1]

D'oh. Good catch.

It's still only three phones though.

I don't know how your getting away with posting such things but an increase from zero is an increase and 2 is many more than 0. I have no horse in this race . I also do not own a smart phone . But next year time permitting I will have a smartphone with OoO x86 dual core at 22nm. It will be the best performing product on the market with excellant battery life..

 

[NTMBK]

I also do not own a smart phone . But next year time permitting I will have a smartphone with OoO x86 dual core at 22nm. It will be the best performing product on the market with excellant battery life..

And I'll have a pet unicorn. Intel are planning on shipping Valleyview silicon in Q4'13, so it's doubtful it'll be in any released phone until Q1'14. And I sincerely doubt they'll get the 2 core parts into a smartphone thermal envelope.

 

[Nemesis 1]

And I'll have a pet unicorn. Intel are planning on shipping Valleyview silicon in Q4'13, so it's doubtful it'll be in any released phone until Q1'14. And I sincerely doubt they'll get the 2 core parts into a smartphone thermal envelope.

Why don't you look at Intel roadmaps . So you got a link for valley view release . from Intel? You better have one . You also need to look at Intels roadmaps .. Your getting away with things others here do not , Pure lies from your mouth and YOU know your lieing because you don't know the truth . We can onlly go by what intel says . What your saying and whast intel says is 2 differant things. I was banned for going after the AMD BD disinformation guru. But he wasn't . I understand how it works here. Theres enough double standards here to fill a dictionary

 

[NTMBK]

Why don't you look at Intel roadmaps . So you got a link for valley view release . from Intel? You better have one . You also need to look at Intels roadmaps .. Your getting away with things others here do not , Pure lies from your mouth and YOU know your lieing because you don't know the truth . We can onlly go by what intel says . What your saying and whast intel says is 2 differant things. I was banned for going after the AMD BD disinformation guru. But he wasn't . I understand how it works here. Theres enough double standards here to fill a dictionary

Right there, at the bottom of the slide, PRQ target Q4'13.

 

[pelov]

Why don't you look at Intel roadmaps . So you got a link for valley view release . from Intel? You better have one . You also need to look at Intels roadmaps .. Your getting away with things others here do not , Pure lies from your mouth and YOU know your lieing because you don't know the truth . We can onlly go by what intel says . What your saying and whast intel says is 2 differant things. I was banned for going after the AMD BD disinformation guru. But he wasn't . I understand how it works here. Theres enough double standards here to fill a dictionary

Or instead of calling him a liar you can search Anandtech and find the answer yourself.

The timing on Valleyview and the Bay Trail platform appears to be late 2013 and into 2014. This seems a bit late, but the netbook/nettop platform isn't really begging for a faster refresh. What really matters is how quickly Intel will bring Valleyview to Windows 8 tablets and a similar setup to Intel's smartphone SoCs. Intel's current roadmaps tell us that we'll see Silvermont based smartphone SoCs in 2013, likely in the latter part of the year. I am hoping for sooner than Q4 2013 if Intel wants to be taken seriously in the smartphone space however. As of now it's efforts have been good, but distinctly not Intel-like.

And it turns out he was right.

 

[podspi]

Microsoft already has an ARM version of Office on their winRT tablets.

And that's what I mean by uninformed opinions. Never mind that I mentioned it above in my posts that you chose to ignore, though you obviously weren't the only one. Quite a bit of that going on around here...

You do know how limited that office version is, right? Something again going back to the "limited processing power" of ARM.

Not to mention WinRT is almost stillborn. Almost everyone picked the x86 version for tablets.

Windows 8 (x86) and WinRT (ARM) have not yet been released. Microsoft does NOT already have an ARM version of Office, and nobody has already picked the x86 version. It feels like this is a debate from two different versions of the future.

BTW, while I am skeptical that ARM is going to push out x86, it is clear that Intel is worried about ARM, with good reason. Any large collection of software that isn't x86 compatible represents an existential threat to Intel's practical monopoly.

And for the record, if the WinRT tablet ends up being $300 ~ $350 as rumored, I plan to pick one up.

Intels Medfield has proven all the naysayers wrong. The OoO Smartphone intel has out in 2013 will be the best performing and have great battery life . So many said X86 will never go in smart phones . NOT true as medfield proves . AT 22nm in the smartphone Intel will be the performance and likely the efficiency leader also. Intel coming from top down seems to be working better than ARMS tring to enter notebook space from the bottom up. . Tablets are differant than phones Performance will steal win with tablets

I do agree the 'x86 isn't efficient enough for phones' refrain is getting tired, but I haven't really heard anybody say that. I think it is pretty clear that Intel's first attempt is poor (performance-wise), but I think some of that has to do with the fact that it emulates some ARM code (I assume they didn't use specially-compiled x86 versions of the benchmarks).

22nm Atoms are when we see the products that are really interesting, they can't get here soon enough. I am kind of hoping Intel forces a 'conroe moment' in smartphones. Would be very exciting...


The thing that really worries me is binary compatibility. I know it isn't great for performance, but I hope moving forward app developers shy away from assembly so that we don't have to worry about the CPU architecture used.

Veering a bit back on topic, I think AMD has some great opportunities available to them. Their willingness to integrate ARM and x86 brings up the possibility of hybrid chips. This might be hard to implement (without Microsoft's help), but a hybrid Jaguar+ARM chip running WinRT, with a 'Win8 Mode' that can run x86-code on-demand (or when plugged in) would be amazing.

 

[Nemesis 1]

Intel's current roadmaps tell us that we'll see Silvermont based smartphone SoCs in 2013, likely in the latter part of the year. I am hoping for sooner than Q4 2013 if Intel wants to be taken seriously in the smartphone space however. As of now it's efforts have been good, but distinctly not Intel-like.

Are we confussing tablets with SoC cellphones. Medfield is out now been out awhile now . Yet the tablet is not out until the 26th of oct . and that is x86 only for windows 8. From your quote it seems andy agrees with me . I llok for a new phone from intel in 2013 @22nm. and 2014 @14nm. Intel did say they were picking up the release of new atom chips . Maybe you should read anandas article about the uptick on atom

 

[NTMBK]

Are we confussing tablets with SoC cellphones. Medfield is out now been out awhile now . Yet the tablet is not out until the 26th of oct . and that is x86 only for windows 8. From your quote it seems andy agrees with me . I llok for a new phone from intel in 2013 @22nm. and 2014 @14nm. Intel did say they were picking up the release of new atom chips . Maybe you should read anandas article about the uptick on atom

Given that he's saying he hopes for it be before Q42013, that implies that he suspects that it won't be.

I am hoping for sooner than Q4 2013 if Intel wants to be taken seriously in the smartphone space however

 

[pelov]

But next year time permitting I will have a smartphone with OoO x86 dual core at 22nm. It will be the best performing product on the market with excellant battery life..

No, because if you look at that slide you'll clearly see the product is slated for Q4 '12 or early '13, and more likely early '13.

As for the pricing issue

Intel, facing notebook players' demand of supplying Ultrabook CPUs at a price 50% lower than original to help increase their profitability, rejected the proposal and is only willing to provide 20% discounts to first-tier notebook players, according to sources from notebook players.

The sources pointed out that Intel's insistence over the high CPU price to maintain its leading position in the supply chain may strongly reduce the effort of the company's counter against its competitors and affect traditional notebook development in the future.

Intel's Oak Trail platform, paired Atom Z670 CPU (US$75) with SM35 chipsets (US$20) for tablet PC machine, is priced at US$95, already accounting for about 40% of the total cost of a tablet PC, even with a 70-80% discount, the platform is still far less attractive than Nvidia's Tegra 2 at around US$20. Although players such as Asustek Computer and Acer have launched models with the platform for the enterprise market, their machines' high price still significantly limit their sales, the sources noted.

As for Ultrabook CPUs, Intel is only willing to provide marketing subsides and 20% discount to the first-tier players, reducing the Core i7-2677 to US$317, Core i7-2637 to US$289 and Core i5-2557 to US$250.

As for Intel's insistence, the sources believe that Intel is concerned that once it agrees to reduce the price, the company may have difficulties to maintain gross margins in the 60% range and even after passing the crisis, the company may have difficulty in maintaining its pricing. Even with Intel able to maintain a high gross margin through its server platform, expecting Intel to drop CPU prices may be difficult to achieve, the sources added.

http://www.digitimes.com/news/a20110815PD216.html

How in the world do they expect to compete with those prices? It's no secret Ultrabooks are way too expensive for what you're getting (which is essentially just a thinner, worse performing, worse battery life and crappier hardware laptop that might get 1/2 the performance for twice the price), it seems they're also applying the same margins to their Atoms in tablets.

Like I said, Intel doesn't hold as much sway over OEMs in the mobile segment as it does on desktops. If they charge too much, they'll simply buy ARM.

 

[Nemesis 1]

And it turns out he was right.

Your going to have to explain to me how he is right when the roadmaps says he isn't

 

[Nemesis 1]

No, because if you look at that slide you'll clearly see the product is slated for Q4 '12 or early '13, and more likely early '13.

As for the pricing issue



http://www.digitimes.com/news/a20110815PD216.html

How in the world do they expect to compete with those prices? It's no secret Ultrabooks are way too expensive for what you're getting (which is essentially just a thinner, worse performing, worse battery life and crappier hardware laptop that might get 1/2 the performance for twice the price), it seems they're also applying the same margins to their Atoms in tablets.

Like I said, Intel doesn't hold as much sway over OEMs in the mobile segment as it does on desktops. If they charge too much, they'll simply buy ARM.

BS This is misdirection . Medfield is out now for Phones were is the tablet version ???? Its not here yet. From your graph. 1 core 2 core and 4 core . The 4 cores are tablet and up only . the 2 core is highperformance phone the single core is intels entree level chip. stronger than anything out today but still intels base model.

 

[pelov]

BS This is misdirection

Yes! Misdirection! Conspiracy! Intel would never tack on extremely high margins on their microprocessors and SoCs! Unheard of!