No one considers the i5 an SoC. Either you are trying to lie about it to make your case, or you are a fool. I haven't decided which.
But I don't suffer liars or fools mildly.
The i5 isn't an SoC, and for you stack your claims on that, makes everything wrong from that point forward.
Intel has barely entered the SoC game and all their SoCs are Atom based. Every Atom SoC in products this year was a combo 45nm/65nm part. Far from Intels state of the art.
But they will be moving to putting their SoCs on leading process going forward. 32nm in 2012 and 22nm in 2013.
Intel is behind in SoCs, but has had it's P4 moment and is waking up in SoCs.
By 2013 Intels 22nm SoCs should be very competitive, that same timeframe as the claim when ARM will be challenging in notebooks.
ARM CPUs still lack the general purpose computing power to be taken as a competitor to Intel Laptop chips which have an order of magnitude more processing power.
A GPU does not make up for the weaknesses in the ARM CPU. GPUs can only help is specific tasks and they need to be custom coded. It is not a general purpose computing solution for the weak CPU.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
J.
You're right that there's no strict definition of an SoC, but Arrandale is not considered an SoC. Intel doesn't call it an SoC; no self-respecting tech journalist calls it an SoC. Intel's SoC has always referred to the Atom line.
as for your point about "ARM socs are actually on par - performance wise - with current intel processors used for example in the MacBookAir," that is just straight wrong.
It has been shown that Kal-El, a 40nm quad core A9 chip clocked at 1.3Ghz is 40% slower than a C2D T7200, a 65nm dual core clocked at 2.0Ghz.
If the highest performing ARM processor out today can't even match a 5 year old Intel C2D, what makes you think it even comes close to Arrandale?
Quote:
Are they transitioning it to the same release cycle as their other cpus as in alternating a new architecture with a die shrink year after year? I'm asking because Haswell = 2013 at 22nm with a die shrink to 14nm the following year. I do hope they examine the issue of component reliability as they continue to shrink things. Expensive unreliable components suck if everything is completely integrated.
Thank you . I was looking for a link with that kind of info. It's nice to see Intel is putting some real energy into their atom line even if it isn't what drives their sales today. Too many people try to extrapolate what ARM could be from what it is. Intel is smart not to wait and see. They obviously don't want to "risc" it and I don't blame them.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
J.
You are very quickly making yourself look like a fool or imbecile. Think hard about what you are saying in your posts.
Thank you . I was looking for a link with that kind of info. It's nice to see Intel is putting some real energy into their atom line even if it isn't what drives their sales today. Too many people try to extrapolate what ARM could be from what it is. Intel is smart not to wait and see. They obviously don't want to "risc" it and I don't blame them.
Yeah, what people don't really realize is that they're comparing 2011 ARM designs to a 2008 Atom. And even then, the ARM designs still fall short in processing power and barely beat atom in power efficiency.
However, even with Intel waking up and actually focusing on Atom, it won't be until 2013 that a new architecture for Atom arrives. By then ARM-based solutions may be so entrenched in the mobile world that, even with similar/superior performance, Intel might not be able to make much headway into this market.
However, even with Intel waking up and actually focusing on Atom, it won't be until 2013 that a new architecture for Atom arrives. By then ARM-based solutions may be so entrenched in the mobile world that, even with similar/superior performance, Intel might not be able to make much headway into this market.
I agree. But it will be enough to keep the OSX/Windows markets it already has.
ARM will essentially keep mobile, x86 will essentially keep desktops/laptops.
Some server workloads are best handles by lots of relatively low performance cores.
Quote:
Originally Posted by majjo
Yeah, what people don't really realize is that they're comparing 2011 ARM designs to a 2008 Atom. And even then, the ARM designs still fall short in processing power and barely beat atom in power efficiency.
Atom doesn't even come close. In any event here is: http://www.calxeda.com/, a reference to a new ARM implementation. Very low power considering the capabilities. What is more interesting is that this is just one example of many custom ARM designs coming to the server market. Each will be a unique approach on the issue.
Quote:
However, even with Intel waking up and actually focusing on Atom, it won't be until 2013 that a new architecture for Atom arrives.
Even then Intel will have one horse in the race. ARM on the otherhand will have many horse custom tailored in various ways.
Quote:
By then ARM-based solutions may be so entrenched in the mobile world that, even with similar/superior performance, Intel might not be able to make much headway into this market.
Intel has a tough fight on its hands. The problem is this, will people be willing to accept lesser performance for other qualities. The calxeda link gives one an interesting perspective of ARM core performance. It gives us a good idea as to what Apple could achieve in the next iPad.
I'm not sure what value it is to compare Intel vs ARM floating point performance. 1000x more flops? So what, few applications use floating point. Integer benchmarks are more interesting.
Another advantage to ARM is die size. You don't need 1.3 billion transistors to build an ARM. Either you get more cores on your machine or get better yield out of your wafers. Unless ARM comes up with a high-performance reference design, though, it won't be competing with the sort of high-end desktops Apple likes to sell. Still, ARM could cannibalize Intel's business on the low end. Look what the PC video card manufacturers did to SGI.
Of course no one is saying that Apple will chuck out the current Intel CPU in favor of slower although higher efficiency ARMs, as they are now.
The point is that the ARM architecture has a pretty straightforward roadmap toward more speed. Notably design possibilities that doesn't "just" involve cranking up the clockspeed or putting in huge high level caches, to overcome the memory bottleneck.
The low power, imbedded market has had huge potential for the last ten or so years. If Intel haven't been able to come up with anything even remotely competitive in all that time, chances are they won't and they have a serious problem, either manegerial or more likely technical.
Apple has shown considerable ability and willingness to do their own IC designs lately.
It would be an incredible advantage to them in so many ways, to control the IP of the CPUs they use, at least partly.
I still think Intel will adapt in some way. They've also been pretty open about their roadmap. Being secretive like Apple wouldn't be much of an option when so many oems depend on them for parts necessary to build systems.
Comments
No one considers the i5 an SoC. Either you are trying to lie about it to make your case, or you are a fool. I haven't decided which.
But I don't suffer liars or fools mildly.
The i5 isn't an SoC, and for you stack your claims on that, makes everything wrong from that point forward.
Intel has barely entered the SoC game and all their SoCs are Atom based. Every Atom SoC in products this year was a combo 45nm/65nm part. Far from Intels state of the art.
But they will be moving to putting their SoCs on leading process going forward. 32nm in 2012 and 22nm in 2013.
Intel is behind in SoCs, but has had it's P4 moment and is waking up in SoCs.
By 2013 Intels 22nm SoCs should be very competitive, that same timeframe as the claim when ARM will be challenging in notebooks.
ARM CPUs still lack the general purpose computing power to be taken as a competitor to Intel Laptop chips which have an order of magnitude more processing power.
A GPU does not make up for the weaknesses in the ARM CPU. GPUs can only help is specific tasks and they need to be custom coded. It is not a general purpose computing solution for the weak CPU.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
J.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
All you have done is repeat ignorant falsehoods.
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
J.
You're right that there's no strict definition of an SoC, but Arrandale is not considered an SoC. Intel doesn't call it an SoC; no self-respecting tech journalist calls it an SoC. Intel's SoC has always referred to the Atom line.
as for your point about "ARM socs are actually on par - performance wise - with current intel processors used for example in the MacBookAir," that is just straight wrong.
It has been shown that Kal-El, a 40nm quad core A9 chip clocked at 1.3Ghz is 40% slower than a C2D T7200, a 65nm dual core clocked at 2.0Ghz.
If the highest performing ARM processor out today can't even match a 5 year old Intel C2D, what makes you think it even comes close to Arrandale?
Are they transitioning it to the same release cycle as their other cpus as in alternating a new architecture with a die shrink year after year? I'm asking because Haswell = 2013 at 22nm with a die shrink to 14nm the following year. I do hope they examine the issue of component reliability as they continue to shrink things. Expensive unreliable components suck if everything is completely integrated.
Yes, see here
Yes, see here
Thank you
Repeating your arguments doesn't make your case stronger. I think I made my point(s) very clear. You're of course free to choose not to understand them.
J.
You are very quickly making yourself look like a fool or imbecile. Think hard about what you are saying in your posts.
Thank you
Yeah, what people don't really realize is that they're comparing 2011 ARM designs to a 2008 Atom. And even then, the ARM designs still fall short in processing power and barely beat atom in power efficiency.
However, even with Intel waking up and actually focusing on Atom, it won't be until 2013 that a new architecture for Atom arrives. By then ARM-based solutions may be so entrenched in the mobile world that, even with similar/superior performance, Intel might not be able to make much headway into this market.
However, even with Intel waking up and actually focusing on Atom, it won't be until 2013 that a new architecture for Atom arrives. By then ARM-based solutions may be so entrenched in the mobile world that, even with similar/superior performance, Intel might not be able to make much headway into this market.
I agree. But it will be enough to keep the OSX/Windows markets it already has.
ARM will essentially keep mobile, x86 will essentially keep desktops/laptops.
Yeah, what people don't really realize is that they're comparing 2011 ARM designs to a 2008 Atom. And even then, the ARM designs still fall short in processing power and barely beat atom in power efficiency.
Atom doesn't even come close. In any event here is: http://www.calxeda.com/, a reference to a new ARM implementation. Very low power considering the capabilities. What is more interesting is that this is just one example of many custom ARM designs coming to the server market. Each will be a unique approach on the issue.
However, even with Intel waking up and actually focusing on Atom, it won't be until 2013 that a new architecture for Atom arrives.
Even then Intel will have one horse in the race. ARM on the otherhand will have many horse custom tailored in various ways.
By then ARM-based solutions may be so entrenched in the mobile world that, even with similar/superior performance, Intel might not be able to make much headway into this market.
Intel has a tough fight on its hands. The problem is this, will people be willing to accept lesser performance for other qualities. The calxeda link gives one an interesting perspective of ARM core performance. It gives us a good idea as to what Apple could achieve in the next iPad.
Another advantage to ARM is die size. You don't need 1.3 billion transistors to build an ARM. Either you get more cores on your machine or get better yield out of your wafers. Unless ARM comes up with a high-performance reference design, though, it won't be competing with the sort of high-end desktops Apple likes to sell. Still, ARM could cannibalize Intel's business on the low end. Look what the PC video card manufacturers did to SGI.
The point is that the ARM architecture has a pretty straightforward roadmap toward more speed. Notably design possibilities that doesn't "just" involve cranking up the clockspeed or putting in huge high level caches, to overcome the memory bottleneck.
The low power, imbedded market has had huge potential for the last ten or so years. If Intel haven't been able to come up with anything even remotely competitive in all that time, chances are they won't and they have a serious problem, either manegerial or more likely technical.
Apple has shown considerable ability and willingness to do their own IC designs lately.
It would be an incredible advantage to them in so many ways, to control the IP of the CPUs they use, at least partly.