Intel Corporation Message Board
- Reply to
arm64 will not be available until 2016...3 years from now! LOL!
by bacbacker •Apr 6, 2013 9:47 AMAll the hype about arm penetrating the microserver market is pure B S. a57 is meant for 16nm finFET, which tsmc has already stated won't be available until 2016 at the earliest. Without 64-bit machines, arm is going nowhere. In 3 years, Intel's Atom-based microservers will be industry standard and by 2016, Intel will be using 10nm technology...forget about it!
-----
ARM based servers are shipping *today*. The only question is how much (if any) effect these ARM based servers have on Intel's bottom line going forward.
As for the tape out of the A57 on TSMC's 16nm process, the news is really about the tool chain/low level IP getting ready for 16nm. We'll see the A57 on 20nm first. -
Good, for I thought you have gone nuts for a second:) (FYI, even when 1 core is used in swift, it still beats 2 threads in Atom for some of those int tests).
Medfield has the same CPU core as Clovertrail and they are clocked the same in Clovertrail+'s case, so on a single core/thread basis the comparison is still valid - swift would still win. Clearly with 4 threads Clovertrail+ should win over all. -
So lets get this right, you are saying that Atom can use mult-threads in it's benchmarks but swift can't?
-
Er, yes it does. Look at the single threaded results and see if yourself. Out of the single threaded tests Atom only beats Swift in 1 - Blowfish.
-
My original posts had the full breakdowns: iPhone 5 Integer 1275, Intel Atom Z2480, 784
-
I've tried to reply twice now, but the posts just get deleted:) Geekbench gives a score of 1500 to the iphone 5 and 1000 for Medfield.
-
In the unlikely event that the Apple "A" chips outperform the new Atom based stuff, Intel BETTER get in bed with Apple...or get ready to sell those excess fabs for 50 cents on the dollar.
-----
Given how much the current Apple chips outperform the current Atom line, I would say it's extremely likely that Apples next gen chips will continue to outperform the next gen Atom. -
SAN FRANCISCO: CHIPMAKER Intel is slowly winding up its Haswell hype machine ahead of what is expected to be a splashy launch in June.
Intel has been revealing high level details of its upcoming Haswell processors for a long time, such as its statement at IDF 2012 that a Haswell 10W TDP part will be pitched at laptops. The firm not only shown off running Haswell silicon at the recent Game Developer Conference but also announced that its HD Graphics 15.31 driver released this week is designed specifically for Haswell processors.
There's no doubt that Intel's Haswell chip will be the big semiconductor product launch of 2013, but it is rare to see Intel give away so many details of its upcoming processor line, given that AMD is doing relatively little to push the firm's third generation Ivy Bridge Core processors to the limit. However Intel's prelaunch announcements show a company that wants to stress the efficiency of its processors above everything else.
...
Andrew Feldman, best known for leading Seamicro and who is currently corporate VP and GM of AMD's Server Business Unit, told The INQUIRER that the cost of developing ARM based processors is so low that chip vendors' biggest customers will be able to subsidise chip development cost and buy the products on a cost-plus basis. Feldman's future might sound extreme, and it is in the present day semiconductor industry, but ARM based processors are considerably quicker and easier to develop and do not require leading edge process nodes to have the energy efficiency that Intel gains with its advanced manufacturing capabilities.
theinquirerDOTnet/inquirer/feature/2258763/customer-designed-arm-chips-will-give-intel-headaches -
[Apparently you don't. You cite how superior ARM is and then when confronted with facts say let's just see. But the point you really don't comprehend is that Intel improvements will come in waves that ARM can't hope to match with its fading fabrication. Here is the bottom line: "...we expect multiple handset manufacturers to jump ship to Intel’s processors..." Maybe you need to give up ARM marketing and head back to engineering and give them a hand since you understand everything...]
You site a benchmark that can't be run on an Apple device and with results that have 'leaked'...
Seriously, nip over to the antutuDOTcom website and have a read about the benchmark suite and what it contains (such as SD read/write speeds). Then search for 'cheating' (Huawei). You'll understand why you can't take this 'leak' seriously. -
[Do you actually read the news?]
----
Indeed, and I also understand it:)
Lets just see if these benchmarks stand up when the device is actually tested for real:) As a side AnTuTu, is made up of 5 separate components and doesn't run iOS. -
[Well, one could go on all day about this but let's take the obvious one. The big fat overheating processor in the "new" iPad. The one that caused it to be so thick. Start with that one, pilgrim, and I'll bring you another. Extra credit for identifying the movie that line came from...]
-----
it's the GPU and the screen power requirements...BTW the GPU that apple is using is *much* more powerful than anything Intel has got on the table. -
Hewlett-Packard's big bet on ARM servers, data center disruption and the company's innovation profile---also known as Project Moonshot---kicks off April 8.
zdnetDOTcom/hp-preps-project-moonshot-launch-april-8-7000013397 -
Don't bother reasoning with ARM fanboys. They think that because anybody can license the instruction set or cores that Intel is doomed.
----
Question for you: why do you think that QCOM and everyone else that takes an arch. licence from ARM still has to pay (significant) royalties on a per core basis? -
Apple is marketing 40nm and 32nm processors and staying clear of 28nm. You would expect that they would be the first in line for 28nm and offer the latest and greatest. Because they don't speaks volumes about their position in the ARM ecosystem.
-----
I suggest you spend a few minutes looking at benchmarks as even with "40nm and 32nm processes" (as you put it) Apples SoC's are *very*, *very* competitive and offer a much more compelling solution than Intel's(for example) SoC's.
Do you really expect any of to believe that Apple wants to push their OS/X business towards ARM64?
-----
Why not? *wink* -
The problem is that I don't believe that in the big picture ARMH really isn't that diversified from Apple. Apple basically started both the smartphone and tablet revolution
-----
Well, it is diversified from Apple. Intel (as reported) is ARMH's biggest customer and it's been repeated many times that not 1 customer is responsible for more than 10% of revenues.
ARM was in phones/PDA type devices before the iPhone launch and is in almost all of the iPhone and iPad competitor products.
-----
The biggest question, can you honestly say or suggest that the problems with Apple have absolutely nothing to do with ARMH or the rest of the fabless model?
-----
What problems has Apple got that has ANYTHING to do with ARMH (or the fabless model)? -
ARM or Atom Based?
----
On HP's recent earnings conference call, CEO Meg Whitman said:
Later in the second quarter, we will be bringing the latest innovation from HP Labs and our Enterprise group to market, the first, commercialized product from our Project Moonshot. We expect this to truly revolutionize the economics of the data center with an entirely new category of server that consumes up to 89% less energy, 94% less space and 63% less costs than our traditional x86 server environment. This is exactly the technological inflection that can fuel the exponential growth of hyper scale computing. To put in that perspective, if just 10 large web services providers switched their traditional x86 servers to Moonshot, they could save a combined $120 million in energy operating expense and nearly one million metric tons of CO2 per year. The equivalent of taking over 180,000 cars off the road for a year. That is a game changer. -
justfine790
----------------
"We also had the chance to pit the RAZR i against the Droid RAZR M. Aside from the benchmark results outlined above, the Medfield entry offered a marginally faster response to most actions. Powering up both phones, the RAZR i landed at its home screen around 12 seconds ahead of the Snapdragon sibling. Loading into the Chrome browser was around the same speed on both, likewise the camera app -- both launching and capture -- was equally swift on both"
------
*aside from benchmarks * which the M won (except javascript)... The point of benchmarks is to remove subjective experiences. So the I loaded the home screen faster? What apps where loaded, what skins where active, what settings on both phones etc etc.
As for battery life, lets see how Intel's SoC's do once they have a competitive GPU on die... -
Nice spin there again....Read justfine790 post again...the main point of article is mass-production of big.Little is delayed until 2nd quarter! we will not see it until June/July period. This is very bad news for arm camp as Intel will have 22nm with integrated 4G LTE by end of year period.
----
So....10 million units, 7 million are QCOM's, so where/what are the other 3 million units? -
I have a Razr M, and spent a lot of time with the Intel powered counterpart, the Razr i.
----
The SoC in the M is the better than the one in the I - it's got a much better GPU and the CPU performance is also better (on all known benchmarks, except for Javascript). -
What do you guys think of this latest development? Could be rehashed rumor, but I think there is some truth to this, especially with announcement that Octa won't be available for the Galaxy S4 in the UK version. This tells me 1) there is some sort of problem getting big.Little to work 2) big.Little will be delayed ...which works to Intel's favor. This could turn into a BIG story if they cannot fix the problem soon
----
Rehashed rumor. Just like the existing S3, there are two versions. One with a QCOM app processor and one with a Samsung processor as they are dependent on the baseband. Some markets get one type (US will get the QCOM version), while others get Samsung's App processor (the Octa), while some markets (such as the UK) get both. Millions of the 'Octa' processors have already been made ready for launch.
As for the fabless model not knowing about manufacturing processes, I suggest you do a google for ARM physical IP and see what you find.