Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple and Intel Have Reportedly Discussed Deal for Production of Future iPhone and iPad Chips
- Thread starter MacRumors
- Start date
- Sort by reaction score
+1!
Apple should do everything they can to not support Samsung. Seems like a deal with Intel on the A series chips would be a good marriage for both Apple and Intel... good name on both sides.
Did you even bother to read at least the first sentence of the article? The focus is on Intel acting as a foundry, not a designer of chips; i.e., Apple gives Intel the design of the chip and Intel produces it. It's not about Intel selling Apple Intel designed chips like the atom or anything else in it's own product lineup.
Hmm I don't know, it's still a sizable dent.
Looking at the iPhone 5 cost breakdown, the processor is $17.50. At 33.4m sales in 2 quarters that equates to $584m.
And that's just for 4 months of one model of the iPhone. Don't forget iPhone 4/4S, various iPad models, iPod Touch 4 and 5, Apple TV etc etc..
That could be cool if Apple used the intel fabrication line, as we might see 22 nm with substantial decreases every 2 years.
I don't think Apple using Samsung as a foundry gives Samsung any advantage except the commerce itself. One advantage to Intel as a supplier is they have bleeding edge technology for reduced nm and also proprietary component designs Apple could license. Due to Intel's non-commodity pricing it would likely be offered on higher end priced units with added functionality.
Rocketman
Rocketman
Oh man this is exciting. Lose the Samsung dependency!
EDIT: Does this mean OSX bootable iPads perhaps?!
How is that exciting?
----------
Why is it sad?
22nm process and 3rd transistors....
There might be more benefits to getting Intel to manafacture Apple's chips than pure transistor size. Intel has developed 3d transistors which leak less at small scales so they might end up with chips that far exceed what other manafacturers can do at that scale.... but would intel let them use that technology on an ARM chip? naaaa.....
There might be more benefits to getting Intel to manafacture Apple's chips than pure transistor size. Intel has developed 3d transistors which leak less at small scales so they might end up with chips that far exceed what other manafacturers can do at that scale.... but would intel let them use that technology on an ARM chip? naaaa.....
People seem to be forgetting that Intel has an ARM architecture license. Not only that but they "inherited" StrongARM from DEC back in 1998 and continued to develop it as XScale before offloading it to Marvel in 2006. So they've got plenty of experience in designing and fabbing ARM silicon.
As long as Apple's own custom ARM design is validate on Intel fabrication process (22nm etc.) then there shouldn't be any problems. Given that Intel are 18-24months ahead of everyone else on process technology any deal would at least guarantee Apple to be ahead rest of posse.
As long as Apple's own custom ARM design is validate on Intel fabrication process (22nm etc.) then there shouldn't be any problems. Given that Intel are 18-24months ahead of everyone else on process technology any deal would at least guarantee Apple to be ahead rest of posse.
Look at the MacBook Pro and air. The pros are getting just as thin so its possible those two markets will probably merge in a few years. Now iOS and OSX will probably combine into a single OS. Some of the features in OSX it seems are moving to a touch interface so the transition will be easier.
Maybe not now. But what's so bad about letting Intel get one foot into the door? Throwing Intel into this arena means more competition all around. And also translates to more choices for chip fab customers like Apple and Microsoft. That's always a good thing in the end.
Nah, remember it's largely thanks to Apple that Intel have gone the way they did in terms of chip design over the last few years. It's Apple that pushed them to really focus on improving the integrated GPU and reduce TDP on their chips. If they hadn't started when they did Intel would not be in a position to release Haswell in a few months and might be struggling more than they are against the incoming ARMy (sorry, that's terrible but I couldn't resist).
More to the point Intel are a generation ahead of most other foundries at this point so it's not a one way street. Yes, Apple have a huge amount of muscle thanks to volume but Intel are in a strong position as well. Add on to that Intel having a vested interest in keeping Apple on board with the Mac and this could work very well indeed for all concerned.
Nope, sorry. Samsung announced a milestone on the way to developing a 14nm process, not the process itself. Specifically that they'd tapped out development parts (i.e. proof of concept). Intel are preparing for 14nm by the end of the year and have already started setting aside fab space for it.
However each division of Samsung is run as a separate profit center which gets graded differently based on the earnings. If you're running Samsung's semiconductor division, you try your hardest to keep Apple as a customer because what you get paid depends on how much your own division made, not how much your colleagues at the mobile division made.
The semiconductor business is the 2nd profitable division in Samsung ahead of consumer electronics and (former) display divisions. However like Apple most of Samsung's profit comes from selling phones.
There's no reason to think Intel wouldn't be good at making mobile chips if they didn't want to stick with their own architecture. If Apple could work with Intel using the ARM-licensed architecture, it'll probably be a great combination. The question is whether Intel is willing to accept the lower margin working as just a fab.
It´s very unlikely that this is going to happen anytime soon. If Intel allows Apple access to it´s fabs for these power-critical devices, they would basically throw away their manufacturing advantage and substancial amounts of investments into new process technologies, fabs, workers and know-how. Intel has just begun to gain some share of the power-aware mobile market for tablets/mobile phones with it´s own products, and doing that, both companies would basically compete with each other.
It is far more likely that Apple will adopt Intel´s SoCs in the future.
It is far more likely that Apple will adopt Intel´s SoCs in the future.
Last edited:
Anybody with a brain knows this is coming. Nobody can match Intel. And only the "genius" analysts thought Apple will use ancient-speed ARM on Macs. Haha. I expect Google and its Android buddies to follow suit, because Intel's progression will make ARM irrelevant in the foreseeable future. Heck, even their slightly updated 45nm chip goes toe-2-toe with current ARM chips.
Ultimately, the best reason to use Intel is to unite iOS with OSX. Imagine both platforms being able to use either softwares.
Ultimately, the best reason to use Intel is to unite iOS with OSX. Imagine both platforms being able to use either softwares.
This might be good for Portland since Intel has an unused Fab plant in Hillsboro that Apple was considering a few months ago.
Could the Apple device being built in the US be the new iPad ?
[edit]
Found the old article here :
http://www.oregonlive.com/silicon-forest/index.ssf/2012/12/oregon_courts_mysterious_proje.html
[/edit]
Could the Apple device being built in the US be the new iPad ?
[edit]
Found the old article here :
http://www.oregonlive.com/silicon-forest/index.ssf/2012/12/oregon_courts_mysterious_proje.html
[/edit]
Last edited:
Actually the more likely pressing question is whether this was about procesors for iOS devices.
I can't find it at the moment by Intel really isn't interested much in fabricating direct competitors to their IP ( CPUs , chipsets , etc. ). Intel doesn't sell FPGA so the Altera and Achronix deals do nothing major to diminish the sales of Intel's other stuff. ( http://www.theregister.co.uk/2013/02/26/intel_altera_fab_deal/ ).
Altera is paying to jump in on paying for the next process shrink. That is a smart move for Intel and Altera.
Intel has said they are open to deals where folks might add something to some of there basic function and get custom chips. Like a future CPU/GPU of intel's making with a custom chipset. Or an agument on a chipset.
For example a special single chip version of Broadwell that had Intel's CPU/GPU/Chipset and apple's flash and touchpad controllers for an almost single chip implementation of a MBA.
Something like that.
Intel is a long way from being a generic fab for all comers.
Folks are sorely underestimating how decent their response to ARM's incursions is going to be.
http://www.anandtech.com/show/6536/arm-vs-x86-the-real-showdown
Intel isn't as far off as the "doom and gloom" from the last couple of years have made them out to be. Windows 8 tablets gather their feet and Intel could easily be back on track.
It´s just a question of time, really. Imagine a mobile device (basically a mix) of a Surface Pro in flexibility and speed (using an i5 core, currently) with drastic future improvements from Intel´s side on power, so it can be build a lot thinner and with more battery life. Windows 8 may look like a mediocre OS now, but it´s only going to get better. MS isn´t purely a software-company anymore. This is a new age for them, too and they have to adjust.
No one else is so perfectly aware of their advantage than Intel is.
If Intel doesn´t allow Apple access to it´s fabs, Apple will always be at least one step behind. They could still make up some of that disadvantage with their software knowledge, but Apple KNOWS they are running into a blind alley.
Intel basically already can dictate what Apple does, Apple needs Intel´s CPUs already. Intel knows they can take that one step further with their new or future SOCs/mobile line-up.
I think it would be Intel's best interest to get Apple interested in low-power x86 SoCs for future iOS devices. Instead of just manufacturing Apple's ARM chips, why not jointly engineer an ARM competitor with a similar electrical footprint? Intel's problem is that it started out strong with the Atom, but wedded the CPU to a woefully out-of-date chipset (which used more power than the Atom CPU itself) with the crappiest possible GPU (the GMA950 and all its successors) and then just gave up on it. If Intel redesigned the Atom architecture to be out-of-order and combined it with a decent mobile GPU (like the A6X's PowerVR 554MP4) at a 22nm fab and Apple's wattage-sipping tweaks, they could have an SoC that beats ARM at its own game.