[Um, because Intel didn't have an LTE modem then. Will you ever have any non-earwig comments on this board ever again?]
Many high end devices have separate modems, there was nothing stopping an OEM pairing an LTE modem with Intel's app processor. Current examples include Samsungs S6, which is paired with a QCOM modem or a Samsung modem depending on the market. Next gen Apple, will have a QCOM modem or Intel modem, again depending on the market.
In short, the lack of a LTE modem wasn't a handicap in winning sockets for the high end.
I see AAPL adding INTC to their next iPhone release.
This is a good example. This win is for the baseband (modem) for some markets, and the modem itself is made on TSMC's 28nm.
That is true generally, and most certainly was true when Intel's mobile parts had process superiority but lacked key communications components needed in the marketplace. What's different this time around is that Intel has largely filled-in those missing pieces with very competitive offerings.
If you are talking about the modem (?), most high end devices have a separate modem (then and now), so why didn't Intel take chunks out of the market with it's process advantage (at 32, 22 and now at 14)?
Here's where I believe you are making a mistake. It's different this time, because the CEO is different. Much more market savvy than the last guy, and knows what the stakes are. You have been correct, but BK will get the designs and pricing right. My opinion of course.
Time will tell, you could be right. But...
Intel isn't facing an AMD, but the likes of Apple, QCOM, ARM etc on the design side, and the likes of TSMC and Samsung on the manufacturing side. These companies have lots of cash and wont sit around twiddling their thumbs.
Not if INTC is able to lower a chip's cost enough by introducing smaller, faster, and more power efficient chips. 10 nm finfet will win INTC lots of mobile business -- by keeping true to Moore's Law.
Why will INTC win lots of mobile business at 10nm?
Do you think the analyst calculated an expected 70% from the recent Samsung 14nm price drop or from an unnamed source? Samsung has declined to comment on yields. How believable do you think the number is?
No idea of how the number was calculated. How believable? So, so:) It was a report on ee times, which gives some credibility. Also, given the size of the rollout of the S6...
So you do not believe in Moore's law? What's up with the paraphrasing my posts? (I know what I've stated.)
Once you've gone below a certain depth in these forums it's hard to see what the reply is actually for.
It's not a case of believing in moores law or not. What are you going to do with those extra transistors? Intel will be fine with PC's and servers, mobility, a different story. It'll be a hard, long slog.
Are you suggesting that INTC doesn't have what it takes to get the designs and pricing right? I find this remarkable, as well.
Some on this board have claimed that INTC has had clear cut fabrication advantages at 14nm, 22nm, 32nm and probably further back. While this is true, what difference has this actually made in getting into other markets (such as mobile)? Why will 10nm be any different?
Intel could have 10nm finfet today but would mean nothing if it doesn't have the right designs at the right price.
“TSMC's monopoly on leading edge has been broken thanks to Samsung's successful ramp of 14nm with current yields exceeding 70%,”
Over on eetimes if anyone is interested.
Those may be true but they're not good excuses for failing to make the right product
If you look at the anandtech it various power measures and it seems like it has a high idle power consumption, dissipating that amount of heat isn't easy for such as small device.
[Well, that's amazing - you know what is going on inside the inner sanctum not only at Apple but at Microsoft as well. Um, where did you read the part about a policy once established never being changed. You do remember that Microsoft said that RT isn't dead and now they say it is. You are going to be as right about the Intel Windows phone as you were about RT. I remember when you used to be technical. Now you just parrot everything we hear from the various ARM marketing departments.
I never knew Microsoft was a marketing department for ARM....
The FACT is that Microsoft said this. Not ARM. Not Me.
Yes, they could change their mind. Have they?
It appears Microsoft was making up this policy as they as go along. In any case It makes perfect sense to allow an Intel-Microsoft smartphone (if it truly exists) to deliver as much value to the consumer as possible, especially if it already has the hardware resources to do it.
I am quite happy to be wrong, but Microsoft was very explicit at the time - no legacy UI for ANY device with a screen size below X.
[Yep, you called it. And Monk said it could never happen.]
Nope, not I. Microsoft.
It's getting a bit silly.
"tri-cluster set up allows for TINY.MEDIUM.HUGE", with quad A53 power optimized for tiny, a performance optimized quad A53 cluster for medium, with dual A72's at 2.5Ghz for huge.
Will have some nice threaded benchmark scores...*cough*
WOW the Exynos costs $29.50 to make? what are their yields like 20%? Hellooooo...
What would you expect the cost to be?
[I understand high cost, low margin and limited power savings/performance boost. You think because the screen is a lot of the cost it changes the overall conclusion??? Remember when I said that if the margins are thin, you have to sell a lot more of them?]
Do you know that the S6 edge has an 100 dollar premium over the normal S6?