You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
And the A9 SoC is.............................. from TSMC!!!! And the modem is QCOM of course. Sorry Intel.
The long-rumored, cost-friendly iPhone SE was finally announced on March 21st of this year. Said to be the iPhone 6s squeezed into the smaller iPhone 5s body, this new iPhone features an appealing price point with flagship-level performance. Some here have even likened it to an old Chevy outfitted with a peppy new Corvette engine. Early reviews of the iPhone SE have been fairly positive; most seem to agree that the 4-inch iPhone packs some powerful internals. We got our hands on an early iPhone SE and will be taking a closer look at those internals to determine how much innovation was packed in between the familiar matte chamfered edges of the iPhone SE.
The teardown confirms that despite persistent rumor chatter to the contrary since December, Intel again failed to get the modem slot. Last year this time, Intel was rumored to have a portion of the iPhone6 modem business. That proved to be false. This year it was the Se. That again proved to be false. Now it's the iPhone7. Is the iPhone 8 next? This is like Waiting For Godot.
I had read that questionable Barclay "short" call, that was premised on that analyst's "back of the napkin" math, related to an event that "could" take place, but prior to knowing whether it "will" occur. My point is a bit different. Rumors of Intel getting part of Apple's modem business, by whatever means, have been ongoing for awhile now. When an "analyst" lowers ratings and price targets based upon rumor, rather than fact, it strikes me as being premature, agenda driven, and irresponsible. I'm in the camp that questions whether Apple will choose to geographically handicap its next flagship device, when competitive Android products are already technologically more advanced. Charging a premium price, may finally require providing a premium product, given the competitive landscape.
Looks like an analyst who wants to make a name for himself by shifting the price. One read, if that, is all they get out of me.
The A9 is no surprise as Apple must keep its BOM down but I had thought they might use Intel to collect the subsidies. Sorry, "contra revenues".
I have a daughter who wants a new iPhone and an iPhone4 I can trade in. Looks as if Apple will get an upgrade!
Intel can't get all of the iPhone business anyway, because their modems don't support CDMA, which is still a requirement in USA. Also, Qualcomm modems have better power efficency and reception.
Intel can't get all of the iPhone business anyway, because their modems don't support CDMA, which is still a requirement in USA. Also, Qualcomm modems have better power efficency and reception.
Presumably Intel was responsible for disclosing they had 1,000 engineers working on getting some Apple mobile business, as well as all the rest of their rumor buzz surrounding their mobile prospects. Being considered as a possible second source, however, is not the same as landing a P.O. Time will tell.
It's not clear to me that Intel is worried at all about the US carrier market. As I've said many times, the US carriers (at least Verizon and AT&T) call the shots, they are using Qualcomm, end of discussion. Every vendor attempting to do something different, including Apple and Samsung, has had to face that reality and change out SKUs accordingly.
Intel DOES care about the Chinese carrier market and has TD-SCDMA support. It appears to be a very conscious decision to provide support for more international LTE bands and TD-SCDMA and not fight a battle they've long since lost over CDMA/EV-DO.
Again, an Apple split between Qualcomm and Intel (Infineon) has precedent. The iPhone 4 had a CDMA version (Qualcomm) and an HSPA version (Infineon).
That said, if Intel is tossing out features and still doesn't provide a power advantage with the XMM7480, that's a problem. It seems every Intel modem announced is "the one" that is going to win them business, and the XMM7480 is the 2016 entry.
I think this is Intel's play to win a slice of the Chinese market, plain and simple.
Don, what's your wag on the Aicha Evans departure? Is it simply Murthy showing the same managerial shortcomings that he apparently exhibited at Qualcomm? Or is there an iPhone7 embarrassment on the way, like we just saw with the SE last week?
I saw that report yesterday. Big strike against diversity. I don't have any inside info, but I do know there has been a lot of Intel exec upheaval since the departure of Renee James and there was another round of mid-layer shuffling in February. Two theories: either this was agreed to and delayed until after MWC and launch of XMM7480, or somebody just made Evans a better offer - she's rumored to be leaving Intel, no word on destination yet. Sometimes Intel execs leave simply because their career path is blocked.
As for the iPhone SE, I'd apply the reverse logic to the above discussion. It's clearly a rightsize move to push US customers who passed on iPhone 6, and they packed pretty much the same BOM inside with changes in PMIC and RF parts. I don't think the Intel modem was ever slated to be inside it - all the rumors I've heard have been iPhone 7.
I refer you to Fudzilla in December 2014, proclaiming from "sources" that Intel got the modem slot for the new 4" iPhone that would be launched in the Spring. That was the persistent rumor from various outlets and blogs, until the teardown of 3/30 established the truth. The SE was to be a test bed for validation of Intel modem performance for the iPhone7.
Still amazes me, how strong Intel's marketing machinery is. For example talks about "the only real 14nm" ( http://i62.tinypic.com/29m6qdw.jpg ) without any results in produced silicon (even in density hardly compares with 28nm HKMG). Or extremely biased POWER8 reviews on andatech (comparing scalar POWER8 performance against Intel Xeons with AVX2, and that same with AMD). And much much more...
So yes, there are "credible" sources says, that Iphone 7 will use Intel chips, and again will be with Iphone 8 (9, 10...). But it is hard to say, whether they are based on reality. But what i know is, that Apple is clearly opportunist company and they will use best available solution for their devices. Like when they used Samsung to manufacture A9 chips, even when there was rumors that that SOC for Iphone 6S will be from Intel.
The iPhone 7 will use a variation of TSMC 16FFC. I think it is called 16FFT. And I'm pretty sure it will use the new QCOM modem. Intel modems still lag in features and power savings. If Intel had a shot at Apple modem business it would have been the low cost SE but that was not the case.
The good news is that anybody can start a media outlet. The bad news is that anyone can start a media outlet. This one really cracks me up:
It's so funny that there are many people posted comments to indicate that they believed this Yahoo fiance story is true! Unless they're fooling around too ...
My guess is Intel itself was the source of the iPhone rumour leaks. Presumably there are only a handful of parties that would know about any plans for an Intel modem chip in an upcoming iPhone, and those would be Intel, Apple, and possibly Qualcomm (if Apple let them know they lost the slot). We know how tight lipped Apple is and it's unlikely Qualcomm would go around talking about how they lost the slot.
And if anyone is actually buying the rumours that the next iPhone SoC would be manufactured at Intel's fabs... I've got a bridge I'd like to sell.
Don, what's your wag on the Aicha Evans departure? Is it simply Murthy showing the same managerial shortcomings that he apparently exhibited at Qualcomm? Or is there an iPhone7 embarrassment on the way, like we just saw with the SE last week?
Do you have a source you can share for Murthy's apparent managerial shortcomings at QCOM? I ask because the only unchanged position seems to be that of Ms Bryant at DCG. Since BK became CEO he has lost $10+billion on mobile and undertaken two reorganisations, both to do with mobile. Now Skaugen has gone as well. If Murthy has got it wrong that's not good, not good at all.
There have been rumours of a "16FF Turbo" process for some time, but no announcement of such from TSMC (and yes, this includes under NDA as far as I know) -- where does your information come from, and do you have any information about what is different from 16FF+ or 16FFC? Or is this just a new shorter name for the "non-shrink 16FFC" that became "official" recently?
There have been rumours of a "16FF Turbo" process for some time, but no announcement of such from TSMC (and yes, this includes under NDA as far as I know) -- where does your information come from, and do you have any information about what is different from 16FF+ or 16FFC? Or is this just a new shorter name for the "non-shrink 16FFC" that became "official" recently?
TSMC discussed this on a conference call saying there will be two versions of third generation FinFETs, one using 7 track libraries (16FFC) and one using 9 track libraries (I think it is 16FFT). The one TSMC is talking about openly is FFC, the other is the one used by Apple. And by the way, Apple did not use 16FF+, they used a custom version of that process as well. Apple gets what Apple wants and Apple wants performance over power savings.
And Yahoo Finance picked it up!