Apple's iPhone modem design is three years behind Qualcomm
Apple intended for the iPhone 15 to use an 5G modem designed in-house, and it has spent billions working to achieve that. Here's why it still has to buy modems from Qualcomm.
Apple is trying to make its own 5G modem for the iPhone
Modems are hard. In 2010, Intel spent $1.4 billion to acquire Infineon, whose baseband chips, were used in the cellular features of the iPhone.
In 2019, Intel sold its modem patents to Apple for $1 billion, and got out of that business. Since at least then, Apple has been aiming to make its own modems, and to do so both to save money, and to stop buying from Qualcomm.
For its part, Qualcomm expected Apple to have produced its own modem by 2024. But then in September 2023, Qualcomm announced a deal with Apple that would see it providing modems up to 2026.
According to the Wall Street Journal, the extended delay comes despite Apple spending billions on the project. It is delayed for multiple reasons, ranging from the expected technical challenges, to the presumably unexpected management problems.
"These delays indicate Apple didn't anticipate the complexity of the effort," Serge Willenegger, ex-Qualcomm executive told the publication. "Cellular is a monster."
"Just because Apple builds the best silicon on the planet," said former Apple wireless director Jaydeep Ranade, "it's ridiculous to think that they could also build a modem."
Neither Willenegger nor Ranade have been involved in the project, which Apple has reportedly codenamed Sinope. Former Apple HR executive Chris Deaver told the Wall Street Journal that the project began in 2018.
That's a point when Apple and Qualcomm were taking each other to court over multiple disagreements, though these were settled in 2019.
At that time, Apple executives were divided over whether to partner with Intel, or to design a chip themselves. Reportedly, Ruben Caballero, then a long-standing head of wireless at Apple, wanted to go with Intel.
However, Johny Srouji, senior vice president of hardware technologies, disagreed and wanted Apple to build its own. Caballero left in 2019 and much of his team were placed under Srouji.
Others were split off into the separate hardware engineering group to work on related issues such as antenna design.
According to the Wall Street Journal, this division of engineering work became a problem. Unspecified sources within the project told the publication that one of the senior managers on Srouji's team had no wireless technology background at all.
Other Apple executives lacked experience with wireless, said former project engineers. Consequently they were setting tight and unrealistic deadlines.
Or they did until late 2022 when Apple began testing prototypes. Reportedly, Apple's 5G modem chips were poor enough that they would have made the iPhone's wireless speeds slower than Android.
People said to be familiar with the tests are said to have estimated that Apple is three years behind Qualcomm.
It's possible that the prototypes were an internal-use-only iPhone SE 4, as some models of that were reportedly manufactured for testing purposes. When that was reported, it was expected that a shipping iPhone SE 4 would feature Apple's modem in 2024, but now that appears to be another unrealistic deadline.
Read on AppleInsider
Comments
5G is a huge collection of different standards and technologies falling under one general umbrella.
A few years ago you could have got away with NSA but now you have to support SA and advances for the immediate future (5.5G) and then be ready for 6G.
Realistically speaking Apple needed to be at the table where the standards themselves are thrashed out.
Without a seat at those tables you will always be playing catch up to a certain degree.
They acquired the Intel division plus accumulated patents. You could argue that got them onto the ladder. Climbing it is another story.
The odds of Apple producing a superior product to anything Huawei, Samsung, Qualcomm, Broadcom can produce are limited simply because those companies have decades of accumulated knowhow and resources.
Perhaps 'good enough' is where they are happy to be at the moment in terms of verticality in manufacturing. I think that's a valid aim but it won't free them of patent agreements.
Beyond the modem and antenna designs themselves, the whole thing was a result of a massive strategic goof. The spat with Qualcomm.
When the Intel plan collapsed it really was a Yikes! moment and although we don't know the details of the deal, it's hard to imagine Qualcomm not having the upper hand in negotiations.
Qualcomm may have had the "upper hand" in negotiations, and has recently seen a reprieve, but it in fact Apple that was visionary enough to add long term options to the agreement, just in case. Should Apple deliver their own modem by 2026, or even later, it is due to Apple being able to generate enough revenue and profits to more than afford that considerable R&D effort.
You know, the same company that set the bar for smartphones some 16 years ago, which nobody saw coming.
https://appleinsider.com/articles/19/06/29/12-years-of-iphone-why-apples-first-smartphone-was-far-from-a-guaranteed-success
Second point: if 'visionary' were even applicable, they would never have found themselves in this predicament in the first place. It was more a case of not having any more cards to play. To Qualcomm it just means one less competitor (and even that term is a stretch) and much more money.
Older western tech?
I hope you realise that NearLink devices are now shipping. LOL!
https://techjaja.com/huawei-nearlink-wireless-revolution/
The real issue is what I outlined in my post. Apple will never reach the top players until it finds a seat on the standards committees which shape the future.
https://www.lightreading.com/5g/3gpp-moving-to-prevent-power-grab-by-apple-others
Apple was first to introduce UWD in a smartphone, and today, I expect every phone to have UWD of some type, but they don't because it isn't as cheap as BT.
Of course, Huawei is able to introduce technology at 400 Mhz of bandwidth, in China that may not be allowed in the rest of the world.
https://www.androidauthority.com/what-is-uwb-1151744/
Tough task ahead to wean off Qualcomm.
I don’t remember your posts, but did you whine all the way through the Apple, Intel replacement process it can’t be done it will never be done impossible, Apple should just give up?
Apple might be faced with another replacement project, and that relates to Unity who struck out, after getting a little limelight on the big stage with their game engine, there are those who suggest Apple should just stay home and not do anything despite the fact that Unity appears to have lost their head, when they came up with terms for their developers that were far beyond anything, Apple has ever done, question is, should Apple trust put their future in AAA games into the hands of companies like Unity, or Epic? or roll up their sleeves and get busy internally.
That said, I was unaware that 3GPP had made changes to the voting system which will even that side out a bit now. The changes are recent and haven't taken effect yet.
Perhaps acquiring intel's 5G modem division is what saw them double their votes.
You were going off about using 'older western tech' so I gave you the perfect example of the exact opposite. Of course, in a thread on 5G and Huawei being one of the major 5G players, trying to make the point you made was already flat on its face before you hit the 'post' button.
And when you say 'western' I suppose you are trying to squeeze TSMC in which is completely laughable.
On the subject of UWB, that is nothing new. It's been around for decades and Huawei has been using it for years now in industrial settings, so saying NearLink is UWB is saying nothing.
Like saying Apple was first to put UWB in a phone.
NearLink is a wireless implementation with over 300 companies on board. What counts here is what it does and how it does it. For example, it uses polar codes. As for cost, I'm not following you. How could Huawei put it in its newest earbuds if cost was an issue? Btw, those earbuds are taking advantage of NearLink to enable Huawei's latest HD audio codec.
Cost used to be a factor but if a pair of earbuds have it....
1.Apple has never made a cellular modem before.
It all boils down to need. Apple needed Intel at the time. Bootcamp was a boon back then.
One thing to remember is that even when OSX was released on PowerPC, it wasn't long before Intel compiles started being built in secret (Marklar).
Everything is context and perspective dependant. That applies here too. If Intel had delivered a decent 5G modem we wouldn't be having this conversation.
It still remains a strategic goof though and of epic proportions.
1. comes back from the dead
2. Turns the music industry right side up with not only an answer to rampant piracy, but a pioneering revolution of the business.
6. Launches an Apple Watch to dominate that industry.
a. THE MOUSE
b. FireWire
c. Thunderbolt -in collaboration with Intel.
1. Apple car…
2. vision pro…
3. Cellular modem. Likely 5.5 or 6g (skate to where the puck is going to be. Not where it has been…)…
not everything apple does succeeds. Obviously, the Apple TV is still “a hobby.” And Apple TV plus could use an injection of enthusiasm. iCloud plus could hold more value, the g4 cube tanked (but the studio seems to be doing just fine…)
it’s a new Apple. Though it’s still finding it’s way again in terms of moving design forward, it’s definitely an engineering powerhouse at the top of the heap with the work ethic, the financial solvency, the drive, and talent to succeed.