Apple's own 5G modem is now expected to arrive in 2025
In a note surrounding other 5G modem manufacturers, Ming-Chi Kuo has confirmed his belief that an Apple-produced 5G modem will arrive at some point in 2025.
The original and second-generation iPhone SE
The rumors about an Apple modem are more solid that rumors that surround Apple typically are. However, what's always been in flux is when it will arrive, and in what product.
On Wednesday, Kuo said that risks for Qualcomm include the Exynos 2400 chip -- and Apple using its own chip starting in 2025.
This is not the first time Kuo has chimed in on Apple's 5G efforts.
Kuo has claimed that an iPhone SE 4-style model with Apple's in-house 5G chip is being made as a prototype. If tests prove successful, that will mean that Apple's own 5G modem works at scale, and so it could be used in an iPhone released in either 2025 or 2026.
"I believe that the mass production schedule for Apple in-house 5G baseband chip will largely hinge on the test results of this engineering prototype," he wrote in April. "Consequently, mass production could commence as early as 2025."
"But if testing falls below expectations, the schedule may be pushed back to 2026 or later," says Kuo.
So whether it's called the iPhone SE 4 or not, or even if it hits the market, there will likely be some model iPhone with the company's own 5G modem by 2025 or 2026.
Read on AppleInsider
Comments
In a recent earnings call they stated that there would be no further material revenues from Huawei. Last week Huawei released its own 5G capable phone.
Expect the same to happen when it comes to losing Apple’s modem revenues.
And by extension, some chip foundry will know since Apple does not manufacture their own ASICs. When/if Apple kicks Qualcomm to the curb, some foundry is going to get a purchase order for tens of millions of 5G modem chips. Most likely that foundry will have been working together with Apple for years, providing assistance with tape-outs, engineering samples, etc.
Apple: "Here's our purchase order."
Qualcomm: "Hmmm, I see you only ordered small quantities of the previous generation modem chips. Would you like any of the new modem chips? They're really great."
Apple: "No, not today."
Qualcomm: "Oh, I see. Okay. Um, thank you for your purchase order."
Apple doesn't just wake up one morning and say, "Let's make a 5G modem chip today." They have been working on this for years probably before they acquired the cellular modem business cast off by Intel. Apple's prototype chips have been side by side with their supplier chips in their labs.
In the same way, Apple probably started working on Apple Silicon M-series chips in earnest since the A8 SoC's debut (first 64-bit SoC widely deployed in a consumer mobile device).
If Apple does go with their own in-house 5G modem with the iPhone 17 lineup (fall 2025), Qualcomm would probably know in spring 2025. Or Apple could choose to debut with another product like an iPad Pro earlier in the year. Either way, Qualcomm will know when an anticipated purchase order fails to appear.
For that, the U.S., et al, will yet again tighten the screws on any incoming semiconductor, AI, or aerospace tech attempted to be exported from the West.
That said, there are some likely benefits: cost savings, fewer royalty payments (cellular technologies are heavily patented), more control, more flexibility with feature choice (which might include jettisoning unneeded technologies), better performance-per-watt (less power), easier integration into Apple's designs, et cetera.
Most likely it would be a combination of several of these plus others I haven't listed.
Remember that Apple is highly motivated to design their own chips, they have done so for the iPhone, iPad, Apple Watch, AirPods, etc. creating custom ASICs for their own use. They do not sell this technology to others. This isn't restricted to SoCs, they have designed ASICs for more specialized purposes. They stopped use off-the-shelf Arm cores and now design their own (they use the Arm instruction set but no longer rely on reference designs). They have included some of their custom silicon alongside Intel CPUs, the T2 Security Chip is one such IC.
Every time Johny Srouji appears in an Apple keynote event, he repeatedly pounds home the "performance per watt" mantra. He has been doing this for years. Personally I believe this is one of the major drivers for Apple's 5G modem development program. No surprise here. There have been benchmarks from the very first iPhone that shows that web surfing over cellular data uses more battery than wifi.
Above all, Apple likes as much as control as possible where it makes the most sense. That's a major reason why Apple dumped Intel CPUs in the Mac product line. They are no longer at the mercy of Intel's product map (and its frequent delays).
Whatever they are working on, it's not a photocopy of an existing Qualcomm product, that's for sure. That's not Apple's style. If they can't offer significant feature differentiation, they really aren't that interested. That's why Apple has abandoned some markets (e.g., AirPort routers, standalone MP3 players) even ones that they formerly dominated.
Meanwhile you might want to refresh your view of what Qualcomm does;
https://www.qualcomm.com/
Meanwhile, the larger problem for Qualcomm is that China's slowing economy is dragging down smartphone sales, and that isn't going to be reversed all that easily.
would hate to see Apple finally launch their own 5g only to see it passed up a year later.
The important point is that it isn't happening anytime soon.
Remember that cellular technologies are industry standards. It's not like the technology magically appears on a chip without warning. Apple cellular modem engineers would consult the same design documents as Qualcomm engineers.
That said Apple is never on the bleeding edge of network technologies. If the first 6G cellular towers go live in late 2028, it might be autumn 2030 when Apple puts a 6G capable chip in their iDevices.
India is growing. Brazil is another growth opportunity for the technology industry. Neither India nor Brazil is fully saturated in the smartphone market.
What's more, is that your reading of the wider situation is iffy at best.
Just days ago you were claiming that Huawei premium phones had sold in the hundreds of thousands, or low millions, when I challenged that crazy claim.
Now you are suddenly banding 40 million units around!
I will tell you this though. Huawei accompanied the Mate 60 series launch with an official note saying that the Mate line alone had topped 100 million units.
Depending on how you evaluate the SoC it will be competitive or not. Obviously, by being based on an older node, it isn't. It can't match the power efficiencies or density of the latest nodes. But then again that wasn't the goal in the first place was it? Of course it wasn't.
But is it (the SoC) , and by extension, the phone competitive in other ways? Most certainly yes.
By your own admission the smartphone market in China is in decline. The pool of buyers is harder to reach for various reasons.
Apple will release its refreshed line into that same market next week, but Huawei, (by your estimates) just sucked 40 million purchase decisions out of that pool! Wow!
That is very competitive.
So much so, that the internet, tech and mainstream media is on fire with news about the phone.
And hardly anything is known about it.
We already know about the knock on effect for Qualcomm (see link included in the article) but let's be realistic. Yes, Qualcomm will lose millions in unit sales from Huawei (and later from Apple) but it never had that business in its plans in the first place. It's a loss but a relative loss. It was just a bonus for which it said 'thank you very much' and filled its pockets.
Huawei sales resulted from sanctions and Apple sales from Apple’s huge strategic error.
I don't know exactly how things will play out in the near term but absolutely everything I've said up to now is happening.
The Mate 60 series is now reportedly 90% Chinese made. A far bigger percentage used to be sourced from US companies.
The billions of dollars that used to flow into the coffers of US companies is now gone - forever.
Not only that. Huawei has rejigged literally thousands of components in its supply chain to remove US technology completely and has created new competitors to those US companies. That's a huge double whammy. Billions lost in revenues that are needed for R&D and those same billions are now being pumped into the R&D budgets of non-US interests. That means Chinese or non-US companies will improve on what they have already created and very likely outperform their US counterparts. For starters the market is much bigger.
And in relative terms, all this has happened at light speed pace. Far faster than you yourself predicted, and the Hawks in the White House are now utterly panic stricken by what this phone represents, even if at present it is only symbolic. Just look at the flurry of comments coming out of the administration.
If it weren't competitive no one would care, would they? They wouldn't give a hoot. The reality is they care a lot. Its potentially their worst nightmare come true.
We already know (per linked article at least) that Huawei will only release Chinese processors going forward. It seems like they have a plan. Or that's what Kuo is saying. He's actually saying more because he's envisaging a sales impact from Huawei on QC that is bigger than than the Huawei phone sales he is hinting at. Does he think that Kirin chips will end up in mid-range non-Huawei Chinese phones?
If we dip into the wacky world of rumours we will see that HarmonyOS will launch on PCs next year and that a rumoured P60 Pro Plus with a Kirin 9100 will launch next week. I'm not going to go out on a limb with those projections but I do know the Kirin 9000s contains a self developed GPU as per some sources and that in AI, Huawei has a upgraded Ascend line which reportedly matches Nvidia's best selling products.
At the very least and as a result of all this, it will be harder for Apple to sell new iPhones in China when compared to market conditions in 2021 and 2022.
There are other headwinds too of course.
At the moment at least, Huawei seems to have hit the ground running with this release. Now the test is real world performance and the user experience. It's basically a completely new phone on a component level.
A balanced viewpoint:
https://interestingengineering.com/culture/huaweis-new-chip-changes-tech-landscape
The fact that we still won't see one until the rumored 2025 date confirms this.
What happened was that Apple had a Yikes! moment when Intel didn't deliver.
This was a strategic error of enormous magnitude. They had got into a massive spat with Qualcomm and effectively put all their 5G eggs in one basket.
When Intel failed to deliver on time Apple had to swallow hard, lick the toad and bear it.
Short term, that meant abandoning the spat with QC, entering into an agreement with them to supply 5G solutions. Mid term, it meant spending billions on buying some of Intel's 5G portfolio of patents and the engineering division behind it.
It was realistically the only way to move forward as other options weren't feasible.
Long term I think it's probably a good thing.
In time Apple will reach parity, for example with the Nvidia, GPUs probably by the M4, M5 SOC’s. For many of those things you mentioned you don’t need to be the best necessarily at the beginning it’s where you end up, for example, the South Korean car industry is a hell of a lot better than those pieces of crap that they made in the 1980s.