2024 Apple Devices

wco81

Ars Legatus Legionis
28,661
Seems the main thing left for this year would be iPhone 15.

Gurman said no new iPads this year but new iPad Pro may come in first half of 2024, including possibly a 14-inch model.

The products I'm interested in, talk of a 5K iMac with M3 processors and maybe new Apple TV 4K at lower price points.


For the second half, it sounds like iPhone 16 may not just be some incremental bump over iPhone 15, which is suppose to have new design as well as periscope options. Apple may be looking to push more tent pole or design features every year since iPhone sales have stalled in the past year.

Also, big changes for 10th anniversary Watch due in 2024 including new band attachment system, according to the front page article.

 

Honeybog

Ars Scholae Palatinae
2,075
For the second half, it sounds like iPhone 16 may not just be some incremental bump over iPhone 15, which is suppose to have new design as well as periscope options.
I’m hoping this isn’t true. I’m due to replace my XS,* but I’m really tired of the 12, 13, 14 design and banking on something interesting to come out in September.

* Primarily due to a damaged display connector, but also because the iOS 17 beta is the first one that makes my XS feel slow the way phones used to feel slow before the 6S. I’ll really miss 3D Touch, though.
 
  • Like
Reactions: Bonusround

dal20402

Ars Tribunus Angusticlavius
7,234
Subscriptor++
Mostly curious about the M3 MBPs and MBAs.

I'm probably going to be going down from my current two-machine life (2017 iMac Pro + 2021 MBP) to a single machine. The MBP would be suitable except that it doesn't really have enough RAM. So I'll probably sell both of the existing machines to get a new one.

Our iPads are all good and we'll be replacing both iPhones in 2023, so I don't have much reason to care about anything other than Macs.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
I think it will be interesting to see what they do with the non-Pro iPhones next year in terms of SoC. Last year and (by all accounts) this year they’ve used the prior year Pro chip. But next year the Pros will be on on the N3E node which is cheaper to manufacture than this years N3B. It won’t make sense for them to use the prior year chip in 2024.

My assumption is that they’ll go back to what they did in the iPhone 13 generation - use the same SoC in both Pro and non-Pro but bin them so the non-Pro has fewer GPU cores and potentially a bit downclocked.
 
  • Like
Reactions: mklein

iPilot05

Ars Tribunus Militum
2,780
Subscriptor
I think if a large-screen iMac dropped I'd be first in line to replace my trashcan Mac Pro. Sure it's running well enough but I'm sick of it being a space heater during the summer. Also it's lost it's software upgrade status and already one OS behind.

The new rumors of an Apple Watch X is also very intriguing to me. I'm rather amazed we've gone this far with barely any change to the physical design of the Watch.
 

Bonusround

Ars Scholae Palatinae
1,060
Subscriptor
I think it will be interesting to see what they do with the non-Pro iPhones next year in terms of SoC. Last year and (by all accounts) this year they’ve used the prior year Pro chip. But next year the Pros will be on on the N3E node which is cheaper to manufacture than this years N3B. It won’t make sense for them to use the prior year chip in 2024.

My assumption is that they’ll go back to what they did in the iPhone 13 generation - use the same SoC in both Pro and non-Pro but bin them so the non-Pro has fewer GPU cores and potentially a bit downclocked.

Not meaning to be contradictory at all, but I genuinely believe Apple is more likely to re-layout the prior year’s chip’s N3B design on N3E and continue fabricating it than the alternative, all in order to keep processor generation part of the plain/pro demarcation.

For years Apple‘s MO has been to grow the product lineup by creating distinct price and feature tiers. Take the iPad lineup. At times the feature distinctions between models are so slight that it’s almost comical. Yet this lineup yields a good product offering in nearly every conceivable price tier.

Having just added ‘processor’ to its iPhone model differentiation list I cannot imagine Apple turning back, even if it costs them some 3nm masks.
 
  • Like
Reactions: mklein

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
Having just added ‘processor’ to its iPhone model differentiation list I cannot imagine Apple turning back, even if it costs them some 3nm masks.
There’s no “re-laying out the A17 on N3E”. N3E uses different design rules than N3B so you’re talking about a fantastically expensive amount of engineering and design work before you even get to fabricating the mask set. And all to replicate last years chip. There’s zero chance that Apple does that.

They may stick with the A17 despite its expense but I’m doubtful. Binning the A18 just seems so much simpler/cheaper all around.
 
  • Like
Reactions: Tagbert

Bonusround

Ars Scholae Palatinae
1,060
Subscriptor
Offering binned iPhone CPUs is a level of complexity Apple will not introduce into the consumer’s iPhone-buying equation. Never gonna happen, IMO.

There’s no “re-laying out the A17 on N3E”. N3E uses different design rules than N3B so you’re talking about a fantastically expensive amount of engineering and design work before you even get to fabricating the mask set.
Are you certain this is so? That TSMC (and/or Apple) could or would not produce tools to aid such a translation? Intel have done it, famously with their backport of a 10nm CPU design to 14nm (11th-gen Core).

And ’die shrink‘ designs are very common, frequently seen with Intel‘s tick phase in their tick-tocking of the early 2010s, and also regularly in the console space as cost-reduced designs are phased in or introduced (see nearly any PlayStation slim).

Besides, according to your own (prior) estimates, the cost of laying out a design on a different process/node would be dwarfed by the cost of the EUV mask anyway, no?
 
Last edited:

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
Offering binned iPhone CPUs is a level of complexity Apple will not introduce into the consumer’s iPhone-buying equation. Never gonna happen, IMO.
But… they did exactly this less than 2 years ago in the iPhone 13 generation. The Pro used an A15 with 5 GPU cores and the non-Pro used the same chip with 4 GPU cores active. And they’re still selling that binned chip today in the iPhone 13 while the iPhone 14 has the A15 with all 5 GPU cores working.

I’m literally saying they’re going to use the 2 years ago strategy instead of the last year’s strategy. How is that a ‘never going to happen’?
 

Bonusround

Ars Scholae Palatinae
1,060
Subscriptor
But… they did exactly this less than 2 years ago in the iPhone 13 generation. The Pro used an A15 with 5 GPU cores and the non-Pro used the same chip with 4 GPU cores active. And they’re still selling that binned chip today in the iPhone 13 while the iPhone 14 has the A15 with all 5 GPU cores working.

I’m literally saying they’re going to use the 2 years ago strategy instead of the last year’s strategy. How is that a ‘never going to happen’?

Sorry, let me clarify: what (I think) is never going to happen is Apple offering binned chip options within the same iPhone model – as they do with, say, the MacBook Air. iPhone buyers will (continue to) select a model and get the (sole) chip that comes with it.

And I see what you were getting at now, given the proposition that they’ll return to a single A-series (but with variations, including RAM and binning) across the various models of a single iPhone generation – exactly as done before. My apologies for misunderstanding. Obviously I disagree and think they’re more likely to maintain the chip distinction across plain and pro by using A-series of different generations.

I’m more interested in a response to the rest of my post, but if not then that’s cool. Cheers.
 
Last edited:

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
I’m more interested in a response to the rest of my post, but if not then that’s cool. Cheers.
Are you certain this is so? That TSMC (and/or Apple) could or would not produce tools to aid such a translation? Intel have done it, famously with their backport of a 10nm CPU design to 14nm (11th-gen Core).
Since Apple is the only company using N3B I doubt that it’s worth TSMCs time to create automated tools for redoing layouts between it and N3E. And the smaller the process node the more expensive everything is, so I’m not sure Intel backporting a core design to 14nm is a good comparison.
And ’die shrink‘ designs are very common, frequently seen with Intel‘s tick phase in their tick-tocking of the early 2010s, and also regularly in the console space as cost-reduced designs are phased in or introduced (see nearly any PlayStation slim).

Besides, according to your own (prior) estimates, the cost of laying out a design on a different process/node would be dwarfed by the cost of the EUV mask anyway, no?
Die shrink designs in consoles are a pretty special case. They’re never to a cutting edge node so they’re using extremely well road-tested tools. And Intel tick tock die shrinks were for their new gen processor. You’re suggesting die shrinking a last gen processor to a cutting edge node for a low end SKU (relative to the Pro). That doesn’t add up for me.

And to your last question, no - I think the engineering/design/re-layout aspects are a substantial proportion of the cost of the masks. I don’t see how it makes economic sense.
 
  • Like
Reactions: Tagbert

Bonusround

Ars Scholae Palatinae
1,060
Subscriptor
Yeah, I agree the Intel backport is not a good compare for Apple’s situation. Moving Sunny Cove to 14nm is a singular example, with the company attempting to automate as much of the migration across (very) different process regimes as possible.

Die shrink designs in consoles are a pretty special case. They’re never to a cutting edge node so they’re using extremely well road-tested tools.

Well that hasn’t been the case. In the previous generation AMD gave console APUs first dibs on the new TSMC node.
(Very likely because this was AMD’s only business thriving at that time, but still.)

And console chip shrinks are a great compare: same design, strong incentive for compatibility, with an unambiguous objective of the same part for less $$.

You’re suggesting die shrinking a last gen processor to a cutting edge node for a low end SKU (relative to the Pro). That doesn’t add up for me.

I’m suggesting Apple are more likely to maintain the status quo and use different generation SoCs between iPhones 16 and iPhones 16 Pro.

You asserted that N3B is so expensive they’ll need want to jump off it by then. Iff that is the case – and the expense or potential savings are so dramatic – then I suggest Apple would rather find a way to cost-reduce the A17 than to re-introduce the flagship A-series as part of their lower-tier new iPhones.
 
Last edited:

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
You asserted that N3B is so expensive they’ll need to jump off it by then. Iff that is the case – and the expense or potential savings are so dramatic – then I suggest Apple would rather find a way to cost-reduce the A17 than to re-introduce the flagship A-series as part of their lower-tier new iPhones.
I mean “have to” is pretty strong. No one is holding a gun to Tim Cook’s head. But N3E is by all accounts going to be substantially cheaper than N3B and they’ll have an N3E part available at zero incremental design cost - just fuse off a GPU cores (which they’ll be able to do because the rumor is they’re going up to 6 GPU cores this year.)

The other aspect of this is that TSMC almost certainly doesn’t want to continue N3B. Apple is their only customer for it - everyone else is going straight to N3E. TSMC doesn’t want to keep that production line open just for the iPhone next year and then keep it available for when that iPhone 16 moves down to the middle of the lineup the next year. TSMC wants to reallocate every machine that is currently working on N3B to N3E as soon as possible.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
Surely the M3 will be on N3B. Will Apple face the same dilemma with that chip over the run of these devices?
My assumption is that yes M3 will be on N3B. But for all the reasons I think the A17 will be a short-lived chip, I imagine the M3 generation will be relatively short-lived and Apple will transition to N3E as soon as possible. Pretty much the only place they’re using an N-1 chip is the low end MBA. I could see them going to 6 P-cores in the M4 and using a 4 P-core variant as the low end MBA, dropping the M3 entirely after a year.

If the move from the M3 on N3B to M4 on N3E is relatively conservative in terms of microarchitectural updates (my assumption given that the M3 will be a substantial update) I could see it launching in fall of 2024. Then the M4 Pro/Max/Ultra in spring of 2025 and the Mac chips could be through their N3B period in a year and a half and Apple takes a breather and returns to the 18 month cadence while TSMC retires N3B completely.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
What’s the over under on whether we’ll see an Apple designed cellular modem in 2024? They’ve had that design team for 4 years now which even given Apples legendary patience is an awful long time for a team that size to go without a product.

And what will be the pipe cleaner SKU that they stick it in? An iPhone? iPad? The Apple Watch? An MBA?

At some point it’s shit or get off the pot time. I’d like to see it in a cellular iPad in 2024.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
What happens when they stop using Qualcomm baseband?

Will they pay licensing to them instead?

Presumably the settlement has provisions for Apple stopping using Qualcomm in their devices?
In terms of competitive advantage, the sooner Apple can stop putting dollars into Qualcomm’s R&D department the better.
 

Bonusround

Ars Scholae Palatinae
1,060
Subscriptor
What happens when they stop using Qualcomm baseband?
Apple no longer buys Qualcomm silicon, paying Qualcomm prices.
Will they pay licensing to them instead?
Yes. Just as they do today with other holders of cellular IP.
Presumably the settlement has provisions for Apple stopping using Qualcomm in their devices?
Sure. Apple already had the ability to end the use Qualcomm; as you'll recall, they began this very process with the substitution of Intel's LTE modems. What Apple sought in its suit and settlement with Qualcomm was better terms for doing so, especially as the world migrated to 5G.
 
Last edited:

wco81

Ars Legatus Legionis
28,661
Gurman dumped new rumors over the weekend, 11 and 13 inch iPad Pros with M3 chip, available in spring or fall of 2024.

At least the 13-inch model will be available with OLED.

Also a new design of the Magic Keyboard. Unknown if current MK will work with the new iPad Pros. Or if the new MK will work with the older iPads.

Supposedly the MK will make it more like a laptop, at least physically.


Well I was considering the MK a few months ago. Glad I didn't buy.
 
  • Like
Reactions: Carhole

Chris FOM

Ars Legatus Legionis
10,001
Subscriptor
Is there a more muddled product in all of Apple than the iPad? I’ll allow the 9th generation as a low-price legacy entry similar to the iPhone SE, but do we really need five active models? Between the 10th generation, mini, Air, and two different Pros it’s a blizzard of salami-sliced features that seem more aimed at justifying a large variety of price points than actually creating meaningfully differentiated products.
 

Bonusround

Ars Scholae Palatinae
1,060
Subscriptor
No, there isn’t. Still, this conclusion is reinforced by the coinciding iPad sales slump – were they selling like hotcakes many would hail the lineup as genius.

I’m sure there’s an interesting set of circumstances behind this. It feels like Apple formulated an ambitious plan to progressively roll out and segment features all under a unified, flat-sided design language, largely beginning with the 2018 iPads Pro. But whether by slip or misstep it didn’t come together, or did so but out-of-order, or a bit of both.

I think I can identify (and explain) the most awkward points of the lineup:

iPad mini: smaller yet… more expensive?
– this has always been the case; the mini has its niche uses (e.g. pilots, industrial applications) and Apple is confident those users will pay the price premium

iPad 9th Gen: why still in the lineup?
– this is the classic form factor adopted by schools, and schools need to replenish hardware b/c kids are monsters
– compatible with lightning-based charging docks/carts; separate non-laminated glass is heartier, more repairable

iPad Air vs. iPad Pro 11”: why so close in the lineup?
– actually, this makes perfect sense to me, closely mirroring the plain/pro iPhone distinction; not an issue

iPad Air vs. iPad 10th Gen: why so close in the lineup?
– this is the ugly one; were I to guess: the 10th Gen is intended as the ‘new iPad for Schools’
– Apple introduced the 10th Gen to begin transitioning away from the prior form factor
– once the 10th Gen (or equivalent) can solely occupy the bottom of the lineup things will feel more balanced

Landscape Webcams or Inductive Pencil Charging?
– feels like Apple Design stepped in it with this one, heheh
– I‘d expect that next year’s new iPads Pro will reveal Apple’s solution to this puzzle
 
Last edited:

cateye

Ars Legatus Legionis
11,760
Moderator
do we really need five active models?

"If you can't dazzle them with brilliance, baffle them with bullshit."

I think it's a sign that there's either conflict within about Apple the direction for the iPad, or Apple has run out of ideas of what they want to do with it as a product other than milquetoast iteration. Hardware wise, they're just such beautiful objects. Home run. But in service of... what? What's the grand vision? Where do these hardware advancements take us that's unique to iPad and justifies its existence, and the existence of so many product levels? No clue.
 

Chris FOM

Ars Legatus Legionis
10,001
Subscriptor
No, there isn’t. Still, this conclusion is reinforced by the coinciding iPad sales slump – were they selling like hotcakes many would hail the lineup as genius.

I’m sure there’s an interesting set of circumstances behind this. It feels like Apple formulated an ambitious plan to progressively roll out and segment features all under a unified, flat-sided design language, largely beginning with the 2018 iPads Pro. But whether by slip or misstep it didn’t come together, or did so but out-of-order, or a bit of both.

[snipped for brevity]

I agree with everything you said (including the snipped portion), and would add one more: the presence of three different models all sharing the 10.9/11” screen size. I’d say the iPad Air does more to distinguish itself from the 10th gen and less from the 11” Pro than the regular iPhone does from the pro, but either way each iPhone screen size has only two current models (ignoring prior-year devices aimed at hitting lower price points). There’s just not enough room to have three models all with the same screen size.

And I’d be genuinely surprised if 9th gen is long for this world. Keeping a legacy form factor around makes sense for now, especially the Lightning connector, but that A13 is closing in on its fourth birthday and the same Lightning port that’s attractive today is gonna feel real dated in 18 months. It’s there to keep the low price point warm while they decide what to do with the rest of the range.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
I think I can identify (and explain) the most awkward points of the lineup:
I agree with all of this. I’d also point out that Apple fans tend to over-index on lineup simplicity as a holdover from Jobs famous quadrant. But the year he rolled that out Apple was selling a little over two and a half million Macs total all year! There is likely an iPad model that sells that well in a quarter.

And when you look at customer sat, it’s off the charts. If people were having a bad/confusing buying experience that would manifest itself as poor customer sat as customers regretted not getting the other model. There’s no evidence of that at all.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
I think it's a sign that there's either conflict within about Apple the direction for the iPad, or Apple has run out of ideas of what they want to do with it as a product other than milquetoast iteration. Hardware wise, they're just such beautiful objects. Home run. But in service of... what? What's the grand vision? Where do these hardware advancements take us that's unique to iPad and justifies its existence, and the existence of so many product levels? No clue.
i dunno - I think with the Apple Pencil, Magic Keyboard and incrementally improving software the vision is pretty clear. This thing will be a more flexible laptop replacement slate over time. It’s heading in the direction of being able to do anything a Mac can + all the things a Mac ergonomically can not. It’s just that Apple is taking a slow and steady approach to building out the software capabilities with an eye towards keeping the UX simple for those who want that and optionally allowing complexity for those who need it.

Pretty much every year for the past decade one or two “you can do it on a Mac but not an iPad” functions made it to iPad. Apple never ever markets it as “the iPad is getting more Mac-like”. And the “now the iPad can do this thing you only used to be able to do on Mac” features are always lost in a sea of new-for iPad and borrowed from iOS features so they never stand out. But they’re there. You can get a dramatically more Mac-like experience on iPad now than you could a decade ago, and I see no reason to assume that the frog won’t continue to heat up over time.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
“Heading in the direction” is doing an awful lot of work here.
He specifically asked for vision. I interpret that as “where it’s heading” not where it is 🤷‍♂️

I mean between Files, Drag and Drop, external monitor support, mousing support, increasing RAM limits, Stage Manager… can anyone honestly compare a circa 2013 iPad and today’s iPad and claim it hasn’t gotten progressively more Mac-like?

You can say “It’s not there yet” and I’d be inclined to agree with you. But claiming there’s not vision, no directionality of progress, just seems blinkered to me.
 

cateye

Ars Legatus Legionis
11,760
Moderator
This thing will be a more flexible laptop replacement slate over time.

I don't think Apple has satisfactorily defined what a "laptop replacement slate" is, though. Or why that's a category that needs to exist, or sufficiently motivated developers to leverage the same power they would have in a laptop in a slate form (the heavy-handed and smothering closed App Store model certainly isn't helping, either).

The iPad is this constant, remarkable evolution of hardware (note the rumors for 2024!) paired with an unbelievably slow and thin trickle of software improvements that barely begin to address the gulf that grows bigger with each product cycle between hardware and software. So if there's a vision for "laptop replacement slate" that ties these halves into a whole, it's certainly not obvious to me.
 

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
I don't think Apple has satisfactorily defined what a "laptop replacement slate" is, though. Or why that's a category that needs to exist, or sufficiently motivated developers to leverage the same power they would have in a laptop in a slate form (the heavy-handed and smothering closed App Store model certainly isn't helping, either).

The iPad is this constant, remarkable evolution of hardware (note the rumors for 2024!) paired with an unbelievably slow and thin trickle of software improvements that don't even begin to leverage the gulf that grows bigger with each product cycle between hardware and software. So if there's a vision for "laptop replacement slate" governing these two things, it's certainly not obvious to me.
I think of it like a series of bridges being built along a narrowing river at the same time. If your house is at the narrow end, you can cross over when the first segment of the bridge is done. If you’re at the wide end there can be ten segments done and the only gap remaining is 20 feet and it doesn’t matter - you still can’t cross over. It doesn’t mean no work has been done just because you can’t crossover yet!

File access, drag and drop, mouse support, Stage Manager are all totally necessary but not sufficient steps toward your workflow. And there are hundreds more of those bridge segments that have been worked on over the last decade. Some of them have a really long time horizon like moving from kexts in system space to dexts in user space that will take a while to play out on the iPad. Others like mouse support are immediately obviously useful.

So I would say that as the hardware gets more powerful, the software is slowly getting closer and closer to bridging your workflow needs… BUT the more powerful the hardware becomes, the more frustrating it is that the bridge isn’t done.
 
  • Like
Reactions: cateye

wrylachlan

Ars Legatus Legionis
12,769
Subscriptor
I understood what you meant. My point was that you can head in the direction of something and be close to achieving it, far away, or laughably distant (After my last jog, I'm "heading in the direction" of being an Olympic athlete).
Sure. And my position is that they’re much closer to Mac replacement than you give them credit for. For very many workflows they’re already there. And for others they’re meaningfully closer than they were even 5 years ago. Everyone uses laptops differently. I haven’t seen my wife do anything on her laptop that she couldn’t do on a modern iPad.

I work 100% in a Remote Desktop so technically I could do 100% of my work from an iPad now that mousing support is in there.