Next-Gen Apple Chip Technology on Track to Reach Production in 2025

[MacRumors]

Next-generation 2nm chip technology destined for future Apple devices is on schedule to begin production next year, DigiTimes reports.

TSMC chip fabrication facilities will begin installing equipment designed for 2nm chip production in April at the earliest. Apple was the first company to utilize TSMC's initial 3nm technology with the A17 Pro chip in the iPhone 15 Pro and iPhone 15 Pro Max, and the company is likely to follow suit with the chipmaker's 2nm chips. With production set to begin in 2025, 2nm chips are likely to make their first appearance in Apple devices soon after.

Last month, the Financial Times reported that TSMC had already demonstrated prototype 2nm chips to Apple ahead of their expected introduction in 2025. Apple is said to be closely aligned with TSMC in the race to develop and implement 2nm chip technology, which will surpass their current 3nm chips and associated nodes in terms of transistor density, performance, and efficiency.

Today, DigiTimes added that the chip supplier is also believed to be evaluating which of its plants will be the first to produce even more advanced 1.4nm chips in 2027. The company began mass production of its enhanced 3nm node, which is likely to appear in Apple devices for the first time later this year, in the fourth quarter of 2023.

Article Link: Next-Gen Apple Chip Technology on Track to Reach Production in 2025

 

[DougTheImpaler]

I guess that blows away any chance that the extremely mediocre upgrade to M3 (especially on the M3 Pro with its reduced performance core counts) was any sort of stopgap due to rumored low N3E yields. Sigh.

 

[coffeemilktea]

I can't wait... scrolling through Twitter and playing ad-filled mobile games are going to SO fast and snappy now.

 

[Biro]

I suspect that, with this production timeframe, I may end up skipping the 3nm era completely. I just bought an M2 Max Mac Studio and an iPhone 15 Plus. I will probably update my iPads this year - likely new M2 Airs. I’ll be good for a while.

 

[Rychiar]

why isn’t 2 nanometer already in the Vision Pro!😤

 

[bradman83]

To be clear, it’s next gen TSMC chip-making technology, not Apple technology. Apple will happily buy the manufacturing capacity to make their chips with it though.

 

[hoorayforhollywood]

So -
M4 Q4 this year - enhanced 3nm. (Sort of M3S, like the M2 was?)
M5 Q4 2025 with 2nm.
M6 2026 with enhanced 2nm?
M7 2027 with 1.4nm.

Is that about it?

 

[giggles]

To be clear, it’s next gen TSMC chip-making technology, not Apple technology. Apple will happily buy the manufacturing capacity to make their chips with it though.

It's chicken and egg with Apple's supply chain, behind every technology and R&D investment one should ask who's paying for it.

 

[bradman83]

I guess that blows away any chance that the extremely mediocre upgrade to M3 (especially on the M3 Pro with its reduced performance core counts) was any sort of stopgap due to rumored low N3E yields. Sigh.

An 18% single core performance increase is hardly mediocre. Yes the Pro was hobbled by Apple’s reconfiguration of the chip to a true mid-range model (which affects multicore only) but an 18-20% year-over-year performance increase is nothing to sneeze at. The performance jump from Intel to M1 was a one-time thing. It’s going to be incremental gains from here on out barring a major microarchitectural shift.

 

[iBluetooth]

I guess that blows away any chance that the extremely mediocre upgrade to M3 (especially on the M3 Pro with its reduced performance core counts) was any sort of stopgap due to rumored low N3E yields. Sigh.

After following Intel throughout the years, I am kind of amazed to see these words "extremely mediocre upgrade" used for the M2 to M3. But, I know after seeing the jump from Intel to the M1, we were hoping for a bigger leap.

 

[jonnysods]

Still loving my M1 Max MBP - she's a rocket. I used to switch out Macs every year or two, but I think I'm going to keep this one for 3-4 more years, its so good at what it does.

 

[citysnaps]

An 18% single core performance increase is hardly mediocre. Yes the Pro was hobbled by Apple’s reconfiguration of the chip to a true mid-range model (which affects multicore only) but an 18-20% year-over-year performance increase is nothing to sneeze at. The performance jump from Intel to M1 was a one-time thing. It’s going to be incremental gains from here on out barring a major microarchitectural shift.

Sure beats Intel's 5% (or was it 3% at one point?) YOY performance improvements when Mac was Intel-based.

 

[Magnus81]

So -
M4 Q4 this year - enhanced 3nm. (Sort of M3S, like the M2 was?)
M5 Q4 2025 with 2nm.
M6 2026 with enhanced 2nm?
M7 2027 with 1.4nm.

Is that about it?

Yeah, looks like that! My next upgrade from 14” M1 Max will likely be the 14” M5 Max then, exciting! 😄

 

[johnny1970]

Next-generation 2nm chip technology destined for future Apple devices is on schedule to begin production next year, DigiTimes reports.

​

So a techie question: 2025 is 2nm, then it will be 1nm. Where does it go to then?

 

[giebe]

Well M1 still works great and will turn 4 this year. So maybe an upgrade will be interesting in 2025 or even 2027. 7 years is pretty good then.

 

[giebe]

So a techie question: 2025 is 2nm, then it will be 1nm. Where does it go to then?

0.7 and 0.5nm

 

[Ivander]

When the time comes I'll measure it just to make sure they don't lie.

 

[thingstoponder]

The company began mass production of its enhanced 3nm node, which is likely to appear in Apple devices for the first time later this year, in the fourth quarter of 2023.

Oh wow I can’t wait.

 

[citysnaps]

Well M1 still works great and will turn 4 this year. So maybe an upgrade will be interesting in 2025 or even 2027. 7 years is pretty good then.

Yeah... I still have an M1-based MBA I use for travel and haven't found it computationally deficient in any manner (compared to my M2 MBP and Mac Studio).

OTOH, I'm not mining bitcoin or decrypting secret Russian encrypted communications with it.

My M1 MBA is still a great computer.

 

[HobeSoundDarryl]

So a techie question: 2025 is 2nm, then it will be 1nm. Where does it go to then?

Change the measure so that things can still get quoted in whole number reductions and start over from some bigger number to work it down. Yes, that means that what's actually happening is fractions of the nm measure that has been used for quite some time now but the average Joe will just see a number printed on a chip and assume the bigger number loosely tied to the smaller architecture is better. There's plenty of measures that already exist to gauge fractions of a single nm.

Simple example: measure price cuts in dollars until you get down to $1. OMG, now what do we do? Change the measure to cents and you can start at 99 and work your way down to 1 cent. OMG, now what do we do? Well if gas stations can charge 9/10ths of a cent, how about we buy 10 more generations by that measure... or maybe make that 99/100ths and then we buy 99 new generations by splitting that single cent. By just those simple examples, we've basically created 99 + 99 or 198 generations to get us to approx. M204 chips in approx. year 2,230.

When you get to 1/10th or 1/100th, OMG, now what do we do? there's always the thousandths measure: 999, 998, 997 to get another 999 generations. And so on. That will get us to roughly the year 3,229 for the M1,203 chips. Tip: I've heard that M1,205 is the one to wait for, so dodge that M1,203 and M1,204 if you can.

There's always a way for marketers. When something actually no longer matters, they'll just jump on something else and make it "important" through messaging.

 

[smulji]

I still have an M1-based MBA I use for travel and haven't found it computationally deficient in any manner (compared to my M2 MBP and Mac Studio).

Once Apple starts to take generative AI seriously, it will be.

How M1 Macs may lag behind

M1 CPUs support ARMv8.5A, which doesn’t support the new bfloat16 floating-point format now widely used in AI. That’s likely to put them at a disadvantage.

eclecticlight.co

 

[roar08]

.

 

[Sorinut]

Once Apple starts to take generative AI seriously, it will be.

How M1 Macs may lag behind

M1 CPUs support ARMv8.5A, which doesn’t support the new bfloat16 floating-point format now widely used in AI. That’s likely to put them at a disadvantage.

eclecticlight.co

Just because they take it seriously doesn't mean I will. I have zero use for AI for anything other than creating excel macros.

I plan to keep my M1 Air for as long as I kept my 2012 MBP; eight years.

 

[kagharaht]

For those of us "that would be me LOL" who are not verse with these NM stuff. I found this read easy to understand.

Chip Nanometer Technology explained, and why the smaller the better

You may have spotted the term nanometer or 'nm' when reading about a chipset or going through a smartphone's specifications. But what is its significance? Does it really matter? Let's check out.

pocketnow.com

 

[NT1440]

Once Apple starts to take generative AI seriously, it will be.

How M1 Macs may lag behind

M1 CPUs support ARMv8.5A, which doesn’t support the new bfloat16 floating-point format now widely used in AI. That’s likely to put them at a disadvantage.

eclecticlight.co

I’m curious as to why people think the demonstrated use-cases of generative AI have consumers clamoring for it. Other than some colleagues with “email jobs” I really don’t get the hype.

Maybe it’s because I’m not embedded in the various platforms for scheduling?

If your tasks are completed on an M1 today, it’s still going to do them without trouble 5+ years from now.