Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Where are those 3nm Apple Silicon products?
- Thread starter hajime
- Start date
- Sort by reaction score
On their way. The earliest anyone is expecting one of these is WWDC, and it's not even clear if they will be based on N3B (I think they will though).
The 3nm chip will be available on iPhone in September with A17 Bionic and on Mac probably in October / November with M3.
Production at TSMC on N3B started December 29, 2022. Just over 3 months.
Still begs the question of where the products are given the recent reporting that no M3 Macs at WWDC. Far too early for A17 production if it’s true that Apple was TSMC’s only N3 customer at that point in time.
Production of these doesn't go from 0-100, it's a ramp up, and initial quality is pretty bad - wouldn't surprise me if 50% of the chips are rejected at first.
It is not like Apple will just all of a sudden start working on the A17 SoC for iPhones, that design has been finalized and test dies produced awhile ago...
The ASi Mac Pro is the Mac that NEEDS released ASAP, Apple NEEDS to complete their transition to Apple silicon, which they are well past their vague two-year window on, WWDC 2023 will mark three years since Tim Cook told us about the transition and the DTK was made available to developers...
TSMC has been mass producing N3B wafers since the end of December 2022, so a solid three+ months now, and Apple is supposed to be the only customer for those wafers...
I think we will see M3, based on A17 cores, at WWDC 2023; these will be made with the N3B process, these will be M3 Ultra and M3 Extreme SoCs destined for the ASi Mac Pro...
The A17 SoCs for the iPhones and the M3/M3 Pro/M3 Max SoCs will be on the N3E process, which should start mass production in the next few months or so...?
The ASi Mac Pro is the Mac that NEEDS released ASAP, Apple NEEDS to complete their transition to Apple silicon, which they are well past their vague two-year window on, WWDC 2023 will mark three years since Tim Cook told us about the transition and the DTK was made available to developers...
TSMC has been mass producing N3B wafers since the end of December 2022, so a solid three+ months now, and Apple is supposed to be the only customer for those wafers...
I think we will see M3, based on A17 cores, at WWDC 2023; these will be made with the N3B process, these will be M3 Ultra and M3 Extreme SoCs destined for the ASi Mac Pro...
The A17 SoCs for the iPhones and the M3/M3 Pro/M3 Max SoCs will be on the N3E process, which should start mass production in the next few months or so...?
And it takes about 3 months from wafer start to chip final packaging. I don’t expect any mature products until fall.
WWDC 2023 will be a full five months since N3B went into mass production, ASi M3 Ultra / M3 Extreme Mac Pro @ WWDC 2023...
The quoted wafer starts was 45,000/month. That gives somewhere around 225,000 wafer starts and 2-3 months of stockpiling dies (make it 2.5 months). With relatively low yields at maybe 200 M2 size dies per wafer that makes 22,500,000 dies by WWDC. For larger dies, divide accordingly but it still is a huge number of chips.
Eh, hardly. I know some people are waiting but they could probably release the earlier M3 gen chips before getting to the MP.
I just hope an Extreme is still on the table.
It appears that TSMC's 3nm yield reaches 63%.
Au, that hurts. Remains to be seen whether Apple will absorb the cost and reduce their margins, roll it over onto the customers (making already very expensive products even more expensive), or find some other way to cut costs.
No idea. But, if TSMC N3B yield was 63%, it would match other rumors. Rumors claim that the yield of the first version of TSMC N3 is poor, so most customers (except Apple) would wait for TSMC N3E.
Analysts Estimate TSMC's 3nm Yields Between 60% and 80%
There are thousands of ways to improve yields.
Apple only sold around 7 million laptops in 4th quarter of 2022, so that would cover more than total units moved easily.
Last edited:
The 63% yield is inline with other reports that I’ve seen. It should increase going forward but that yield doesn’t bode well for using N3B early for larger dies. The estimates for M2 sized dies was around 350 dies per wafer. A 63% yield puts that at about 220 good dies per wafer. Two months of manufacturing at 45,000 wafers per month is still a staggering number of chips. About 20M.No idea. But, if TSMC N3B yield was 63%, it would match other rumors. Rumors claim that the yield of the first version of TSMC N3 is poor, so most customers (except Apple) would wait for TSMC N3E.
Analysts Estimate TSMC's 3nm Yields Between 60% and 80%
There are thousands of ways to improve yields.www.tomshardware.com
Does it make any sense?
wccftech.com
Apple’s A17 Bionic Could Switch To TSMC’s N3E Process In 2024 For A Lower Production Cost, But It Will Come At A Drawback
One rumor claims that the A17 Bionic for the iPhone 15 Pro models will switch to N3E process from TSMC starting next year
wccftech.com
Does it make any sense?
View attachment 2216369
Apple’s A17 Bionic Could Switch To TSMC’s N3E Process In 2024 For A Lower Production Cost, But It Will Come At A Drawback
One rumor claims that the A17 Bionic for the iPhone 15 Pro models will switch to N3E process from TSMC starting next year
Yes. Given the reports of low-yielding N3B, it's unlikely Apple designed A17 to push the limits of that process. Apple knows they need millions of chips. Apple can lower their target clock, say from 3.65 GHz to 3.5 GHz to make A17 suitable for N3B and E.
It only makes sense to use expensive and low-yielding N3B for iPhone. Everybody knows September is iPhone month. But everything else including Mac and iPad doesn't have an expected date. Those can wait until early 2024 for chips fabbed on N3E.