The future is very unclear. A16 certainly LOOKS like it was not on the roadmap, it was just an A15 boosted as much as possible by N4 rather than N5, and the real next design (the one that gives us IPC improvements, not just process improvements) is targeting N3B.
So will M3 be based on that next design or will it be based on N4 and the A16? Who knows??? I think we are still at the stage where both covid and the Mac Apple Silicon transition were one-time events that generated all sorts of one-time delays that it's not yet possible to talk of any sort of Apple "pattern" in how long it takes to move a design from phone to mac to high end.
Seems likely N3B for the M3 and that hardware accelerated RT GPU that was canned for the iPhone 14. Another year's delay to the hardware RT featuring next gen GPU architecture would be a sore letdown, as would M3 being on 4nm.
I agree on that RT on next gen GPU is really needed and would be a letdown on M3 if it wasn't there. However, I don't know that the Mac lineup really needs to be on the very latest node. N4 would still be a process improvement and fine for the next generation. Phones need the bleeding edge efficiency. The A17 has to be on N3.
Yeah, I've seen the rumors and obviously would prefer them to go with either 3nm process for M3. I'm just saying that I don't think it's actually necessary. The assumption would be that N4 would be more cost effective given that it's a more mature process with higher yield rates. They'd still gain efficiency and performance improvements. Obviously, moving to N3 (or N3E) would bring even greater improvements.
If they are shipped late 2023 or early 2024 that's plenty of time for good yields on whatever flavor of N3. TSMC showed a graph late last summer comparing yields of N3E in various stages prior to the start of risk production and they were quite a bit higher than N5 yields during comparable - around 80% for SRAM, mobile and logic test chips three months prior to the start of risk production.
If they were already hitting 80% in August 2022 they should be well into the 90s by the year+ timeframe Apple wafers would enter mass production.
Apple is (unfortunately) heavily reliant on TSMC. The first thing they will push out is the iPhone 15. The TSMC-3nm has been in the works for awhile, and has been delayed by 1-2 years due to the pandemic and chip supply issues. The upshot is that when TSMC does ship it out, it will come out in volume, as all of the work has been done and there are little to no yield issues to complain about.
So the iPhone 14 (which was delayed) was eventually cancelled, and we got the iPhone 14 instead. No that is not a typo you read that right. The "iPhone 14" we got is basically an iPhone 13. The iPhone 15 in Q3 2023 will come with the A17-Bionic chipset using TSMC-3nm lithography and a new architecture that Apple will base on ARMv9. So yeah, big upgrades incoming.
I know the yields are so good that Apple might also launch the Macbook Air with the M3 chipset for (Christmas season) Q4 2023. There's also the iPad Pro, but that was refreshed sooner, most likely the third device to come with M3 around Q2 2024. The Macbook Pro has a healthy profit margin for Apple, and were delayed by a decent while, still they won't come with M3 Pro/Max till much later around Q4 2024. The Mac Mini only just got updated and it's not the best money maker, so that will be on the tail end around Q1 2025. With the iMac not having a great track-record, they will get an M2 refresh soon (Q2 2023), then get an M3 launch very late as well around Q3 2025. The Mac Studio likely will not get an M2 Ultra chipset, and will be the last to get upgraded to the M3 Ultra around Q3 2025.
The delays are due to TSMC struggling with their original 3nm target. That's why there is now a new, less dense 3nm node that everyone will be using instead and why they don't expect much revenue out of the current n3 node. If the original N3 was ready for high volume Apple would've probably jumped on it to stamp out the competition like they did with the first run of 5nm.
Amazing stuff. being on 3nm in 2023. I remember working on chip designs on 130nm and 90nm back in 2003 thinking how incredible that was. We've made crazy strides in 20 years. Wonder what process we'll be on in 10 or 20 years.
Well 130nm and 90 nm were allmort 130nm and 90nm while 3nm is something like 30 to 40nm in reality so the advancement is not as high as sales man claims…
Things get even wonkier with TSMC's 2nm as they switch to GAA so there is even more of the transistor in 3 dimensions. The nm terms are more marketing than anything else these days which Intel acknowledged with their new names.
If this is what you are arguing, a 130nm has a gate length of around 130nm while a 3nm chip has a gate length of about 16-18nm. The 30-40nm is pulled out of your ass. If you are talking about the pitch, then at the 130nm node, the pitch wasn't 130nm either...
Its close to reaching its limits. at 3nm my guess is its just 10-20 atoms. could they go to 1nm? it will be remarkable if they do. but you see how close they are to the limits.
Oh no, it's not 3nm. "3nm" or "5nm" today is no longer a true indication of transistor size. Actual size is around 30nm. It's more to do with marketing instead.
Back then when we had only planar transistor, yes. But today we have finfet, gaafet etc so the size is just more of an estimation of planar equivalent with lots of assumptions made.
There is no standard of how you define 3nm or 5nm. It entirely up to the company.
"There is no standard for defining 3nm or 5nm. It is entirely up to the company."
There is some semblance ot standard as established by IMEC. It would be bad for the company's image if it's too far off IMEC's expected specs for that node...
N2 starts ramping up in 2024, not 2026. This is a quote from the earning call: "our N2 technology development is on track, actually is better than what we thought. We have very good progress recently, and our risk production will be in 2024 and volume production in 2025. "
Unless they have pulled it in, IIRC they have been talking about H2 2025 for volume production. N3 was H2 2022 for volume production, which ended up being December, with customer shipments / revenue not happening until 2023.
So saying "2026" is probably not unreasonable. It is certainly 2026 from Apple's perspective unless it gets pulled into H1 2025.
Risk production is not the same as "ramping up". That's just completely wrong. Volume production in 2025 means an actual product that contains chips under this node would appear in late 2025 or 2026.
It is indeed an incredibly long run-up to release of a product. I've been in R&D for 25 years now. The processes/devices I work with daily won't be a product on the shelf that you can buy for 6-8 years. We tend to think of tech as constantly evolving and changing, and that is true, but there's also so long-standing plans that have to come to fruition for that to happen.
Current progress is becoming unsustainable and progress will slow down quite significantly. TSMC has been pushing out new nodes every 2-3yrs. But price has been going up as well. Companies are passing this cost to consumers. We have reached a point where consumers are no longer upgrading regularly and rejecting latest hardware due to cost (Ryzen is a good example, 5000 series are still selling very well due to much lower platform cost).
Another point is that not everything requires the latest process.... There arent that many chips that requires even 7nm, let alone 5nm or 3nm.
"There arent that many chips that requires even 7nm, let alone 5nm or 3nm. "
from your mouth to God's ear.
"Bear in mind the majority of chips in the things you use are not state-of-the-art 10, 7, or 5nm parts – they're using larger, older nodes. Fabs turning out 14 and 5nm chips aren't helping products that just need 130nm microcontrollers, 45nm embedded processors, and so on.
It's fortunate, then, that TI is investing in its capacity to churn out analog and embedded electronics using 45nm to 130nm process nodes."
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
30 Comments
Back to Article
Threska - Tuesday, January 17, 2023 - link
Well High-NA: EUV for their next generation.https://www.youtube.com/watch?v=en7hhFJBrAI
lightningz71 - Tuesday, January 17, 2023 - link
I suspect that TSMC referring to HPC for N3(B) is referring to the M2/M3 product families.name99 - Tuesday, January 17, 2023 - link
M2 (and Pro and Max) are on N5.A16 is on N4.
The future is very unclear. A16 certainly LOOKS like it was not on the roadmap, it was just an A15 boosted as much as possible by N4 rather than N5, and the real next design (the one that gives us IPC improvements, not just process improvements) is targeting N3B.
So will M3 be based on that next design or will it be based on N4 and the A16? Who knows???
I think we are still at the stage where both covid and the Mac Apple Silicon transition were one-time events that generated all sorts of one-time delays that it's not yet possible to talk of any sort of Apple "pattern" in how long it takes to move a design from phone to mac to high end.
tipoo - Tuesday, January 17, 2023 - link
Seems likely N3B for the M3 and that hardware accelerated RT GPU that was canned for the iPhone 14. Another year's delay to the hardware RT featuring next gen GPU architecture would be a sore letdown, as would M3 being on 4nm.techconc - Wednesday, January 18, 2023 - link
I agree on that RT on next gen GPU is really needed and would be a letdown on M3 if it wasn't there. However, I don't know that the Mac lineup really needs to be on the very latest node. N4 would still be a process improvement and fine for the next generation. Phones need the bleeding edge efficiency. The A17 has to be on N3.tipoo - Wednesday, January 18, 2023 - link
Looks like 3nm is happening on M3, but that still leaves the question on N3 or N3Ehttps://www.macrumors.com/2023/01/17/macbook-pro-2...
techconc - Wednesday, January 18, 2023 - link
Yeah, I've seen the rumors and obviously would prefer them to go with either 3nm process for M3. I'm just saying that I don't think it's actually necessary. The assumption would be that N4 would be more cost effective given that it's a more mature process with higher yield rates. They'd still gain efficiency and performance improvements. Obviously, moving to N3 (or N3E) would bring even greater improvements.Doug_S - Thursday, January 19, 2023 - link
If they are shipped late 2023 or early 2024 that's plenty of time for good yields on whatever flavor of N3. TSMC showed a graph late last summer comparing yields of N3E in various stages prior to the start of risk production and they were quite a bit higher than N5 yields during comparable - around 80% for SRAM, mobile and logic test chips three months prior to the start of risk production.If they were already hitting 80% in August 2022 they should be well into the 90s by the year+ timeframe Apple wafers would enter mass production.
Kangal - Friday, January 20, 2023 - link
Apple is (unfortunately) heavily reliant on TSMC.The first thing they will push out is the iPhone 15. The TSMC-3nm has been in the works for awhile, and has been delayed by 1-2 years due to the pandemic and chip supply issues. The upshot is that when TSMC does ship it out, it will come out in volume, as all of the work has been done and there are little to no yield issues to complain about.
So the iPhone 14 (which was delayed) was eventually cancelled, and we got the iPhone 14 instead. No that is not a typo you read that right. The "iPhone 14" we got is basically an iPhone 13. The iPhone 15 in Q3 2023 will come with the A17-Bionic chipset using TSMC-3nm lithography and a new architecture that Apple will base on ARMv9. So yeah, big upgrades incoming.
I know the yields are so good that Apple might also launch the Macbook Air with the M3 chipset for (Christmas season) Q4 2023. There's also the iPad Pro, but that was refreshed sooner, most likely the third device to come with M3 around Q2 2024. The Macbook Pro has a healthy profit margin for Apple, and were delayed by a decent while, still they won't come with M3 Pro/Max till much later around Q4 2024. The Mac Mini only just got updated and it's not the best money maker, so that will be on the tail end around Q1 2025. With the iMac not having a great track-record, they will get an M2 refresh soon (Q2 2023), then get an M3 launch very late as well around Q3 2025. The Mac Studio likely will not get an M2 Ultra chipset, and will be the last to get upgraded to the M3 Ultra around Q3 2025.
iphonebestgamephone - Thursday, January 26, 2023 - link
Provide sourceswhatthe123 - Wednesday, January 18, 2023 - link
The delays are due to TSMC struggling with their original 3nm target. That's why there is now a new, less dense 3nm node that everyone will be using instead and why they don't expect much revenue out of the current n3 node. If the original N3 was ready for high volume Apple would've probably jumped on it to stamp out the competition like they did with the first run of 5nm.webdoctors - Tuesday, January 17, 2023 - link
Amazing stuff. being on 3nm in 2023. I remember working on chip designs on 130nm and 90nm back in 2003 thinking how incredible that was. We've made crazy strides in 20 years. Wonder what process we'll be on in 10 or 20 years.haukionkannel - Wednesday, January 18, 2023 - link
Well 130nm and 90 nm were allmort 130nm and 90nm while 3nm is something like 30 to 40nm in reality so the advancement is not as high as sales man claims…FreckledTrout - Friday, January 20, 2023 - link
Things get even wonkier with TSMC's 2nm as they switch to GAA so there is even more of the transistor in 3 dimensions. The nm terms are more marketing than anything else these days which Intel acknowledged with their new names.Threska - Friday, January 20, 2023 - link
Problems problems.https://www.tomshardware.com/news/materials-breakt...
mattbe - Tuesday, January 24, 2023 - link
If this is what you are arguing, a 130nm has a gate length of around 130nm while a 3nm chip has a gate length of about 16-18nm. The 30-40nm is pulled out of your ass. If you are talking about the pitch, then at the 130nm node, the pitch wasn't 130nm either...sharath.naik - Wednesday, January 18, 2023 - link
Its close to reaching its limits. at 3nm my guess is its just 10-20 atoms. could they go to 1nm? it will be remarkable if they do. but you see how close they are to the limits.escksu - Wednesday, January 18, 2023 - link
Oh no, it's not 3nm. "3nm" or "5nm" today is no longer a true indication of transistor size. Actual size is around 30nm. It's more to do with marketing instead.Back then when we had only planar transistor, yes. But today we have finfet, gaafet etc so the size is just more of an estimation of planar equivalent with lots of assumptions made.
There is no standard of how you define 3nm or 5nm. It entirely up to the company.
mattbe - Tuesday, January 24, 2023 - link
"There is no standard for defining 3nm or 5nm. It is entirely up to the company."There is some semblance ot standard as established by IMEC. It would be bad for the company's image if it's too far off IMEC's expected specs for that node...
ABR - Thursday, January 19, 2023 - link
They don't even try to say it's 3nm any more. It's "N3". Of course they want you to make the association, but that's on you then.lefty2 - Tuesday, January 17, 2023 - link
N2 starts ramping up in 2024, not 2026.This is a quote from the earning call:
"our N2 technology development is on track, actually is better than what we thought. We have very good progress recently, and our risk production will be in 2024 and volume production in 2025. "
Doug_S - Wednesday, January 18, 2023 - link
Unless they have pulled it in, IIRC they have been talking about H2 2025 for volume production. N3 was H2 2022 for volume production, which ended up being December, with customer shipments / revenue not happening until 2023.So saying "2026" is probably not unreasonable. It is certainly 2026 from Apple's perspective unless it gets pulled into H1 2025.
lefty2 - Wednesday, January 18, 2023 - link
dude, he clearly says 2025. It didn't get pulled in, it was always was 2025.https://seekingalpha.com/article/4569491-taiwan-se...
mattbe - Tuesday, January 24, 2023 - link
Risk production is not the same as "ramping up". That's just completely wrong. Volume production in 2025 means an actual product that contains chips under this node would appear in late 2025 or 2026.Search Engine Optimization - Wednesday, January 18, 2023 - link
Runways on this tech are so looooong. This does help us plan for our own cap ex for devices!JKflipflop98 - Thursday, January 19, 2023 - link
It is indeed an incredibly long run-up to release of a product. I've been in R&D for 25 years now. The processes/devices I work with daily won't be a product on the shelf that you can buy for 6-8 years. We tend to think of tech as constantly evolving and changing, and that is true, but there's also so long-standing plans that have to come to fruition for that to happen.Threska - Friday, January 20, 2023 - link
Who knew a crashed UFO could yield so much. :-)FunBunny2 - Friday, January 20, 2023 - link
Who knew a crashed UFO could yield so much. :-)"I want to believe"
-- Fox Mulder/1993
escksu - Monday, January 23, 2023 - link
Current progress is becoming unsustainable and progress will slow down quite significantly. TSMC has been pushing out new nodes every 2-3yrs. But price has been going up as well. Companies are passing this cost to consumers. We have reached a point where consumers are no longer upgrading regularly and rejecting latest hardware due to cost (Ryzen is a good example, 5000 series are still selling very well due to much lower platform cost).Another point is that not everything requires the latest process.... There arent that many chips that requires even 7nm, let alone 5nm or 3nm.
FunBunny2 - Monday, January 23, 2023 - link
"There arent that many chips that requires even 7nm, let alone 5nm or 3nm. "from your mouth to God's ear.
"Bear in mind the majority of chips in the things you use are not state-of-the-art 10, 7, or 5nm parts – they're using larger, older nodes. Fabs turning out 14 and 5nm chips aren't helping products that just need 130nm microcontrollers, 45nm embedded processors, and so on.
It's fortunate, then, that TI is investing in its capacity to churn out analog and embedded electronics using 45nm to 130nm process nodes."
https://www.theregister.com/2022/01/26/texas_instr...
aren't those the nodes that we were supposed to get for all that high endurance 3D NAND?? just kidding.