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.
Apple Likely to Debut Next-Gen 3nm Chip Technology in 2022 iPad Pro
- Thread starter MacRumors
- Start date
- Sort by reaction score
N3 is looking like a nice jump in transistor density and scaling from N5 and N5P. They even managed a decent 20% scaling improvement to SRAM which is notoriously difficult to shrink and Apple uses a lot of to help keep the cores fed.
I wouldn't be surprised if I saw yet another doubling of my TSMC stock, they just announced they are building Intel chips now. So that's AMD, Intel, Nvidia, Qualcomm, Apple all fighting over TSMC's silicon, with Apple I'm sure having secured the lions share.
I wouldn't be surprised if I saw yet another doubling of my TSMC stock, they just announced they are building Intel chips now. So that's AMD, Intel, Nvidia, Qualcomm, Apple all fighting over TSMC's silicon, with Apple I'm sure having secured the lions share.
Don’t you think at this point the OS is holding back the iPads capability and not the silicon inside of it?
I think a lot of us want the Mac apps for the iPad. Logic, final cut etc.
Maybe a better file system. I think that the iPad is going in the right direction though!
It doesn't matter what Apple puts in a Mac, until it has the IO bandwidth, reliability, and uptime of an IBM Z-series mainframe, it's just a toy. 🤪
Apple and TSMC are so far ahead of other chip designers and manufacturers it’s kind of scary. Shout out to Johny Srouji and Apple’s entire chip team, the incredibly talented group of people over at TSMC, and everyone else involved. I 100% believe that ARM will and should be the future of computing.
And to the very petty Intel, how does it feel to see this “lifestyle” company leading the way in your industry? You have 52 years of chipmaking experience vs Apple’s 11 and Apple is putting you to shame. To shame.
And to the very petty Intel, how does it feel to see this “lifestyle” company leading the way in your industry? You have 52 years of chipmaking experience vs Apple’s 11 and Apple is putting you to shame. To shame.
Last edited:
The (2022) iPad Air strikes me as the more likely candidate for this 3nm "A16" than iPad Pro.
One expects the 2022 iPad Pro would be updated with "M2" based on the (presumed 4nm) A15 SoC.
One expects the 2022 iPad Pro would be updated with "M2" based on the (presumed 4nm) A15 SoC.
Darth Tulhu
macrumors 68020
Darth Tulhu
macrumors 68020
I can accept all of your points, but Apple simply doesn't agree with you on having a cursor-based OS on a hardware form-factor designed for touch as a PRIMARY means of interaction.
Cursor operation on the iPad is an addition, and it's meant to be secondary.
It's that simple.
So how is making the iPad faster going to improve my youtube and netflix alike experience? The iPad doesn't seem like it could do much of anything else.
Wasting their chip lead? I see that Intel is going to be on TSMC 3nm as well. It'll be interesting to what kind of perf lead Apple maintains. No matter, it's good for users.
I'm sure those chips will be in future macs as well.
Darth Tulhu
macrumors 68020
It won't.
Of course it can. If you really want to know (and not just rage against the machine) then here you go:
iPad Pro - Why iPad
iPad is so versatile, it’s more than up to any task. Get things done in both fun and powerful ways with iPad.
www.apple.com
You can already have an iPad with macOS on it as a DMG file, and thats about as useful as macOS would be on an iPad anyway.
We'll soon see how good Apple's designs really are when Intel has products using the same TSMC process technology.
I doubt it, given the emphasis that their new CEO is putting on manufacturing and the billions of dollars they are investing. The plan seems to be to use TSMC to tide them over until they get their own process technology back on track (similar to Apple using Qualcomm modems until they can get their own modems based on the former Intel designs up to par). Of course, if the plans don't pan out in a few years it might still happen.
If you look at the numerous "laugh" reactions to ooans' post, I think you'll see that he was making a joke.
You are misunderstanding how this might work.
Look at the A10X, that's the relevant example. The A10 shipped on 16nm, the A10X shipped (a few months later) on 10nm. You could view this as Apple wanting to get that SoC out ASAP (and it would be cheaper and lower energy) on 10nm, or you could view it as a preliminary run of 10nm, testing that the process behaves as expected on a real product, with no catastrophe if the sample A10X's don't work well and have to be delayed a few months.
In other words IF 3nm looks like it might be ready slightly sooner than expected, once again the schedule could look like
- 2021 September A15 iPhones on 5nm+ (began bulk production ~May 2021)
- 2022 September A16 iPhones on 4nm (began bulk production ~May 2022)
- 2022 ~Dec M3 iPads/Macs on 3nm (began bulk production ~Sept 2022)
Apple doesn't want to delay iPhones to hit 3nm, neither does it want to take the risk that 3nm might be delayed a few months. But it makes perfect sense to schedule the M3 generation of SoCs on 3nm. They don't have an expected date, so they can handle a delay if one arises AND they ship in much smaller volumes, so you don't need the full 4..5 months of production to build up September inventory that you need with iPhones.
10% processing boost is a game changer? I don’t even think the 25% boost in battery will make much of a difference either - the current model lasts way longer than I need to go between charges.