Releasing it with a M3 would not sound right for such a high end device. iPad sales are going down, they need the magic of AI or something, putting in again a M3 will not work. And yes it will be N3E, and M3 chips still will be made but so what? It really makes perfect sense for a high end tiered device to get a M4 and not a low tier MacBook Air not Pro (cause that would eat others).
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.
M4+ Chip Generation - Speculation Megathread [MERGED]
- Thread starter GoetzPhil
- Start date
- Sort by reaction score
I would be very sad if the M3 can't do AI at least to some extend, as I just got one.
I think Apple would rather use the long wait since the last models to get the M3 iPad out, yell about AI, and then when the M4 come in fall, release another iPad Pro update early '25, and yell "AI" even louder and how it does a bazillion more LLM calculations compared to the M3. Coupled with the next iOS/iPadOS release, people will be happy, and Cook richer.
I think Apple would rather use the long wait since the last models to get the M3 iPad out, yell about AI, and then when the M4 come in fall, release another iPad Pro update early '25, and yell "AI" even louder and how it does a bazillion more LLM calculations compared to the M3. Coupled with the next iOS/iPadOS release, people will be happy, and Cook richer.
I don't think you need to worry about that. I'll guess, at the very least, the on-device Ajax LLM is going to run well on the A14/M1 and above. Basically anything with a 16-core Apple Neural Engine NPU. It will run even better on A15/M2, A16, and A17/M3, but those advances are incremental and architectural compared the jump from 8-core to 16-core NPU.
A18/M4 might go to a 32-core NPU? Seems possible, maybe even likely. Instantly doubles NPU performance, just like the M1/M2 Ultra does. But there are tradeoffs. You're probably taking something away when you do that. So I don't know. I don't think it is a sure thing. Might have to wait for N2 and Nanosheet transistors?
Last edited:
How can it be "again" when the current iPad Pros have the M2 SoC...?
The unifying feature of many of the posts that insist that M4 is coming now is they all think it will be on N3E. I imagine Gurman also believes that.
Ever since Intel revealed that Lunar Lake is being produced on TSMC N3B, pretty amazing news really, the ignorance surrounding N3B ("poor yields" "too expensive") has been further exposed. Which anyone paying attention to Apple's behavior already knew, but the rumor just refused to die. It's true that N3E is designed to get better yields than N3B, but that is by design. People seem to believe that means N3B is a failure because it hasn't met expectations for yields or it has cost more than expected, neither of which is correct.
I can't say I can guess what Apple is or is not doing with the M3 Ultra, we will see, but I can say that whatever it is or is not, it's not because N3B is a failure. It's true that it is a dead end, but again, that is by design. Indeed, N3B can be seen as the peak of FinFET (FINFLEX) technology, with N3E as a step down from that summit, back onto the path toward the future, toward GAAFET (nanosheet transistors) and backside power rail. That's why the two most-advanced, most-efficient families of SoCs are using it, from two of the largest companies in the world, and by the end of this year every major OEM will feature high-end laptops with N3B silicon in them.
I feel the M4-series of SoCs will be using the N3E process, and what better way to introduce this than a monolithic N3E M4 Ultra SoC...!?!
But the iPad Pros will most likely not skip the M3 SoC, so N3B M3 SoCs for the iPad Pro seems most likely, time will tell...!
Interesting fact: M4 starts with only 3 performance cores; if you want 4 of them you have to buy the models with 1 or 2 TB of storage.
N3B has better performance than N3E, but if Apple has tweaked the CPU, GPU and NE cores even more, that would improve M4's performance even on the "slower" process.
Not unexpected since this is the first N3E SoC for Apple so there will be yield issues leading to chip-binning and cheaper binned SoCs would be used for the cheaper configurations.
But N3E supposedly would be slower right? And N3P isn't finished I would assume?
Still, one would almost think it’s N3P but not seen anything it was ready.
They used Affinity Pro as a CPU test. It’s unlikely that GB6 will see similar improvement. I would expect M4 to be only tad faster than M3 on most tests and slower in single-core.
If that is the case I cannot see them waiting until the end of the year to upgrade the mac line with this. Rather within the next few months, to bring the expensive N3B to an end quickly.
iPad Pro
iPad Pro features the M4 chip, 11-inch or 13-inch Ultra Retina XDR display, Wi-Fi 6E, 5G, and support for Apple Pencil Pro.
www.apple.com
The M4, should be faster no matter what in that fanless design,if indeed Apple is doing proper thermals inside...so even in current thermal design with M2 should be faster than the former M2
At peak and after on sustaining performance, thermals helps, hope this will translate with the next iphones pro as well
Hoping Apple is not lying
At peak and after on sustaining performance, thermals helps, hope this will translate with the next iphones pro as well
Hoping Apple is not lying
Why would you assume that the Mac line will use M4? Next laptops could jump straight to M5. I could see M4 in the desktops though. It is possible that the N3E can be clocked higher than the M3.
N3P is an optical shrink of N3E that will offer better performance or lower power consumption (choose one) and higher transistor densities, but it is not yet in general production (it is expected in 2H 2024).
A17 and M3 were likely meant to be a short-term interim step to both validate the 3nm process and crow about being "first to 3nm" with A18 and M4 being the "mainstream" 3nm SoCs for Apple products.
As such, I do not expect M5 until next year and possibly even 2026 (after A19 ships in the iPhone in September 2025).
There must be an M4 Mac coming at WWDC. The iPhone 15 Pro (A17 Pro) and now iPad Pro (M4) have large NPU bumps compared to the previous chip generations (~35-38 TOPS vs ~15-18 TOPS in the A16 & M3). If Apple are going to push AI at WWDC, they will need some kind of Mac with the latest NPU otherwise MacOS will struggle in AI tasks compared to iOS/iPadOS. My money is on a Mac Studio with M4 Ultra but who knows? Perhaps we could see a wider Mac line-up with M4-related chips. Given MacOS won't roll out until October 2025, I think it's more likely that an M4 Ultra Mac Studio is shown at WWDC to demo the AI capabilities of the next MacOS, then just as MacOS launches, we get the M4 range of MBPs to also take advantage of the latest NPU.
Well I would think if they at least release the Mac Mini with M4, Studio with M4 Pro/Max would boost the sales and the MacBook Pro later this year. Hence I got the M2 Max Studio, wasn't gonna buy the M3 Max but the M4 Max would be tempting.
N3E has better performance and efficiency than N3 (not by much on either, but it is better) and supposedly much cheaper. It is simply less dense than N3.
