14” MBP and 16” MBP disappointing launch

[UBS28]

Seeing that people already have to wait 2 months for the M1 iPad Pro for shipping and this is only the pre-order, the 14” and 16“ MBP could be even worse since it will be based off a new chip (instead of the M1 that is already in mass production).

So if the M2X 16” MBP gets launched in Q3 2021, it might only become available for purchase in Q4 2021 and available to the masses in Q1 2022 possibly.

 

[Kung gu]

The M2X for the 14" and 16" won't come this year, the M2 needs to happen before M2X comes out.

Ming Chi Kuo a reliable leaker said the 14" and 16" is coming out in Q3 likely with M1X not M2X because M2 has been not released yet.

 

[deeddawg]

Apple said up to two years for the transition.

JMHO, expecting anything faster than that is likely to lead to unnecessary disappointment.

 

[UBS28]

The M2X for the 14" and 16" won't come this year, the M2 needs to happen before M2X comes out.

Ming Chi Kuo a reliable leaker said the 14" and 16" is coming out in Q3 likely with M1X not M2X because M2 has been not released yet.

The A15 will launch soon with iPhone 13 which is M2 minus a few cores. So the M2X could be possible without the M2.

It doesn’t make sense to go for the M1X if the foundation of the M2X is already there with the A15.

 

[Fomalhaut]

The M2X for the 14" and 16" won't come this year, the M2 needs to happen before M2X comes out.

Ming Chi Kuo a reliable leaker said the 14" and 16" is coming out in Q3 likely with M1X not M2X because M2 has been not released yet.

We don't know whether the next chip will be "M1X" (M1 with more cores) or, as rumored, a second generation "M2" based on the same core-architecture as the A15, which will probably start production by June/July (in preparation for an iPhone 13 release in Sep/Oct).

It's certainly possible that the next generation Mac chips could arrive before the iPhone chips, because they don't have to produce as many, and it allows the fabrication to ramp up to larger volumes.

I would expect the ramp-up for mass production in a fabrication plant would take at least 2-3 months, so it would make sense to start to hear rumors in April/May for an M2 chip that might be announced in late June at WWDC and released in July/August. So we could see an M2 MacBook Pro towards the end of July.

Of course, things may be moving more slowly than this, and what we will see at WWDC is a larger M1 variant. We'll see!

 

[Fomalhaut]

The A15 will launch soon with iPhone 13 which is M2 minus a few cores. So the M2X could be possible without the M2.

It doesn’t make sense to go for the M1X if the foundation of the M2X is already there with the A15.

I agree. If the A15 will be released in Sep/Oct, it means it is probably nearly ready for the fabrication facility to start gearing up for it at scale. The M2 is a lower volume SoC based on the same A15 core architecture and could be part of the same rollout.

It would be a bit disappointing for a Mac released in July/Aug to still be based on the M1, and not to include the advances of the A15.

Unless Apple are staggering the release of Mac SoCs to trail the same generation of iPhone chips, I would expect an M2 and A15 to be within 2-3 months of each other. It could be the M2 that comes first, or vice versa.

 

[UltimateSyn]

The M2X for the 14" and 16" won't come this year, the M2 needs to happen before M2X comes out.

Ming Chi Kuo a reliable leaker said the 14" and 16" is coming out in Q3 likely with M1X not M2X because M2 has been not released yet.

N5P chips for the MacBook Pros are in production right now, according to Nikkei. The fact that they’re N5P instead of N5 implies a generational change to M2. There is nothing in the SoC Rulebooks that states that Apple could not start with their higher-core-count chips of this generation (let’s say a 12-core M2 aka M2X, whatever you want to call it) and then produce the lower-core-count variants (8-core M2, 6-core A15, etc.) afterwards. Just because they’ve gone from less complex -> more complex in the past doesn’t mean they need to continue to follow that pattern moving forward.

 

[quarkysg]

N5P chips for the MacBook Pros are in production right now, according to Nikkei. The fact that they’re N5P instead of N5 implies a generational change to M2. There is nothing in the SoC Rulebooks that states that Apple could not start with their higher-core-count chips of this generation (let’s say a 12-core M2 aka M2X, whatever you want to call it) and then produce the lower-core-count variants (8-core M2, 6-core A15, etc.) afterwards. Just because they’ve gone from less complex -> more complex in the past doesn’t mean they need to continue to follow that pattern moving forward.

Apple could also get TSCM to use the N5P process to fabricate more M1 based cores SoCs with higher core clocks or better thermals, or both. It could also be based on the next core (i.e. A15) architecture. I guess we should not have to wait long to find out as I think new higher end Macs will be introduced during WWDC '21 next month.

 

[Kung gu]

All I am saying is don't expect to see M2X as the M2 has not launched

 

[UltimateSyn]

Apple could also get TSCM to use the N5P process to fabricate more M1 based cores SoCs with higher core clocks or better thermals, or both. It could also be based on the next core (i.e. A15) architecture. I guess we should not have to wait long to find out as I think new higher end Macs will be introduced during WWDC '21 next month.

Can they just use a new process with the same M1 architecture - plug and play like that? I feel like the architecture and process change would occur at the same time and this would be marketed as a new generation… But I do not know much about chip design…

 

[UltimateSyn]

All I am saying is don't expect to see M2X as the M2 has not launched

A knowledgeable user on this forum with connections in the industry stated that he doesn’t believe there will be any X marketing in the M-Series lineup, but rather they will distinguish by giving specific core counts (e.g. 8-Core CPU & 8-Core GPU M1, 12-Core CPU & 16-Core GPU M1, etc.). This makes a ton of sense for a few reasons:

• There are no existing products with the X or Z chip-variants right now. It seems they are doing away with that nomenclature. X and Z has meant ’the same as the baseline chip with more cores’ for a while now but that’s when A-Series chips didn’t proudly have their core-counts boasted. Now in all the tech specs you can clearly see how many CPU and GPU cores each product has.
• There are likely to be many variations of the M-Series chips. There are already two with the existing products, which keep the same name and are distinguished by core counts (those binned 7-Core GPU variants in the MBA and iMac). Are they going to slap a new letter on there each time a new variant comes up? There will probably be half a dozen or more CPU/GPU combinations across the MBP / iMac Pro / Mac Pro / Mac Pro Mini and I’m sure nobody wants to see the M2G.

My point there is that they wouldn’t be releasing the M2X before the M2, which I agree wouldn’t make sense from a marketing standpoint. They would instead be releasing the 12-core M2 before the 8-core M2, and it would likely be beneficial for them to put that breakthrough performance in their highest-end machines first.

 

[quarkysg]

Can they just use a new process with the same M1 architecture - plug and play like that? I feel like the architecture and process change would occur at the same time and this would be marketed as a new generation… But I do not know much about chip design…

I remember reading somewhere that TSMC claims the N5P is compatible with N5, so probably just a slight tweak to the design should be sufficient.

 

[UltimateSyn]

I remember reading somewhere that TSMC claims the N5P is compatible with N5, so probably just a slight tweak to the design should be sufficient.

That’s interesting! I didn’t know that I just think it would be weird if chips within the same generation were on different architectures because it would set up a vicious cycle -> ‘M2X’ would need to get 4nm while M2 would still be on N5P, etc. Makes more logical sense to me for the cores to stay the same within a generation and just have different core count combos.

 

[cardfan]

Seeing that people already have to wait 2 months for the M1 iPad Pro for shipping and this is only the pre-order, the 14” and 16“ MBP could be even worse since it will be based off a new chip (instead of the M1 that is already in mass production).

So if the M2X 16” MBP gets launched in Q3 2021, it might only become available for purchase in Q4 2021 and available to the masses in Q1 2022 possibly.

Many people got 1400 checks. Plenty incentive to start preorders early.

 

[Kung gu]

A knowledgeable user on this forum with connections in the industry stated that he doesn’t believe there will be any X marketing in the M-Series lineup, but rather they will distinguish by giving specific core counts (e.g. 8-Core CPU & 8-Core GPU M1, 12-Core CPU & 16-Core GPU M1, etc.). This makes a ton of sense for a few reasons:

• There are no existing products with the X or Z chip-variants right now. It seems they are doing away with that nomenclature. X and Z has meant ’the same as the baseline chip with more cores’ for a while now but that’s when A-Series chips didn’t proudly have their core-counts boasted. Now in all the tech specs you can clearly see how many CPU and GPU cores each product has.
• There are likely to be many variations of the M-Series chips. There are already two with the existing products, which keep the same name and are distinguished by core counts (those binned 7-Core GPU variants in the MBA and iMac). Are they going to slap a new letter on there each time a new variant comes up? There will probably be half a dozen or more CPU/GPU combinations across the MBP / iMac Pro / Mac Pro / Mac Pro Mini and I’m sure nobody wants to see the M2G.

My point there is that they wouldn’t be releasing the M2X before the M2, which I agree wouldn’t make sense from a marketing standpoint. They would instead be releasing the 12-core M2 before the 8-core M2, and it would likely be beneficial for them to put that breakthrough performance in their highest-end machines first.

It's simple really anything before A15 launch will have M1 based arch, if the next Macs launch after A15 then it have arch based on A15. Since Kuo is expecting the MacBooks to come in Q3, so thats either gonna be in July or August(not September cause its for iPhones) so I am expecting a M1 based MacBook 14" and 16"

Now if the Macs launch in November then expect A15 based SoC on the MacBook pros

 

[Metrosey]

I just want a 16 inch MacBook, as long as it has MiniLED and 32GB RAM, gimme gimme gimme. I’d eat a car if it came with 120hz too.

 

[leman]

Seeing that people already have to wait 2 months for the M1 iPad Pro for shipping and this is only the pre-order, the 14” and 16“ MBP could be even worse since it will be based off a new chip (instead of the M1 that is already in mass production).

It seems like the main reason for iPad delays is the display. It may or may not be a limiting factor for the new Mac laptops.

So if the M2X 16” MBP gets launched in Q3 2021, it might only become available for purchase in Q4 2021 and available to the masses in Q1 2022 possibly.

Why are you guys so set on all the M2X etc. nomenclature? We didn't get M1X yet, did we? Who says that Apple is going to reuse the naming scheme they have used with iPads in the past for the Mac?

It's simple really anything before A15 launch will have M1 based arch, if the next Macs launch after A15 then it have arch based on A15. Since Kuo is expecting the MacBooks to come in Q3, so thats either gonna be in July or August(not September cause its for iPhones) so I am expecting a M1 based MacBook 14" and 16"

While I really admire your confidence is saying these things, I don't see how any of this follows. There is no law that states that iPhone and Mac chips have to be manufactured in a certain order.

 

[JPack]

Even if we ignore the highly credible Nikkei Asia report indicating M2 silicon is already in production, the fact that iPad Air launched with A14 before iPhone 12 already shows there’s no specific order that Apple Silicon needs to follow.

 

[Significant1]

Even if we ignore the highly credible Nikkei Asia report indicating M2 silicon is already in production, the fact that iPad Air launched with A14 before iPhone 12 already shows there’s no specific order that Apple Silicon needs to follow.

It was announced before iPhone 12, but released at the same time.

 

[scottrichardson]

Just to be clear on Apple’s CPU nomenclature...

A”n” chip = base design (used in iPhones and other devices)

A”nX” chip = same core / pipeline design as base chip, just with more cores on the GPU/CPU and slightly higher clock. (Used in some iPads and other devices where more voltage isn’t as problematic)

In other words an A15 and an A15X are the same chip but the X has faster clock, and more silicon for more cores.

It’s entirely possible that Apple could release an M1X that is the same chip just with added silicon for additional CPU and GPU cores with a higher clock.

It’s also just as possible that Apple will focus on a newer, more optimised core design with the A15 and use it as the basis for the next Mac chip, the M2.

With that all in mind, it’s entirely safe to assume that the M1 is actually an A14X with a different name.

 

[leman]

In other words an A15 and an A15X are the same chip but the X has faster clock, and more silicon for more cores.

It's not the same chip, it's the same microarchitecture.

It’s also just as possible that Apple will focus on a newer, more optimised core design with the A15 and use it as the basis for the next Mac chip, the M2.

What is equally possible is that both the A15 and the upcoming Mac prosumer chip (whatever it will be called) will be based on the same microarchitecture, but will diverge more strongly from each other than M1 did from A14. Basically, instead of basing "M2" on "A15", Apple could base both "M2" and "A15" on a "generation X"

With that all in mind, it’s entirely safe to assume that the M1 is actually an A14X with a different name.

Kind of, but then again not really. M1 contains a bunch of logic that makes it suitable for the desktop (like the IO controllers etc.), and there are also some GPU differences (A14 GPU needs two cycles to carry out a FP32 operation where M1 needs only one, but it can also be that this functionality has been deliberately disabled in A14 to save power).

 

[w4rm]

N5P chips for the MacBook Pros are in production right now, according to Nikkei. The fact that they’re N5P instead of N5 implies a generational change to M2. There is nothing in the SoC Rulebooks that states that Apple could not start with their higher-core-count chips of this generation (let’s say a 12-core M2 aka M2X, whatever you want to call it) and then produce the lower-core-count variants (8-core M2, 6-core A15, etc.) afterwards. Just because they’ve gone from less complex -> more complex in the past doesn’t mean they need to continue to follow that pattern moving forward.

This also makes sense from a marketing perspective. By associating people of high performance with the M2X first, it makes M2 easier to sell.

Rather than starting with a M2 and having that associated with a lower end product then come to a higher end product as a M2X.

This new pattern is what Apple did with iPad, people generally see Mac’s as higher performance so by bringing that processor to a iPad it benefits from the M1 branding.

No one but tech nerds like us know about A14 being the same generation of M1. So of course it doesn’t matter if Apple releases the higher end version first.

But that might just be me being hopefully because I really want a MacBook Pro 14 inch for WWDC, with a next gen processor.

 

[Fomalhaut]

All I am saying is don't expect to see M2X as the M2 has not launched

not yet, no. But it might launch in July. It’s also possible, as mentioned above, that it starts with a larger M2 (or “M2X”) and then progresses to a smaller one in greater volumes.

 

[scottrichardson]

It's not the same chip, it's the same microarchitecture.

What is equally possible is that both the A15 and the upcoming Mac prosumer chip (whatever it will be called) will be based on the same microarchitecture, but will diverge more strongly from each other than M1 did from A14. Basically, instead of basing "M2" on "A15", Apple could base both "M2" and "A15" on a "generation X"

Kind of, but then again not really. M1 contains a bunch of logic that makes it suitable for the desktop (like the IO controllers etc.), and there are also some GPU differences (A14 GPU needs two cycles to carry out a FP32 operation where M1 needs only one, but it can also be that this functionality has been deliberately disabled in A14 to save power).

Many thanks for the additional info and clarification! I did realise after I posted that there ARE a few elements of the M1 that are specific to the M1, but felt my post mostly covered the X vs non-X variants well enough. Thank you for the clarification on the term 'microarchitecture', it's what I meant, but couldn't think of the correct term!

 

[cmaier]

I remember reading somewhere that TSMC claims the N5P is compatible with N5, so probably just a slight tweak to the design should be sufficient.

It’s always more work than you think. For example, even if the design rules are the same, if the transistors switch faster at the same voltage (which, of course, is the point of the improvement), that can cause what we call “mintime” or “hold time” violations. Essentially, race conditions can occur which case the chip to malfunction. So even a “compatible” process can cause quite a bit of modeling and physical design work. Of course, a lot of that can be mitigated by careful planning.