I never said that M2 cannot be a better version of the M1. All I'm saying is that an M2, with 20-30% better performance compared to M1 would not be enough to satisfy the pro market.
Apple needs a different class processor to address the pro market.
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's 'M2' Next-Gen Mac Chip Enters Mass Production, Expected to Debut in Redesigned MacBooks Later This Year
- Thread starter MacRumors
- Start date
- Sort by reaction score
Apple needs a different class processor to address the pro market.
Apart from the black edge to the monitor, the previous 2008 24in. design in my opinion is better than the new 24in.
but incorporating thinner new 24in. iMac but based on the older design, it would not now need the black edges or white or black bezels at all as the screen is kept in by the wrap around casing.
but incorporating thinner new 24in. iMac but based on the older design, it would not now need the black edges or white or black bezels at all as the screen is kept in by the wrap around casing.
Attachments
This sounds about right, with possibly different names.
As has been already proposed, the M-number will most likely refer to the generation in the same way that A13, A14, A15 refer to the yearly generation of iPhone/iPad SoCs, each with incremental improvements.
An M2 in its "base configuration" will only be incrementally faster (maybe up to 20%) than an M1. This is what will go into next year's iPad Pro, MBA, low-end MBP14 and iMac.
There will be another 1 or probably 2 levels for each generation - with naming yet to be defined. Maybe it won't have a specific name but just be "12-core M2" , "16-core M2" etc.
The question I have in your 2021 lineup is what would differentiate the MBP14 and MBP16? Would they both have the same 12-core M1 with improved GPU? It's conceivable, but I would expect some processor differentiation between the 14" and 16" MBP, so there may be two variants of the next chip (whether it be a bigger M1 or new M2).
Personally I'm hoping for the new MBPs to be an M2 plus more CPU & GPU cores, to take advantage of the generational improvement as well as much better multi-core CPU and GPU performance.
I agree. A 4+4 core CPU/8 core GPU M2 will be great for next year's entry level machines, but is not what the high-end MBP and iMac market is expecting.
If there isn't at least a 12-core MBP with 12-core GPU at a minimum, I would be surprised and disappointed, and would not be buying.
I absolutely agree on "Real professionals would not stipulate any number of cores". It's irrelevant. I am using the number of cores, RAM and GPU as a realistic way to scale up the performance of the SOC in a way that leads to at least 2x performance, which in turn benefits professionals.
The definition of professionals is people who get paid for their work. If the computer is faster, many professional workflows will be faster. The professionals running those workflows will in turn be more productive and earn more.
And that's what professional machines are all about: allow professionals to be more productive and earn more (and also potentially make the work more enjoyable).
Apple has the chance of building a 12-core or even 16-core variant of the M1 that will unlock the kind of performance that professionals will benefit from at a 60W to 90W TDP, which perfectly suits a 16-inch MacBook Pro or a 30" iMac Pro. And sell these machines at a considerable premium, that professionals will gladly pay in exchange for such dramatic performance increase.
Well, the M1 is the lowest end of Apple's Mac oriented SoCs. There'll only be faster and more powerful Mx SoCs going forward.
I agree on everything.
To have the low-end and high-end Mac SOC be upgraded with the new architecture every year (like the iPhone SOC) genuinely sounds like a dream to me and I hope it will be like this.
It's definitely doable from a technical point of view and it would make computer enthusiasts like me VERY happy.
That said, what I realistically expect from Apple is to "optimize" as much as possible to keep R&D costs down or focus the energy of the Apple Silicon team on fewer projects at a time.
What gives me the impression that we won't get a low end and high end Mac SOC with the new architecture every year is that, while it was technically possible to make, Apple chose to skip the A11X or the A13X SOC for the iPad Pro.
I ended pulling the trigger but for a different reason in the end. I was initially planning to replace my Macbook but opt to wait for the M2. However, my family doesn't have a PC at this moment so I bought a Mac Mini for them. Who knows, if they like it, they might opt to sell the PC and keep the Mac Mini.
But cores and RAM DO affect performance. And that's why some professionals, whose work is dependent on those specs, DO care about them.
Even in Apple-only discussion, you see lots of people here posting that they are waiting for 32GB RAM and higher.
Your paragraph is one huge No-True Scotsman post.
After the latest leaks it seems that Apple is planning to remove the current 13-inch MacBook Pro (M1) from the product line and replace it with a lower-spec 14-inch MacBook Pro with 3 USB-C ports, only 2 of them being thunderbolt ports.
This, to me, suggests the following lineup of MacBook Pros for 2021:
MacBook Pro 14-inch
8-core M1 on the base model (2 thunderbolt ports)
12-core M1X with midrange GPU on the high end model
MacBook Pro 16-inch
12-core M1X with midrange GPU on the base model
12-core M1X with faster GPU option on the high end model
I wish the base model had the M1X and not the M1, but I don't think it's realistic given the leaks and the fact that Apple needs something in the middle between the MacBook Air and the M1X 14-inch MacBook Pro to maximize profits (they are gonna sell the low-end 14-inch M1 version to people that want pro features such as better screen, battery life etc., but are not interested in faster than M1 performance).
Literally me when someone says M1X in this forum.
With the introduction of the M1 in the iPad I believe it's safe to say the days of X chips are over.
With the introduction of the M1 in the iPad I believe it's safe to say the days of X chips are over.
Could be.
I think everybody here (me for sure) is using M1X name just as a placeholder to identify a different class of SOC dedicated to pro Macs, but still based on the A14/M1 core architecture.
Could also be named the P1. or X1. Or maybe the Z1? (lol that sounds a lot like xeon).
I like the name X1 so much better. The whole idea of an (?)X/Z chip really rubs me the wrong way, especially after the pathetic half baked A12X. One extra GPU core? Are you kidding me with that?
Baby steps. Let's aim for say, non-gimped I/O and display support, before we aim for 64-cores?
They've had plenty of time to perfect this. If they can't do it, let another company achieve it instead.
I think you are both right. People who rely on their machines to achieve a given task with some required performance level will look closely at whether new offerings can meet these requirements.
This *often* means comparing specifications and predicting likely performance. I do it all the time (professionally). Benchmarks are helpful, but real-world results from others are even better, provided they are reliable and well-documented.
The world of fixed physical hardware is a tough one, and reminds me of the "olden days" when you had to size physical computers that cost tens of thousands of dollars, often with little more than manufacturers specs and promises. These days a lot of infrastructure is virtual and elastic, so you measure and resize until it "fits" the workload. Having an idea of what performance you can expect from processor X with Y cores and Z memory at frequency F is very useful as a starting point, however.
On the subject of memory, I have measured my requirements, and determined at 16GB is often a bit of stretch and 32GB would be more "comfortable". I have a 64GB machine, and found I never use that much memory, so that was wasted money (for my purposes). As such I would be happy with any Apple Silicon release that supports at least 32GB of fast memory.
What is the "this" in your sentence referring to?
I'm talking about the removal/reduction of TB3 ports, and reduction in display support in M1 models released so far. I understand that they're mostly consumer/entry level machines, but they're still gimped in I/O compared to their immediate predecessors.
The "pro" models need to at least be able to maintain the same levels, before we start worrying about having huge core counts.
I still use a 2011 Sandy Bridge quad-core Mac Mini and bought an M1 Mini in December. The M1 is in a different league to the 2011 Mini, so there's no need to wait for your old ones to die!Still waiting for either of my MacMinis (2010 & 2012) to die - then I can justify buying a new one with an M chip. To be honest the way they are going, M4 will be about
That said, it is noteworthy that a 10-year computer is still quite usable for all-day development and productivity work. It feels slow after the M1, but it's by no means sluggish in absolute terms.
That's only half-right. The M1 MBA and MBP replaced the entry-level Intel models that also had only two TB3 ports and up to 16GB of RAM, so no reduction there. However, the previous Intel models could support 2 external displays, so it is true that they lost that capability.
I also wouldn't say that the capabilities were deliberately "gimped"; the limitations are simply part of the design compromises that just about every engineering project has to accept due to budget, power, manufacture or other limitations.
So why are they finding it so difficult to implement display support? Even ancient QuickDraw Macs seemed to work well with multiple screens. One company has got an 80 core ARM chip running already.
