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.
Craig Federighi and Greg Joswiak Discuss Apple Silicon Transition, Lack of Boot Camp Support, and More
- Thread starter MacRumors
- Start date
- Sort by reaction score
Apple said that most developers would update their games to run on Catalina, which never happened. Only online games made the transition, and some just abandoned the Mac. Furthermore, the games Apple showed in the presentation were old games that were running in low resolution and fidelity in an effort to hit frame rates…and they still stuttered.
I agree - I bet this will see a push on the server market at some point as well - if these ARM chips can gain traction and prove better performance density - I can see the worlds of AWS and Microsoft deploying these across their clouds to help create the next generation of compute.
At the end of the day, x86 is an instruction set - just like ARM - and what Apple has proven so far is that the ARM instruction set on the device isn't a limiting factor for what people want and need to do on a daily basis as both cover the majority of use cases that are needed, and for those use cases that are not covered those are the areas of innovation.
Massive profit margin increases and amazing leap in the wealth of Apple executives.
gruber's ad reads have always been painfully dull compared to his fellow podcasters and have always been especially bad when standing live in front of a crowd at his annual WWDC interview event... it is what it is. i give total credit to him for putting it up front because it only takes seconds to skip it entirely if you don't want to listen to it and get to the interview instead.
I suspect MathWorks, the makers of MATLAB, going back to the mid 1980s, are all set to go from working with Apple. Can't wait!
"but it will affect Apple in certain markets, specifically the Engineering Professional market, Macs are not good as they are for my line of work and now they will be further from attractive."
Macs have been great for my line of work, for decades.
So? If their CPUs are "all that and a bag of chips", they owe their shareholders maximized profits on their investment. Even though they're doing amazingly well, they're obligated to do better and increase thier share value if they can. Or is Apple somehow exempt from capitalist principles?
To your point, I suspect that thier CPUs aren't actually much better (if at all) and they're banking on "integration" and "ecosystem" to sell a "good enough" product. But as the product hasn't actually changed materially (most consumers won't know/care or assumed the machines were already full of "Apple Silicone"), where will the uplift come from?
Either:
- the CPUs are orders of magnitude better, in which case they're leaving money on the table, or
- the CPUs are about the same or worse, this move alienates some percentage of the technical community, and won't really result in sales uplift because it hasn't changed anything ...
As the developer units find their ways into people's hands, we'll start hearing about the performance merits of their chips and see what's what. NDAs won't prevent the truth from coming out.
Not sure what you mean, Windows 64 bit is running on ARM! It is just the x86 emulator in Windows (e.g. the equivalent piece to Rosetta 2) is currently only supporting x86 32 bit. This still makes 80% of Windows application runable under Windows on ARM and thus also under MacOS via Virtualization.
[automerge]1593113715[/automerge]
Except of course, that the A12z chip from the iPad has no relation to the upcoming Mac chips as far as performance is concerned.
If you came to me and said this then I would tell you to clear out your desk and turn in your Intel Mac laptop.
[automerge]1593113969[/automerge]
You build and test on the OS you're deploying on. If you don't then you're lousy at your job.
Well, just a thought: What do you think about an external X86/X64 CPU as a coprocessor in an external case. Just like an external GPU. 🤷♂️
The developer units are still running the iPad chip so really we need the first ARM Macs to see what we are in for.
What platform do you deploy to? If the consensus on here is it doesn't matter then you really are not doing your job properly.
Intel have the Xeon Pi that was or still is on a PCIe card.
Sure, all the lousy developers, developing for iOS...
Ideally there would be no variables, I mean why not deploy MacBook Pros in all data centers-, that way the client and server match 100%!
You don't have any realistic technical issue here- you just like hearing yourself whine I think.
Well you can setup AWS code build to build and deploy to test every time you commit to get past the ARM thing on Mac. As long as macOS runs docker and there are images for things like ElasticSearch it won't matter too much.
"Platform" includes more than the cpu. In this particular case, the CPU is probably the least important aspect.
But since you're in know-it-all mode, you might not be doing your job properly if you think developers go straight from their local machine to live production, same CPU architecture or not
May as well just buy a separate box and share a monitor/keyboard with a kvm switch at that point, no?
The only solutions for the first problem are increasing the market share of Macs to make it worthwhile for companies to port, or increase the market share for Windows on ARM making it worthwhile for companies to port to that. This switch brings the possibility of both. Arm Macs will be more competitive with Intel-based Windows from a performance standpoint, and with specialized hardware like the Neural Engine, will be much more interesting than they have been for some classes of AI applications. On the other hand, there will very quickly be a much larger base of ARM-based laptops/desktops, which may make porting Windows apps to Windows on ARM.
As I have asked others, what specific applications do you run (both native that and Windows only) for which you are concerned?
Apple will no longer have to pay Intel/AMD’s R&D, marketing, administrative costs and profit margin on one of their most expensive components. Their profit margins have remained remarkably consistent over the years, meaning they have 4 choices for their savings:
- Increase their margins.
- Lower their prices.
- Maintain their prices, but increase performance.
- Some combination of 2 and 3 (or of 1, 2 and 3).
Given that Metal has been pretty much a requirement for 2-3 years, any company that has not made the move, is not investing anything in the platform. That clearly means that Apple’s current strategy was not working for them. This is a new strategy. I have no guarantee that it will work, but it seems highly likely that it will not be worse.
We have seen a million posts where someone compares some hacked together machine, with no support or warranty that is faster (for some thing on Windows) and/or cheaper than Apple’s products. Being on Apple’s own silicon means that they once again have a chance to offer genuinely different systems, that have the potential to have real advantages. Macs on Intel/AMD will be inherently more expensive as they pay around the same underlying cost of components, have to do their own hardware design and have to pay for all their OS R&D for a system with under 15% market share.
With these machines, for the first time ever, they have the chance to offer products that no one can match. Whether they will succeed at that remains to be seen.
Last edited:
but it does. the A12z is essentially a 2-year old processor, related to the A12 which has been since been improved upon with the A13.
There seems to be no indication that an A13x would not be possible, or that an A14 is not happening on the iPhone 12. Apple's development trajectory is relatively stable.
so, extrapolating, Apple likely has an A14X model* that could be made for a new ipad. and that model could possibly be extended to a macbook or a desktop.
the mac minis with the A12Z are essentially as if we were looking at development machines with Intel 8-th gen processors today without the knowledge that 9th and 10th exist and that 11th is in the works, with the additional knowledge that Apple's processors have not been stagnating in terms of performance unlike Intel's.
so that's what the relation with the upcoming chips is.
* by model I mean a framework to build an X-processor for an iPad starting from a base iPhone processor, using the same tech, but given the different constraints in the hardware.
