No More Intel Inside, Apple Plans To Use Its Own Custom-Built Chips in Mac (bloomberg.com)
Apple is planning to use homegrown custom-built processors in its Mac line of computers, ditching Intel, the processors by which powers Apple's current line of computers, Bloomberg reported on Monday. The company could make the switch to its own chips as early as 2020, the report said. From the report: The initiative, code named Kalamata, is still in the early developmental stages, but comes as part of a larger strategy to make all of Apple's devices -- including Macs, iPhones, and iPads -- work more similarly and seamlessly together, said the people, who asked not to be identified discussing private information. The project, which executives have approved, will likely result in a multi-step transition.
The shift would be a blow to Intel, whose partnership helped revive Apple's Mac success and linked the chipmaker to one of the leading brands in electronics. Apple provides Intel with about 5 percent of its annual revenue, according to Bloomberg supply chain analysis. Intel shares dropped as much as 9.2 percent, the biggest intraday drop in more than two years, on the news.
The shift would be a blow to Intel, whose partnership helped revive Apple's Mac success and linked the chipmaker to one of the leading brands in electronics. Apple provides Intel with about 5 percent of its annual revenue, according to Bloomberg supply chain analysis. Intel shares dropped as much as 9.2 percent, the biggest intraday drop in more than two years, on the news.
I think it used to be. Now it's people pretending to be those things.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
This has been in the works since the start of the LLVM project.
A few years back, Apple and the LLVM project made the announcement that code compiled for x64 with CLang was finally able to run unmodified on the ARM architecture. By compiling into an intermediate language, Apple has made it possible to write code that should run unmodified on any LLVM platform so long as all libraries are present to support it and that the code doesnâ(TM)t depend on hand written assembly or code which needs direct access to the stack for the platform ABI.
With the transition from PPC to x86, a lot of transitional APIs such as Carbon were introduced. Also, the principle of fat binaries were made common place in such that each application or framework could be compiled for two or more platforms. Consider that Apple had Yellow Box running in house on x86, PPC and Sparc.
Over the years, Apple has progressively deprecated any API which was too tightly bound to a single architecture one by one. All code not compiled with LLVM has been slowly killed off. The App Store on IOS and MacOS have set restrictions as to what system calls could be made. Most performance oriented libraries such as QuickTime have been altered, enhanced, etc... to slowly eliminate the need for hand written code. Apple has bullied developers into never writing Mac targeted compilers and instead focused them on compiling to IL or Swift/Obj-C first.
Just like Microsoft has been trying to universally move to .NET for a retargettable platform, Apple has moved to LLVM.
There is no technical reason why Mac couldnâ(TM)t run on ARM today. Iâ(TM)d imagine Apple has had Mac OS running on an iPad Pro for some time. The main difference would probably be the type of SSD they employed.
Performance wise, current Apple chips should have more than enough CPU to handle tasks at least as well as the m3 chips in the Mac Book. 4GB or RAM should be enough for most users as well. PCIe for M.2 storage should be a trivial change for Apple. And Apple has already said they are preparing their own GPU core. I would expect that GPU core to be comparable to Intelâ(TM)s from the beginning. Unlike other GPUs, OpenCL and even most of OpenGL are optional as Apple will dictate the OS graphics API. Of course they already have a strong enough following among game developers that if they cut corners, the developers will suck it up and continue.
What most people mention is a problem is that Mac has a huge dual boot audience. I would expect an agreement with MS or Amazon to happen to push cloud based virtual desktops. Many enterprises get security by using Mac because malicious Mac software doesnâ(TM)t tend to screw with virtual machines. So they deliver the enterprise desktop on a VM and let the user mess with their Mac however they want.
What I expect to really shake things up will be an announcement from Apple to support Windows for ARM as an application/subsystem. Then I expect to see Microsoft support their x86 emulator possibly with acceleration on Mac. Unlike Transmeta. Apple working with Microsoft could easily make their x86 JIT perform better than real hardware. This has to do with how branch prediction, pipelines and cache work.
I honestly donâ(TM)t see anything particularly amazing about this other than the long time it took to get here. Apple must have assessed that the lost business will be offset by the profits gained. Of course, I have been hoping to buy a new Mac Mini this year, my 2012 model is getting old. If Apple releases something âoerespectableâ for $500 or so, Iâ(TM)m in. I only need it for testing Mac builds.
I agree though, that this is probably Apple trying to extract some sort of concession from Intel, be it pricing, input in, or influence on, the feature set or direction of development, or all of the above. The threat of a switch to ARM may seem more credible than the threat of a switch to AMD, perhaps, but either seems incredible to me.
https://www.eff.org/https-everywhere