Might be interesting to see what their crack tech of processor designers do with x86, given the performance they managed to squeeze out of the ARM arch.
Do you have a link for the Bloomberg article?
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 Plans to Ditch Intel and Use Custom Mac Chips Starting in 2020
- Thread starter MacRumors
- Start date
- Sort by reaction score
Do you have a link for the Bloomberg article?
Beating a dead house, are we? It's actually sad that some people are so willing to dump Intel, for hopes of a dumbed down, toyOS to replace MacOS?
Imagine having to get used to displaying only one or two apps at a time, or reloading of Safari tabs every time someone checks candy crush. This just sounds like a terrible idea, and many users are expressing their doubts.
Imagine having to get used to displaying only one or two apps at a time, or reloading of Safari tabs every time someone checks candy crush. This just sounds like a terrible idea, and many users are expressing their doubts.
Many of them have designed x86 before.
But x86 requires a license from intel, so unless intel is agreeable to all this I don’t buy it.
[doublepost=1523021644][/doublepost]
The os has nothing to do with the processor architecture. Any code that you can write that runs on x86 can be written to run on ARM.
Sure you can run it on both, but do you honestly think future ARM chips can even beat today's Intel midrange chips, in 2020? If it can't come close to, or beat Intel performance, then it's literally like taking steps backwards.
If they had went with AMD on the other hand, I'm pretty confident this thread wouldn't have reached 59 pages, with many upset users that are ready to jump ship.
Yes, I am positive arm designs can beat intels mid range chips. Because I designed many CPUs (exponential PowerPC x704, ultrasparc 5, AMD k6+, AMD opteron, among others) and I understand how these things work.
Apparently you will be able to as of later this year when iOS apps run on the Mac. Makes far more sense to emulate arm apps on x86 then x86 apps on ARM.
Wow. Theres a difference between "I don't want all Mac application development to be bottlenecked by iPhone development" to running iOS apps on the Mac without completely destroying the utility of the Mac.
I am kidding. I just find all of the toy references from the legacy folks amusing. Your points are valid, we just have a different opinion.
They already have two platforms, good god, it's like the last 10 years and 1.5+ billion devices produced (and 1B in activity) never happened...
[doublepost=1523023923][/doublepost]
It could beat it TODAY if they give it enough heat headroom, there is nothing magical about what Intel is doing.
When the A11X hits, we'll see how far they are now from this mythical "midrange" (sic) desktop on something that runs on a few watt.
Two platforms that run different operating systems. Very different scenario.
The current price of manufacturing custom A11 chip is $32 per unit. Does that sound expensive to you? Even if the i9 Equivalent processor from Apple costs $200-300 it would still be cheaper than any i7-i9 whose cost range from $1000-1600+, Both AMD and Intel manufacture Ryzen and i7-i9 in the range of $150-$200 per unit. That is all the manufacturing cost then they add other costs and fat distributor margin to it and you have a product that 6-7 times more expensive meaning with 700% profit margins that the actual cost. So i think even if Apple doesnt have much economies of scale they would still be able to manufacture it. Apple makes its profit from the end product, not on the component cost.
Last edited:
Its a lot like the printer companies and their ink cartridge racket.
You may be correct from the technical standpoint but there should be a reason why all the companies which designed the CPUs on your list lost to Intel (in terms of CPU design/adoption). It may be different this time but I would not bet on it.
*Development computer
My phone is a computer. So is the smart watch on my wrist. I do plenty of paid for work on my phone (email, messages, camera, to do list). Most people aren’t coders.
The reason they “lost” had nothing to do with that. The Wintel alliance was a tough nut. And I dispute your premise. AMD *beat* Intel - every x86-64 you buy from Intel is based on our work at AMD. Intel’s 64 bit solution was Itanium. We destroyed it with AMD64, and Intel was forced to copy it and take a license.
Intel was able to improve the performance of x86 by translating each instruction into a bunch of μops and feeding those into the long pipeline – essentially, they hid a RISC processor inside the Pentium. From what I see, it looks like ARM A73 design (upon which the A7-11 chips are based) also uses μops to get the most out of a long-ish pipe.
So the underlying mechanism is the same FAIT&P. All Apple has to do is make an instruction stream interpreter for x86 and set it up alongside the A73 interpreter. How well it would work is uncertain, but it would provide a better result than emulation.
Of course, the phrase “all Apple has to do” greatly understates the undertaking. If they are not already 75% there by now, it would probably not be happening before 2025.
So the underlying mechanism is the same FAIT&P. All Apple has to do is make an instruction stream interpreter for x86 and set it up alongside the A73 interpreter. How well it would work is uncertain, but it would provide a better result than emulation.
Of course, the phrase “all Apple has to do” greatly understates the undertaking. If they are not already 75% there by now, it would probably not be happening before 2025.
I agree, just being facetious. I for one can’t wait until the crowd that thinks that way fades into obscurity.
Someone in this thread mentioned earlier that the strategy of DEC Alpha64 was the most excellent and it could emulate 3 different instruction sets simultaneously all within the same processor.
I have. I remember them fairly well. They had a pretty good mobile processor for their day.
But I am not talking about code morphing. ARM and Intel both use μop translation, and ARMv8 has more registers than x86-64, so it seems like building a real-time hybrid might be a realistic possibility. There was rumor of a Taiwanese company that had a hybrid (32-bit ARM/64-bit Intel) a few years back, so it probably can be done effectively.