Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
It would be mildly intriguing to build the ability to read/run RISC-V code into their existing ARM-based CPUs (basically as a hardware front-end processor / translator), though I’m not sure what benefit it might bring to their long game. But, might be moving chess pieces around to handle some future possible scenario.

Maybe something oddball like governments decree that smart cars must run RISC-V code so they can all run some government approved safety modules.

Would make much more sense to just have Rosetta 3 do code translation.
 
Well, this was inevitable considering how much Apple hates Nvidia (who bought out ARM) over its graphics chip debacle while back. They do not want to pay Nvidia a single cent going forward and going with open source method will likely bring more chip innovation for the company.
I think Steve Jobs was bigger on retribution. Apple of today just decides they will no longer rely on or partner with the company. Nvidia left Apple holding the bag on defective parts, so Apple isn't going to use Nvidia parts.

Of course, it probably has about as much negative impact today due to Apple being several times larger than 10 years ago.

That said, I don't think Apple gives two craps about what Nvidia does with Arm in terms of Apple hardware, because they have a perpetual agreement to use the Arm architecture regardless of what Nvidia does. They probably care quite a bit more about the market impact, e.g. does Nvidia start to make it easier or harder for other handset manufacturers.
 
  • Like
Reactions: CarlJ
Honest thought -

I'd like to see a study on who (private companies or the government) does a better job at investing in 'social well-being' on an equitable basis.

There are ills on both sides:

- government bodies (like NASA) allocating money in a manner detrimental to their goals.
- private companies doling out spending in a self-serving(?) slap-dash(?) manner.

I wonder which one is least worse... and I'm sure we can do better.
 
  • Like
Reactions: CarlJ
I think the competition had a much more realistic spin on this:
Some are speculating that Apple may be in the early stages of planning a move away from ARM to RISC-V for future generations of A-series and M-series chips, in order to avoid having to pay royalties.

This is extremely unlikely. Switching architectures is a massive undertaking, involving years of planning and preparation. If Apple had intended to switch to RISC-V chips, it would almost certainly have done so before switching its Macs from Intel to ARM.
This would be true if RISC-V chips were ready for prime time 5-10 years ago (when rumors and leaks of Apple experimenting with MacOS on ARM started coming out). My understanding is that RISC-V is just finally getting to the point where it is viable to start playing with the idea of transitioning. If this transition is going to happen it is probably 10 years away - there’s no way Apple could stay with Intel that long.

Companies like Micro Magic and SiFive have produced chips that seem to indicate that RISC-V is more power efficient than ARM. Maybe things like AirPods or the Cortex-Mx CPUs in AirTags would be good examples of what might transition first.

 
  • Like
Reactions: jz0309
This would be true if RISC-V chips were ready for prime time 5-10 years ago (when rumors and leaks of Apple experimenting with MacOS on ARM started coming out). My understanding is that RISC-V is just finally getting to the point where it is viable to start playing with the idea of transitioning. If this transition is going to happen it is probably 10 years away - there’s no way Apple could stay with Intel that long.

Companies like Micro Magic and SiFive have produced chips that seem to indicate that RISC-V is more power efficient than ARM. Maybe things like AirPods or the Cortex-Mx CPUs in AirTags would be good examples of what might transition first.


Hard to know what to make of the micro magic numbers. Give me today’s Fab nodes and tell me to make a single core, single thread, super simple CPU that gets around a third of the performance of an M1 but clocks at 5GHz, and you can bet it would be pretty power efficient - that core is so simple that you can carefully place every gate by hand. My gut tells me that if Apple wanted to target only efficiency and didn’t care about actual performance, they could achieve at least as much with Arm.
 
  • Like
Reactions: CarlJ
It just hit me, Apple isn't investigating this for ANY of their existing products, they're very-likely doing so for their Apple Car effort, which offers lots of clean-slate opportunity to transition to & use a new ISA.
 
BSD+Creative Commons licensing will work perfect first in the Apple Car as the main SoC. Apple will have been developing for RISC-V at least for the past five years in-house. At least that's how we didn't back when I worked there and earlier at NeXT.
 
It just hit me, Apple isn't investigating this for ANY of their existing products, they're very-likely doing so for their Apple Car effort, which offers lots of clean-slate opportunity to transition to & use a new ISA.

I can’t say for sure, but if it were me I’d be doing everything in my power to minimize the amount of “new” in the Apple Car.

Let’s switch to RISC-V! Can we not? Well, yeah, but… Then don’t.

From the outside, with AC shrouded in mystery, it’s tempting to connect everything to it, but from the inside there’s way too much complexity to manage. There’s already too much ”new”. They won’t be trying out any extra new tech just because they can, they’ll be using as much old tech as possible because they have to.

It’s not clear exactly what Apple Car is, right now— if it’s an OS for OEMs, then maybe, but if it’s an actual car that Apple plans to produce then everything in me would be screaming “stop changing **** until what we have works!”.
 
Companies like Micro Magic and SiFive have produced chips that seem to indicate that RISC-V is more power efficient than ARM. Maybe things like AirPods or the Cortex-Mx CPUs in AirTags would be good examples of what might transition first.


I wouldn’t draw too many conclusions from those microbenchmarks. It’s not that difficult to build a power-efficient ALU, but it doesn’t mean that the resulting CPU is going to be any good. I could not find any information about the Micro Magick product. We don’t even know whether it’s an in order or an out of order design. I wouldn’t be surprised if it’s a very simple CPU that does reasonably well in simple arithmetics, but will immediately bog down if you start dealing with cache misses and indirections.
 
Apple has an ARM architecture license.
They were original investors is Advanced RISC Machines.

They go as far back and the Newton with their use of ARM processors.

They don't pay much in the form of licensing is my guess.
 
  • Like
Reactions: cmaier
Once again, these savings won’t get passed on to us customers, prices will continue to go up. 🤦‍♂️
they must develop the processes. There won’t be savings for years after implementation.
If everyone ends up on RISC-V, where’s the competitive advantage for any of them?
Same as now. They’ll have a full stack. And with RISC-V, they won’t owe extra cash.
 
  • Sad
Reactions: LFC2020
I expect this is just something to keep in their pockets in case NVIDIA gets hold of ARM and starts gouging the prices or leveraging licenses to try to press Apple to take their GPUs.

Apple certainly still works on x86 kit and maybe even PPC just in case they come back into contention again down the line. They ran OS X on Intel from day one behind the scenes.
 
I expect this is just something to keep in their pockets in case NVIDIA gets hold of ARM and starts gouging the prices or leveraging licenses to try to press Apple to take their GPUs.

Apple certainly still works on x86 kit and maybe even PPC just in case they come back into contention again down the line. They ran OS X on Intel from day one behind the scenes.

Whatever license deal apple has, there’s little chance nvidia can do anything to change it at this point. As one of the three companies that formed the original Arm joint venture, you can bet that Apple’s rights are sewn up and not subject to whatever Nvidia may decide to do in the future.

Apple also is not doing anything at PowerPC at this point. If, for some reason, their own design team falls down and they can no longer compete with…um… IBM I guess?… then there are lots of other places Apple will be able to get Arm designs. And if, instead, it’s the stellar advancements made by Global Foundries that make this hypothetical future PowerPC such a great deal, Apple can just switch from TSMC to Global Foundries.

By designing the CPUs themselves, and in a world where every cutting edge fab is a contract fab (including, it looks like, Intel going-forward), there’s no reason for Apple to worry about PowerPC (and there is zero inherent advantage of PowerPC over Arm. I’ve designed PowerPC CPUs - they are a pain in the arse).
 
  • Like
Reactions: CarlJ
This line of thinking makes no sense. As already done by other manufacturers, it might be very cost effective to design image stabilization processors and other peripherals using risc v rather than paying up for an ARM license for each one. But, for main processor, Apple is doing a huge effort with Apple Silicon now, which is based on ARM Architecture, associated with all effort from partners, software companies recompiling everything, Rosetta 2 outstanding performance (which probably took up to a decade of hard work to be achieved)...

Probably is time to stop hiring teenagers to write tech articles.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.