Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
$pple is just a cynical company that is so greedy and selfish and never wants to give to others unless they get something in return. 200B in the bank and act like a troll.

Ah yes, no better way to start a day but with a senseless complaint about Apple. My compliments.
 
  • Haha
Reactions: amartinez1660
$pple is just a cynical company that is so greedy and selfish and never wants to give to others unless they get something in return. 200B in the bank and act like a troll.


 
Apple is looking for an engineer that specializes in RISC-V

"An engineer" isn't going to single handedly replace a decade of Arm development. Maybe they already have an ongoing development in the labs that they're trying to add to, but even then RISC-V is relatively immature versus Arm.

I think Apple specifically refers to the new machines as "Apple Silicon" and goes way out of their way to avoid mentioning Arm except when absolutely necessary because they want the freedom to deviate from the Arm architecture when they choose to. That said, I don't think we're about to see a wholesale swap to RISC-V anytime soon.

The usual Apple way is experimentation over incredibly long time horizons. At some point we'll find out they put a RISC-V in a magic mouse or something to test it out. Then a support core somewhere in a Mac. Little by little they'll gain experience and confidence. Eventually we'll find out they've had MacOS running on it since 2010 and were just waiting until all their benchmarks were met before releasing to the market.
 
  • Like
Reactions: amartinez1660
Possibly fine for Apple to port its own mostly low end applications (browsers, email clients and word processors). To be seen how many developers are then willing to port their code to yet one more (and very niche) architecture...
 
"An engineer" isn't going to single handedly replace a decade of Arm development. Maybe they already have an ongoing development in the labs that they're trying to add to, but even then RISC-V is relatively immature versus Arm.

I think Apple specifically refers to the new machines as "Apple Silicon" and goes way out of their way to avoid mentioning Arm except when absolutely necessary because they want the freedom to deviate from the Arm architecture when they choose to. That said, I don't think we're about to see a wholesale swap to RISC-V anytime soon.

The usual Apple way is experimentation over incredibly long time horizons. At some point we'll find out they put a RISC-V in a magic mouse or something to test it out. Then a support core somewhere in a Mac. Little by little they'll gain experience and confidence. Eventually we'll find out they've had MacOS running on it since 2010 and were just waiting until all their benchmarks were met before releasing to the market.
Even back in the days of Nextstep before it came to Apple and became the basis of MacOSX, Jobs had it running on multiple processor’s. Since Jobs returned to Apple then it’s been universally understood that Apple actively compares multiple types of hardware based processing running different OS environments to compare and learn from all that. So yes they likely already have several ongoing teams looking at anything that might offer possibilities in their labs. Good for them learning from all the possibilities as evident with the Apple Silicone examples. :cool:
 
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.
The deal has not gone through and UK SoS has received guidance to oppose Nvidia (Delware reg.) acq. of ARM Ltd (UK registered).

I'm sure this is a pure money play. Even a single dollar per ARM device license (and assume a license per CPU core + T2 and other chips) runs at a couple of dollars / device.
x 200M devices per year (phones + laptops) that adds up.
 
That is nothing more than a clickbait article when Apple just begun transitioning to ARM architecture from x86 and chance of making another transition to RISC-V as of now is pretty much zero. That is probably for an accelerator controller and job posting already indicating as much.

"This is to support the necessary computation for such things as machine learning, vision algorithms, signal and video processing. Push the state of the art in low level computation and drive them towards energy efficient and high performance implementations by tightly integrating software and hardware."
 
Apple doesn’t use any of ARM hardware IP, so they only pay for the use of architecture (ISA), which probably makes their license costs insignificant compared to R&D costs. I bet ARM makes much more money of Qualcomm and Samsung who actually use ARMs hardware designs than they make off Apple.
Apple, Samsung and Qualcomm all hold architecture licences, which allows them to use the instruction set in their own designs, and provides a test suite to make sure the chips still run regular ARM code, even if the licensees make tweaks.

This is the most expensive licence ARM offers, and Apple is probably thinking that if they don’t use the reference designs then that’s a lot of money when they’re doing all the work.

They could change to a different instruction set and 99.9% of their customers wouldn’t notice. (As long as virtual machines and container setups like Docker make sure they’re compatible).
 
I guess it’s probably good to keep their options open, lest they get bottlenecked by someone else’s roadmap viz Intel Macs. That’s probably what this is about at this point more than anything, especially as Arm currently underpins their golden goose the iphone… though I wonder why not then go the whole hog and create their own instruction set?
Agreed. There are two explanations for this (possibly both) - one, Apple is considering the potential for using RISC-V in some ancillary devices, or two, they’re keeping their options open for the long game.

They hate getting stuck in some corner instead of moving forward, because their plans don’t fall in line with someone else’s capabilities or whims (Intel, IBM, but also others). It makes them want to own the whole process, end-to-end, so they can set the pace. And this is understandable.
 
  • Like
Reactions: Falhófnir
So this article confirms absolutely nothing.
Pretty much.

I've designed with RISC-V. It's a good research platform. Their cache coherency (Tile-Link) is a joke. Either ARM or Arteris (nCore) is better and since RISC-V requires a gasket to get the AMBA/AHB protocol not really compatible with the rest of the world. On the other hand their are a couple of RISC-V implementations with native support.

I can't see Apple switching to RISC-V for their main processor architecture, but for embedded and watchdog processors absolutely.
 
  • Like
Reactions: CarlJ
What do you mean? Apple is heavily invested into open source, their platform is in fact built on open source initiatives. For example, virtually every modern browser today traces its codebase to WebKit, an open source browser engine developed by Apple. LLVM, one of the most used compiler frameworks, was built on Apples money.
Agreed that Apple both uses and supports open source (even though that doesn’t fit with a lot of people‘s narrative of “Apple is a perpetual villain”), but to be clear, WebKit started as a fork of KHTML and KJS, from the KDE open source project. Yes, Apple has poured a ton of effort into it since, and arguably made the web a better place (standards-wise) in the process. And LLVM and Swift are lovely examples.
 
Apple, Samsung and Qualcomm all hold architecture licences, which allows them to use the instruction set in their own designs, and provides a test suite to make sure the chips still run regular ARM code, even if the licensees make tweaks.

This is the most expensive licence ARM offers, and Apple is probably thinking that if they don’t use the reference designs then that’s a lot of money when they’re doing all the work.

They could change to a different instruction set and 99.9% of their customers wouldn’t notice. (As long as virtual machines and container setups like Docker make sure they’re compatible).

Given that ARM Ltd was a joint venture of apple and two other companies, I wouldn’t assume apple is paying full freight here.
 
I mean, Macrumors could have at least posted the most relevant information.

Apple is currently looking for experienced programmers with detailed knowledge of the RISC-V Instruction Set Architecture (ISA) and Arm's Neon vector ISA for its Vector and Numerics Group (VaNG) within its Core Operating Systems group. Apple's VaNG is responsible for developing and improving various embedded subsystems running on iOS, macOS, watchOS, and tvOS.

This is interesting because without this piece of information I would have thought Apple is trying to save pennies in those embedded ARM cores ( not the the ones used in CPU ) such as U1, W1 etc. But this is being specific to VaNG. And RISC-V is certainly a great fit for this use case with their Vector extension instructions. ( If you want to understand this I recommend searching for SIMD and Vector ) It will be interesting to see how all these fits together.

The embedded ARM cores aren't always 100% Apple ones. Apple licenses work from other folks for embedded jobs. If those other folks are starting to switch over to using RISC-V then Apple would need to cover that if they want to do highly custom drivers/interfaces to those targets.

RISC-V in DSP and signal processing isn't getting deep traction because of the "vector machine" approach in the RISC-V standard set. There is also flexibility to put in specific instructions for narrow DSP problems into the implementation.

If Apple is picking 3rd party embedded implementations because they are cheaper because that 3rd party can sell their solution at a lower cost and still make a profit.... then it hasn't particularly escaped Apple chasing pennies via cheaper component costs. Especially, where Apple is the major external force pushing for downward pricing on the components.


There was lots of low level 'dot the i's and cross the t's' work that Intel pragmatically did for Apple because coasted on EFI/UEFI/BIOS boot and baseline driver design already written to interface from x86. If Apple is substantively changing how to talk to the devices then they'll need their own team to help replace that baseline driver work.


P.S. there are lots of embedded chips that have "someone else's cores " embedded inside somewhat anonymosly. CEVA cores are in most of Intel, Samsung, MediaTeks modems.

"..Of the latter, there’s only Qualcomm and most recently HiSilicon (the in-house modem IP introduced in the K960), whereas most other major players over the last couple of years have used CEVA IP. Samsung, Intel, MediaTek, and ..."

Maybe Apple is kicking CEVA out after they acquired Intel's implementations, but is almost certainly going to cost them substantive time to market. But the notion of Apple kicking every embedded 'CPU' core out of every chip soldered to or licensed from integration into an Apple product with an Apple deisgned from scratch CPU core is loopy and probably totally unnecessary.

Apple is going to have to have some folks on staff to deal with cores that they don't do themselves. Can't be a 'down to the metal" systems provider and ignore everyone else's stuff at the same time.
 
I thought Apple paid a one time sum in order to have use ARM instructions in their chips indefinitely? I didn't think they were paying annual fees?

Everyone is looking at RISC-V and I think it's a good idea, but I don't see why Apple would feel the need to switch away from ARM, perhaps this is more a speculative R&D thing. RISC-V is a long, long way off from competing with ARM and x86.

They can basically do what they want with ARM. They can no doubt create their own custom instructions if they felt like it, I don't see how this really makes much sense. ARM has big advantages over x86 like efficiency, parallelism etc but what advantage does RISC-V have over ARM other than being open source?
 
I could see Apple transitioning to RISC-V in the future but that is probably 10+ years off. Smaller embedded cores could maybe transition sooner (I believe that Western Digital and Seagate are both using RISC-V some of their hard drive chips).

Edit: 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.

 
Last edited:
"An engineer" isn't going to single handedly replace a decade of Arm development. Maybe they already have an ongoing development in the labs that they're trying to add to, but even then RISC-V is relatively immature versus Arm.
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.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.