Monday February 25, 2019

Linus Torvalds Comments On Apple's Potential Move to ARM

Rumors about the Mac's theoretical transition to ARM have been swirling ever since the original iPhone came out, if not earlier, but those rumors got particularly heated last week when both Bloomberg and Axios published reports claiming that the Cupertino-based company could make the switch sooner rather than later. While this would undoubtedly be a huge shift for Apple alone, the bigger question is what it would mean for the rest of the industry. Some think that ARM SOCs from a slew of companies could take over x86, Intel and AMD-based laptop, desktop, and servers in the future, while others say its not even within the realm of possibility.

On the Real World Technologies forums, Linus Torvalds, the creator and lead developer of Linux, decided to chime in. In his usual "warm and fuzzy" style, Linus said that the x86 ecosystem is entrenched, and that even if ARM server chips that offered a substantial performance and power advantage over x86 counterparts came out tomorrow, they would have a tough time gaining any significant market share. However, he does acknowledge "that the ARM laptops may make this all work out on the development side. Whether from the PC side ("WARM") or Apple cutting their laptops over." Having native ARM systems sitting in the homes and workplaces of developers is supposedly critical to the platform's adoption, meaning that low-power ARM laptops could be a stepping stone to the proliferation of higher-power ARM designs. Thanks to cageymaru for spotting the posts.

News Image

That said, I still strongly suspect that any "hyperscaling" 64-128 core system will be very underwhelming. It may be just good enough on Specrate style "nothing shared" benchmarks, and yes, you have the traditional network processor loads etc, but I think people always underestimate how big of a jump it really is from "look, we know how to make 4 cores work" to "look, we can cram a lot of cores on a die" and then to actually having something that truly scales. You simply need many many generations. And even then it's hard. So says a lot of mostly dead companies. Right now, ARM doesn't have even a single generation of server parts out, and they are pushing the hyperscaling story? Does that really make sense to anybody? (Yes, yes, I realize it makes tons of sense to all those people who already believe in the "sea or cores" fairytale despite not having any evidence of that ever working. But there's no arguing with delusion)

Discussion