On purely a hardware level, it seems much more likely to me that an x86 chip with x86-class performance will be put in a smartphone long before an ARM chip with x86-class performance can be shrunk.
Absolutely fascinating article on why mobile apps are slow. The reason x86 would come mobile first is because CPU fabrication became massively expensive. x86 CPUs would fit a smartphone within the next under-20nm process, which necessitates fabs that cost a couple billions to build today and that only Intel has (Samsung too but they don’t makes CPUs).
The rest of the article highlights the big problem with memory and GC. But GC is used a lot, everywhere. That’s the thing, it’s been invented to produce code faster it’s a design decision. It’s a problem right now with mobile based on ARM.
But will it be in three years with x86-class performance in any device? Because GC doesn’t impact x86 CPUs as much and helps programmers tremendously, it might be better: mobile x86 + GC = robust apps for users, easier work for developers.
Nevertheless, developers need to get better with performance. There is no way we can deal with a sluggish, unresponsive digital life whatever language, platform you guys use.