"Intel didn't invent FinFETs, they productized them first because they were supplying gazillions of chips to the PC market."
[Supplying "gazillions of chips to the PC market" had nothing to do with it. They produced them first because they put ten years of development into it.]
"Now it's the turn of the foundries to do the same with the mobile chip market. FinFETs are new technology but not rocket science and all the foundries have the best scientists working on them. They have time and they have money."
[Amazing how many factual errors you can put in two sentences. FinFETs are rocket science. That's why it took ten years of development. The foundries don't have the best scientists working for them. Otherwise they woudn't be a constant two to three years behind Intel in R&D. They don't have time. You yourself just said it takes 4 years on average for each new node. Well, it only takes Intel half of that and FinFET is not going to be the average node development for ARM. Don't expect to see any ARM FinFET for years and years from now. Too little, too late as Intel will own mobility by then. The money is another issue as only TSMC and Samsung can keep up and we're not too sure about TSMC any more. No matter how many times you say it, the fabless model is broken. Simply because it is fabless and doesn't provide money for fabs. It leaves it up to the foundries to find the money and more and more of them are dropping by the wayside unable to come up with the billions for construction. And even if they did, they don't have the technology. Game over...]
It's not that the foundries are broken, it's that they aren't structurally or economically equipped to produce leading edge technology. ARM promoters that continue to argue this point and claim the foundries will match Intel's ability in manufacturing sound completely clueless and it further undermines their credibility.
The ARM camp, along with its foundries needs to reconcile the fact that their role is to produce trailing-edge technology and be happy with it. It's not a bad business because there are a lot of products that don't need leading-edge, it's just not the products and markets Intel is targeting.