Intel's strategy for low power was to make up for their relative inexperience in low power design with their advantage in fabrication technology. At 32 nm with Medfield, that failed. Unless they do something drastically better with Silvermont and by that I mean their improvement needs to be far better than ARMs own improvement going into the next generation, Silvermont is going to fail too. I would go so far as to say even if they are drastically, better, they'll fail because of higher price, absence of customization and the commitment of the big two to their in-house chips. Therefore in the long run, contrary to Intel's touted fabrication advantage, they are expected to fail -- it's a failed strategy. A more successful strategy would have been not to sell XScale in 2006 and bet the farm of low-power IA.
What's a fail are all your faulty assumptions and statements and they don't just apply to this post ! Medfield has no fabrication advantage over its rivals and it not a failure, it is just starting out and selling well in the markets it has been released with businesses specifically seeking it out in the UK because of the performance/power and Intel brand. Intel can control its price at any time if it wants marketshare as it can make *performance* chips cheaper than any foundry, notice a $200 chromebook has a 32nm Sandy Bridge Core Celeron 867 in it which was priced $134 at introduction so obviously Intel can vary its prices to suit market situations. Sure Intel is not going to win Samsung or Apple but what about the other 50% of the market ? You are just an ARMH investor trolling and just causing mischief here and it isn't even subtle lol !