>> Chip vendors face higher costs and complexities due to tighter margins, new processes, and materials at 20 nm and beyond, they say.
Absolutely because chip manufacturing is not an exciting business model where billions of investments go obsolete because you want to make a new generation of products. Unless there is scale, this may be the new normal. However, when many players exit, the remaining will have good time.
The key question is - does it really matter with this roadmap if people cannot make money doing it? We need to be working on intelligient systems and not fixated on Moore's law which helped Intel but may not be guiding it to make nimbler systems for the mobile world.
Traditionally, going to lower node means better performance at equal or lower cost but dynamics is completely different now. Especially during development the cost of mask is a gating factor for engineers already and management is asking to keep the cost of development under check.
Not sure i agree with conclusion though: It hints that 3d nand continues moore's law. But according  , "Industry experts believe 3D NAND had yet to compete against the current planar technologies in terms of cost structure.".If that's true, we can't yet be inspired with nand regarding the continuation of moore's law.
As we unveil EE Times’ 2015 Silicon 60 list, journalist & Silicon 60 researcher Peter Clarke hosts a conversation on startups in the electronics industry. Panelists Dan Armbrust (investment firm Silicon Catalyst), Andrew Kau (venture capital firm Walden International), and Stan Boland (successful serial entrepreneur, former CEO of Neul, Icera) join in the live debate.