

|
Via: NY Governor | News Archive
| Tags:
AMD,
Intel,
IBM,
GPU,
CPU,
TSMC,
GlobalFoundries,
APU,
GloFo
|
|
@ Joel "I recently read an article on another website that you work for, Extremetech.Com, : "Is 14nm the end of the road for silicon chips?" where the author talks about the difficulties of moving towards a smaller die shrink and the limitations Companies might face after 14nm (specifically Intel who has advanced the most). So I would think that All those leading chip developers joining force can help them more efficiently adjust to a new process but also the research and development will help Intel break that 14nm wall, right?" http://www.extremetech.com/computing/97469-is-14nm-the-end-of-the-road-for-silicon-lithography |
|
Even worse news for AMD it seems. |
|
My nerd says YAY! But the other part of me is: "what about the other guys?" |
|
Optimus, Akumu, Jonation: This sort of research addresses common problems faced by everyone--to a certain extent, it's a win/win/win for all involved. The question here is "What bets are going to work out best?" Here's an easy example of that. TSMC is moving towards using 3D chip packaging with through-silicon-vias (TSVs), where chip components are connected vertically. Intel is talking up 3D chip technology through the use of Tri-Gates (3D transistors). Both companies are claiming power consumption benefits and performance enhancements. The question of which one is "better" (assuming both companies successfully implement them) may depend a lot more on the company's goals than on the technology at hand. Sometimes, things end up a wash. AMD moved to immersion lithography before Intel, for example, but both GlobalFoundries and Intel have ended up adopting both immersion lithography and double-patterning. Pushing below 14nm is likely to be more of a steep slope than a brick wall. It's one thing to say "That's the end of scaling," but no one has come up with a credible alternative tech, yet. |
|
"Got It" |