Looking at the tech intel's 14nm appears to be as good as most other companies 10nm. Intel's 10nm when it launches next year will most likely be as dense as other companies 7nm. Intel's claim that their 10nm will perform worse is vague. Does that mean their 10nm will consume more or as much power or frequencies will be lower? I don't think this is a shock as 14nm performed worse than 22nm when it came out. Intel had to strap a bunch of cache on 14nm when it came out to make it competitive. I don't think Intel has anything to worry about in 2018 or 2019, they can always add cache or find other ways to reduce overall power like adding vega or including an on package ARM chip(apple). AMD is a tiny company and doesn't have any process and solely rely on third party foundries. To think these third party foundries have found a way to get large x86 processor's on 10nm/7nm and push frequencies beyond Intel is a joke. At best AMD could see themselves in a position where they can produce chips with better density but, if they don't have the performance to back that up it won't matter. AMD won't win any race making chips with better density unless they have the frequencies and performance to back it up. Remember that AMD lead Intel in every single price\performance category desktop\workstation\server in the early 2000's and still never managed to get more than 20ish percent market share. At AMD's peak they were offering Opteron server's with more cores and overall server costs 20-30% less than Intel servers and still companies mostly avoided them as they were scene as budget and not fit for enterprise. I admit that Epyc looks amazing in the server space and for someone who monitors\maintains server farms in 13 different counties I'm looking forward for the first Dell AMD servers to hit the market. I fear that like last time I wont be able to order them because of the brand.
Facebook
Twitter