InvestorsHub Logo
Followers 29
Posts 25865
Boards Moderated 0
Alias Born 09/11/2002

Re: chipguy post# 119960

Tuesday, 06/18/2013 3:02:20 PM

Tuesday, June 18, 2013 3:02:20 PM

Post# of 151823

Seattle: 8 and then 16 core SoCs?

The latter has to be a dual die MCM. I can't believe AMD would
commit resources to develop two separate ARM devices/masksets
for an untested market.


That sounds reasonable to me. The problem with AMD is that they're late with this product, and still on 28nm - which for an H2'14 launch means they will go up against 14nm parts from Intel.

Just from an intellectual starting point, even if ARM's v8 architecture is amazing, and AMD does a fantastic job integrating their new fabric - it's kind of hard to make up for 2 process nodes of difference.

But in terms of judging the power efficiency of ARM's A15 cores, and knowing that A57 includes only incremental improvements that are meant for 20nm in order to reach really good power efficiency ratings - and realizing that AMD's recent roadmap change is a dive-and-catch tactic to make up for the fact that their own cores are not very power efficient - most of us should be able to extrapolate and expect these new "Seattle" products to be a generation behind and fully uninteresting to micro-servers once Intel gets established with Avoton - let alone the 14nm compaction.

Like I said before, AMD can make themselves appear like they are on track for leadership when it comes to PowerPoint, but when it comes to actual implementation, AMD's record speaks for itself.
Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent INTC News