Moore's Law at 50

Discussion in 'The Pub' started by tjontheroad, Apr 20, 2015.

  1. tjontheroad

    tjontheroad Supporting Member

    Messages:
    5,983
    Joined:
    Oct 10, 2014
    Location:
    New Jersey... YOU GOTTA PROBLEM WITH THAT?!?!
  2. GAD

    GAD Wubbalubbadubdub Silver Supporting Member

    Messages:
    14,303
    Joined:
    May 22, 2008
    Location:
    NJ
  3. tjontheroad

    tjontheroad Supporting Member

    Messages:
    5,983
    Joined:
    Oct 10, 2014
    Location:
    New Jersey... YOU GOTTA PROBLEM WITH THAT?!?!
  4. GAD

    GAD Wubbalubbadubdub Silver Supporting Member

    Messages:
    14,303
    Joined:
    May 22, 2008
    Location:
    NJ
    It just seems that way in some parts on NJ. :)
     
  5. JWDubois

    JWDubois Member

    Messages:
    6,469
    Joined:
    Aug 16, 2009
    Location:
    East TN
  6. tjontheroad

    tjontheroad Supporting Member

    Messages:
    5,983
    Joined:
    Oct 10, 2014
    Location:
    New Jersey... YOU GOTTA PROBLEM WITH THAT?!?!
    That and extraneous junk running in the backround
     
  7. GAD

    GAD Wubbalubbadubdub Silver Supporting Member

    Messages:
    14,303
    Joined:
    May 22, 2008
    Location:
    NJ
    Software gets slower faster than hardware gets faster.
     
  8. Pietro

    Pietro 2-Voice Guitar Junkie and All-Around Awesome Guy

    Messages:
    16,410
    Joined:
    Feb 17, 2007
    Location:
    Maineville, OH
    I escaped from New Jersey...

    I don't remember any guy named "Moore"...

    Wait, I went to HS with a guy named Moore.

    He didn't smell funny... (maybe I didn't notice because I smell funny).

    [​IMG]
     
  9. fr8_trane

    fr8_trane Member

    Messages:
    6,234
    Joined:
    Jan 20, 2005
    Location:
    New Brunswick, NJ
    Moore's law has stagnated.... but ended? I dont think so.

    Moores law may be specifically tied to the # of transistors you can fit in a given space but ultimately its also about increased performance

    3D chips, quantum computers and replacing silicon with graphene and/or other meta materials all will put Moores Law right back on track in the near future.

    http://www.top500.org/blog/life-beyond-moores-law/

    http://spectrum.ieee.org/semiconductors/devices/the-status-of-moores-law-its-complicated

    http://www.cnet.com/news/moores-law-the-rule-that-really-matters-in-tech/

    "The most probable outcome is that silicon technology will find a way to keep scaling, some way continue to deliver more value with succeeding generations," said Nvidia Chief Scientist Bill Dally.
     
  10. RickC

    RickC Gold Supporting Member

    Messages:
    7,000
    Joined:
    Feb 17, 2005
    Location:
    Boston, MA
  11. GAD

    GAD Wubbalubbadubdub Silver Supporting Member

    Messages:
    14,303
    Joined:
    May 22, 2008
    Location:
    NJ

    Really, my problem with the whole Moore's Law thing is that multiprocessing is trumping number of transistors in a given space. Processors got faster and faster and faster, and not there's more of them and more of them while the speed of each doesn't increase much. Yes, we're getting faster, but for different reasons.
     
  12. SammyTickler

    SammyTickler Active Member

    Messages:
    46
    Joined:
    Apr 19, 2015
    I took Moore's Law to predict our increased knowledge/ability etc due to each evolution of computing power.
     
  13. RickC

    RickC Gold Supporting Member

    Messages:
    7,000
    Joined:
    Feb 17, 2005
    Location:
    Boston, MA
    Not sure I follow what you mean. Processors are still transistors. We're cramming more (Moore? :)) of them into a given space, but also becoming more clever in what we do with them. Multiprocessing is a good example of that. Moore's Law strictly speaking just talks about numbers of transistors/gates, but we like to extrapolate that idea onto things like performance/processing power, and you need the cleverness to make proportional increases there too.

    /rick
     
  14. Roark

    Roark Member

    Messages:
    8,657
    Joined:
    Oct 19, 2006
    Location:
    sasnaK
    S'mores Law at 50. I'm actually 55, but...

    [​IMG]
     
  15. tjontheroad

    tjontheroad Supporting Member

    Messages:
    5,983
    Joined:
    Oct 10, 2014
    Location:
    New Jersey... YOU GOTTA PROBLEM WITH THAT?!?!
    Now that's an "advancement" :)
     
  16. GAD

    GAD Wubbalubbadubdub Silver Supporting Member

    Messages:
    14,303
    Joined:
    May 22, 2008
    Location:
    NJ
    I actually recant part of my complaint. I had always thought that Moore was talking about the number of transistors per square inch, though it seems over the years many people (myself included) have injected their own thoughts over what he actually wrote, which was this:

    If, as in this Wikipedia article, the common meaning of Moore's Law is, "over the history of computing hardware, the number of transistors in a dense integrated circuit has doubled approximately every two years", then multi-core processors would be considered a single "dense integrated circuit".

    If we're talking about the number of transistors within a single square inch or centimeter, then we're pushing up against size limits where it will take a major breakthrough to shatter that barrier.
     
  17. Roark

    Roark Member

    Messages:
    8,657
    Joined:
    Oct 19, 2006
    Location:
    sasnaK

Share This Page