1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

100 GHz Transistor

Discussion in 'General Hardware' started by Pehla, Feb 18, 2014.

  1. Pehla

    Pehla

    Joined:
    Mar 29, 2012
    Messages:
    349 (0.35/day)
    Thanks Received:
    72
    Location:
    brčko dc/bosnia and herzegovina
  2. micropage7

    micropage7

    Joined:
    Mar 26, 2010
    Messages:
    6,144 (3.55/day)
    Thanks Received:
    1,429
    Location:
    Jakarta, Indonesia
    the title is just when??
    please make clear the title not just when
    http://www.techpowerup.com/forums/threads/forum-guidelines.197329/

    personally i would like to see better performance per watt, higher Ghz not always means better performance
     
    GreiverBlade and the54thvoid say thanks.
  3. Drone

    Drone

    Joined:
    Sep 1, 2010
    Messages:
    2,910 (1.85/day)
    Thanks Received:
    1,684
    Silicon-germanium (SiGe) transistor runs @ 800 GHz. Graphene is overhyped
     
  4. Aquinus

    Aquinus Resident Wat-man

    Joined:
    Jan 28, 2012
    Messages:
    6,851 (6.48/day)
    Thanks Received:
    2,459
    Location:
    Concord, NH
    Good luck creating a circuit with low enough clock skew to drive an IC at speeds that high. The higher the frequency that more sensitive the circuit is to clock skew. It's not a matter of generating these high frequency clocks, it's a matter of being able to use them. Also, higher frequency signals attenuate faster than lower frequency signals, so size of the IC is a factor as well.
     
    GreiverBlade says thanks.
  5. Solaris17

    Solaris17 Creator Solaris Utility DVD

    Joined:
    Aug 16, 2005
    Messages:
    17,413 (5.10/day)
    Thanks Received:
    3,714
    Location:
    Florida
    Its easy to do with one transistor its a diffirent story with billions close together

    I mean alot first off while we probably have the tech to do it we will need to study the effects of electron migration heat output voltage bleed drive strength just to name a few.

    Basically you would never take even an old arc like core 2 and directly translate this with it. it would not be the same CPU
     
  6. Blue-Knight

    Blue-Knight

    Joined:
    Dec 9, 2013
    Messages:
    474 (1.26/day)
    Thanks Received:
    101
    Location:
    -22.809962, -43.018649
    How old are you?
     
  7. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    903 (1.88/day)
    Thanks Received:
    358
    Location:
    Cumberland Plateau
  8. Blue-Knight

    Blue-Knight

    Joined:
    Dec 9, 2013
    Messages:
    474 (1.26/day)
    Thanks Received:
    101
    Location:
    -22.809962, -43.018649
    And they can put fake information as well...
     
  9. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    903 (1.88/day)
    Thanks Received:
    358
    Location:
    Cumberland Plateau
    So you think he would lie in his profile and then decide to tell you the truth? LOL
     
  10. Blue-Knight

    Blue-Knight

    Joined:
    Dec 9, 2013
    Messages:
    474 (1.26/day)
    Thanks Received:
    101
    Location:
    -22.809962, -43.018649
    Why not? I would.
     
  11. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    903 (1.88/day)
    Thanks Received:
    358
    Location:
    Cumberland Plateau
    Yes, you are good at being illogical.
     
    Blue-Knight says thanks.
  12. BiggieShady

    BiggieShady

    Joined:
    Feb 8, 2012
    Messages:
    1,041 (1.00/day)
    Thanks Received:
    360
    Location:
    Zagreb, Croatia
    [​IMG]
     
    Bansaku, Aquinus, FX-GMC and 1 other person say thanks.
  13. lilhasselhoffer

    lilhasselhoffer

    Joined:
    Apr 2, 2011
    Messages:
    1,678 (1.24/day)
    Thanks Received:
    1,039
    Location:
    East Coast, USA
    This is a pointless.

    One transistor operating at 100 GHz, means absolutely nothing. Demonstrate a network of 10 transistors that can consistently perform any sort of operation and I'll bite. As it stands, this tech is a decade or more away from being incorporated into anything consumer related. It's nice to see this done, but by the time this actually gets out of the lab and into something relevant the CPU will likely no longer exist. SOCs will be the norm, and our devices won't need separate GPUs, CPUs, and other components. Silicon is dead, long live Carbon? Or perhaps, the CPU is dead, long live the CPU.
     
  14. theoneandonlymrk

    theoneandonlymrk

    Joined:
    Mar 10, 2010
    Messages:
    3,414 (1.96/day)
    Thanks Received:
    573
    Location:
    Manchester uk
    10 Million points folded for TPU
  15. yogurt_21

    yogurt_21

    Joined:
    Feb 18, 2006
    Messages:
    4,452 (1.38/day)
    Thanks Received:
    598
    Location:
    AZ
    article was written 4 years ago. If it was meant to be the death of the silicone based computer it failed miserably.
     
  16. Scheich

    Joined:
    Dec 20, 2005
    Messages:
    245 (0.07/day)
    Thanks Received:
    20
    Its not their fault, intel is still selling their ancient x86 stuff and will prob for the next 200 years
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page