Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Diamond Computers?

  1. Oct 20, 2003 #1
    I see in an article on computers that because of faster operating
    speeds and more transistors per chip, the heat generated will be
    too much for silicon processors to handle thus a need for something
    that will and they believe man-made diamond based computer chips
    are the answer. Is this the future of computers? Faster and faster
    computers with more storage space and faster methods of downloading
    combined with smaller and smaller boxes to hold the hardware? Or
    is another form of computing ready to make everything obsolete?
    What are your thoughts?
  2. jcsd
  3. Oct 30, 2003 #2

    I think the new super computers will be a form of light manipulation with alternating colors
    for the binary code. data will be stored on hard drives the difference will be bus. speeds as fast as the processer! imagine that.
  4. Nov 7, 2003 #3
  5. Nov 7, 2003 #4


    User Avatar


    The future of creation is widely agreed to be a combination of nano-tech, molecular electronics and quantum computing.

    Diamond based processors are likely only to be a stop-gap measure.
  6. Nov 10, 2003 #5
    I read the samething in a book I'm reading. It states that man made diamond films are being created as semiconductors to replace silicon semiconductors.

    I think that with the implementation of diamond film semiconductors, we might see systems using fiber-optics to get information from the storage medium (whatever that may be in the future) to the RAM and then to the processor and back to the RAM at the speed of light, so BUS speed won't be an issue to worry about anymore, all you would have to do then is catch up the rest of the PC to those speed (if they haven't already reached those speeds).
  7. Dec 7, 2003 #6
    This is really cool

    I think I read something in USA today about HP, who made an awesome silicon chip that had switches based on the silicon particles, whether or not they were left and right handed. Does that make sense to someone smarter than I? Also, are diamonds easier to dissipate heat or something? What about conductivity? Anyways, back on to HP, they have made these chips that can transfer info extremely fast, and they are infinitely small, so the switches turn off and on due to their right/left handedness. The handedness i think they were speaking about is the spin, if I am not mistaken. If I am, plz correct me. Has anyone else read this article, or heard anything about it? I find it fascination.
  8. Dec 17, 2003 #7
    I totally agree, the nano-tech will be super fast because it would be a perfect machine, built atom by atom. And hopefully in the near future they will be able to stabilize quantum computing, it will change the world.
  9. Dec 23, 2003 #8
    i thought that dimonds cant be
    melted only cut to shape
    its the hardest substance in the world isnt it?
  10. Jan 5, 2004 #9
    I believe that quantum computing, plus the actual manipulation of light and the miniturisation of computers(actually the 2 I just mentioned leads to this) will be the next step.

    About diamonds, I'm not to sure about. But, if I'm not wrong, physicists recently were able to stop light in it's tracks using a ruby. Though this kind of manipulation is in it's infancy, I think it has great potential by being able transfer information at incredible speeds. I guess the same could be said about diamonds...
  11. Jan 5, 2004 #10
    Yes, another is ready to make them obsolete. No heat.
  12. Jan 27, 2004 #11
    Re: diamonds??????????

    Correct me if im wrong, but is this referring to something along the lines of a hologram? Like a holographic hard drive?
  13. Jan 29, 2004 #12
    diamond why not almonds

    i think that diamond thing is really about synthetic diamond makers and diamond miners trying to sell more diamonds. why not make a computer using something cheap. maybe frozen almonds
  14. Jan 29, 2004 #13


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    Re: diamond why not almonds

    Or what's left over when they make Chicken McNuggets.
  15. Feb 5, 2004 #14
    This is a great article about the diamond age :

  16. Mar 8, 2004 #15
    I think proteomics will create computers to a very excellent specification. You see. If proteomics is refined to a complete formula. With no mystery. The ideas of Superconductivity will be improved and perfected as well. Since medicine and Superconductivity are engineered together.

    So. Proteomics will usher in Superconductivity at some unknown point, I think.

    So. We will be able to understand Quantum effects from the Superconductor theory that proteomics will build. And a T.o.E. will be better understood. Maybe even perfected.

    So. The Quantum science will be very strong, Superconductivity will exist, and proteomics will be understood.

    What kind of A.I will these sciences build ?
    And what kind of computer hardware will be possible ?

    All I know is those chicken mcnuggets sound pretty nice right about now.
  17. Mar 17, 2004 #16
    Hi everyone,

    I was wondering about one thing. If diamond is going to replace silicon (even if it's for a short period), how are they going to make diamond chips. I know that with silicon they use something like ultra violet rays. Is that also possible with manmade diamonds, or are they going to use some other technique, or are the chips getting bigger again.


    BTW. This is my first post, and I must say, I really like this site :smile:
  18. Mar 24, 2004 #17
    This is how they make the diamond chips

    they grow the diamonds from carbon, but during the growth, they leave grooves that the wires can later be printed into.
  19. Apr 2, 2004 #18
    Long time lurker, first time poster. A few thoughts.

    The key problem currently inhibiting diamond semiconductors is the inability to grow large single crystals. Silicon can be grown into large (300 mm) diameter cylinders that are up to 2 meters long. These cylinders are enormous single crystals with very few defects that are then cut with a laser into thin wafers. These wafers are then used to make microchips.

    The key is single crystal (the more or less infinite repetition of the atomic layout without distortion). My analogy for single crystal versus polycrystal is a city with the streets are laid out in one big grid pattern versus a city with lots of little grids joining at odd angles. Those awkward intersections in the latter really jam up the traffic, right?

    If you were to slice wafers out of something that is polycrystalline, the eventual device usually doesn't work. That's because electrons flow quite nicely in regular single crystals, but when they hit the boundary between two crystals they crash and burn because the periodic regularity of the crystal is broken. These electron crashes become heat.

    As the Wired article notes, there are people trying to grow large single crystals of diamond. The jewelry market is an obvious first step because a relatively small single crystal of diamond is quite valuable. You need to grow diamond crystals at least 25 mm big before you can start to slice them into wafers to make microchips. So why not sell metastable (NOT forever!)carbon to starry-eyed lovers in the meantime? :tongue:

    Once you have the crystal, the rest of the manufacturing process is not radically different than silicon. Any changes are reasonable, predictable, and surmountable. The hardness of diamond isn't too much of an issue because you don't really have to remove the diamond to make a microchip. Like silicon, the insulation and metal layers lie on top of the semiconductor, not in it, so no channel cutting is required for the wiring.
  20. Apr 26, 2004 #19
    the spin, thats it, youre talking about the magnetic moment or spin orientation hopefully used to logic elements. yes the theory works, but its probably a decade or more away from reality.

    using diamond or sapphire is a logical step in silicon devices, but todays PCs are facing the speed of electricity as a bottleneck, how long does it take a signal to pass one foot of a conductor? the thicker the better; but as the feature sizes of silicon shrink, the interconnect lines get too thin and their resistance increases. Without adding copper to the process, I dont believe older CHMOS processes could make >1 Ghz processors.

    Disk drives are stuck at about the speeds they have today, access time in the industry hasnt changed much since the 1970s, density has skyrocked, rpms are faster, but they wont speed up much that Im aware of;

    if that kind of solid state physics is fun, please join the IEEE Electron Devices society or read their publications.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook