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

Limits on technology

  1. Jul 15, 2009 #1
    According to Ray Kurzweil, human intelligence (both biological and non-biological) will saturate the universe and cause it to "wake up" sometime within a few centuries. Some have also speculated that exponential technological progress could lead to the development of sub-Planckian "ontotechnology" which can bypass or even change the laws of physics.

    Do you think that there are fundamental limits on technological growth? Do you think that any limits that we see are just the result of our limited intelligence? I mean, after all, once upon a time, we thought it was impossible to go into space.
  2. jcsd
  3. Jul 22, 2009 #2
    Perhaps we didn't think it was impossible, but rather vastly intractable. A person should be able to visualize a long enough ladder that would take one to space, which would be physically consistent. Such a thing was not ruled out by what they knew and by what we know about physics.

    We know however that there is a rigid and proven limit on an object's ability to accelerate to a certain speed. Knowing this we can confidently say that doing so is impossible. There are usually alternative approaches to solving our problems so this is probably not a big setback in the grand scheme of things, but nevertheless an evident impossibility.

    More likely than not there are ultimate limits, but as far as I can tell not limiting enough to stop us from achieving mostly everything we wish to achieve.
  4. Jul 22, 2009 #3


    User Avatar
    Gold Member

    Well, it can't bypass the laws of physics, since whatever it did would still be within the bounds of physics... But I can see it changing existing laws.
  5. Jul 23, 2009 #4
    Right, so couldn't sufficiently advanced technology, say, tweak some parameters in order to stop the expansion of the universe, allow for spontaneous creation of energy, or enable wormholes or FTL travel?
  6. Jul 23, 2009 #5


    User Avatar
    Gold Member

    In a word: sure.
  7. Aug 12, 2009 #6
    As T.H. White once wrote, "That which is not forbidden, is mandatory!"

    Unless there's a physical law that makes it impossible, we can potentially achieve it technologically.
  8. Aug 12, 2009 #7


    User Avatar

    Staff: Mentor

    What about the laws we just don't know about yet?
  9. Oct 6, 2009 #8
    I think any progression of this type would still be considered a discovery. We would developer a way of manipulating principles we discovered were already there.
  10. Nov 8, 2009 #9
    Seems like it would be a lot of work to overcome deltaX*deltaP >= hbar/2

  11. Nov 19, 2009 #10
    Scientific models are incomplete. There is always room to add or change things.
  12. Dec 6, 2009 #11
    Can you elaborate on this? What exactly do you mean?
  13. Dec 7, 2009 #12
    To what extent is the exponential growth of technological progress due to the increased ease of communication and collaboration by groups of professionals. The advancements based on the collaboration of many is far greater then the sum of the individuals.

    Is there a limit whereas additional collaborators begin to add less then the sum of their individual labor(necessitating that the management and communication of the group is unbounded).

    The management and operation of groups of 'experts' seems to be the key to continuing the exponential growth of technology.

    On a side note, check out Moore's law. It states that the computing power(or transistor size) will double every two years, but it's likely to end once transistor sizes hit the minimum theoretical size.
  14. Dec 9, 2009 #13
    I don't believe it states either of those things. If I recall correctly, it quite simply suggests that the number of transistors in an integrated circuit grows exponentially. The root of annual growth is merely speculative. Initially Moore' said it was 2, then sqrt(2), then some people decided to average the two predictions to somewhere in the middle (~1.6), or a doubling every 1.5 years. The last of these appears closest to the truth, although looking at the last four generations of Nvidia GPUs, for example, transistor count has been doubling every consecutive year. If TSMC has it's way, next year's 28nm GPUs will once again double transistor count making that 5 years straight. If transistor count has in fact been doubling every two years historically, then this rate sped up and would suggest it is not fixed afterall.

    Far from it. Circuits today are still planar. Once they hit their limit of miniaturization, they will expand in three dimensions. Perhaps stacked planar circuits at first and later natively built in 3 dimensions. Theoretically, a modern die would be able to incorporate a million layers of transistors without significantly increasing in size. This is because the vast majority of it today, in thickness, is just empty silicon. At today's rate I can easily foresee at least another three decades, increasing performance a billion-fold.
  15. Dec 10, 2009 #14
    I think heat is what currently prevents stacking, won't they have to modify the materials substantially, or design a more efficient cooling method?
  16. Dec 10, 2009 #15
    As a result of power laws, with respect to voltage and frequency, a processor would be able to operate at half-clock while using 12.5% of the energy. Double the processors and you can maintain the same performance while using a quarter of the power. Repeat step as necessary.

    Future designs could very well operate at current frequencies and be comfortably stackable.
    Last edited: Dec 10, 2009
  17. Dec 29, 2009 #16
    I think Ray Kurzweil is overly optimistic. Fast computers allow one to search a few layers deeper down a large tree not infinitely far down the tree. Some times experimental results are required (and faster) than "pure thought". Growing bacteria will be just as slow a process for a super computer as it is for a human. Fast chips will be useful tools but they will not allow us to transcend the physical plane and become pure spirits of energy with infinite knowledge, intelligence and power to manipulate the physical plane. No Daniel Jackson.
  18. Dec 29, 2009 #17
    Human have 10^15 synapses with a cycle time of 100-1000Hz so 10^18 synapses per second. In a package of under two liters with a power under 50 watts. I would say the physical limits of the world allow huge advances in human computer technology before we hit any fundamental limits. Your PC is currently processing about 10^11 bits per second so about seven orders of magnitude to go if a synapse only stores one bit of information (which I think is too low a value to use).
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook