much more sneerclub than techtakes

  • scruiser@awful.systems
    link
    fedilink
    English
    arrow-up
    5
    ·
    4 days ago

    even assuming sufficient computation power, storage space, and knowledge of physics and neurology

    but sufficiently detailed simulation is something we have no reason to think is impossible.

    So, I actually agree broadly with you in the abstract principle but I’ve increasingly come around to it being computationally intractable for various reasons. But even if functionalism is correct…

    • We don’t have the neurology knowledge to do a neural-level simulation, and it would be extremely computationally expensive to actually simulate all the neural features properly in full detail, well beyond the biggest super computers we have now and “moore’s law” (scare quotes deliberate) has been slowing down such that I don’t think we’ll get there.

    • A simulation from the physics level up is even more out of reach in terms of computational power required.

    As you say:

    I think there would be other, more efficient means well before we get to that point

    We really really don’t have the neuroscience/cognitive science to find a more efficient way. And it is possible all of the neural features really are that important to overall cognition, so you won’t be able to do it that much more “efficiently” in the first place…

    Lesswrong actually had someone argue that the brain is within an order or magnitude or two of the thermodynamic limit on computational efficiency: https://www.lesswrong.com/posts/xwBuoE9p8GE7RAuhd/brain-efficiency-much-more-than-you-wanted-to-know