• by mikewarot on 3/26/2025, 8:27:50 PM

    I'm an old guy who has an idea I'd like to see adopted before I go, the BitGrid.

    Von Neumann's architecture for computing greatly increased the possible complexity of computation, but at a huge hit to performance. I believe eliminating the memory compute divide is essential to getting exaFlops to the masses. Waiting for memory access is the root cause of much grief in high performance computing.

    Also, like knowing many languages helps programming, learning another computation architecture offers alternatives that might make other problems easier to solve.

  • by fuzzfactor on 3/26/2025, 7:54:44 PM

    >If you are solving a real problem, you would have already helped someone who needs your solution,instead of being at HN.

    OTOH, if you are capable of solving more than one problem, you have probably already helped numerous people in addition to taking a regular look at HN as it evolves.