Truly personal computation
Bill Gates wanted to put a desktop computer into every home
+1
–
–
I want to do some thing similar: I want everyone to use their computers to their full potential.
I want my CPU to be at 100% CPU trying to find solutions to my problems.
想法 Using the CPU efficiently
This idea is ideas on how to use the CPU efficiently, to use CPU to solve problems and use the resource what it is capable of
項目 Parallel multithreaded actors without locks
Implement efficient scalable multiple thread communication
+[想法]
+[項目]
是的，這些天似乎比好主意如何使用具有更多的計算能力。 1969年的夢想家知道他們將使用額外的計算能力是什麼：他們將嘗試解決哪些物理方程式，他們將要運行哪些模擬。 1999年的人們擁有SETI @ HOME。但是，2020年的人們在某種程度上可以充分利用計算機的不足，從而保持計算機的性能。
所以，是的，讓CPU一直在做有用的事情會很好。那會是什麼呢？
Yeah, there seem to be more compute power these days than good ideas how to use it. The dreamers of 1969 knew what would they have used the extra compute power: what physics equations would they try to solve, what simulations would they have wanted to run. The people of 1999 had SETI@HOME. The people of 2020, however are somehow OK with that under-use of compute, keeping computers cool.
So, yeah, it would be nice to have the CPUs doing something useful all the time. What could that be?
[+]
中央計劃經濟。
它是如此複雜，它涉及無限的迴歸。所以你需要一個爬山算法，你需要避免局部最優並找到全局最優。
如果我這樣做，你就這樣做。但是你會對我做出這樣的反應來回應那個。
經濟很複雜，人也很複雜。如果我們可以捕獲一些信息並對其進行計算會怎樣。該模型必須與現實相匹配，我懷疑找到一個模型會很困難，它可能是神經網絡的解決方案——找到經濟運作的函數。
Centrally planned economy.
It's so complicated and it involves infinite regress. So you would need a hill climbing algorithm and you need to avoid local optimum and find global Optimums.
You do this if I do that. But you would react to me doing this in reaction to that.
Economics is complicated and people are complicated. What if we could capture some information and compute on it. The model must match reality and I suspect finding a model would be difficult and it might be a solution for neural networks - finding the function over which the economy works.