Moore's Law
From Simulism
On several pages on this wiki, Moore's Law was mentioned at some point. We therefore decided to dedicate a page to this subject.
Contents |
[edit] About Moore's Law
[edit] Doubling processing power
Moore's Law is a well-known law. It states that every 2 years, the processing power of a computer approximately doubles (wikipedia). So far, Moore's Law has been surprisingly accurate. Every now and then the law is challenged by 'limits of technology', for example "a computer can't possible become faster than x because we can't make smaller chips than y". So far, all these challenges were overcome by new technologies, and Moore's Law still applies.
Interesting point: Moore made his statement in 1965 and he stated that it would probably be true 'for the next 10 years' but was a bit uncertain of the years after that; it's now 2006, the law still applies.
[edit] More than double
Computing power can be measured in many ways. You can take the amount of computing power a single pc has, or the total computing power in the world, etc. Moore 'normalized' his statement in terms of pricing. He argued that computing power for 'minimum component costs' would double. Roughly this means that for the same price, in 2 years you can buy a computer with twice the processing power or, in 2 years the same processing power will cost you half.
But if you look at the total computing power in the world, we are more than doubling. First of all, we are connecting computers together into clusters. 10 clustered machines have more computing power than each of them individually.
The second reason we are more than doubling is that the size of computers is becoming smaller and smaller, which makes it possible to add embedded computers to almost anything. Phones, tv's, dvdplayers, games, watches, dishwashers, almost anything has a microprocessor these days. So probably in the future the total amount of available computing power will increase faster than Moore's Law predicted.
[edit] How much processing power would we need to simulate a human, earth or the universe?
Once we know how much computing power is needed, we can more or less calculate at which point in time we will be able to run sufficiently realistic simulations. As one data point, supposedly the known universe contains about 10^80 elementary particles. (todo: this section needs to have some numbers and predictions.)
[edit] Moore's Law and Simulism
If computing power continues to increase according to Moore's Law, then 'The Simulation Argument' will eventually be proven correct. It's only a matter of time, and because of the 'doubling' every 2 years, we will see it coming closer exponentially. The reasoning here is simple. Right now, we are able to simulate something simple. In 2 years, we will be able to simulate something twice as complex. In 20 years, we will be able to simulate with 2*2*2*2*2*2*2*2*2*2 = 1024 times as much complexity as today. 40 years will bring us 2 to the power of 20 = 1 million times more power than we have today.
Let's focus on the human body. Suppose we are able to simulate a complete human being in, say year t. How long would it then take before we are able to simulate the world population of approximately 6.000.000.000 people? Simple equasion: 2^x = 6 billion, gives us something between 32 and 33 for x.
This means that from the time we will be able to simulate one person, to the time we will be able to simulate the entire population, will take only 33 times 2 years = 66 years.
[edit] Moore's Law and Stacking of Simulations
Some are of the opinion that when inside a simulation the inhabitants become capable of running simulations themselves (Stacking), the simulation could be terminated because it would require too much processing power. Moore's Law however is an interesting counterargument to this theory. Suppose for a moment that we would be living in a simulation (that's what this site is all about after all). It took as millions of years before we developed the technology to run simulations. When our simulation started, the person or people running our simulation already had that capability. This means that by now (assuming for a moment that time in our simulation and on the 'outside' runs at the same speed) they have billions and billions more computing power. That is why I think that stacking will not be a problem, as long as Moore's Law applies both within and outside of the simulation.
