Subcultural engagement for the 21st Century...
Barbelith is a new kind of community (find out more)...
You can login or register.

Underground > Laboratory >

The Universe in a Box. A Computer Box.

Our Lady Has Left the Building

08:13 / 02.06.05

From The Guardian.

From New Scientist.

I haven't got much further than 'wow' yet. Newsnight had a very good piece on it last night, with the adventures of the head programmer having to rewrite the OS as it was doing the calculating because of the problems with the program needing more memory than was anticipated.

What does make me wonder is we have no real way of juding it real time against the universe for accuracy, what with the cosmic scale of time being so long and all. I'm also thinking about how they code things they don't understand yet, like the possible existence of Dark Matter maybe being the reason for the acceleration of the universe's expansion. If we don't understand parts of how the universe works, how important is that to the integrity of the program?

Next up, a quantum computer that creates Warren Ellis' snowflake...

09:24 / 02.06.05

I'm also thinking about how they code things they don't understand yet, like the possible existence of Dark Matter maybe being the reason for the acceleration of the universe's expansion.

Surely the code has a variety of parameters that they can change and see what effect it would have on the development of the universe. What they must be doing is encoding certain laws of physics and allowing each run of the program to have different input data. In fact, if they are sensible, the input data should also allow the tweaking of the laws of physics that they are simulating. From reading the articles, it seems that the way they are using it is to see if and how known objects form and generate reearch hypotheses from that.

09:26 / 02.06.05

“The British, German, US and Canadian astrophysicists in the Virgo consortium, led by Volker Springel of the Max Planck Institute in Germany, report in Nature today that they already knew the so-called "initial conditions" of the universe.
These were imprinted in the cosmic microwave background radiation, the embers of the Big Bang, when the universe was only 400,000 years old.“

I didn’t realise that we did know the ‘initial conditions’ of the universe.

When I read this I have to confess I thought it was a joke – or one of those faked-up stories. This quote didn’t help: “We let it churn away - in fact we shut down all science in Germany, we excluded all German science for a month while this very large machine ground away”.
There seems to be some kind of boot-strapping going on here, doesn't there? If we have all the information needed to run the simulation, do we need to run the simulation? Isn’t the simulation just telling us what we told it?

Don’t get me wrong, I love the idea – all very ‘Counterfeit World’, but I suppose I thought simulations like these would necessarily be very blunt tools that could represent what we know, but never really tell us anything new to any degree of certainty. But I’m a bit hard of understanding.

09:39 / 02.06.05

Isn't it that the dynamic evolution, if you will, of the universe subject to laws and initial conditions isn't at all clear. You can write down a partial differential equation, which tells you how things work, but it won't straightforward tell you how things develop. I think.

Darumesten's second variety

09:43 / 02.06.05

I love the idea, even if it can seem a bit pointless to try to run the simulation with so many data we don't know about the universe .. but even this can be useful, as the simulation results can point the diferences between the scientific model of universe ( the simulated one ) and well, the real universe .. don't you think ?

09:51 / 02.06.05

I suppose I just didn’t realise that we could model the universe anything like accurately enough. If we can do *this* with a computer, how come we’re still so rubbish at forecasting the weather?

09:58 / 02.06.05

Chaos theory, probably. In the universe simulation, I bet they are only interested in extremely coarse information. They aren't looking at where black holes actually are, they are looking at where black holes tend to form in a repeated run of a simulation based on plausible physical laws.

This is like predicting that the UK has a moderate climate, which is of course a useless piece of weather forecasting. You don't want to know that tornados are very uncommon in England, you want to know if it will rain on Wednesday. Different problem entirely.

alejandrodelloco

02:08 / 03.06.05

Ok, this totally reminds me of the very end of Neal Stephenson's essay, In the Beginning Was the Command Line. They need to release the code so people can make their own universe boxes. Then we will see some really cool "universe art".