# I don't understand

1. Feb 6, 2012

### absurdist

How can someone studying theoretical physics for their undergrad degree get a phd in computational fluid dynamics? Does it still mean that they are studying theoretical or applied physics?

Also in what sense can computational physics help you understand more about the nature of the universe (how does it relate to theoretical physics)?

And finally why do people always say that science and math give us a model of the universe. I always though that what we call laws and theories are what govern the workings of the universe.

And finally whats the closest subject from theoretical physics that overlaps with chem.engineering or vice versa?

2. Feb 6, 2012

### Uniquebum

Computation helps you pin point an issue in your theory. Suppose you have a theory that just isn't quite working. You throw the model and numbers on a computer and you can clearly see the part where things start going wrong and what maybe needs to be thought again.

Science - other than math - is not perfect. It's all just approximation. As an example, let's look at simple diffusion.
$j = -D\frac{d\rho}{dx}$
Which tells us the rate of diffusion (j) based on diffusivity (D) and variation of concentration ($\frac{d\rho}{dx}$).

Now the important part here is the diffusivity. It's a "constant" and every system has it's own. However, question is where is this constant coming from, why is it there? If you'd look the system starting from the very atoms and create an equation for the system based on that, the D would disappear but even computationally this would be insanely hard, and in macro systems, just impossible. So, it's just a model on how the system works - approximately.

Anyway, just an answer from my part to the 2 middle questions.

3. Feb 6, 2012

### Snicker

1. Sure, it usually is not a big deal. I mean, Ed Witten was a history major as an undergrad!

2. Computational physics is a very important tool for understanding the universe. It helps you translate the data from experimental physics to something that may be useful to theoretical physics.

3. Mathematics is not magical number-crunching. It is the study of precise patterns. Natural science is built on the hypothesis that patterns exist in the universe. In physics, these patterns just happens to be more sublime. Also, we call our physical theories models and not laws because that is what science is -- a collection of models that has yet to be falsified. You can treat them as ultimate laws if you wish, but that is pseudo-religious.

4. Quantum mechanics.

4. Feb 6, 2012

### absurdist

Thanks both of you!

@Uniquebum From that perspective the contribution of science and math to the behavior of that system lies in that constant right? That constant is just putting a value based on experimental data into an equation for that system. Unless I am missing something, there is no qualitative understanding about the nature of the universe but rather for application purposes and it sounds more like engineering.
@Snicker : Ed Witten, MAN how did this man manage to end up getting a degree with the help of a nobel laureate.Right if he can do it ....hmmm
So are computational methods just a substitute for (exp.data) for comparisions b/w experimental and theoretical data?
I see how computational nat.science play a major role but it would never play a central role in getting a more qualitative understanding of our universe, I GUESS.

If science is based on hypotheses and is only a model of patterns in the universe yet to be falsified, then whenever we use say newtons laws we never say that we are 90% sure that the answer represents reality or something like that. Unless you treat these as ultimate laws or atleast until you reach the point of no disproof such as laws like the 2nd law of thermodynamics, how will we ever get close to getting a universal theory of everything to explain the existence of the cosmos?

I thought you would’ve gone with statistical mechs but I am not even sure if theres anything left to discover theoretically in it is there?

5. Feb 6, 2012

### mal4mac

... who just a happened to have a leading expert in General Relativity for a father.