My dad and my physics teacher have (independently) advised me, as a CS major, to minor or double-major, in physics.
Do you think this will really help me get ahead? I want to end up doing something with computer/video games, not writing scientific software, so I think basic general college physics (required for CS major already) is enough.
Theory is not required for programming, that's what I meant. Because abstract concepts does not apply to the real world.
OP said he would like something to do with computer games, so he doesn't need to know advanced Physics theory as most doesn't apply to real-world situations because of over complex variables, he just need the basics and advanced math.
Name:
Anonymous2007-05-24 2:31 ID:2uqYqMTn
>>43
Oh. Yeah, knowing the difference between a boson and a fermion is not going to be all that useful to a game programmer. I agree.
However, many common devices and natural phenomena are fundamentally based in quantum mechanics, esp. anything involving optics or semiconductors. Read Feynman's QED, he talks about this stuff in detail without hardly employing any math at all. Very awesome book.
Basically I don't understand how you can say advanced physics does not apply to the real world, because that is against the very definition of physics! TaoZenElegantWuLiUniverse "physics" notwithstanding.
Any attempt to model real life situations with Physics results in over complicated methods and formulas which no average students would hope to understand because of mere factors.
Advanced Physical theorem is explained in an "assumed" state where multiple influenced is eliminated.
Example? We would never see the blue shift with normal objects since they cannot travel near the speed of light, even though relativity applies to them, we cannot notice a change.
What I mean by Physics do not apply to the real world is because it is merely an explaination of the natural universe according to our point of view.
where multiple influenced is eliminated. What I mean by Physics do not apply to the real world is because it is merely an explaination of the natural universe according to our point of view.
please to learn the english language.
normal objects ... cannot travel near the speed of light
wrong.
Name:
Anonymous2007-05-24 10:48 ID:2uqYqMTn
>>45
Those "over complicated methods" exist because they work. Transistors, lasers, MRI, solar panels, electron microscopes etc... try designing any of those without knowledge of modern physics!
Feynman has a few examples of "everyday" phenomenon in the book I mention. The one that I remember most is the diffraction grating. It cannot be explained without quantum mechanics, and microscopic diffraction-grating-like structures are responsible for the reflective properties of many everyday objects. Loads of stuff about glass and lenses, too.
Relativity is not as obviously useful as other things - it is an important correction to Newtonian mechanics that is essential for getting the right numbers in some cases. Although pime taradoxes are fun too.
You worry about real life situations being too complex to model mathematically because of many interactions happening simultaneously. In that case I recommend you study thermodynamics, where you learn to build up sophisticated statistical models from basic Newtonian mechanics. If you are instead complaining that physicists tend to disregard realistic factors sometimes to narrow the discussion... well, if they didn't do that nothing could get figured out! Fields of engineering are spawned out of advancements in physics, and in those fields people can concern themselves with how to make the models most practically useful. Physicists are mostly only concerned that they are correct.
Name:
Anonymous2007-05-24 12:30 ID:AnkFewkX
>>1
It'll probably be a waste of time. You've wasted enough time already at uni: unless you went to a really good one, like MIT, chances are that what they told you was not spectacular, not deep, not bleeding edge, and not even SICP. As a CS graduate, I advise you to study what you need for yourself. I recommend:
1. Read SICP. Then deal with languages, grammar and automata. Play with the stuff.
2. Be sure you have the right tools for the right job. This means C (standard base, low-level stuff, lots of open sauce), either Python, Ruby, Lisp or Haskell (for actual work, to get things done), and some shell or Perl scripting (for everyday stuff).
3. Lunix. Hack your way with it, it's worth knowing your OS.
4. Porn. Lots. You'll need this.
Name:
Anonymous2007-05-24 19:55 ID:2uqYqMTn
>>48
Standard American computer science curriculum is fucking depressing. You have some good advice, but the average student will reject it because all they want to do is learn PHP/Java/C# to get a dead-end job writing crappy web code. (or gamez lol) They can't entertain abstract concepts and view their professors as spacey incompetents who simply couldn't hack it in The Industry. So their professors simply don't give a fuck about them.
Name:
Anonymous2007-05-25 5:43 ID:5Y8v7YKp
>>48
That was my way to say I've read SICP, though.
Name:
Anonymous2007-05-25 5:53 ID:Q9GrRFe8
Implementing physics in programs is fuck easy. Let me illustrate:
Each object has an x, y, and z coordinate. Per frame it exists in a particular x,y,z coordinate. To model forces acting on that object, we need two attributes for each coordinate: current velocity (v), and change in velocity per frame aka acceleration (a).
For each frame, using x as an example, do this:
x.v = x.v + x.a
x = x + x.v
When forces act on an object, you have to modify a and v accordingly. Velocites are reduced by division, and directions are reversed by a sign change. So, say object A and B collide, you can model that like this:
A.[coordinate].v = - (A.[coordinate].v / 1.5)
and same for B
You might want to divide by a factor dependent on the objects weight or mass, so you could define that as well, a w or whatever.
Gravity is easy too. Just do something like this per frame
[object].y.v = [object].y.v + .05
for .05 substitue a weight factored value or other value that works depending on the types of objects
I missed a word and mis-spelt one, whoopy fuckin do.
>>47
Whilst I agree with everything that you say, my main point is that advanced Physics isn't in play at the moment due to limitation of processing power. But maybe soon, these concepts can be implemented. Classic Physics is "realistic" enough to be used in current games without a performance hit.
I'm directing the argument towards basis of whether these concepts are useful for games.
>>51
Another OpenGL noob... go back and read your tutorials.
Name:
Anonymous2007-05-25 11:26 ID:G9ZMuSVa
>>53
It ain't got shit to do with how the shit is rendered. I'm just talking about object interaction, shitface.
I missed a word and mis-spelt one, whoopy fuckin do.
i'm wondering what word it is that you missed that could fix What I mean by Physics do not apply to the real world is because it is merely an explaination of the natural universe according to our point of view.
that sentence made my brain have to switch to quirks mode in the middle of parsing your post. you wasted about 30 milliseconds of my life. that is unforgivable.
you also claimed that normal objects cannot travel near the speed of light.
Name:
Anonymous2007-05-26 1:22 ID:l9D/pwpD
I currently doing a physics/comp sci double major in college and have found so far that recruiters of all sorts are very impressed when you tell them what you're studying.
Name:
Anonymous2007-05-26 2:07 ID:KuEnsUVz
If you want to get a head, surely biology would be more suitable. unless you want like a robot head.
sure, if you want to do physics programming for games. but if not, no sense in specializing in it.
Name:
Anonymous2008-11-01 0:14
A post got deleted on /prog/ ? I don't habeeb it
What could it have possibly been?
Name:
mario developer2008-11-01 11:14
u need a physics degree and a maths degree in order to program mario brothers. yes, mario brothers is _that_ complicated. very advanced physics engine.. i mean think when marrio hits his head on a block ... thta takes a lot of processing power, most likely prrocessing power harnessed by using haskell.
so, to sum it all up, learn haskell and opengl in haskell and _remember_, the goal of gfx is to minimize the # of triangles that u draw to the screen. if youre drawin 5000 triangles, then thats not good.... you need to draw less.. thats why 2d games are so fast, but 3d games are slow...
Name:
Anonymous2008-11-01 13:38
listen to >>70, drawing 3d graphics is basically a minimization problem of triangles. haskell really helps here because it can perform the computations in paralel automatically
Name:
Anonymous2008-11-01 16:24
but what if I want to use squares instead.
Name:
Anonymous2008-11-01 17:05
>>70 i mean think when marrio hits his head on a block ... thta takes a lot of processing power, most likely prrocessing power harnessed by using haskell.