Ask HN: Which areas of math are practical to programming/algorithms and why?

There are lots of practical uses. For example, the billion dollar mistake [1] is painfully obvious when viewed through the lens of type theory.

Software such as Coq can be used to write verified (provably correct) software and can be used by mathematicians for proving theorems.

[1] http://lambda-the-ultimate.org/node/3186

You said "Have you tried interviewing at game development companies"

These are two separate things. It's like saying "Most people can learn how to fill out their tax forms on the fly. You don't need anything special you can just read up what to do when it comes time to." and then you coming back and asking "have you ever applied for a job at H&R Block? You need to be well versed in the tax system to be a full time tax filer"

This is not at all what I was referring to.

The question at hand was "Which areas of math are practical to programming/algorithms and why?" Would you say that game development is the representation of the entire programming field? No it's a small portion of our profession. If you want to do that professionally round the clock then you need to prepare for what you will be doing in that field but that doesn't mean that if you want to get into programming or algorithmic thinking that you should learn game development. That's just one small element.

> Honestly I'm not sure if I'm be able to handle a modern graphics programming jobs unless I studied and practiced constantly for at least six months straight

What you have to ask yourself is what is it that I'm practicing for this job

You didn't learn the maths first and said "you know game development looks good let me go there" and sit down and start to learn how to program and think algorithmicly. You sat down and said "I need to learn everything that directly applies to this field" which in most cases (for rendering) is vectors, frustum view angles, Eular and maybe Polar coordinates, an understanding of how to use a quaternion, what a matrix is. The remaining, and much more important part in this equation, is understanding how to interface with OGL/DX/Vulkan. This is what a graphics person does best, not math.

It boils down to what our field is: understanding, in great depth, the systems of computation and how to interface with them in a structured, maintainable, terse, and fast manor. Math is a cart and programming pulls it to the goal post.

http://forums.xkcd.com/download/file.php?id=29015&sid=060c11...

But I have no others to go on to compare.

The basics: + - x /

Generally speaking, any vector or array can be thought of as a matrix (not that it is important).

Some problems and domains involved many operations on [multi dimensional] matrices. Where some math knowledge is advised.

Then there is "graph" both theory and practice. It's usually classified as math but has little to do with mathematics. I'd rather consider that as a unique separate domain with many practical and theoretical applications in other fields.

Forget about "crypto". Nothing of that is needed for programming. You'll use some algorithms but you'll never have to write or invent them.

In a theoretical sense. "Functional [programming]" is a branch of mathematics.

----

Overall. Math is seriously overrated and has nothing to do with programming.

If you're into programming. Maths are simply not required to be a programmer. Don't worry about that.

There are some domains (maps=graph, 2D/3D=geometry, variousstuff=matrices) in where programming regularly require some maths. The maths part is always a very specific subset that can usually be learnt alone.

If you're into abstract mathematics. There are some theoretical stuff that can be worked on which are very far from practical programming.