Is it even worth going to uni or college for game development...

is it even worth going to uni or college for game development? what skills do you learn there that you can't get elsewhere?

Other urls found in this thread:

pixiv.net/member_illust.php?mode=medium&illust_id=60897270
twitter.com/NSFWRedditVideo

how to be poor

Hell no

Go to college to learn real shit, learn game development on the side. There are free resources for it.

To be fair these skills will be used in your career as a game dev.

like general programming?

>career
>game dev

Shut up and post more giantess

>is it even worth going to uni or college for game development?
no
>what skills do you learn there that you can't get elsewhere?
literally none

...

Does anyone else feel sad that you'll never be able to fuck a sexy 2D girl ever in your life?
I know I do

BIGGER

anything is possible through science
hey theres a reason to go to college
i saw a bigger one once but it was 2big to save

Become a marketer for a game OP, youll earn more money and say that you had a hand in developing the game.

Degrees and formal education help but most employees look for experience and portfolio work. Especially in art.

Whatever you do, pick something you want to do and specialise. 'Game development' courses are complete horse that'll waste your time with entry level stuff you can learn off the internet in a weekend.

On that note, self-teaching is a very good thing to do. I think too many courses stretch stuff out for months and teach you bad habits in time management but that's just me.

scratch that

I'm a Comp Sci junior here taking a Game Dev class as a major elective class. Do that, don't go solely for game dev. Comp Sci will make it easier

Networking
Remember, firm handshake (unless is a faggot company)

...

So what's the deal with the mega mermaid

You don't meed to go to college for game dev, go for Compu Sci. Also, you should really only make games if you have a passion for computing. If not, then follow Game Maker or Unity 5 Tutorials for a few months and start small. You'll learn really useful things there.

If you want to make games go to college for computer programming and mathematics.

Rent or steal a book on photshop and maya/3ds max

pixiv.net/member_illust.php?mode=medium&illust_id=60897270

my problem is i try to juggle learning music, art, and programming to the point of burning out real fast. school's probably way worse than that cause you've got no way of dropping a course (unless you like throwing money away), which i think makes you hate the shit you wanted to learn in the first place.
once you attach a mark to something, it becomes work instead of a hobby

>not using game maker softwares
>not using tutorials
>not cranking out series for steams
>not using kickstarter
It's like you don't want to make moneys in games

>game development

Not a real degree, even at a semi-respectable university.

The degree game developers really look for is software or computational engineering. Computer science could work if you build a good portfolio by doing co-op or developing freeware.

whats the difference between software engi/ computational engi/ computer science?

the only purpose of going to a game dev school is to attempt to 'kick start' your drive/ambition. you're forced to work on projects, learn things, and fail.

here's the thing, to get a job in game dev you need to be a self-starter. you need to be always working on little projects, building things, which is the only way to learn. those people get jobs without any formal education credits.

everyone else is the type of person who says they have so many great ideas, and maybe some of them get put on paper every now and then. these people will get into level editors, unity, unreal, or whatever from time to time and work on something for a bit and then abandon it. 99% of those people will never get into game dev because they don't have the drive. the remaining 1% do have potential that could possibly be unlocked by going into a structured environment where they're forced to work on things and essentially 'learn' how to sit down and just work on stuff.

"Software engineer" is a fancy name for programmer

Computational engineering is a fairly new field that focuses on developing simulations and artificial intelligence. Upside to doing this degree is that it is HIGHLY employable in almost any field (defense, chemistry, medicine etc) so if there is a downturn in gaming you can find work anywhere else. Downside to this degree is that it's a very competitive program in most schools and you have to either work really hard or be really smart to get in, much less graduate from this program.

Software engineering is pretty self-explanatory, you develop software or specialize in certain aspects of software like U.I or programming logic gates. You normally have to pass first year engineering to get into software engineering.

Computer science is an all-encompassing field that ranges from coding, to security, to server maintenance. Easiest to get into and you can apply straight into it from highschool.

Not all post secondary institutions offer these programs and sometimes they don't follow the above criteria in program and course content. Do your research.

Computer science is straightforward programming pretty much.
Officially, software engineer is a rung higher than your typical programmer.
The software engineer would at times manage them and are in general, responsible for more engineering-type activities like making the design documents and planning.
However, as the industry has evolved, the role of software engineers have become less defined and significant.
Programmers make a lot of the documentation and do a lot of the testing now, subsequently they changed the job name to Software Engineer.
t. nearly complete BS in software engineering
Basically this.

>computational engi

is this another name for data scientist

I don't know, what's a data scientist. All engineers and scientists deal with data in an analytical sense.

programming, yes

design, no, all they teach you is how to be so racist and obnoxious that you can't even hold a job at bioware

I didn't even know game design was even a real degree. Sounds like some DeVry level shit.

You can learn any skill you want without going to college, but it will probably be harder to get a job that applies that skill if you don't have the little piece of paper from a school that says you know the skill.

This basically.
I guarantee that 90% of the people on this board could figure out how to make a better game than most AAA+ devs if they just had the knowledge and motivation.
It's basically worthless because you can pretty much figure out what good design is just by playing vidya and analyzing what sort of shit is fun and what isn't, or on a more 'advanced' level watching a bunch of those autistic analytical videos on youtube that go into way too much detail about how games are designed.

>could figure out how to make a better game than most AAA+ devs if they just had the knowledge and motivation

having the knowledge is one thing.

having the money needed to fund such a game is where 99% of this board is lacking.

no. data scientists are more like statisticians, they understand programming but only enough to design analysis pipelines. they are generally concerned with soundness, correctness, accuracy etc.

data/computational engineers design and build pipelines and backends. these are the guys concerned with correctness (i.e. bugs), robustness, performance

it's the difference between the guys who use Hadoop and the guys who write Hadoop. all these terms are not really standardized and will probably change in 5 years as people become more specific in their descriptions

t. CS PhD

No user, learn to code, get server certs, network, storage, etc ... Those aren't too hard to learn and give you a nice job

Why does a fish need such big breasts?

>Computational engineering is a fairly new field that focuses on developing simulations and artificial intelligence.
both these things have been around for many, many decades, they're generally called computational science in America

again, these terms are incredibly nondescript. if two CS people meet they will talk about their work in much more specific terms. for example, my field is systems and databases and specifically graph data analysis frameworks

an undergrad getting into the field should think in much more broad terms, i.e. "i want to learn to program, i want to learn statistics, i want to learn machine learning"

You will go in wanting to make your own, rad game and come out, if you're lucky, making an interface for a mobile app or scripting code for a feature in someone else's game at best.

It's good to dream, but know when to draw the line.

>Computer science is straightforward programming pretty much.
>Officially, software engineer is a rung higher than your typical programmer.
this really depends a lot on the institution.

broadly, the study of computing can be scientific insofar as you do theory/modeling (e.g., algorithms and complexity theory), or empiricism (e.g. designing and benchmarking high-performance systems)

a good CS program ought to emphasize these things over programming, whereas a software engineer ought to learn programming AND ENGINEERING, i.e. the design of programs

of course there's plenty of lower-quality institutions that only teach programming

the good news is the quality of online courses for computer disciplines is insane. once you have the basic skills you can learn pretty much anything from coursera or whatever. i know many, many traditional scientists who switched to computer science with only basic programming skills, by taking online courses

programming is probably going to be the easiest part for me; Java is a bit confusing at times but problems are easier to solve the more they recur, whereas i can't come up with songs for shit no matter how much i fiddle around in a DAW. don't know shit about music theory or structure.
thanks for all the replies btw, i'll go into comp science and figure it out from there

to store oxygen, obviously. didn't you major in biology?

It's as someone said. Don't go to college for game design. Go to learn general shit like event handling and event loops, which are core for GUI applications in general. And learning how things like stack frames work, which you can apply to your programming logic as well. I mean once you start understanding the internal workings of linked lists, etc, and how to work with pointers and memory management in general, you kind of get the gist of how a game would work.

The only advanced topic to game logic is how to work with graphics, but even when it comes to 3D, just take math classes, like linear algebra and vector calculus are really essential.

Game courses are if you just want to bullshit around and then get put in a stupid group project just so you can get some teamwork experience, but trust me, you're going to hate that. Just go to a game design club or something for experience if you really care about that.

the thing about programming is that in 4 years if your interests change or you want a higher paying job or you want to go to grad school and be a scientist, you can do those things

you don't have to stay in CS forever or even necessarily get a CS degree, just knowing how to program *well* (i.e. years of experience) is insanely valuable right now

>The only advanced topic to game logic is how to work with graphics
performance optimization, specifically the data path, can get pretty brutal if you allow it

AI is a badly neglected topic in gamedev. i mean the average AAA AI can't reason about pathing on an elevator

game engines are not generally beautiful pieces of software so there's plenty of radical improvements you could make with a little research. graphics just gets the big bucks

You're talking about algorithms and analysis, which I believe is a required course in most CS curriculums anyway. That course alone should give you an idea of why you need to optimize your logic, at least at the higher level. Then there's the system level, and many people would be impressed by how many actual instructions of assembly it takes to do certain operations, and say "HOLY SHIT I THOUGHT x = 2+2 WAS JUST ONE LINE OF CODE".

Sure, if you want, you can take a graphics programming class. I did since I had a free elective. It's not what I aim to do, but it's decent enough knowledge to get me started if I cared to move in that direction. I think my compilers course gave me more insight on approach to programming, though.

Does anyone know any great math books that i later apply to game programming?

I'm a complete idiot when it comes to mathematics

Why go through such effort. Most math isn't going to be easily applicable to programming. Everything you'd need to know about programming is typically learned by learning each respective language. There's nothing you'd universally gain from math books under the context of applying it to programming.

Just know alot of computer science because that helps with making games

New software is makeing people make games with low effort put into them due to being so streamlined but even then it isn't the same

Having a degree shows that you can start something and finish it .

Thanks user, that was the main thing stopping me

Depends on what you want to program since the only "advanced" math you need is in physics and graphics. Otherwise you would focus on boolean logic and learn simple proofs and DeMorgan's theorem.

...

I work as a video game developer.

It's okay,

Program to your heat's content, user. It's the way of the future.

How did you enter the industry, user?

optimization is not just minimizing your complexity and instructions. computations stall on the data path (i.e. load and store), which is 100x slower if it's not correctly serialized

the classic example is:
for x; for y; a[x][y] = something
versus
for y; for x; a[x][y] = something
the latter example may be strided awkwardly and stall horrendously

game engines generate a lot of ad-hoc dynamic data which they then try to stream into many different systems for e.g. collision testing. when the data is incorrectly marshalled you get brutal data stalls. in particular this is one of the key reasons that modern game engines switched to entity-component system designs (like Unity, although Unity is crap in other ways). jonathan blow literally wrote a programming language for The Witness with semantics to handle these issues

just in general, game engines manipulate a lot of data and want to do things like buffer and debuffer assets, call malloc in real time code, etc. game engines don't stall on computational instructions, they stall on data pipes

U SHOULDA USED A POP HARRY

Right now I'm doing Web/Programming and its garbage. You'll go to class for 2 hours out of the week and have 20 hours of homework time. The kids who take this program expecting to use it to make video games drop out faster than any other program because it's not as fun as they thought it would be.

yes
contacts, ask professor for jobs and letter of recommendation or have one of his friends to link you a job

But I didn't really do much, or is this sarcasm?

Anyway, seriously, what do you aim to program? If you're just making a simple game, even simple math will do since you're not working with 3d. And even then there are libraries and most people will just tell you "not to reinvent the wheel" in that case and just learn how to use the library. Of course, you need to learn why things are setup as a scene graph or stack to push or pop things like transformation matrices, if you at least want to understand how to use them properly, which probably can come just as second nature from using them a lot, but trust me, you can save yourself the time and grief by looking up the concepts first. And THAT'S how you avoid reinventing the wheel.

Now you're just talking about prop delay and potentially multithreading, which was inclusive in systems level stuff, as systems level courses will tell you about those stalls you bring up. Same with memory collisions. That's why I ...well, that course is REQUIRED for CS majors anyway.

Don't do gamedev in uni.
Go for something specific, like programming or animation and then branch out.

how good at maths do you have to be to actually program?

I'm not bad at math, I just hate em

every programmer is on a countdown until the moment they get pissed off and take a basic 3-week online course on linear algebra

the question is how fast your countdown is

Why do fishes have tits if they lay eggs?

Do they too produce milk?

My countdown was that lin algebra was required for me to graduate. So was calculus 3 which covers similar topics, so I had no choice.

>how good at maths do you have to be to actually program?

Average, up till integrates and derivatives.

Mostly it comes to patience.

>fishes
>not mammals
>has mammaries?
Ofcourse not.

(non-ironic) Thank you for the response, I'm learning a lot from everyone posting here

whats the difference between, computer science, software engi and system engi?

Pretty much by accident. My first job during university was at a games place and they wanted me back.

Regarding calculus, not at all. I hated it too. The whiz kids who do lots of graphics stuff are usually good at math, but that part of CS never really interested me much anyway.

Definitely learn linear algebra (vectors, matrices, etc).

Specifically for game development? Fuck no. But programming, 3d modeling and UI mapping are all in demand.

Especially 3d model rigging.

>fish milk

jonathan blow hasnt used his language for anything, the witness was made in c++, even though it could have been made in unity and halved the development time, and data-driven programming is a meme

>system engi

what the fuck is that? never heard of it

Has anyone thought about making games then stopped because it was too hard?

>Now you're just talking about prop delay and potentially multithreading
i was really just talking about data bottlenecks, and giving a trivial example, but yeah if you add multi-threading you can make it a lot worse

you will be told about these issues in Systems, but if you want to *solve* them, well, efficient computing on amorphously-structured data is an open research problem. optimizing the data path can very easily devolve to an NP-hard problem like graph partitioning

frankly data bandwidth problems are a bigger issue in high-performance computing right now than parallelism. people incorrectly intuit that computation play a bigger role than data stalls. actually data stalls are fucking brutal if you quantify them

Software engineer is basically CS but with a heavier emphasis on requirements engineering, ethics, and other engineery things.

Seconding this

depends on the programming and medium, odds are pretty good you're going to be copying a formula from somwhere anyways. Logic is far more important than any other aspect of programming.

The program will only do what you tell it to do. 90% of your battle will be trying to figure out everything you need to tell the program it needs to do.

why are people learning web programming? Its honestly a waste of your time.

data-driven programming is a meme

i'm not talking about data-driven programming, i'm just talking about how much (quantifiable) time you spend on data stalls in a typical complex software system

if you download some graph and run some simple-ass computation on it like BFS or PageRank, you can easily see a 1.2x-2x speed up for a randomly-numbered graph (i.e. a randomly-indexed array) versus a degree-sorted graph, with zero change in code. i've benchmarked this on a gold-standard implementation recently

I've started making games and then stopped cause making UI is incredibly dull.

Learn programming or modelling or some shit and study game design yourself.

Yes

Community college fag here, is it even remotely viable to finish my 2 years here and just go for a cheap house/apartment and a decent job?

In the c++ std queue pop is a void return.

Dunno, I've heard its on demand in shit world countries, but the pay is pretty low desu.

Programming (specially web) is not that freaking hard, wonder why most people are pretty slow or dont even try it.

2 years in?

also Jai wasn't a data-driven language, and Unity uses an ECS to organize the data path just like i mentioned

i bet if you married the Unity ECS to a column-slab allocator you could solve 90% of Unity's performance issues

you can afford an apartment on fast food salary if you get a roommate or two, yes

You're asking the wrong place.
Sup Forums doesn't know jack shit about video game development.

That's not really true. It's mostly just pic related.

Well, some people dont like to think too much. Also they need to be somehow skilled even at basic maths.
the 80% ppl from high school are bad at those things because they prefer to study something easier, arts or become a scum of society.

but web programming is something a freaking high school student can even archieve.

>is it worth going to uni or college for game development?
No. No one looking for a programmer or 3D modeller/animator will even give a second glance to something like a Uni degree. The only thing they'll care about is your portfolio, which is just a collection of projects you've made yourself.

You can get the same level of recognition and qualification that you would get from a game development degree by just sitting at home and building your portfolio while also practicing your craft. In fact by the end of the same period of time that the degree would take you would probably be more qualified from just building an extensive portfolio by yourself.

Game companies literally do not care about what degrees you have, they only care what you can prove to them that you can actually do.

not in Scandinavia...

It's nice when people actually consider that stuff, but most people won't even be working at that level to have it in mind. I mean a lot of people are doing javascript or Ruby or something, and when you throw something like that at people who are really basic, they start to get all existential about it.

I know it's a little off topic, but where's the best place to start if I want to get into programming?

oh yeah, i am definitely not saying that people ought to care about this

what i'm saying is that you can make almost any problem this hard by caring about it enough. the games industry cares about graphics enough such that graphics programming for games is really hard. but the games industry doesn't put much money into real-time optimization or AI, so those features of games are usually not pushing the envelope like graphics

you get what you pay for. developers will pay $140k for a graphics programmer but they won't pay the asking price for an AI or HPC programmer