• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

Game Developing

Cold_Ankles

Gargoyle
WFTO Backer
Dec 10, 2011
372
115
315
26
South Australia
#1
I have no experience making games (well except for this one time) but it is something I would like to get into, does anyone have any suggestions besides try and make a game? I had a look at the UDK documentation today and it sounded looked interesting.

This one time I had to make a game as a university assignment: 19/20.
 
#2
There's no one way to learn how to program. Some like to just dig in and get started, some prefer watching tutorials and following along, it's all dependant on the person. If you've got no programming experience, then try out a few different languages and see what you like. I'd recommend getting Unity free because you can try a few different styles of language; C#, JavaScript & Boo.
 

MrWahloh

Dwarven Worker
May 6, 2012
47
8
20
27
The Netherlands
#3
You can follow courses. But there's no real all-in-one course that learns you how to create a game.

Things you need to know to create a game:
3D Modeling, Rigging, Animating, Concept Art, Texture Art, Programming, Game/Level Design, Project Management.

You can find allot of books about those subjects, but you should find the one that interests you the most and focus on that one.

And of course you can go to a school that focuses on Game Design, that's what I'm doing now.

There's one problem though, you really gotta show that you're good at it to get a steady job and in The Netherlands there aren't that much Game Studios so you'll have to be prepare to move to a different country for a job.
 

Keleben

Blood Imp
Jun 17, 2012
1
0
0
31
#7
My personnal experience (without any serious game finished, but due to lack of motivation/time and no tools) would say, if you don't have studied programming i advice you game maker, wich permit you to achieve a lot of things in 2d without knowing programming. If you want to go 3d, it include easy to use and weel documented 3d rendering function, with poor capacity, but that's not really a problem since you don't have access to hight level 3d assets.

If you know programming (throught university studies i mean, no some random tutorial on your free time) then search on hight level library for graphic engine and, go to objects oriented language as java (and i can suggest lwjgl, wich is very powerfull, and no that complicated as you stay away froms shaders ) and read some articles on game architecture. don't be afraid of 3d engine, they're no much complicated than 2d in fact (but assets producing time explode, i personnaly stick to minecraft like very basic models or 2d sprites as in doom 1 ou wolfeistein to compensate) .

If you know programming, once you've got the basis, it's pretty similar to any other type of program, with often simple tests and algorithms, but numerous, and runned 60 time per second.
 

LFW

Priestess
Dec 15, 2011
125
31
210
25
Tauranga, New Zealand
#9
Screw around with SDK's or Map Making programs, like TES Contruction kit for Oblivion or Skyrim, they generally contain all the tools you need to make ANYTHING from within the parameters that the game itself provides.

It may not be a great idea if you want to make a game from scratch, but it can help in easing that learning curve you hit when you first start out

I once tried to replicate the terrifying Shalebridge Cradle level from Thief: Deadly Shadows into Half-Life 2, I built a replica of the building, created audio cues (the music that plays when you first enter, and the door banging part) added 9 Headcrab Zombies with an insanely large amount of health to the second section and started on the 'turn the generator on' objective when I just... stopped.

Unfortunately, I lost the files to the map :(

Still, I learned a lot from it about level design (despite following preset blueprints), how tiny details (like the 'pulsing' light) can severely alter the atmosphere and the use of math counters and triggered events.

Find a game with an SDK, use it to find out how the developers put a certain part of the game together, then try to make something unique for yourself. It's a great experience.
 
Aug 5, 2012
54
5
5
UK
#11
In this day and age once you're qualified you can work from home provided you have the relevant equipment and bandwidth, on some cases you can work by commission.
For starters it's best to decide one area of game development that you want to get into, for example, map and level design, creature and character design, graphic modelling, sound and music, programming, scripting, legal, performance art (for cut scenes and motion capture) then work on that area, take courses and practice your area until you are good enough to take small jobs, some of these can be found online, when you have a reputation bigger jobs will follow and you can get experience to branch into other areas.
 

Cold_Ankles

Gargoyle
WFTO Backer
Dec 10, 2011
372
115
315
26
South Australia
#15
Essentially what I'm doing is demaking a game. Kinda like what Gang Garrison 2 is to Team Fortress 2. The game in question is Crimson Skies. I want to demake it into a 2D shooter. Currently setting up the multiplayer aspect is giving me grief. What I really need someone to do is make 16-bit models of all the original aircraft as well as a rocket, an aerial torpedo, a zeppelin, and a blimp. Complete with textures. Still game?
 
Jul 13, 2012
155
72
235
35
#16
IMHO, I would recommend not dealing directly with a game engine if you are looking into programming. You learn a lot more fundamentals by building something from the ground up. If going for a 2D game, building a basic 2D game engine and basic tools to make it easier to build the game. If going for a 3D game, going towards something like OpenGL (my personal preference). You can get content rendering pretty quick. Something like this: http://www.quintessencegame.com took maybe a month from an engine standpoint. And that was having to build a custom exporter for Maya models. The rendering side was done in a few weeks. Will this compare in that time to UDK or CryEngine? Not at all. But given time, practicing shader development and other techniques used like deferred rendering, you'll keep making steps to get to that point.

The problem with using a third party engine first is that if you are planning to make game design a career, you will not always use the same engine. Sometimes you'll be making a custom engine for a game because the engines in the market just don't fit the game design. Or the engine you go to is completely opposite to the one you're using. By building your own, you understand the inner workings of an engine. And hell, you might find that you can accomplish something better with your own engine with smaller games.

And keep goals low. Multiplayer should be a stage two task. Build Pong. This is a very straightforward game, but still requires AI design for the computer paddle that can be controlled by difficulty. Then two player Pong on the same system. Then two player Pong through networking. Then Pac-Man where you start getting a bit more complex, rudimentary pathfinding on the ghosts. Then continue working your way up through different game milestones to 3D.

I would not, for my first game, remake a larger scale game until you can comfortably make a bug-free Pong or Pac-man game. Accomplishing those tasks, you can then pretty much do anything as the processes are the same for larger games, they just have a helluva lot more moving parts, better quality assets, etc.
 
Jul 13, 2012
155
72
235
35
#18
Of course. And by all means, definitely work on a project as it is the only way to learn the processes. :)

All I am saying is that there is going to be a lot of subjects in game design that you will overlook jumping into an engine first that in a second project, you'll be having to learn all over again. Especially using something like Unity which is most definitely not an engine many AAA companies are using currently, though it is getting there in time.

A lot of what I had to learn was concepts like proper optimization so you can keep a game, even 2D, at a consistent and respectable frame rate that goes beyond simply building the game logic. Do you couple all classes together, or use a central message/event systems to communicate between classes? Do you multithread areas like AI of enemy units if there are many on the screen (this is a sore subject in Unity as you cannot multithread anything that touches Unity functionality)? Do you utilize inheritance for most of the game functionality, or more advanced data paradigms like factories and facades which may give you more control? Multiplayer opens up another huge area that requires a lot of finesse to get right (TCP vs UDP, normally UDP is better for a game as its quicker but requires more work as packets are sent out of order). It's much easier to learn these concepts with a very simplistic game, test them out and expand on them until you have the game working and have a firm grasp of the advanced topics.

Something like Crimson Skies, even if a 2D game, would not be something I honestly would build first even if dumbed down. Maybe if dumbed down to Asteroids level that could be a goal. Then once that was working perfect, the mechanics would naturally progress to more advanced gameplay.

Just stuff that was drilled in my head for the last 12 years dealing with game design, and getting my degree at Full Sail. :)
 

Cold_Ankles

Gargoyle
WFTO Backer
Dec 10, 2011
372
115
315
26
South Australia
#19
Asteroids level is what it really is going to be first, since I'm ditching multiplayer and just getting a singleplayer balloon shoot sorta thing going. After that, add more planes and then eventually multiplayer.
 
Jul 13, 2012
155
72
235
35
#20
Well if you need any help, don't hesitate to send me a PM. Can give you any pointers on coding if you run into anything. :)
 
Top Bottom