How long would it take?

Networking & Multiplayer
#1

I am a new programmer and I would like to know how long it will take me to learn how to program a game like Runesacpe?

Thank You.

#2

I haven’t played it, but know it does have 3D and networking, so you’ll need to learn those specialisations as well as regular game programming. It also has quite a bit of content (world design, character models) which take a fair time to produce, as well as requiring artistic flair. Usually several people work in collaboration to produce a game of this magnitude in an acceptable timescale (a year or two).

Alan

#3

Thanks, if I where to make it in 2d would that cut out a lot of the work and is there any softwere that can help me make a game like this?

#4

2D would make the graphics much easier. In particular knocking out neat 2D sprites is much easier than doing 3D models. If you go for a tile based design, there are tile based editors out there. e.g. Mappy http://www.tilemap.co.uk/. You still need to figure out how to make the game a RPG rather than just a shoot em up though.

#5

You should start with a very limited feature set. Create a GOAL that you will hit, don’t change the GOAL much during development, and when you hit the GOAL, polish it up and call version 1 done. Along the way you will get ideas and inspirations that you can use in the future.

If you are inspired to keep working on your game, create a new GOAL, and repeat. Otherwise start a new game. Any work done on a small game will NOT be wasted work. You will be learning important techniques that you can re-use.

For instance I might choose a goalset of:

  • large tiled 2d maps
  • arrow key movement
  • enemy sprites that close in and attack
  • an attack key to attack in your current direction
  • character hitpoints
  • character level
  • character damage per attack
  • all character & enemy sprites have only 2 animation frames (normal & attack)
  • temporary powerups that you can walk over to pickup

Or whatever else you like.

My point is this: You need to know what features you will have in your game, and STICK TO THE LIST. Otherwise you will constantly be coming up with ‘good ideas’ and trying to implement them, which is called a RAT HOLE. More often than not, programmers get stuck down rat holes and never come up, then the project gets abandoned.

For instance, during development you might have the ‘good idea’ to replace your sprite animations with quake MD3 models. So off you go writing loaders, keyframe interpolators, etc… Then guess what the models don’t really look nice enough without support for various specular & bumpmapped shaders, and wow wouldn’t shadows look cool, nevermind the fact that you’ve never implemented shadows before and have no idea how long it will take, etc…

Hopefully I’ve gotten my point across. FEATURE CREEP is the bane of all game development, both professional and amateur…
You need to evaluate each feature you’ve chosen and estimate how long it will take you. If you don’t know how long a particular feature will take, then that feature is a RISK. Too many RISKS in your feature list and you are in trouble.

Enjoy!

#6

4-8 years of hard work and you’ve got a inferior clone up and running.

#7

That’s a bit disappointing since it only took Andrew 2 years, working on his own.

EDIT: although RS 1 was amusingly silly - you had to beat the shopkeeper to death with your fists because he was the only monster around (apart from a rather hardcore sheep)

#8

But he was probably not a complete noob at programming when he started? But as you say, even getting the first version up will take years.

#9

Multiplayer internet games are harder and more complex then LAN games.

LAN games are harder and more complex then single player games.

You must learn to crawl before you can walk, grasshopper, and walk before you can fly.

Furtehr discussion probably belongs in “Newless CLubies.”

#10

Yeah, he was already a good programmer - but the “real” version only took 2 years to write, and was done using java 1.1.x - nowadays there are a lot of shortcuts you could use, lots of places where you wouldn’t have to waste time writing code (for instance, nowadays you have OGL bindings and don’t have to faff about writing your own triangle rasterizer). Also, when you have something to clone, it saves a vast amount of time. I was only really pointing out that 4-8 years is probably unfair - 2-4 would be more like it IMHO.

I guess the real issue with RS is whether you just want the game or the content - RS’s content must be many many man-years of work by now. Writing quests, making textures, animating models, coding item-logic, all that takes a lot of time. With 4 years you could get a lot of that done in tandem with code cutting, so that you could certainly get a substantial proportion of the game cloned, but it wouldn’t be the full monty.