I’ve been working with DirectX lately to try and build an action-RPG I’ve designed which I’ve dubbed Crystal Genesis and I’ve noticed something with I’m sure you’re all aware of; DirectX is built around modular components to implement game components such as sound, graphics, input/output etc. I was thinking has/should something be made to standardize the way games are designed. I mean, all games share common components and the way these games are made conforms to genera’s such as an RPG for example will allways have characters, enemies, a Über strong storyline etc and below that all fantasy-RPG games have things in common just as all sci-fi-RPG games have things in common. Now in theory could a kind of structure be made on a high level* principle to make genera game design standardized while retaining the ability of genera combination and development? (*High as in the core game design must have X to belong to this genera template but you can arrange the lower level parts of the design any way you want or use a low level system template to adapt a low level game system from.) Personally I believe this could be achieved, but by no means easily created, but would require severe editing to stop clone games, not to mention the complete mine field of intellectual property rights and that. Badly made a construct like this would be disastrous but done right and it could make game design a piece of pie. Opinions? Siolis

I don't think it could nor should be standardized. We already have some components which are fairly standard, like most people creating a game engine will write something commonly referred to as a "renderer", everyone knows roughly what a scene graph does and a main loop is even part of the most simple application.

But upon that, there are a number of very different architectural approaches, all quite effective, eg. the message based system (which I strongly dislike ;)), a constant update loop or an event based architecture. The way to navigate these constructs is by knowing the parts with which systems using these architectural patterns are built, design patterns.

The main question here probably still is, what do we gain from it ? Would it become easier to build game engines ? Could I re-use my mods written for UT2004 in HalfLife 2 ?

-Markus-

I think you may have got me wrong, I don’t mean programming patterns or programming at all, I purely mean game design wise.

If you look at the sheer architecture of games you can find patterns between them and any good game design book specifies the components of an action game, an RPG game, an fps game etc. What I’m saying is, why don’t we have standards when it comes to how to design a game like we have for what we use to build one, case in point is DirectX which gives us abilities to do things which are either to complicated or don’t need to be repeated for each game we make. Transferred to game design I’m saying we should have kind of free to access spec templates for designing games which can be edited to mimic game designer made systems such as the meteria system of final fantasy 7 BUT with our own edits to make a new kind of "Magic Crystal, Magic Usage Limiter".

To make it clear, in nearly all RPG's you have weapons, how the player uses weapons, how long weapons can be used for, how you equip them etc is a game design system which in itself interacts with other game design systems, in this case the weapons equip screen would be a components of the GUI and in the same region as the players item bag, both of which are designed game systems.

Why don’t we have standards of these game design systems which can be accessed by designers making new games to combine together, edit and make new systems of?

The similarities between games can easily be seen, ideas such as those can’t be copy writed (I know that for a fact) so why isn’t there a program, a database, a web site with these systems. Why isn’t game design modular in this day and age?

Siolis

Quote:

Original post by Cygon But upon that, there are a number of very different architectural approaches, all quite effective, eg. the message based system (which I strongly dislike ;)), a constant update loop or an event based architecture. The way to navigate these constructs is by knowing the parts with which systems using these architectural patterns are built, design patterns. -Markus-

Hi Markus, could you perhaps link me to a more detailed technical discussion of the various approaches?

Thanks!

Check out Staffan Bjork and Jussi Holopainen new book: "Patterns in Game Design" from Charles River Media

I perosonally love it. Check em out at www.gamedesignpatterns.org.

I think you will be pleasently surprised by what they have done. :)

Thats what i plan on doing with my game, crystal genesis. One half of the game will be a fixed storyline where i lead the player and the other half is a free roaming type adventure where you go and do anything you want withen game limits and then their will be a mod to that where you can make a map and customized everything on it.

I take it in your view this would be a good way to do things?

Siolis

The movie industry has something like this. For example, there's a word processor and story development program called "Dramatica" which pretty much all script writers use.

Quote:

Original post by Anonymous Poster Well, I have also been anxiously waiting for standardization in game engines, but for a different reason. If there is a standard that includes everything necessary to make a game of any genre, then it would also be possible to have a single game with ALL genres mixed together in a seamless gameworld.

And you could use HTML to select which genres to use...

Quote:

Original post by hplus0603 The movie industry has something like this. For example, there's a word processor and story development program called "Dramatica" which pretty much all script writers use.

So that's why they've become so lame and predictable.

Quote:

Original post by Anonymous Poster Yes, but I was thinking more along the lines of melding the genres together... ALL of this could be accomplished with a universal "master engine" as soon as the computer technology catches up to support that much going on at any one time. As the original poster said, all of these gametypes already exist & have similarities in their engines, it's just that nobody is consolidating them...

You could make it now but theres a few problems,
This master engine would be expensive to develop and you would only get to make a couple of games with it before it started to look and feel "last generation", also developing the engine at this level would meant most games made with it would have visable simalaritys in graphics and play control