| Game Engines; Properties, Reasons and Capabilities |
|
When you usually think about a game engine, you usually think about the graphics and visuals of a computer game. Great game engines such as the one used for Gears of War, the Unreal engine, and the one used for Bioshock... which also happens to be the Unreal engine, "wait that can't be right, they look totally different!" (explained later). So when we usually think about graphics engines we think about games with eye candy and blistering detail, such as the two already mentioned, but game engines are a much larger portion of in game development and are a part of the decision making process through the entirety of the game's creation.
Think of a game engine as 1/3 chunk of creation, inside this fairly large section is a large number of features and purposes, hard to be too specific as certain games require different features, but in general most games need a way of rendering their 2D/3D environments, they might also need a physics engine and collision detection for the moving objects, and a way of characters or NPCs acting realistically. So inside the box of "game engine" we have all these features and functions which are required for our game to operate properly, figuring out these features is obviously put forward before the developer creates the engine. But the main point to remember is that game engines are a larger part of the actual game than you usually assume, and even very basic games, such as Frogger, have some sort of game engine which operates its collision features and renders its simplistic animations.
Without boring you too much with the development process it's good to know some of the developers decisions, so in the future you might know why a particular title which seemed to boast new technology and visuals in its trailers doesn't look anything like the actual game. Firstly, that announcement trailer which looked "mind blowingly awesome" on first glance most likely wasn't in-game, of course the producers don't want to tell you that, they just let you assume it is. Basically, if the publisher has a fairly big percentage of cash for marketing, which is very likely for highly anticipated titles, and then they can put a portion of that into producing a pre-rendered cinematic or two. Pre-rendered is to run the game and then record it, so when you're watching it you're not actually running or playing the game... it might sound obvious, but that piece of CGI (Computer Generated Imagery) you're watching would never run on your Xbox 360, Playstation 3, Wii or even most people's PCs. This segment of great looking footage is created on a very fast and expensive computer designed for video editing and creation in the development studio. They cost a tidy sum to produce and require special hardware and software to run them (you ever wonder why they're so short?). However, there have been exceptions e.g. the Halo 3 announcement trailer, but that didn't really look like the full game did it? All that presented, was a demonstration to show what the Xbox 360 was capable of visually, quite jaw dropping to watch, as long as it wasn't doing much else.
Now, a developer typically has one of two choices when obtaining a graphics engine (depending on the budget, time and requirements obviously, but let's ignore that for now) they can either write their engine from scratch or get permission to use another developers graphics engine. Reasons will differ for the choice, if the developer has a lot of unique technological requirements (HDR lighting, advanced physics, ragdoll physics etc) then it's probably best to build the engine from scratch as you can build around these requirements, however, if the game has pretty standard requirements or the developers have a lack of expertise, then it would be much less of a time constraint to just use another company's engine, e.g. the infamous Unreal engine.
Due to the advancement of the Unreal engine, it has become more of a set of tools merged into a basis of an engine, probably writing out the motives to building your own game engine from scratch. Originally the engine had just been used for 1st person shooters, by Epic games (The creators of the Unreal engine); the Unreal series and the Unreal Tournament series. This engine has been licensed out to many other companies allowing them to have an already formed engine, plus the options to adapt it without having to delve too deep into the foundations of the code. They also have three engines to choose from all with their own feature sets (Unreal Engine 1, Unreal Engine 2, Unreal Engine 3 and with 4 on its way). What also makes this engine so popular is its portability to other systems, it can support the majority of computer operating systems along with all the seventh generation consoles, plus the Xbox and PS2. The scope is vast for future ports and allows for every game designer to create their own graphical style, and that is why we see games looking so contrasted using this engine, like Gears of War and Bioshock.
There are major benefits from building your own engine in the long run, one of the core reasons of why it's such a big decision. If you've built your own game engine you can then build your own identity, it seems more independent for a developer to have their own engine to play with and it avoids the fuss of signing up licensing deals. It will also allow a developer to license their own engine to other companies potentially earning more profit and showing off the popularity of your graphics engine, but is it always worth the extra development time? The answer is, certainly not.
But back to an earlier question, why do those early in game trailers look nothing like the finished product? This is could be for a number of reasons, but to pull the camera back further, why don't all of our next gen games look pristine with the latest graphics technologies? The main reason for this is most likely due to the balance of budget and purpose. The game's purpose, whether it's meant to be totally reliant on interactivity and gameplay or whether it's aimed at authenticity and immersion, obviously the latter would demand the more sophisticated visuals and so more money and time would be put into that section of production. Of course within purpose there's the audience, it's not a coincidence why the Wii can't support the most advanced visuals, it's that Nintendo sacrificed the option as it wasn't entirely necessary for its major audience, the casual gamer.
As for titles which start off looking great, well the reasons are very similar, as production goes on the developers may find that stability is more important than the current quality for this type of game, or, the content they've put in has used more than the estimated amount of memory available and so visuals may need to be reduced. Developers will generally start off with the best looking visuals they can muster, but as development goes on when they run into a problem, the graphics will be chiseled down for optimization purposes. And this may seem pretty cheeky, but at the peak of its visual quality, the developers would generally be more than happy to share their work for early screenshots and gameplay trailers as it builds a better impression, so it also becomes a form of marketing tactic.
So what's the purpose of all this? The main purpose of this is to give you a brief insight to how a lot of developers think and their reasons for some their motives, but more specifically about the graphics engine. Why do so many of the 3rd person/ 1st person developers choose the Unreal engine over any other? Why don't all games look as good as their early screenshots? And of course, what is a game engine and how important is it in the development process? The most important point to remember after reading this though is I hope you'll stop and think about why a game looks like it does before complaining about it, as there are many other reasons to bring into consideration and not all the time is it due to sheer laziness on the developer's behalf...just most of the time.
Posted By: Lee Burton
Date: 2008-09-06
|
