Pre-Rendered and Real Time Graphics - 3D Game Production Blog Task #2

Pre Rendered

Pre-Rendered graphics is a very simple concept that introduced very high level graphics into relatively weak systems by using either one or multiple frames of images instead of rendering everything in real time. When it was first popularized in the 90s, pre-rendering was a much bigger deal than it may seem today, as back then only a couple polygons would be able to be rendered.

One of the first games to pioneer a form of this, and a personal favorite of mine, is Donkey Kong Country.

Released in 1994, it pushed the SNES to it's limits. The technology behind it was straightforward. But the impressive part of it was that the developers at Rare were able to get all of the 3d detail and colors down into a set of 2D sprites and only a couple of colors. This gave the game a level of graphical detail that far surpassed other SNES games like Super Mario World.

Other examples of pre-rendered game graphics utilized to a far higher potential were in Resident Evil and Final Fantasy. Both of these games also incorporated animated parts on top of the pre-rendered backgrounds to create an even better effect.

This type of graphics however, has a very obvious limitation in that the camera cannot be moved freely. The background is just a series of images, and so the ways to use them in games is limited. They still exist in modern games however, most commonly in cutscenes.

Pre-rendered backgrounds are not exclusive to video games and are used far more commonly in 3D films. Same as games, pre-rendering allows for very high quality models and backgrounds to be able to be shown on systems that would be unable to render them in real time. Films take this a step further to be able to create near-photorealistic CGI nowadays however. Some notable films to use this are Toy Story or Shrek. Films like Monsters Inc. that had characters with fur especially benefitted from being pre-rendered.

Real Time Rendering

Real time rendering is a type of rendering where individual polygons are rendered via the graphics processor in real time. This type of rendering is far more costly, but it allows for total freedom of the camera, and allows for much more open environments. It also allows for player interactivity and input responses that would not have been possible without it.  

As I showed already in the pre-rendered graphics example, pre-rendered backgrounds were used to accomodate 2D and 3D real-time characters even back then. The player character is the core of our input as the player, so that makes sense.

One of my favorite examples of real-time rendering being used to it's full potential is with my favorite game of all time, Half Life 2. Different from it's predecessors like Quake and DOOM, that only faked a 3D effect, Half Life 2 was fully rendered in real time.

Being able to look around fully in this 3d environment and interact with physics objects, as well as characters, was only possible because of the advancements that real-time rendering had created. The player is able to make a multitude of decisions, such as throwing a barrel at a group of enemies, engage in cover to cover combat, explore intricate 3D levels in all three directions unlike DOOM. Dynamic enemies that respond to the player's actions would also been impossible to implement in any other way.

Basically, real-time rendering is what gives video games the feeling of being in control, the decision making. Pre-rendered graphics meanwhile are able to display were high quality images at low cost as long as the perspective doesn't change.