You loaded it, you might as well read it

What’s the next step in 3D games?

By Dave Menendez
Thursday, October 30, 2003, at 5:35 PM

Summary: Wolfenstein 3D, Doom, and Quake all were quantum leaps in terms of the worlds they could simulate. Is there still room for improvement?

I’ve been playing No One Lives Forever and it got me thinking about the advances in world modelling in 3D games. There have been at least three quantum leaps in 3D games, the first being the addition of texture mapping, seen in games like Wolfenstein 3D, the second being the ability to place walls at arbitrary angles and locations, seen in Doom, and the third being the ability to slant walls and floors and to have bridges, seen in Quake. (Marathon falls into the Doom category, although its model is able to place a room over another room, even though you can’t see one from the other as in Quake.)

Quake has been out for a while—long enough that Id Software has released the source code for others to play with—and the advances since then have mostly been in non-geometric areas: more complex environments, better scripting, and more visual effects like transparency, shadows, and reflection. These are all fine things of course (I recently shot a fellow in NOLF and he tumbled down a ramp—much cooler than just collapsing where he stood), but I can’t help wondering what the next big leap in geometric modelling will be.

One possibility is arbitrary rotation. I think games right now can rotate elements around a vertical axis, but that’s just one of many ways for things to rotate. I don’t think any engines today can do things like the rotating tubes you see in fun houses. (Or maybe they can. I’ve been away from gaming for a while.)

Another possibility is solid modelling and mutable structures. Or, without the made-up jargon, the ability to fire a rocket at a wall and actually blow up the wall. Sure, we leave bullet holes today, but those are just decals stuck on the wall: no matter how many times you shoot in the same place, you’ll never puncture the wall. (I’m pretty sure this hasn’t been done on a large scale. If you know differently, please let me know.)

I’m not saying we need this, mind you. At this point, you can probably do more for making single-player games interesting by improving scripting and AI and moving away from the “kill everything in sight”-style of interaction. (Again, NOLF is a good example of this, with plenty of innocent civilians around who you have to keep safe and bits of humorous dialogue between the background characters.)