The Future of 3D Gaming - Sid Shuman?

BoardBonobo

My hat is white(ish)!
Veteran
Found this on the nZone thingy that Nvidia have running. It's an article by Sid Shuman (Sid's Human :? ) about the future of 3D Gaming.

The Future of 3D Gaming

by Sid Shuman

With recent advances in 3D technology, it's an appropriate time to discuss the future of 3D gaming. Not just the obvious improvements, like soaring polygon counts, ultra-high resolution textures, and sharper anti-aliasing and filtering methods. Let's think big and focus on the technologies that will revolutionize 3D gaming in the coming years.

Unified Dynamic Lighting

Of all the graphical technologies on the horizon, unified dynamic lighting is one of the most promising. Currently, most games use light maps—translucent 2D circles painted onto environment surfaces—to approximate the lighting effects of, say, a flaming rocket screaming down a gloomy corridor. But with graphic detail soaring in recent games like Unreal Tournament 2003 and Halo, this crude approach to environment lighting is about to outlive its usefulness.

Enter per-pixel environment lighting. Instead of rendering shadows and lighting as static paint jobs, per-pixel lighting technology (first introduced with NVIDIA’s GeForce card series) calculates lighting on a pixel-by-pixel basis. (Hence the name). The best aspect of global per-pixel lighting is that it's directly related to gameplay immersion—it's not just a flashy new effect or a slightly crisper image. For decades, smart filmmakers have used careful lighting tricks to stimulate the emotions of viewers. Once per-pixel lighting becomes standard, it should usher in a new era of video game immersion. And with the upcoming release of top-tier games like Doom III and Deus Ex: The Invisible War, gamers will soon have an opportunity to judge this fledgling technology firsthand.

Dynamic, Deformable Architecture

Despite the best efforts of game designers, completely deformable environments remain a sort of Holy Grail for 3D gaming. Though a few older titles like Duke Nukem 3D and Red Faction have allowed players to blow holes through walls and destroy furniture, technology limitations have limited these attempts. For a glimpse into the future of deathmatch games, imagine launching a grenade into a mine shaft and crushing your unsuspecting opponent under a massive cave-in. Cool, huh? But don't get all hot and bothered just yet—this technology won't make an appearance for several years, at the earliest. But things might be different today had a little game called Prey ever been completed.

Ah, Prey. Way back in 1996, 3D Realms announced that this first-person shooter would use an experimental rendering method called “portal technology.†Alas, for reasons unknown, 3D Realms eventually canned both Prey and the portal rendering system. And that's a damn shame, because Prey's portal technology still sounds exciting today, especially compared to the aging (and less dynamic) Binary Space Partition (BSP) system found in most modern first-person shooters. Besides allowing for completely dynamic and deformable environments, portal rendering would have enabled some nifty spatial and optical tricks. In 1998, game journalist Jason "loonyboi" Bergman described his experience with Prey's portals: "[we] looked up to the ceiling and we saw through the glass window the water, and through the water the sky, and through the sky the side of a spaceship waaaaay up there." Woooah.

But portal technology represents only one possible solution to the problem of static environment architecture. In time, designers will invent other ways to breathe more life into 3D environments.

Expansive, Detailed Environments

For years, game designers have had to choose between creating small, highly ornate environments or massive, low-detail landscapes. The real world, as we all know, makes no such compromises. But with the rapid advancement of 3D technology, game designers are making fewer and fewer graphical concessions. Through brute polygonal force and the careful use of shaders, Unreal Tournament 2003’s vast outdoor arenas are a first step toward this ideal scenario.

Other games have attempted to skirt the tradeoff between environment quality and size in more novel ways. The first two entries in the Delta Force series, for example, rendered outdoor terrain with voxel (“volumetric pixelâ€) technology. In Delta Force's sprawling landscapes, voxels rendered realistic rolling foothills and waist-high grass—effects that are highly impractical using polygonal technology. But unlike polygons, voxels aren't accelerated by current 3D hardware. Due to disappointing speed performance, the developers of the Delta Force series later ditched the voxel landscapes in favor of more traditional polygonal environments. Voxels have some notable advantages, but only time will tell if they will become practical for use in 3D gaming.  

…To Infinity and Beyond

Though many of the technologies highlighted here rely on high-tech graphical advances, experienced gamers know that eye-popping visuals will never replace quality gameplay. And ultimately, critics will judge Doom III based on its gameplay merits, not its juicy lighting effects. But used properly, each of these featured technologies will enhance, not replace, the nuts-and-bolts of video gaming. And this is how it should be.
 
Ah. . . sweet memories of voxels. . . :) Put to great use in Bioware's first game: Shattered Steel -- the first 3d game I saw to have deformable outdoor terrain.
 
I may be a bit dense, but what's the relationship between portal technology (ie occlusion detection based on "visibility through doorsteps") and deformable environments ?

And if you go into voxels, Outcast was, IMVHO, way better than Delta Force...
 
CorwinB said:
I may be a bit dense, but what's the relationship between portal technology (ie occlusion detection based on "visibility through doorsteps") and deformable environments?

w/o knowing the author original thoughts here, i think the relationship is implied by those deformations leading to visibility changes. BSP threes are pretty static - deformations are ok as long as no tree nodes are broken (yes, there exist dynamic enhancements to BSP techniques allowing for nodes re-arrangements, but their occlusion efficiency drops with the number and type of re-arrangements), whereas portals are way more flexible with regard to visibility changes.
 
darkblu said:
w/o knowing the author original thoughts here, i think the relationship is implied by those deformations leading to visibility changes. BSP threes are pretty static - deformations are ok as long as no tree nodes are broken (yes, there exist dynamic enhancements to BSP techniques allowing for nodes re-arrangements, but their occlusion efficiency drops with the number and type of re-arrangements), whereas portals are way more flexible with regard to visibility changes.

Ok, I get that. Such a system of dynamic environment would be impossible to do with static BSP. Still, I think the author makes a direct link between Portal technology and dynamic environment which isn't justified.
 
Back
Top