This page is a short informal treatise on the kinds of future games I'd like to see created.
To recap on my general research intentions: side effects of computer games & VR especially in terms of high realism, video & image manipulation and fakery, impact on society of advanced technologies, uses and ramifications of high-realism VR in the fields of education, training, entertainment, military (eg. torture), legal problems, etc.
Many people are doing or have done research into VR side effects, but they all concentrate on physiological issues (motion sickness, etc.) Nobody is publicly conducting research into long term perceptual side effects which my BSc dissertation tackled - these include fear, fright and panic, dreams, after images, flashbacks, behavioural changes, obsession, etc.
My earlier study was not detailed. I just wanted to show that the effects occured, which I did. So what's next? Well, think of it this way: no one has the faintest idea what happens to a person's world view and perception if said person plays a computer game for hours and hours and hours. More importantly, no one has researched the implications of increased realism in computer games. Military organisations are bound to be researching this for obvious reasons (they're heavy users of advanced simulators), but they're not about to make that information public - which is a pity, because such a move would save an awful lot of time (public medical research will eventually find out the same things anyway, so why not save some time and lives and release results earlier).
What if you could play a game that was visually as good as your eyes can see? Or at worst indistinguishable from real life? (the two are not the same). Imagine a VR Aliens game, etc. Far fetched? Not at all. Commercial high-end systems that can generate these kind of images in real-time will exist by 2003 (certainly military ones will). By 2012, you'll be able to buy a games console that can do it (certainly by 2015). Note that these estimates are simple exponential improvements (ie. constant order of magnitude improvement every 3 to 5 years), even though real technology trends are double-exponential (the rate of change has a rate of change); further, they do not take into account sudden technology leaps (eg. quantum technology) which are more likely to occur as time goes on, although it's possible that a delay may occur for a short while as Moore's Law kicks in within the semiconductor industry.
My hardware graphics research has focused on SGI, who design the most powerful graphics systems in the world. The Nintendo64, which was designed by SGI, was about half as good (in polygon blatting terms) as their low-end workstation of the day, plus or minus the odd hardware effect or two, though its design was based on earlier high-end technology (RealityEngine). All one has to do is then work out how long it takes for high-end compute levels to reach low-end desktop, then onto the console market. Mathematically, this requires 4 generation changes, ie. 16 years, plus or minus 4 years. Not very far away at all. In late 1996, I was told by one designer of their high-end graphics system that SGI wanted their, "current high-graphics capabilities to be in a games console within 5 years", ie. 2002. This did indeed happen, albeit in a very roundabout way. NVIDIA based its original Geforce technology on SGI's InfiniteReality, though very much feature-reduced (which was logical - IR has many things games do not need). Microsoft's XBox uses a later generation of NVIDIA's Geforce graphics.
SGI's direct involvment with consoles, a la the N64, didn't work out quite as expected. With revenue coming from console sales, continuing price drops inevitably meant SGI did not make very much money from the arrangement. Meanwhile, the people at SGI who worked on the N64 left to form ArtX, and they then supplied the technology for the GameCube. These and other events have led to SGI technology being incorporated in all of the modern consoles in one way or another, even in the PS2 in the form of a MIPS R5900 main CPU. Consoles beyond the PS2, GameCube and XBox are more likely to be original designs though, since SGI has moved away from such embedded/consumer markets, selling off MIPS in the process.
Back to side effects. There are no laws, regulations or even public awareness about these issues at the moment (and here I've only discussed the realism/side-effects issue; consider the manipulation and fakery of image and video data on a wide scale, with the technology available to anyone), so 16 years is not a long time given how long it takes for society to take on board new concepts.
Doom hit the headlines when it came out, but this has not really happened with the Nintendo64 even though the realism levels in games like Turok are much higher. New Scientist once had a very good article on the power of the latest high-end graphics systems at the time (SGI Onyx RealityEngine2) and what they could do (this was around 1994), but they've never done a follow-up article even though the systems have become 100 times faster since then. Because no one bothers to tell the 'general' public anything these days, I doubt people will sit up and take notice for quite a while. Given the amount of time it takes to research this kind of thing, create a debate, come to decisions and implement laws, it's something that ought to be looked into now.
In the long term, I want to help create the most advanced computer games ever made. Hundreds or thousands of players, extreme realism, enormous freedom levels, etc. If you want to wipe out an entire planet with a Carbon/Iron fusion bomb, eliminating dozens of other players and ships in the process, I want that to be possible. If you want to team up with 20 other players and become the most hated pirate force in the game world, I want that to be possible too. Some of the technologies involved (eg. networking & communication) don't even exist yet except in research labs, but I'm aiming for about 2005 to 2010 to begin my focus on advanced game creation (if life can't be fun, it ain't worth living, eh?) At present, I'm merely making a living on my own, doing SGI stuff, and thus not yet in a position to proceed with anything more concrete. But I still keep an eye on developments, eg. the release of Doom3, the approach of Stalker, etc.
So, research aside, the rest of this page contains a collection of real-time high-end graphics images that are from real simulations which exist now (ie. simulations designed sometime between 1994 and 1996 on either RE2 or InfiniteRealty systems). The intention is to demonstrate what can be done already and to give you some idea of what games consoles are going to be able to do in the future (for these images, we're talking around 2003 - perhaps sooner - for the games console equivalent). I have movie files too, but not enough disk space at present to include them. Note that I'll be putting up some more up-to-date images soon.
Oh, the images shown above were all, at the time of creation, non-real-time examples (see films such as 'The Lost World', 'The Relic', 'Dragonheart', etc. for more examples), but they have been included because I have a saying: if it isn't real-time now, it will be some day.
(link is to same-size image)
(link is to same-size image)