Abstraction and VR


post cover

I distinctly remember around the year 2007 my older brother built a brand new PC. I recall this experience so vividly because I was to receive his older machine to play my own games on. That aside, I do remember him being particularly excited to try out his new GeForce 8800 GT on a game called Crysis. Crysis was a first person shooter with open world, stealth and cinematic elements set on an Island off the coast of North Korea. This sounds like your typical shooter, however, its main appeal was aesthetics. Crysis was a new milestone in fidelity, featuring realistic lighting relative to the time of day, bright new shaders, and water that looked good enough to drink or even dive into. It pushed new hardware to its limits and was the literal benchmark for insane PC builds years after. This was the main appeal of buying a new graphics card back in the day. Sure you could render cool, complicated art using Maya and other programs (as I told my parents countless times), but the real underlying reason was to play games with much more complex graphics. But is this really a valid reason to designate a piece of software as good? Do better graphics really equate to a more intimate, engaging experience?

This is something that I’ve wrestled with throughout my career as a 3D artist: the balance between realism and simplification or stylization. On the one hand you have easily effective means of communication, and on the opposite, a flashy Baroque-era amount of detail and excess. On the one hand you have something that might be unappealing at first but deeply satisfying (Minecraft for example) but on the other you have games that sell millions of copies just because of how a skin shader simulates light accurately, games that might have horrendous gameplay or be a complete waste of money. However, I don’t want to pretend that those two categories are somehow divided by a mile-wide chasm. But if they are, that chasm would be called the Uncanny Valley. Often in the pursuit of realism artists and animators lose sight of what makes a model look human and end up creating something that breaks immersion immediately. What was a $100,000 character or robot becomes the laughing stock of the internet or potentially the fuel of techno-dystopian nightmares.

Appearances are only millimeters-thick in terms of substance (they truly are infinitely thin in terms of 3D faces). It’s like an elegant cake that tastes of manure or an extremely beautiful woman who only takes and never gives. Graphics are only the marketing that draw an audience in, and are a win-win for companies who invest millions in realism (and for hardware manufacturers who make even more from upgrade sales). EA doesn’t really care that much if players get their money’s worth in gameplay time. As long as the graphics are a large enough gimmick to sell copies initially, they’ll make a return on investment. If that sort of trend in rising fidelity is allowed to continue, games will continue to degrade in terms of fun, kinesthetic innovation. This is why Virtual Reality (VR) development is concerning. Anything that would normally run fine on a 1080p monitor will run about half as well on a VR headset, since there are now two cameras/screens where there were previously one (this can increase to four or more times as slow if real time reflections are integrated). When VR becomes commonplace, PC game sales, hardware sales, and everything else in that ecosystem will grow considerably from people upgrading. And where there’s exponential growth, investors and board members are going to congregate. Not that a thriving industry is a bad thing, but bubbles can be disastrous, especially when built on nothing more than complex visuals. We’ve seen it happen before when Atari was flooded with incomplete, corporate movie-licensed games, and from my perspective we’re already in a bubble considering how similar a lot of the top titles are today (Call of Duty and a lot of Clancy games for example).

So what sort of action can dilute this highly volatile games market? You might initially say indie games, which is fair but they’re a small section of the market, excepting Minecraft. I recently got to try out Google’s new Tilt Brush app on the HTC Vive. It was incredible because it leveraged the medium and the new technology seamlessly. This is what needs to be pursued: fulfilling mechanics and great user experience above legions of useless extra content and detailed graphics. Not to say visual accuracy is something to be condemned on its own, it just shouldn’t be the sole selling point of games or VR apps. I can’t wait to see what sort of other exclusive VR content gets created since it has the most potential to be the lowest bar of entry into video games. However, the price point and the current CG climate is working counter to that end. VR in practice is incredibly intuitive since it only requires the player to move and grab objects as they would in real life, but if software companies concentrate on an overabundance of detail, framerate drops and input lag will increase motion sickness and force consumers to buy even larger hardware that they shouldn’t need.

I’m in favor of more abstract visuals because they are quicker to create, more effective in terms of emotional connection (think Pixar or cartoons), and they don’t strain the hardware, making the experience seamless. The pursuit of realism is a plateau and we’ve pretty much reached the limits of practicality within our level of technology. Of course there will always be enthusiasts, but our efforts as developers should be focused on substantive mechanics and meaningful stories. VR is opening up a door into brand new forms of expression and art style, and it would be a waste of time and potential to continue pursuing a goal post that is always moving out of our grasp. Why be constrained to the ground as a soldier on the battlefield when you could be exploring the galaxy in your own space ship or acting as a phantom, observing some interesting story unfold before you. Style and creativity is more important to games and VR as a medium than visual detail ever will be. I hope that the major games producers start to realize this and begin supporting promising younger developers when it comes to VR and games as a whole.