One of the first things I thought about when the Oculus Rift concept was introduced on Kickstarter was.. what kind of hardware would you need to drive a device like this?
In practical terms, this is to be the first real evolution of a display device since CRTs were bundled with the first IBM PCs. Think I'm exaggerating? PC displays have changed over the years, of course, but it has essentially been a passive display which has grown in screen size, resolution and quality; while changing in monitor weight, size, power consumption and backlight technologies. Other than a relatively small market for touchscreens on PCs, this is the first time a display has an input function as well. It is also, by far, the most difficult to get right. No other input device has the potential make you physically sick, if they don't get absolutely right the first time. You could make it look all kinds of shiny, have amazing futuristic shapes, with very fancy names and market it to kingdom come, but nothing is going to overcome the lack of interest, if wearing it for a few minutes makes you dizzy or worse, makes you puke.
The video card needed to fully push this device is going to be the new benchmark once it comes out. Of course, as screen densities spike over the years, the strain on a PC's graphics system will multiply.
A VR headset's huge field of view and with no perceptible lag during head tracking is both its strongest as well as its weakest point. Few, if any, products will as unforgiving in its first impression. You'll either get it right the first time or blow it for good. This, probably, is Palmer's worst case scenario.
The basic difference between all previous ways video performance / quality standards were judged until now; versus what should become the minimum standard on a VR headset - is that the intuitive nature of the device will make people compare it with real life objects rather than a computer generated model on a flat screen in front of them.
The immersive nature of the device taps into such a basic part of the human brain, that allowing the wearer to suspend disbelief is a extraordinarily easy. When people believe that what they are looking at is real, however briefly, they will compare it with real life objects. It can't be helped, it's instinct.
Computer generated objects, especially in a dynamic game environment, have a long way to go before being convincing in a VR environment. Polygon count will have shoot through the roof for one. Textures will have to be good enough to allow players to lean forward and take a closer look while maintaining the illusion of the gameworld. Lighting and shadow effects will be even more critical. Anything that breaks immersion will have to be huge no-no. This will force an unprecedented attention to detail. Considering the data processing ability jump will need to be in orders of magnitude, I'd guess that a new way of handling VR graphics will have to invented, because I doubt graphics hardware will be able to scale up quickly enough. Graphics chips have been seeing incremental increases over the past few years, but I'd guess this will fall short when rendering VR scenes. Of course, one option around this hurdle is to focus most of the optimizations on the Unity engine, which seems to be the go-to environment for VR at this time.
I've played hours of Skyrim and thoroughly enjoyed it, but never I have I glanced at a shield and thought 'why does that round shield has jagged edges', because no matter how compelling the title, my character could move and look around purely based on mouse movement. In real life, I need to turn my head to look around, so immersion is broken at a very basic level.
:)
In practical terms, this is to be the first real evolution of a display device since CRTs were bundled with the first IBM PCs. Think I'm exaggerating? PC displays have changed over the years, of course, but it has essentially been a passive display which has grown in screen size, resolution and quality; while changing in monitor weight, size, power consumption and backlight technologies. Other than a relatively small market for touchscreens on PCs, this is the first time a display has an input function as well. It is also, by far, the most difficult to get right. No other input device has the potential make you physically sick, if they don't get absolutely right the first time. You could make it look all kinds of shiny, have amazing futuristic shapes, with very fancy names and market it to kingdom come, but nothing is going to overcome the lack of interest, if wearing it for a few minutes makes you dizzy or worse, makes you puke.
The video card needed to fully push this device is going to be the new benchmark once it comes out. Of course, as screen densities spike over the years, the strain on a PC's graphics system will multiply.
A VR headset's huge field of view and with no perceptible lag during head tracking is both its strongest as well as its weakest point. Few, if any, products will as unforgiving in its first impression. You'll either get it right the first time or blow it for good. This, probably, is Palmer's worst case scenario.
 |
| The Oculus Rift - Consumer Version 1. Still no price, pre-order date or a release date :( |
The basic difference between all previous ways video performance / quality standards were judged until now; versus what should become the minimum standard on a VR headset - is that the intuitive nature of the device will make people compare it with real life objects rather than a computer generated model on a flat screen in front of them.
The immersive nature of the device taps into such a basic part of the human brain, that allowing the wearer to suspend disbelief is a extraordinarily easy. When people believe that what they are looking at is real, however briefly, they will compare it with real life objects. It can't be helped, it's instinct.
Computer generated objects, especially in a dynamic game environment, have a long way to go before being convincing in a VR environment. Polygon count will have shoot through the roof for one. Textures will have to be good enough to allow players to lean forward and take a closer look while maintaining the illusion of the gameworld. Lighting and shadow effects will be even more critical. Anything that breaks immersion will have to be huge no-no. This will force an unprecedented attention to detail. Considering the data processing ability jump will need to be in orders of magnitude, I'd guess that a new way of handling VR graphics will have to invented, because I doubt graphics hardware will be able to scale up quickly enough. Graphics chips have been seeing incremental increases over the past few years, but I'd guess this will fall short when rendering VR scenes. Of course, one option around this hurdle is to focus most of the optimizations on the Unity engine, which seems to be the go-to environment for VR at this time.
I've played hours of Skyrim and thoroughly enjoyed it, but never I have I glanced at a shield and thought 'why does that round shield has jagged edges', because no matter how compelling the title, my character could move and look around purely based on mouse movement. In real life, I need to turn my head to look around, so immersion is broken at a very basic level.
:)
No comments:
Post a Comment