More resolution = more revolution

"eidiazcas" wrote:
ok, I got a simple question, well, not so simple but I'm curious about, why don't they use some kind of mirrors and two screens instead of using one screen and divide it in two by the 3d tec.

As soon as you move into mirrors and prisms, you immediately start drastically increasing the cost of components and engineering on the device.

The whole point of the kickstarter and the current design is that commodity hardware has advanced to the point that you can do something like the barrel shader at almost zero computational cost compared the rest of a typical 3D render. Combined with the cheap and easy availability of high resolution LCD technology and accelerometer technology being driven by the mobile computing sector, it's an ideal time to create something like this.

Coming from the other direction, exactly what problem do you think would be solved by multiple screens? If it's resolution issues, we'd be better off just upping the DPI, resolution and quality of the current split screen design than drastically changing it.

"jherico" wrote:
... The whole point of the kickstarter and the current design is that commodity hardware has advanced to the point that you can do something like the barrel shader at almost zero computational cost compared the rest of a typical 3D render. ...

A side-by-side split screen is also much easier to drive with commodity hardware than a pair of screens.

thanks for your answers, I got another one, ¿why haven't they used active 3d tecnology? ¿the price or the energy needed?, this could be a solution to the resolution issue becouse the image would change with a certain frecuency for each eye having no effect on the efective resolution, anyway I know that there has to be a strong reason for them not to do it.

"eidiazcas" wrote:
why haven't they used active 3d tecnology?

Active 3D technology implies using LCD shutters to display an image to only one eye at a time, presumably in order to allow you to use the whole screen resolution for each eye. This is problematic though.

Right now the distortion introduced by the lenses means that the center of the LCD is distorted to near the edge of my field of vision. In order to perceive more of the right half of the screen in my left eye, the lens would need to introduce less distortion, thus using more pixels. However, this means you'd need even more pixels on the left side of the screen to provide the full field of view you get right now. For the left eye, the left side is more important in providing a full field of view, so using active 3D to get a small additional amount of resolution in the very center of my vision doesn't seem worth it. The area between the eyes is already pretty high resolution/low distortion because it's not very far off axis.

Other ways to use 'more of the screen' for each eye would inevitably involve off-axis use of lenses for prism effects, or even worse, asymmetrical lenses. This brings you right back to much costlier parts and engineering. Leave that nonsense the Sony and it's $500 and over crowd.

At first I could really see the screen door effect but then I worked my way through it. The thing is when we first put on our HMD for the first time we take in everything about the hardware. We see its flaws as we focus on the black lines and then we look past them. We see the whole picture that works together despite the minor separation. You desire that picture more than you want the lined effect. After awhile just like how the brain ignores the nose on your face, the separation dwindles and becomes a lesser problem. There is nothing wrong with higher resolution, as I feel a higher resolution can really help cut down on motion sickness, but there are so many other aspects to VR that are being missed by focusing too much on the resolution of the screen.

How about designing the rift around the concept that the screen is changeable/upgradable? As higher resolution screens become cheaper, you can simply buy a new screen for your old rift.

"virtualeks" wrote:
How about designing the rift around the concept that the screen is changeable/upgradable? As higher resolution screens become cheaper, you can simply buy a new screen for your old rift.

There's no way to positively predict the dimensions of the next display that people are going to want to use. Oculus would have to come up with some kind of universal adapter system, which would potentially require significant development time and cost.

Interchangeable screens would be nice, but it's likely that the tracking technology, form factor and fit, and mechanical features of the Rift will evolve over time as well. Those are things that you're also going to want to upgrade along with the display, but they can't make all of these things interchangeable. I think it makes more sense to manufacture only fully integrated units that aren't individually upgradable. It's simpler and cheaper.

However, I think a good compromise would be to make the Rift easy to take apart, so that people who want to retrofit new displays can do so with minimal risk to the device.

.

There's no way to positively predict the dimensions of the next display that people are going to want to use. Oculus would have to come up with some kind of universal adapter system, which would potentially require significant development time and cost

If you look at the tear down of the Rift the LCD is 70%+ of the unit hardware. I think it would be much easier to upgrade the tracking hardware since its a small board with a USB connector. The Rift electronically is a Monitor with a high update refresh rate plus tracking board plus some special optics. The Realtek display driver chip in the rift was designed to be used inside LCD monitors. The sourcing of the new LCD panel is a key factor to the success of the unit in future.

The tracking board to making is very important to how consumer version will evolve. I believe that is the part that needs to be interchangable. If a DK2 is released with 1080p board that reduces the screen door effect is sourced I can't see that hardware changing very much after that point.