Does size really matter?
Posted: Thu Dec 18, 2014 11:29 am
I'll have to open by saying that, thus far, I've just been pottering around with this stuff on my own so I have no idea if what I'm thinking about is already well-understood and been discussed.
Thing is, I've got this nagging feeling that size, screen-size that is, really does matter, in a single-screen HMD at least.
For software to generate an image with perspective, it has to know what the user's POV is and have a "vanishing-point" for objects to be drawn from and is the point that you are, theoretically, looking toward as objects appear in front of you.
If you have, say, a 5.6" 16:10 display and, for the sake of argument, the vanishing point is in the centre of each half-screen then, as it turns out, the vanishing-points of each half-screen would be around 61mm apart, which means that, with an IPD of 65mm, each eye can look at the vanishing-point of the image in a fairly natural manner, as if the focus of your eyes is converging on an object in front of you.
Conversely, if you have a 7" 16:10 screen, and you're looking at the same image, the vanishing-points would now be 76mm apart and your eyes would either focus on a part of the screen that isn't the intended vanishing-point or each eye would have to look in a slightly different direction (think Marty Feldman) to focus on each intended vanishing-point.
By way of an exaggerated example, imagine if you made a HMD using a 15" screen.
You bolt it in front of a pair of lenses and the result is that you're going to to be staring at the extreme inner-edge of each half-screen and the vast majority of the image is going to be way out in your peripheral vision rather than in your direct line of sight.
I can't help wondering if Oculus went from using a 7" screen in the DK1 to a 5.7" screen in the DK2 simply because a high-quality 5.7" screen was available OR because they realised that a 5.7" screen (coincidentally, using a 5.7" 16:9 screen means that the centres of each half-screen are almost exactly 65mm apart) allowed for images to be viewed more naturally?
Course, there'd be nothing to stop somebody from using software to adjust the image so that the intended vanishing-point IS in an optimised position but I can't help thinking that viewing an image on a screen that's a different size to what was intended is likely to detract from the immersion and, possibly, lead to eye-strain or motion-sickness.
Ideally, it seems like you'd want software that would allow you to enter your screen size and your IPD so that it could generate an image where the perspective is all tuned to suit your own requirements.
Thing is, I've got this nagging feeling that size, screen-size that is, really does matter, in a single-screen HMD at least.
For software to generate an image with perspective, it has to know what the user's POV is and have a "vanishing-point" for objects to be drawn from and is the point that you are, theoretically, looking toward as objects appear in front of you.
If you have, say, a 5.6" 16:10 display and, for the sake of argument, the vanishing point is in the centre of each half-screen then, as it turns out, the vanishing-points of each half-screen would be around 61mm apart, which means that, with an IPD of 65mm, each eye can look at the vanishing-point of the image in a fairly natural manner, as if the focus of your eyes is converging on an object in front of you.
Conversely, if you have a 7" 16:10 screen, and you're looking at the same image, the vanishing-points would now be 76mm apart and your eyes would either focus on a part of the screen that isn't the intended vanishing-point or each eye would have to look in a slightly different direction (think Marty Feldman) to focus on each intended vanishing-point.
By way of an exaggerated example, imagine if you made a HMD using a 15" screen.
You bolt it in front of a pair of lenses and the result is that you're going to to be staring at the extreme inner-edge of each half-screen and the vast majority of the image is going to be way out in your peripheral vision rather than in your direct line of sight.
I can't help wondering if Oculus went from using a 7" screen in the DK1 to a 5.7" screen in the DK2 simply because a high-quality 5.7" screen was available OR because they realised that a 5.7" screen (coincidentally, using a 5.7" 16:9 screen means that the centres of each half-screen are almost exactly 65mm apart) allowed for images to be viewed more naturally?
Course, there'd be nothing to stop somebody from using software to adjust the image so that the intended vanishing-point IS in an optimised position but I can't help thinking that viewing an image on a screen that's a different size to what was intended is likely to detract from the immersion and, possibly, lead to eye-strain or motion-sickness.
Ideally, it seems like you'd want software that would allow you to enter your screen size and your IPD so that it could generate an image where the perspective is all tuned to suit your own requirements.