The 3DS Under a Microscope
The 3DS can sometimes seem like a disorganised pile of features cobbled together and thrust uncomfortably into one package. If there’s one thing that holds it together, it’s this: every bit of the experience is designed to convince you that you’re holding a little piece of magic.
My first couple of days with it were littered with oohs and aahs and evoked a sense of wonder that has always been Nintendo’s great talent. When I decided to look at the 3D screen under my pocket microscope, I half expected to have a wonderful land of pixel-pushing pixies revealed to me. What I got instead is this:
Side by side view:
The 3DS screen when viewed through a microscope. The image on the left is with 3D turned off.
Mouse over to switch between the images [Ed - due to the popularity of this article, the mouse-over image may take a few moments to load]:
So what are we looking at here?
Well, in the first picture we have something very much like what you’ll see if you place any modern LCD device under a microscope. When you think about it, it’s rather impressive in itself that we now have ready access to electronics capable of coordinating hundreds of thousands of, well, microscopic lights in the right way in order to reproduce an image.
The second picture is where it’s at, though. It’s a remarkably simple design in many ways. The lines you see on the picture are the parallax barrier. Parallax is a word describing how the apparent relative position of things in the foreground and background change when looking at them from different viewpoints (in this case, each eye).
To demonstrate this, try a quick experiment: close one eye and then hold both index fingers up vertically, putting one a short distance in front the other so that the far finger is obscured by the near one. Keeping the fingers in place, then switch eyes, and you should now be able to see both fingers.
That is basically how a parallax barrier works. Placed just slightly in front of the LCD screen, when held at the correct distance it ensures that each eye is only able to see certain pixels on the screen. The image presented to each eye is of the same scene from a slightly different angle. The brain interprets this as a 3D image because it performs the same trick when we look at stuff in the real world. Try closing one eye again, then opening it. Ever really noticed how much flatter things look with only one eye opened, or how it takes a moment for the double image to resolve into a single, 3D image when you open both again?
So there we go: Nintendo were telling the truth. No pixies, just pixels and simple-but-impressive technology. But have I ruined the magic by looking too close? Well, to borrow the words of Richard Feynman when confronted with a similar objection: “It only adds. I don’t understand how it subtracts.”
There was a NEOGaf thing on this I think, But you have explained it wonderfully. To the customers asking me, however, I will continue to say “dunno.”
I knew some of this, but this was put very concisely. Should I receive inquiry from tech deficient relatives, I will refer them to this instead of giving myself a head ache.
So I moused over the image, but it still doesn’t look in 3D. I think Nukezilla is broken.
@John Kershaw: Nah, cuz it’s zoomed in 100 times you have to be 100 times further away for the 3D to work. :D
Something that occured to me some people might be wondering about – the barrier in the 3D image looks kind of blurry, and that’s because my microscope was still focused on the main LED screen. The barrier actually looks like solid black bars, but the focus is sensitive enough at 100x magnification that you can’t focus on both the screen and the barrier in front of it at the same time.
Pete, great article and fascinating pictures.
Peter: can God create a 3DS so technologically advanced that even He can’t focus both on the screen and the barrier in front of it at the same time at 100x magnification?
Retarded article. In 3D mode, the microscope sees half the pixels from it’s perspective. This half of the pixels is the image meant for one human eye; if the viewing angle of the microscope was changed to simulate what the other eye might see at normal viewing distance, it would see the other half of the pixels. Two different pictures, one meant for each eye, are displayed at the same time on the interlaced pixels, the 3D effect comes from the positional difference in these two images, not any ‘parallax effect’ or whatever the poster is attempting to describe.
I do beleive, Dr. Wario, that the positional diffence resulting in a 3D effect, is called ‘Parallax’ to those in the know. Even the basest research on wikipedia should have gleaned that information.
I can’t wait until NZP and I can filter this out.
No way, Flanksy! Dr. Wario would never lead us astray. If Peter hadn’t been such a special snowflake, he would not only have used two microscope angles, but also have thought to move the microscopes back 50 feet, so that we could make out the full (now 3d!) image! Duhhh!! 9_~
Just to address what Dr. Wario is saying, briefly:
The effect is indeed produced by something called a ‘parallax barrier’ (that is its proper name). The barrier works because it is placed a little way in front of the LCD screen, and owing to the effect known as parallax it obscures a different column of pixels from each eye, leaving the hardware free to render a different angle to those alternating columns.
At that level of zoom you’re not going to see what one eye is looking at, ever, because it’s much too close. The correct pixels are only blocked when you’re in the sweet spot of around 30cm or so away from the screen, hence why the viewing angle for 3D is so small on the device. That’s also most likely why the pattern you see is not consistent in the second image, even though my microscope was flush to the screen.