I had to check the date when the news came through. No, it wasn’t 1st April. And yet there is a press release telling me that Nintendo are launching a handheld machine capable of 3D effects without the use of special glasses.
My initial reaction was not positive. With the DS, DS Lite, DSi and DSiXL all having surfaced over the last handful of years, my patience for another iterative take on their handheld was wearing thin. If the Nintendo 3DS, as it is to be temporarily known, was to be just another minor hardware improvement brought about for no other reason than to give Ant & Dec something further to discuss with their cabbie friends, then any interest I may have had would be quashed instantly.
But then a colleague showed me this…
Available in Japan through DSiWare, the downloadable service for the DSi, Hidden 3D Pictures! makes clever use of the hardware’s camera. By tracking the player’s head it is able in real-time to change the angle at which the scene on display is viewed from, allowing for an illusion of depth within the DSi itself. You can literally crane your head to one side to look around a corner, or duck down to see what is hidden above the screen’s limits; you are the game’s camera, controlling the viewing frustrum.
If this seems familiar to some of you, you may have seen Johnny Chung Lee’s head tracking application using the Wii’s interface which also created the illusion of 3D on a standard television. The principle is the same, except rather than the Wii tracking the relative position of the Wii-mote compared to the sensor bar, it is the relative position of your head to the camera that is of interest to the DSi. However, with the advent of Natal’s full body tracking and the wand based movement of the PS3, this technique is not just restricted to Nintendo consoles.
The key to realising the possibility of this representation of depth is accessibility. And not in regards to the public, but the developers. On the three main home consoles and their respective motion controls, retrieving a relative position of a controller or a person is straightforward as it has been built accessibly into the API. Natal programmers call the equivalent of Player.GetJoint( Head ), and the position of a player’s head will be presented to them for use in calculating the angle and offset from the camera, likewise with the wands. If the 3DS does work as we speculate then such image processing will have to be done as standard and not rely on each development studio rewriting the same code to track a player’s head. Anything less and the experience will differ greatly from game to game as implementations and their effectiveness will vary likewise. This may sound obvious but you’d be surprised what some platform holders omit.
For 3D, the downside to this approach is that it will only ever look coherent to a single person, as only one body can ever be the reference point. On a home console a split screen experience utilising such tech is likely to end in motion sickness, but on a DS when in the vast majority of cases on a single person is ever intent on what is happening on screen, this is not really an issue. You will have a pocket 3D world for you to peer into and gorge your eyes on.
But does it truly need new hardware for such a thing? The example above is running on a DSi so whilst possible you can imagine that as the primarily 2D DS/DSi is asked to shift into rendering complex, dynamic, 3D scenes the processor might come under considerable strain. Even if this head tracking approach is not taken and the technology is instead based on any of the number of full 3D screens that are in the pipeline, more juice will be needed to cope with the complexity.
And my gut is telling me that it will be more likely to be one of these 3D screens that operate sans glasses rather than any of the approaches seen above. Nintendo has made its recent money on accessibility and so something as tricky as image recognition and body tracking could see massive barriers to entry. Anything from an eccentric patterned wallpaper behind the user’s head, a huge hair do or dim lighting conditions could plague such a solution, and as soon as you’re head is not recognised then the whole illusion is shattered. This is a area where the definite nature of the PS3’s glowing wands prove superior.
Of course we’re going to have to wait a further three months for the details on the project. Bar the proclamation of portable 3D gaming and the promise of backwards compatibility with the DS and DSi, details were scant. Speculation as to just what the method for 3D generation will be could easily keep us busy until E3 in June, but if it’s a choice between image recognition on one hand or cutting edge screens on the other then there’s a balance to be struck between accessibility, reliability and cost. Each has been key to their recent resurgence with Wii and DS, but can they keep the balancing act going?