Video games have recently taken an interesting turn. Up until now, developers would drive the industry forward by bragging about things like which console could render the most realistic sweat. But with the profound success of Nintendo’s Wii and its accompanying method of input, developers are quite suddenly less interested in graphics, and jumping on the motion control bandwagon instead.

Both Microsoft and Sony are now following suit with their own takes on motion control. Microsoft has demonstrated a radical new device called Project Natal that requires no controller whatsoever, but instead determines the spatial location of your whole body using a pair of cameras, which can be used to manipulate the game world in 3D, and in real time, Minority Report style. Sony’s offering is a bit more conventional, using an EyeToy as a camera and a new wand controller to provide tracking data for the games. (This works in the opposite way of the Wii, which has a camera in each controller which track a stationary infrared source.)

But my problem with these devices are not their revolutionary nature, or even what kinds of games could be derived from them. In fact, I’ve seen several fantastic examples of motion control done correctly in gaming, and have my full support for those types of applications. But what I have a problem with is the method of control itself, and how the closer certain games get to emulating real life, the less entertaining they become.

There’s an important distinction here – I’m not saying that motion controls automatically make a game less fun. By sacrificing a binary input with one that has a lot more variables, gaming is slowly being replaced by a simulation. Back in the days of the NES (and even earlier, really) there were simple buttons that usually corresponded to simple commands. You push “A”, Mario jumps. You hold right on the directional pad, he runs unsurprisingly to the right. From the perspective of the game console accepting these commands, the difference is pretty much night and day. Mario’s either jumping, or he’s falling down a bottomless pit (who built those anyway?).

When the analog stick was invented, there was suddenly an entire analog range between “standing still” and “full sprint” when it came to moving characters around a 3D world. Although perhaps strange at first for some people, it soon became commonplace, and it worked because people still had full control over their games. If you wanted to stop Mario from running into a wall, you’d just let go of the analog stick, and it would snap back to its resting position. This wasn’t a pointless addition, either. In Metal Gear Solid for the PlayStation, the difference between slowly sneaking along a corridor and running down it as fast as possible was the difference between successful infiltration and painting a giant target on your head. This was soon followed by multiple analog sticks, analog shoulder triggers, and even analog face buttons in the case of the DualShock 2 controller. Even with all these additions, the use of analog made sense, and the gamer was still in direct control of the game.

How quickly can you switch between "stirring" gestures and IR pointing?

But as we now look at the current generation of game consoles, a trend is becoming increasingly clear: while more “natural” control schemes offer a deeper level of immersion, they are also starting to be misused in ways that truly harm many games’ entertainment value. One example of this is Cooking Mama: Cook Off for Wii. On its surface, it looks like a game that could be very fun, as you swing the Wii Remote around to perform all the various steps of cooking a delicious meal. But in reality, the motion detection code used in the game is so unrefined that more often than not it will mistake your gesture as something else, causing all kinds of chaos as you try to correct yourself to make the game happy before the timer runs out. (Actually, part of the reason this particular game is so hilarious to watch other people play is because the controls are so bad.)

But this is really the crux of the motion control dilemma: whether through fault of hardware, software or some combination of both, motion controls in games simply misunderstand the intent of gamers too often, making the experience less immersive each time it happens. Gamers should not have to change how they’re playing the game just so that the console can figure out the correct command – it should just work.

Try pulling off this trick in mere seconds

Another example of bad design in motion control involves SSX Blur. This is another Wii title that has an excellent idea on paper, but ultimately fails in its execution. In the game, you build up a charge by successfully landing various snowboarding tricks, then when the meter is full you can perform an ubertrick by holding the “A” button and performing a gesture with the Wii Remote. But wait – after actually spending some time with the game, I came to realize that it’s not a simple gesture you’re asked to perform, but instead the tracing of a downright complicated shape with the IR pointer – a task that is near impossible to fulfill in the span of just a few seconds before you perform a catastrophic faceplant in the snow.

The other issue, of course, is adding motion control for the sake of having motion control. There are plenty of games out there that have this problem, most frequently at the behest of game publishers for marketing purposes. It’s new, it’s a gimmick, and it will sell more units. I’ve seen plenty of fighting games demonstrated where one of the mechanics involves shaking the Wii Remote as violently as possible to perform a special attack. Not only do I consider this completely unnecessary in the scope of the game, but it also wears out the gamers’ arms to the point of not wanting to play. How is that supposed to get people to play your game more?

Just as expensive as a real fish tank, except you'll never get wet

Taking all this into consideration, it certainly makes me wonder if upcoming devices like Project Natal will become even more counterproductive in the quest for immersive gameplay. There are just so many more factors for it to  contend with compared to a traditional controller, such as variations in the room environment, lighting, body shapes, speed of the motions being performed, and even how a “tennis swing” would look slightly different for pretty much everyone on the planet. Adding all these up, the console has no choice but to rely on a lot of guesswork, which in turn leaves more room for error. And error-prone games are not very fun to play.

It’s not all bad news though. As I mentioned before, there are also many shining examples of motion control done well, and these do give me hope for the technology. Some of the highest-rated games on Wii like Super Mario Galaxy, The Legend of Zelda: Twilight Princess, and Metroid Prime 3: Corruption all use motion control in subtle, but effective ways. There are occasional gestures as well, such as using Samus’ grapple beam by casting the Nunchuk forward – but it’s not overused, and recognized nearly every time.

While I do think the technology will eventually mature enough to a point where motion control is accurate, reliable, and easy to use, we aren’t quite there yet. For that reason I think that many current games would be better off either investing a lot more time refining their controls until it becomes a worthwhile addition, or avoiding it entirely. Not every new game needs motion controls to be fun, just like how every modern game doesn’t need cutting edge graphics to be deemed entertaining. But that’s another blog post entirely.