Kinect Died in the Uncanny Valley

The problem was always that Microsoft took the Kinect too seriously. It's very futuristic, sure, but it's also just inherently silly. No matter how hard you try, the Kinect will always only be capable of approximating an action the same way any other controller does.

There needed to be more games like WarioWare: Smooth Moves that literally demand you let go of inhibitions and just embrace the silliness of it. Like, look at that picture at the top of the article? Who wants to run in place in front of their TV to simulate running? I can run in place without the Kinect, or just go outside to run for real. Make me do something silly and fun instead.

Kinect Sports has two minigames in it that are fantastic at that: one has you returning as many ping pong balls as you can with both your hands as paddles, and one has you dodging around the play area to avoid volleyballs thrown at you. In both games, if you're just kind of self-consciously half-assing it the way most people do with motion games, you're going to lose immediately and I'm going to call you a lame stick-in-the-mud. You have to be willing to move.

And therein lies the problem: most Kinect games don't require you to move. They want precise tilts and hand movements so you can stand in place and look cool and that shit just doesn't work on Kinect. It doesn't work well with subtlety. It's not a controller. It's a camera.

Worse though, is that Microsoft took Kinect so seriously that even Kinect Fun Labs is lame. They should have opened that up to indie developers, made a lot of bite-sized, experimental minigames to inspire other developers, and ensured the whole experience would always be free of charge. But nope.

So yeah, I agree with @patrickklepek. It's a bummer that Kinect is basically dead in the water now. It's smart for Microsoft to jettison it since they had no idea what to do with it and they need to compete with Sony something fierce, but it's disappointing nonetheless. I was just playing that Kinect Sports dodging minigame (I think it might be called Body Bash or something) last week and I still had a blast.

Don't you think that a working version of motion tech is essential for really immersive VR? The fact that you can't see a keyboard or controller while wearing VR goggles really begs for integration of accurate motion control. Of course it is possible to navigate using a controller, as we don't often actually look down at the controller while we are playing, but being able to use our voice and movements to play would be so much more appropriate to a virtual world.

My dream for the next set of consoles in 10years time is that they all come with integrated motion and VR. Kinect and its like still has a big role to play - it is just very early. I think we will look back on today's motion control the same way we look at the old Atari consoles.

I think part of the problem is that developers were so dead-set on discarding the controller in favor of the Kinect rather than use them together. I think the Kinect-controller combo would have made for some really interesting experiences.

Examples:

1 - Update Road Rash. Allow the throttle and steering to be controlled with either the left or right thumbstick/trigger/shoulder button. Combat would be controlled using the Kinect. Swing your arm to whip someone with a chain. Kick your leg to boot an opponents bike into oncoming traffic. The game would be a blast, and its because of the combined control scheme. Think of the alternative. Leaning forward or back to control speed would have to rely on the Kinect to sense depth accurately and quickly, which it can't do, at least not for a bike racing game. Holding your arms out straight in front of you is an equally shitty control mechanism and makes you want to stop playing after only a handful of 4 minute long races. Anyone who played the Podracing section of Kinect Star Wars knows this.

2 - What about a game that actually came out for Kinect? Fable: The Journey was a blast to play. With a few simple gestures you could whip a Balverine up into the air, launch a handful of spears into him and then rain fire down on all his friends. Thanks to the Kinect controls some fairly simple combat could feel really engaging and fun. The problem is people still wanted to explore the world of Fable: The Journey and Kinect controls prevented this. Sure, you could steer the cart and sometimes that meant choosing a direction at a fork in the road, or whether or not to stop for an optional encounter, but that's a far cry from previous games. A combined control scheme could have fixed this. World exploration and interaction could be handled with a controller and combat scenarios with a Kinect. You'd have to find a way to handle the transition between the two, but it's a hell of a lot better than having extremely limited exploration.

I know that a few games used the Kinect's mic to supplement the game play, but that hardly uses the technology to it's full potential. I'll admit, it was pretty cool to shout squad commands in ME3, but that could have been accomplished with a headset.

I can't help but think that with a slogan like "You are the controller" Microsoft is responsible for pushing developers away from a combined control scheme. And if that's true it's a real shame. I can't be the only person that sees the potential. Instead we got a slew of dance and fitness games.

Sorry about the rant.

As repeated a couple of times: Mimicking running in place to tell a game "I am running" or hopping in the air to tell a game "I am jumping" is the definition of "uncanny valley". The more closely one tries to mimic running and jumping for a game without actually running or jumping, even with the game showing all indications you are running fast and jumping over stuff, the more you instinctively know you really aren't running or jumping over anything.

I think the kinect and it's flaws aren't related to the uncanny valley. It's an interesting analogy to make, but to say "The uncanny valley is when technology is able to almost mimic reality" is a very imprecise definition of the uncanny valley. The uncanny valley is the idea that there is "a dip in a graph of the comfort level of humans as subjects move toward a healthy, natural human likeness described in a function of a subject's aesthetic acceptability."

Kinect doesn't cause more revulsion by getting closer at reading a user's movement, and it's a very odd connection to make that seems based on a shallow understanding of the uncanny valley concept.

Talking about cool things the Kinect could be used for, here's Minority Report but in real life, made by some fellow students at Imperial in just 48 hours.

If Kinect ended up being like that youtube video I would want it. But it didn't.

Wasn't terribly surprised by the announcement. I think consumers are, at this point, expecting this kind of retroactive dismissal of previous declarations, e.g. used games, always-on, Kinect. I think if Microsoft had entered the console market for the first time this year, the product would have found more success solely due to the lack of the 360's shadow bearing down on the Xbone, but unfortunately Microsoft followed up a great product with one that fell below most expectations, and has winded up paying for it.

The big clue for anyone, including Microsoft, that something was amiss with Kinect as the fact you had to clear a large space in your living room to use it. Even with the Wii, you had to shuffle people around so the player was standing in front of the sensor bar but actually rearranging your living room to have a chance of getting the thing to work is a stupid and unrealistic requirement.

Great write up!

Thought this idea was dumb in 2009 when the 360 was my primary gaming device and, to my mind, Microsoft was in a position to dictate the future of gaming I was excited to see what they were going to show at E3. Five years of them trying to cram crap I don't care about down my throat later and the last game I played on a console was Mass Effect 3.

There's no point in thinking Kinect can turn itself around anymore. It's had year after year to catch on. People know about it. They don't want it. All Kinect is in the end is one part of the story of how Xbox was on top of the gaming world and then shot itself in the face repeatedly.

Not entirely sure that Patrick understands what the Uncanny Valley is. :)

Ahhh ok I see now, thanks. The article suddenly makes more sense to me.

While I see where you're coming from, I can't get behind this argument of none of it being Kinects fault and lying down so much of the blame at developers feet. It's bad and flawed so take it as is, is not a good enough excuse to whitewash it.

Great article @patrick, refreshing change from. the dancing around the flames of almost every other site out there.

Getting so tired of the close-minded, numbers focused nature of 'enthusiast' gaming.

I believe the point is that instead of embracing its limitations and doing fun, inventine things with its technology, Kinect tried to do and showcase 1 to 1 interpretation of movement, which is something it's not equipped to do.

Microsoft made the same mistakes Nintendo made when trying to market the Wii U "Gamepad". MS gave NO reasons to include the Kinect; 6 months after the Xbone release and there's no games that show off why the new Kinect is worth it. Like Nintendo, MS needed that "Wii Sports" to show off the Kinect's capibilities. As of now, it's just a novelty that Patrick is a little too attracted to lol.

I don't understand the uncanny valley reference/connection at all? Can anyone explain to me what Patrick is talking about in that regard? (I understand the concept, just not as it's implied here)

All I know is Microsoft need to come to E3 with a very clear branding of what the Xbone actually is at this point, and sell that Idea clearly, confidently and convincingly.

At this point I feel like they have backed on everything they originally tried to sell. I'm worried Microsoft is going to find themselves in some sort of Wii U position where they backpedal to appease the hardcore, only to find they lose both audiences.

Turning their back on Kinect is a terrible tragedy, but then again if they are never going to manufacture or inspire anything compelling for it - they might as well stop supporting it, really we are lamenting Kinects failure as a whole and not just MS scaling back focus from it.

I feel in a lot of people minds when you factor in the original kinect and it's promise, this adds up to a failure of a reboot that was a failure (at least critically). It feels like Microsoft has been putting it's eggs in the Kinect basket for a long time expecting it to sort of pay itself off magically. I think the disappointment comes from seeing such a wasted promise.

I also feel they sold the Xbone with Kinect being the main component making it any different from PS4. To now show a lack of support for that, feels like a slug to early adopters who bought that dream.

@patrickklepek typo in paragraph after the Kinect hack videos 'developers who wanted o try'

great article though, as usual :)

I went out of my way to get a xbox one without a kinect lol. As it turns out now so will everyone.

I bet Kinect 5 will be great. It will probably be a VR headset by then :D

I feel sorry for those who believed in all this bullshit Kinect crap.

From day one it looked like a gimmick that was trying to fool stupid people. I am really glad it is dead now, and people can concentrate on real games.

I once modified a pair of led hardware glasses into a wii lightbar to do the wii version of that VR head tracking exercise. It was cool, but it's way more impressive when you record the effect than when you witness it. Recordings have zero depth perception, and human vision has some amount of depth perception even with only one eye open. If you have a wiimote and your computer has bluetooth capability, you can do this by taping a wii lightbar to the bill of a baseball cap.

[1/3] Also: multipart twitter messages are dumb.

[2/3] Twitter is a wasteland and watching someone try to form a thought or an argument there in multiple parts is like watching a straightjacketed lunatic run into the wall of his cell over and over.

[3/3] balls

As much as Microsoft oversold it, I doubt that many developers or consumers would be enthused by them announcing it with a more realistic portrayal. "It detects your movements, kind of! But only as long as you're in a properly lit space and the correct distance from the camera! And I mean, the controls aren't nearly one to one and..."

internet rage...you are the best. From the vocal minority at sonygaf; and Microsoft's apathy.... A group of people who had mock rage, and never were going to buy the system, to the developers who never showed a reasonable use of the tec. I salute you.

@patrickklepek: IF and when VR takes off and when more people would have it in their own homes, could we see a revival of the Kinect perhaps?

I think that's over thinking it, much more basically, the Kinect is too laggy and inaccurate. It really does not work at the most basic things it should do, it doesn't live up to its most basic technological promises, and everybody involved tries to pretend like it's not totally busted.

It was an "emperor has no clothes" kinda scenario where all the people making games for it, and microsoft, were marketing an experience that simply was impossible with all that lag and glitchiness. But gamers saw through it right away, moments after trying it.

Real talk, Kinect was a shitty, half-baked piece of hardware that Microsoft put out solely to cynically capitalize on the Wii motion control fad* and the only good thing Kinect ever contributed to the world of gaming is Giant Bomb Quick Looks of Kinect games. Good riddance. (Although I will miss those Quick Looks.)

*Yes, I said fad.

The Kinect had potential. As Patrick points out, Microsoft didn't frame the technology and its potential in the right way.

Buying patents isn't proof that a product is currently in development. It just means that they've claimed ownership of specific tech ideas that they may or may not find a viable use for. And outside of Xbox, Microsoft's forays into hardware have not been particularly successful. They're already behind the game on wearable tech, which in itself is still a niche market in its infancy. There's no guarantee that any of it is going to take off or that even the more prominent devices like Google Glass will find commonplace acceptance.

@hailinel: MS bought 150 million in patents, in relation to AR, and wearable are a market that is more than likely going to be a good place to be for a company like MS.

Eh, sure. I just feel there's a separation between games as interactive physical experiences and a form of relaxation, where you slump in front of a TV or monitor with a controller or mouse and keyboard. Minimal movement - maximum control.

The other is the jumping around, shouting and moving kind of gaming. It's a different experience, and might be enabled by whatever motion control, voice activated gimmick that's popular for the day.

And personally, if I reach a point where I want to play virtual ball toss, I'd rather actually go outside and toss some balls. Not digitally, in my living room.

Been saying it for the longest time, developers should not use the Kinect as a controller, it should be used as an accessory to enhance immersion and experiences.

The 360 kinect definitely had some untapped potential. I mean, look at what this company has made.

basically they provided low cost motion capture to animators and indie developers

http://ipisoft.com/

cant wait to see what they can do with the xbone's kinect

@clonedzero: I can see where you are coming from, but James (the writer) is a developer and a teacher (I think at Digipen,) so I think he at least somewhat knows what he is talking about.

@metalsnakezero: I kind of hate those smug bastards. Everytime i watch one of those videos it comes off as smug preachy bullshit. Maybe thats just me though?

Extra credit made a video about how it is a very uncanny valley of motions(or input controls).

From the video it is a very a human interaction problem and to get around it we need to think what motions do work before the idea of a game, not the game then motion.

Hey Patrick.

It feels good to say this.

Great article.

As long as I keep shouting at my tv Kinect will never die!

The problem with motion control is obvious: SYNTAX

These two generations of motion control devices have highlighted that people don't yet know how to interface with a device that is reading a fairly arbitrary input.

I remember playing SSX Blur for the Wii when it came out, and a lot of players complained about the lack of control over the tricks (drawing shapes in the air). Yet with a small amount of practice you can learn what the motion controller is actually trying to read (or what the developers have designed as the correct "activation" of a move) and I was pulling off moves with no problem at all.

Developers need to stop trying to shoehorn the device into a "perfect simulation" scenario, and instead recognise those flaws and ONLY design functions that have an almost 100% success rate. THEN they need to figure out a way to teach people those inputs so that players don't suffer from the frustration of a, say, 60% success rate.

It's almost like Microsoft is giving us a regular sedan with all-wheel-drive and assuming that because it uses all wheels, it can go over rough terrain like a 4WD truck.

Motion control is not dead. Microsoft just suck at communication.

Quite simply, the Kinect use limits the game genres and possibilities. You can't turn back or you won't see the screen anymore, just that eliminates : FPS, TPS, action/adventure, football, fighting ... And even if asvertised I'm sure a driving simulator wouldn't respond as well as a controller : you need extreme precision when you turn or accelerate or brake.

And what about controlling your TV? Damn get a remote/controller it'll be 10 times easier.

I don't know that I agree it wasn't Kinect's fault. Oh, I know there are limitations. And I completely agree that Harmonix showed that the key is knowing how to make a game within them.

But the onus of those limitations is partly Kinect's fault. Sure, there are plenty more, like it's going to be hard to get gamers to play Kinect games for hours on end like they do controller games, simply because it's more tiring to play Kinect games.

In the end, it's kind of the party game problem again. The gamers who buy the most games, the "hardcore gamers" to use a phrase I hate, want games that are deep and complex. They don't want party games, they want a game they can be challenged by and master.

And it's just supremely hard to make a game that can be challenging and have a lot of depth and still run it with Kinect because of the fidelity limitations.

That means Kinect not only requires knowing the limitations of the controls, but also the limitation that you will have to reach an entirely new audience, because the audience that makes most successful games successful is going to spurn your game.

Wii did it. It absolutely nailed the "mile wide and an inch deep" needed to turn a lot of non-gamers into buyers. But it's supremely hard, Kinect didn't manage it, Move didn't manage it and Wii U didn't manage it either.

In the end, the product's price profile and appeal profile means you are aiming at a very small target. And MS didn't hit that target.

Even if the existence of that document is accurate, you have to remember that it's also years old. There is no confirmation that the project it's associated with is even in the works anymore. It might never have progressed beyond the R&D stage. And even if Microsoft is still working on AR glasses, there's no guarantee that the tech will live up to the promises of the initial documentation. I mean, just compare Microsoft's promises of Project Natal with the reality of Kinect. One is sunshine and rainbows of a perfect, fluidly working technology. The other is a crapshoot of spotty voice recognition, imprecise motion detection and feedback, and restrictions of environmental necessity. The reality was never even close to the initial promises.

I don't really care about the marketing aspect of it, what people imagined looking at it was to perform an action and that action taking place in the game, and that seems to have never come true to a satisfying degree. If Happy Action theatre and Dance Central managed to cover that up, good for them, but those games were "do something for a few minutes and we'll hide how stupid you look/we'll make it more explicit". That's not fun for me save for messing around for a few hours/looking at you guys mess around for a few hours. Buttons are more useful for every genre, and no new ones cropped up. So for me, it's hard to feel sad about the Kinect, or any motion controllers. At least in the way you intended, Patrick.

I thought the people who want these kinds of games don't really get many games. A lot of people bought a Wii, but that thing was collecting dust ages before the Wii U arrived. I don't think those people are going to buy a ton of Kinect games, when they're just an aside you do with a group of friends you have over maybe once or twice instead of watching a movie or playing singstar or whatever. I guess I don't have any statistics to back this up, but I don't think Nintendo would make their new controller this much closer to a classic controller if the wii was still doing great for them and all their new customers kept buying games for it.

Oh please! Let me introduce you two: Rider meet horse, horse meet rider.

Oh please! Let me introduce you two: Rider meet horse, horse meet rider.

Oh please! Let me introduce you two: Rider meet horse, horse meet rider.

Oh please! Let me introduce you two: Rider meet horse, horse meet rider.

I think one big problem is they keep trying to get traditional gamers on board instead of shooting for new audiences. The Wii didn't care if Call of Duty people were on board, it just was what it was and sold to who was interested. Kinect kept trying to force itself into Mass Effect or use the Fable name to try and get me to like it. I was never, ever, going to like it, but the secret is that you don't need me.

The whole video game industry has a problem with budgeting and marketing toward people who want the experience you're making, rather than some mythical "gamer" they have in their heads.

The xbox one's Kinect is going to find it's use in experimental and new technologies as well as VR for years to come.

I was just never interested in kinect to start with. Nothing would have changed my mind. Sure they did voice commands on some later games only but had those on some rainbow six games on the first xbox. Could have done those without a kinect on the 360

Sorry, I don't buy it. Microsoft isn't "partially" to blame. I've seen plenty of articles over the past few days that have blamed both consumers and developers for the death of Kinect, but the blame lies right at Microsoft's feet. You have a device that had great potential as an enthusiast / hobbyist piece of hardware and they squandered it by marketing it as something it's not.

Both developers and consumers bought into the promise of the Kinect, took their lumps, learned their lessons and moved on. Microsoft making the decision for everyone that no, we're not done with the Kinect, instead of letting the market decide was incredibly arrogant and nearsighted on their part. They've got no one to blame but themselves.