Microsoft: "Kinect CPU Usage Down to Single Digits."
Microsoft's Kinect (formerly Project Natal) motion control system launches today into a market that's just recently seen the debut of the PlayStation Move. Prior to now, there's been some discussion of whether or not the XBox 360's CPU could handle both Kinect and the demands of modern games; CPU usage in Kinect-capable software was claimed to be in the 10-15 percent range. According to Alex Kipman, Microsoft's incubation director and the brains behind Kinect, the company has managed to reduce the motion controller's demand for cycles down to single digits.
According to Kipman, Microsoft gamers who use Project Natal will never have to deal with slower game performance or stuttery playback due to the performance requirements on the device. As he explains: As we create games, you can think about the platform as a set of paints and paintbrushes...What Kinect brings to the table is a new set of paints and paintbrushes, it broadens the palette and allows you to do different things. Not all features are created equal, you can totally imagine a game that's using practically the entire of the Xbox 360 and still uses identity recognition. You can have a game that uses a small vocabulary of voice recognition that will still have pretty much 100 per cent of the processing."
One of the most interesting comments Kipman makes is in regards to recent releases like Call of Duty: Black Ops. "as much as we like to talk about bits and percentages, you take a game like, I don't know, Call of Duty: Black Ops - there's a significant amount of processing, be it CPU or GPU, that still remains on the table."
The reason that quote is so interesting is that it doesn't match up with what the game industry generally likes to emphasize. Listen to developers talk, and you'd be excused for thinking that while older titles could scarcely take advantage of multiprocessing on the PS3 or XBox 360, modern titles are able to milk those capabilities for all they're worth. The truth is much more nuanced. Multicore gaming has improved, but parallel programming is inherently difficult and subject to diminishing marginal returns. In a way, both Kinect and Move could be seen as a solution to a question that's doubtlessly nagged both Microsoft and Sony: "What do we do with the processing cycles that very few games are ever going to be able to use?"
It's a motion-sensing bar of joy that may or may not actually be fun.
Intel's solution to this issue, for the curious, is Turbo Boost. As CPU core counts continue to grow, Santa Clara continues pushing single-thread performance through the introduction of a feature that doesn't actually require them to guarantee you anything. Since all Turbo Boost speeds are above the minimum they specify, customers have no grounds to complain if they don't get the speeds they want. As tricks go, it's a great one.
Back to the topic of Kinect itself, we're curious to see if gamers will embrace Redmond's movement controllers as readily as they seem to be embracing Sony's. Of the two, we're far more dubious about Kinect for the simple reason that it offers no buttons and does not seem amenable to the simultaneous use of a controller. It's certainly possible to make great games that don't use buttons, but there's no guarantee that Microsoft has found a formula for translating complex game controllers and behaviors into button-less functionality. It's one thing to move a character by pantomime, another thing entirely to contemplate inventory management and weapon-switching in a fast-paced game. God help anyone who ever tries to circle strafe and aim simultaneously.
According to Kipman, Microsoft gamers who use Project Natal will never have to deal with slower game performance or stuttery playback due to the performance requirements on the device. As he explains: As we create games, you can think about the platform as a set of paints and paintbrushes...What Kinect brings to the table is a new set of paints and paintbrushes, it broadens the palette and allows you to do different things. Not all features are created equal, you can totally imagine a game that's using practically the entire of the Xbox 360 and still uses identity recognition. You can have a game that uses a small vocabulary of voice recognition that will still have pretty much 100 per cent of the processing."
One of the most interesting comments Kipman makes is in regards to recent releases like Call of Duty: Black Ops. "as much as we like to talk about bits and percentages, you take a game like, I don't know, Call of Duty: Black Ops - there's a significant amount of processing, be it CPU or GPU, that still remains on the table."
The reason that quote is so interesting is that it doesn't match up with what the game industry generally likes to emphasize. Listen to developers talk, and you'd be excused for thinking that while older titles could scarcely take advantage of multiprocessing on the PS3 or XBox 360, modern titles are able to milk those capabilities for all they're worth. The truth is much more nuanced. Multicore gaming has improved, but parallel programming is inherently difficult and subject to diminishing marginal returns. In a way, both Kinect and Move could be seen as a solution to a question that's doubtlessly nagged both Microsoft and Sony: "What do we do with the processing cycles that very few games are ever going to be able to use?"
It's a motion-sensing bar of joy that may or may not actually be fun.
Intel's solution to this issue, for the curious, is Turbo Boost. As CPU core counts continue to grow, Santa Clara continues pushing single-thread performance through the introduction of a feature that doesn't actually require them to guarantee you anything. Since all Turbo Boost speeds are above the minimum they specify, customers have no grounds to complain if they don't get the speeds they want. As tricks go, it's a great one.
Back to the topic of Kinect itself, we're curious to see if gamers will embrace Redmond's movement controllers as readily as they seem to be embracing Sony's. Of the two, we're far more dubious about Kinect for the simple reason that it offers no buttons and does not seem amenable to the simultaneous use of a controller. It's certainly possible to make great games that don't use buttons, but there's no guarantee that Microsoft has found a formula for translating complex game controllers and behaviors into button-less functionality. It's one thing to move a character by pantomime, another thing entirely to contemplate inventory management and weapon-switching in a fast-paced game. God help anyone who ever tries to circle strafe and aim simultaneously.
