[fmstephe]

I am very curious about the use of a stripped down VM manager that creates and destroys multiple OSes.

"One of the machine’s operating systems boots up and is “created” when the machine turns on, Multerer said. Then it stays operating while the machine is on, and then disappears when the machine is turned off."

"There’s a third operating system as well that handles other kinds of services, such as the TV services. Switching between these operating systems quickly gives the system its instantaneous feel, Multerer said."

It isn't immediately obvious to me why we would want to use multiple operating systems? Is this a kind of super strong separation of processes? Why aren't standard processes sufficient? Is it done for reliability?

I am not criticising, I am just confused and this indicates that there are concerns here that I am not familiar with. Anyone with any insights?

[berkut]

I think it's so that the Game VM can stay constant, and game developers can develop for that for the next ~6 years knowing each XBox will have exactly the same behaviour, whereas the third one for TV services and internet can be continually upgraded and updated, without affecting the game one.

[threeseed]

So it seems as though there is basically a Hypervisor OS and two Virtual Machines. One for TV and one for Gaming. Given that TV and Gaming have vastly different resource requirements and use cases it isn't surprising that you would optimise for each use case.

Also it wouldn't surprise me if Microsoft in the future releases a cheap AppleTV competitor with just the TV VM and a cheap gaming only XBox with just the Gaming VM.

[upthedale]

I think its well known that Microsoft are working on the former, a non-gaming Xbox (save for casual games perhaps), primarily designed for films, TV and music. Unfortunately I can't go digging for sources now, and a cursory search is swamped by results of this new announcement

[zokier]

From technical standpoint I'd guess that separating the game and "everything else" to different OSs allows tighter control of resources, ensuring that games run with consistent performance. It might not be completely unlike the approach taken by RTLinux <http://en.wikipedia.org/wiki/RTLinux>;

[fmstephe]

That does seem like a reasonable explanation. It would be great to hear from someone in the know though.

Couldn't follow that link, due to the '>'

http://en.wikipedia.org/wiki/RTLinux

[adventured]

Maybe it separates the development teams in a way that they didn't have to rely on the whole potato being literally integrated. Microsoft, like many huge companies, is known to be hyper territorial. With the computational resources available, this might have been both a pragmatic and easy choice. Each group perhaps wouldn't have to be so careful about crashing the rest of the system, and it might make it easier to develop and update each segment individually.

This might also make it easier to rip pieces off later, or add large pieces later, without needing to do a comprehensive integration across the Xbox One system, while keeping everything in its own sandbox.

[sycren]

Doesn't this also mean that everything is sandboxed in terms of security?