D
Deleted member 11852
Guest
Seconded. We have several platforms that use this model and it proves highly advantageous when it comes to maximising CPU and memory resources across the primary (high level) operating systems.Because that adds complexity to the Hypervisor. Part of the Host OS job is to provide communication channels between the game and system OS, and manage hardware resource allocations. Adding that to the Hypervisor adds a huge surface area for possible Hypervisor attacks. In general, you want the Hypervisor to be as small as possible.
Although conventional wisdom is that more layers is bad for performance, that view tends to be echoed from people who've used systems where OS and hardware resources are managed and re-managed at multiple levels. If you have a virtual OS hierarchy where each system is cognizant the others above it, as I'm assuming the Xbox One is, not only is the overhead minimal but because you have a better picture of a piece of code in the overall system at a multiple levels, debugging and optimising can be much easier across the whole system.