[sonnym]

I think this very much oversimplifies the costs. From a hardware standpoint this is probably mostly correct, but the problem is the sheer amount of research and talent behind it is not remotely inexpensive. I recall Carmack saying, about a year after he joined Oculus, that he thought getting the latency down would be straightforward (sorry - I can't find a source), but it turned out to be a much harder problem than he expected. And, when Carmack is stuck, I cannot but believe there are intricacies that I could not hope to understand at play. The fact that they have been unable to ship a consumer version after all this time, I think, corroborates that this is a much more difficult problem than just throwing some hardware together and calling it a day.