The technology on-the-go has to catch up to cloud streaming before I can see this taking off. But I'm interested to see how a powerful 5G phone like the Samsung Galaxy series for example could handle it. Will be awful for battery life though, and I don't see many casuals/commuters going for it - they're happy playing cheap games likes Candy Crush for 10 - 20 minutes at a time.
Agreed, it has to catch up, but I don't think the advances in streaming and lag-reduction will allow the business model to leapfrog traditional gaming.
Let's imagine that we reduce the input lag by 80%, we find tricks to compress and stream textures better, etc
What would prevent these same techniques from being used to improve
local game performance? Nothing. So the streaming services reduce their lag and improve their textures, but the same concepts get used on console / PC games and improve them as well. I am sorry for oversimplifying a complex service, but at the end of the day, it's all hardware, the main difference being that one is under your TV and the other is a virtual instance on a server 350 miles away, routed through your local internet provider's network hardware.
Yeah, we can definitely improve it to be "good enough", and I'm all about the market viability of "good enough" products, but if the best-case-scenario cloud won't catch up, then the mobile performance is never going to catch up either.
The unspoken truth about this whole situation is that
gaming has been dumbed down significantly. A lot of traditional genres cannot work on streaming due to the lag, but if you're satisfied with aim-assistance and low intensity demands, then streaming will be fine. The average CoD player won't notice but the average fighting game player will.