I don't think it's all that controversial to claim that newer architectures will effectively give certain chips "flop for flop advantages" over older chips. The Wii U is kinda proof of this. I very much doubt we can be certain of the exact percentage in a console environment (compared to on PC) but when we don't know any more it's not a bad place to start comparisons.
As for the 3SM dream, I'm starting to think there's really no chance of that. What would Nintendo truly gain by increasing their production costs by that much? An extra ~20-40% GPU power? Is that really something they're all that interested in? Especially when GPU functions are as scaleable as they are?
I think it's far more important that they've customized the CPU configuration to improve that part of the SoC. Getting an extra ~40% out of the CPU would be a much better improvement as that can definitely be a barrier in even being able to run certain games.
How about everyone relax and have a
nice cup of refreshing orange juice.
Y'all got a couple weeks left.
But I just brushed my teeth...
EDIT: Also regarding Takashi's tweets a few pages back, he may be talking about newly published Japanese patent applications. I don't know why the patent application we saw the other day would say anything about open-world capabilities or screen resolution, but a newly published JP patent might have certain embodiments where the screen resolution is 1080p and the console is capable of 1440p or whatnot.
Just like with the US patent from the other day, these would just represent potential embodiments and not be any indicator of what the Switch product winds up being.