You call it gimmicks, they call it new forms of gameplay .
All in honesty, I think that this is some kind of vicious circle (or chicken and egg) situation. Nintendo went with the wiimotes, the first "gimmick", because they couldn't compete in the hardware depantment and had to try something different with a weaker machine. The problem is that nowadays we don't know if they keep doing that because they think that they need that differential to offset their hardware limitations, or the hardware limitations come because they focus too much on the incorporation of some gimmick.
In any case, I'd like to think that Nintendo is smart enough to realize that with some DLSS capable hardware, they would no longer need to develop their games with two resolutions and setting in mind. Just one set for the handheld mode and then turn the DLSS on when hooked to a TV. That is an improvement that even them would appreciatte.
In any case, Nintendo's decision still depends on what Nvidia offers them because, right now, the only SoC from Nvidia that's already in production and that could be a somewhat viable option for a console, without counting on the Tegra X2, is the Tegra Xavier. The good news is that Xavier uses the Volta architecture that was the first to include Tensor Cores, but the bad news is only has 48 1st gen Tensor Cores, which would be hardly enough to do the task, let alone think about ray tracing.
So well, we're in a situation where, as far as we know, Nvidia doesn't have a SoC for Nintendo with the features we would like to see and, even if they did, we don't know if Nintendo would choose it over a cheaper, but with limited capabilities, chip.
Please excuse my bad English.
Currently gaming on a PC with an i5-4670k@stock (for now), 16Gb RAM 1600 MHz and a GTX 1070
Steam / Live / NNID : jonxiquet Add me if you want, but I'm a single player gamer.