Ok, the first thing is for a company to make customers and money. What good will it do for Nintendo to be held back by the likes of WII U's userbase ? What's more is that is there anything in Nintendo's future software pipeline for it to generate a reasonable amount of revenue ? Nintendo can release as many games as they want on the WII U but what's the point if they just keep performing poorly ? 

The second thing that I want to make clear is that a GTX 970 is manufactured on a 28nm process node, not on 45nm. (Of which 28nm will be the cheapest one for a long time.) Just because a GTX 970 will have aged doesn't mean that the future of microprocessors are going to get any cheaper than it is now. You tell me, what other options are there for effective cost reductions in chip manufacturing ? 

I don't think the Wii U's userbase would be holding back a new console by 2016, I just think it would just not be developed by then. In other words, you can't roll our what you don't have

About the chip, I tought you where talking about the one of the Wii U (whose chip is still produced in 45nm) when talking about lowering costs, my bad.

I do know that the GTX 970 is made in 28 nm. I am however not sure if 28nm will stay cheapest for as long as you think. First rumors about an Radeon 390 (X) being made in 20nm are starting to arrive, meaning that finally someone seems to have gotten the 20nm production under control (probably TMSC, maybe GloFo). IF thats true then 20 nm might not be that far away after all.

And there is another way to reduce costs, but that one is a bit special: bigger wafers. Fabs use 300mm today and 450mm wafers have already been developed, but not yet put into action (and will probably only come after 14/16nm and FinFETs are also implemented due to high initial investments). The cost reduction is however less pronounced than one due to a Die shrink

28nm prcoess node WILL stay as being the cheapest node until next generation lithography and multipatterning technologies like EUV or DSA arrive but that won't happen for a very long time. 

Yes, production of 20nm wafers are going by but it WON'T reduce costs compared to 28nm cause of the fact that multipatterning is required and that drives up the costs of wafers exponentially due to the fact that extra masks are required for the exposure step in lithography. 

The development of 450mm wafers have been HALTED and the gains are questionable for the most part. There maybe some saving when multipatterning is involved however there's absolutely no savings with respect to the light exposure. The costs of replacing tools to suit the production of 450mm wafers are also very high so in the end it may not even be worth it ... 

This figure is a few years old, but it might still be accurate to some point. And it quite nicely shows that Nvidia basically expects the same thing as you. But they also still expect the lowering of production costs per transistor by around 10% per year. And smaller nodes also allow higher clock frequencies, which would further increase their computational power per dollar. I won't expect big jumps but cost savings of around 20% per year still seem likely.

My proposed console on page 3 of this thread was based on the assumption of a reduction of manufacturing costs for the chip components by this 20% per year and the availability of the 16nm node in the middle of 2017, reducing the requirements for cooling of the system.

Tough one. I'd release a New U or a U Advanced or a Ui as an incremental update in 2015. A refined wii u style gamepad with a slightly more ergonomic design and better battery life. Hardware would be slightly enhanced wii u specs, backwardly compatible and most games will run on both. I'd patch all major wii u games to run at a higher res and/or framerate on the new system.

It could be that from now on the console will only ever needs an incremental updates once very few years rather than a big (Next gen) step every 6 or 7. It could keep software support for the Wii U for 7 years for all major releases and far longer for any games that don't require the higher specs.

I am not a fan on fusion only because it wold limit the console capability. Unless they can build some sort of engine to redo games on the fly scaled down to work on the handheld only device. 

Overall close to what I want. Agree about replacing DVD drive. I rather go back to the old cart (well flash cards these days). That way the save files live on the game itself.

As for expansion. I rather they just have internal free slots. Or like previously mentioned in other threads go modular. so you can rip out one bit and replace it with another better bit (mobile phone makes are experimenting with this. 

They could start with 2/3rd the specs you outlined release it cheap and then when PS5 launches release module updates bundled with some of their most wanted IPs. 

The funny thing about the "Nintendo Fusion" rumor is that it detailed a handheld and a home console, not an hybrid of both:

http://vr-zone.com/articles/rumor-nintendo-fusion-nintendos-next-gen-hardware-solution/70439.html

Considering previous comments from Nintendo about merging handheld and console space, I suspect they may just go with the next generation of ARM processors for their next console with a medium-to-high power GPU. The fact that the PS4 and X1 can manage to produce the games they have with such low power X86 processors suggests that by the time Nintendo releases their next console, it may actually be feasible to just use a low power ARM based CPU. Maybe even a custom version of Tegra without the power restrictions (although historically Nintendo have used AMD... who are also looking to adapt ARM in their chips).

That way, they could share console and handheld games or at the very least, have an easier time developing for both systems in parallel.

1. Ditch the tablet.
2. Ditch region locking.
3. Powerful than PS4.
4. Introduction of new controller which will be similar to Dualshock 4.
5. Introduce achievement / trophy.

why would they need one similar to dualshock 4? The pro controller is better feel in your hand. Jus tneed analog triggers.