Will read on my spare time.

I think this is kind of interesting, Kimishima says they're throwing away the concept of hardware lifecycles having to be certain amount of years. I've said before this line of thinking is outdated:

This means that our product lifecycles are not going to last for a set number of years, but will be
flexible enough to change when required by changing consumer needs.
In general, this is the sort of thinking we want to adopt for all our hardware development. We want
to have flexible hardware cycles where the launch of new hardware sets off the development of the
next hardware that will respond to consumer trends.

They also want to sell multiple Switches per household. 

I think we are definitely going to see some changes in Nintendo's hardware model, it's not going to be the same ol', same ol' and Nintendo will sell more hardware as a result of that. Having one basic system for 5-6 years and not refreshing it IMO creates lost sales in the back half of your life cycle, you only really get 3 peak selling years out of 6, which is quite frankly not a great business model. 

Also sounds like no price cut until after they sell 10 million according to that QA anyway. 

They asked fairly intriguing questions this time around? Applause.

There won't be a price cut any time soon. Even if sales drop off, I believe they'd rather sell fewer systems at a profit than more at a loss. Anyway as long as they are barely able to keep up with demand, discussion of price cuts and new models are somewhat premature.

My guess is Nintendo has some kind of contractual obligation to sell 10 million Switch systems to Nvidia/TSMC (the manufacturer).

They got a cheap price on those Tegra X1s on the basis that they would minimum order 10 million of the 20nm Tegra X1. That price is fixed. Once they get past that, I think other options will open up for them.

Would make a lot of sense. Nvidia had a chip that they couldn't find vendors for on the 20nm node no one else is really using, so they gave Nintendo the chip for cheap on the condition that they have to minimum order a certain amount. 10-12 million sounds about right as that would be a minimal risk for Nintendo. 

In the time after fiscal year 2017 is over (ending March 2018) I think you will see a new Switch model(s) using the 16nm Tegra X2. 

Nintendo didn't project 10 million at first though, one would think they would project that amount of sales if they are contractually obligated to order that many units.

However, I guess you could still be right. Especially considering the mentioning of production budgets dropping after 10 million consoles is reached. Could be hinting at a new partnership for parts, as you say.

I think 10-12 million could was the contractual obligaion not just for one year, but for the lifespan or maybe like 3-4 years. It's minimal risk to Nintendo but still some risk since it could have sold like Wii U (no one can be 100% sure beforehand). 

But if Nintendo can clear that entire inventory in just 1 fiscal year, I imagine they could be happy, and they could be free to dump the older Tegra X1 at 20nm (which is a really weird production node) for a newer, more power efficient chip (hello 6 hour battery life).  

Just a guess though. My guess is Nvidia gave Nintendo a nice price on the Tegra X1 but insisted on a fixed order amount that was gaurunteed. They probably had bigger plans for the Tegra X1 and couldn't find vendors for it, maybe they made a deal with TSMC on the 20nm production node and TSMC is saying "dude, what the fuck we built this factory line to make your 20nm Tegra chips, and you're not bringing us any vendors to make chips for". Enter Nintendo. 

I do think Nintendo's strategy with 3DS will be quite interesting from now on. GBA sold fairly strongly in North America even as DS took off because there was a low-income market Nintendo managed to attract. Nintendo must be looking at global totals for 3DS and wondering how they can tap into the markets that bought GBA and DS at this point in their life-cycle, especially given North America sales in excess of 40 million for GBA and just over 60 million for DS (near 60 million for DS in Europe, too). By way of comparison, 3DS is just over the 20 million mark in both North America and Europe.

It's not inconceivable we'll see a mixture of 'late' localisations coming over from Japan, as well as continued support using existing game engines and assets. If the rumoured Superstar Saga remake is true, we're already seeing further evidence of that. We might also see further use of the X/Y engine, with Pokemon Stars hitting Switch and 3DS (or even an additional gen 1 remake next year), or the Link Between Worlds engine being used for a Link's Awakening 3D, or a new handheld title. Nintendo's software development practices have undergone some tweaks to keep content coming, with re-use of engines and assets becoming more common. Fire Emblem is a great example of this, with the Awakening engine being used to create the three Fates versions and now a remake of Fire Emblem Gaiden. But we've also had several Wii U, N64 and Wii ports on 3DS.

The early signs on Switch - cross-platform Zelda, cross-platform Fire Emblem Warriors, re-use of the Splatoon assets/engine, port of Mario Kart 8 DX, rumoured port of Smash, rumoured Pokemon Sun - point to a continuation of this trend. Continued cross-platform development of remakes and remasters from older systems would be an easy way to provide further first party support on Switch and 3DS, even as Nintendo's higher end titles (like Splatoon 2, Super Mario Odyssey) remain exclusive to Switch.

Saved for a read later on the bus, thanks Rol.

Very interesting. Definitely no price drops in the near future, but given the way it's selling that's no surprise.