I was talking about the software side of that, development for that kind of hardware solution could be complex and very unappealing for third parties.

How?  One game, two settings.  Simpler than PC (infinite hardware configs and settings) and the same as XBox and PS are going to be once gen 8.5 systems release.  One game, two settings.

Can we just stop with the "developers can't make games for two different specs OMG!".

Every freaking platform is doing this now.

PC. Playstation (PS4 and PS4 Neo). XBox (XBox One and Scorpio).

We are past the point of debate on this, this is happening and developers are going to adjust to it, and that's basically the end of story on that. Games are going to run at multiple different settings from now on, that's that. NX is doing nothing special or terribly different if it offers an option that basically does the same thing. 

It's not the same as on PC or the other consoles because those are independent machines. PS4 Neo/Scorpio won't be add-ons that connect to the current PS4/X1, they are stand-alone machines with closed hardware.

The PS4, and the same goes for MSoft and its machines, has CPU "A", GPU "A" and "A" amounts of RAM, PS4Neo has CPU "B", GPU "B" and "B" amounts of RAM. Devs have to develop for two different machines with specs "A" and "B".

NX, in the case that has extra hardware on the dock, does not have one set of hardware but two, and devs will have to work with those two sets of hardware at the same time, together.

As before, main NX has CPU "A", GPU "A" and "A" amounts of RAM, NX Dock has CPU "B", GPU "B" and "B" amounts of RAM. Devs, in this case, will have to develop for "A" and "A+B", and that's where I see the problem, because this is something that doesn't happen even on PC.

It's just two hardware configs at the end of the day, there is no discreet "B" configuration in your example, because the SCD doesn't work without the main unit. 

It's likely just literally double the same chip anyway (doubt Nintendo is going to spend a ton of money on a custom chip for a SCD add-on), whereas PS4 Neo and Scorpio are basically entirely different GPUs (Scorpio is also a new CPU, memory, etc. on top of that). The PS4 Neo GPU is basically a Polaris for example. 

Developers will adjust. Because that's their freaking job. 

They are paid contracters, if you work at a McDonalds it's not like you have a say if McDonalds HQ says they're introducing a new sandwhich or salads that will take more of your time to make as an employee. Tough shit. That's the business and you are contracted to do your job. 

Every major game pretty much from now on will require multiple settings for Playstation, PC, XBox, and probably NX too. That's what the industry will be going forward. No ifs, ands, or buts. Even mobile devs basically are required to make multiple versions of every app (iOS and Android at least). 

That's how it goes. 

Seriously, you're really just stretching now.  Just make A+B=C if that helps you understand it.  Companies will have A and B settings for XBox games and A and C setting for NX games.  So very, very , simple.

I actually think Nintendo's setup will be easier for developers than Sony and MS' setup.

Nintendo will likely just use the same exact chip if they go the SCD route or something similar to what Apple does with the Apple A9 (for iPhone) and Apple A9X (for iPad).

Very easy, they're not going to spend a ton of money on two seperate chip designs.

Sony/MS by doing what they're doing are changing things up much moreso, the PS4 Neo is basically a Polaris chip, a chip that didn't exist when the PS4 launched. The Scorpio is basically entirely different hardware full stop.

Wouldn't surprise me at all if making NX games at different settings is much easier than what Sony and MS are asking of devs (not to mention some games like RE7 also needing to be optimized for VR).

In general I think the reaction to NX will be a net positive from devs. In the past to support Nintendo, a company like EA had to support the Wii U and 3DS too for example, and that probably was a pain in the ass too. Now they can just make one version, make a few settings changes, and offer it to the entire Nintendo buyer base, this can all be handled by one team instead of two. 

Developers don't have to adjust, specially for a from Nintendo, for which they haven't worked for in two gens.

And you're missing the point I'm trying to make. If PS4 were a car, PS4Neo would be another, more powerful car. You know how to drive both of them, but you can't drive both at the same time. With the configuration we're discussing here, NX would be like a motorbike and the dock would be another bike, and you're trying to ride both of them at the same time. 

Sure, Nintendo will provide software tools to help development, but it will still take more time than usual to adjust, and that means devs complaining about being hard to develop for and publishers talking about being expensive.

But you're still dependant of two sets of hardware. It's not A+B=C, it's more like A+B=AB.

You are right. You aren't a computer expert.

FLOPS aren't FLOPS.
You literally have different types of Flops.

You can literally halve your Flop speed by using one type of Flop or double it with another.

If you disagree with this, then that's your circus, not my problem, but I would prefer to have accurate information so people can build accurate opinions on the information we actually have.

The thing about Pascal and Tegra is that, Pascal was an evolutionary transition from Maxwell, all the big fruit-picking efficiency gains, nVidia did with Maxwell.

The reason why Pascal on the desktop was so impressive was because nVidia wen't from 28nm down to 16nm FinFet and reworked the chips to drive up clockrates, this cannot happen with a Pascal based Tegra as Tegra isn't being taken from 28nm to 16nm FF.

I don't doubt that a Pascal based Tegra is going to be fairly potent, for a mobile chip, but it will still fall far short of the Xbox One in capability and it will unlikely be the same kind of jump we saw with the transition to a Maxwell based Tegra.

I'm well aware of what you said. I just assume you were being sarcastic, otherwise what's the point in arguing in the first place?

The thing is though, if nVidia pulls off code morphing in Denver 2, it might be fully compatible with x86.
Thus if people's rumor/assumptions of a complimentary computing device via a dock is to be believed, it wouldn't matter if it was being driven by AMD hardware.

I don't think you could look at the consoles and think of them as Desktop PC hardware equivalent (Aka. Neo as Polaris).
The Neo will likely be based around AMD's Graphics Core Next 4.0 hardware, like Scorpio. Thus on a technicality they are both derived from the same graphics technology, which is also the technology that will be pushing the next generation of APU's on the PC.

The main question is how many GCN pipelines will it have, and the clocks?

The positive reaction from developers is a good sign, but didn't we also have positive reactions from developers prior to the Wii U launch?
Just food for thought.

You just keep repeating yourself, regardless of what people say has anything to do with the value of flops.