By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Nintendo Discussion - What If The NX Console Is Portable Too?

Miyamotoo said:
potato_hamster said:
Miyamotoo said:
potato_hamster said:
Soundwave said:

I guess another way of doing it is what if Nintendo made a different console for different regional tastes?

I'm going say Nintendo chooses to be a little bold and uses AMD's 14nm FinFET process which is supposed to be firing on all cylinders by next year. So lets assume 70 GFLOPS/watt.

NX Pocket Handheld - 350 GFLOP. 960x540 4.88-inch LCD screen. $199.99. Standard Nintendo option, good for kids, people who want a DS/3DS successor. 3GB RAM. Cheap screen but does the job. 

NX Mobile Console (Japan) - 600 GFLOP (on battery); 900 GFLOP (plugged in). New Console Concept. Has a 1280x720 7-inch LCD screen. Can stream wirelessly to the TV via HDMI receiver (sold separately). Form factor may look like a Wii U controller or maybe a Surface tablet (kickstand display, play with controller). Not designed for pockets, but easy enough to take in a bag or carry from room to room. 6GB RAM. - $299.99 MSRP

NX Home Console (US/EU Markets) - 2TFLOP console (@28 watts), 1TB internal HDD, your standard Nintendo console. Games run at the full 1080P resolution for TV. 8GB RAM. About the size of the OG Wii (no disc drive). $299.99 MSRP.

All three versions could be sold in all markets of course, just the focus in the US would be the home console, in Japan the mobile console is the console made for Japanese tastes, and you have the standard Nintendo portable option for the typical kid market, budget parent, and the gamer who values portability/pocket-ability.

The only thing is I don't think the NX Pocket would be able to run all games (though at 350GFLOPS for only 540p render is pretty beastly still), but it would be able to run most third party games with scaled down effects and probably all Nintendo games at the lowered resolution, plus virtual console games and perhaps Android app ports. Ideal for getting kids with budget strict parents into the NX ecosystem and playing Splatoon 2/Mario Maker 2.0/Dragon Quest XI, then later on they can start bugging mom/dad for one of the console versions. 

More of this complete and utter nonsense. Games don't just "scale" like you think they do. It's not like PC games where you can just make the game run decently in on a variety of hardware specs and just keep driving up the minimum requirements until the game runs okay. That is not how it works. The specs don't change.  Video games cannot "just scale" on consoles. It never has and it never will.

But let's just assume it does.

Game engines still have to be optimized for each hardware spec, or mode. Every single one of them. Games have to be tested for each hardware spe, individually. Instead of each developer requiring one dev kit, they now need three. Now instead of taking an hour to make a simple adjustment and test it on PS4/XB1/Wii U, they need to test it on PS4/XB1/NXA/NXB/NXC/NXD Wonderful! Awesome. Now the developers need to spend even more time testing things before their code is submitted. Did I mention this process can happen hundreds of times per day? Or at least it did. You just took 1/2 hour to test to make sure your code didn't break the build and turned it into an hour-1.5 hour process. Never mind the added cost and time needed to test the game.

Do you ever want to see a third party game on a Nintendo console ever again? Because the cost of developing for all those different skus and modes drives development cost through the roof, and I do mean astronomically high.

This will never happen, unless you want the NX to fail harder than the Virtual Boy.

Actualy we already have one intesting example, Monster Hunter Ultimate 4 that works on New 3DS but also on 3DS too, difference are better textures (and probably better FPS) on New 3DS.

We already have games that are developed for 5 different platforms, PS3, Xbox 360, PS4, Xbox One and PC, and we have some games that are developed for 8 totally different platforms (like Lego Jurassic World), all above mentioned plus Wii U, 3DS and Vita.

Very important thing about NX (and probably all future Nintendo hardware) is that it will have same architecture, so that means same development kits, same assets, same way of developing and programing, litarly devolpers will devolp in same time game for NX handheld and home console very easy and fast, incomparably easier than when they devolped games for PS3, Xbox 360, PS4, Xbox One and PC.

Congrats on the Monster Hunter example. I take it you've also never made a video game before? They did a similar thing with PSP models. Guess what? developers had to pretty much treat both of those settings as two different platforms, and as a result, most developers didn't really bother with the added performance boost. It jiust wasn't worth the hassle. There's also texture differences between multi-platform PS4 and Xbox One games, and it requires making platform specific coding and testing on each platform to do such a thing. This is literally an example of the additional cost I am talking about. They didn't just make it work for 3DS or New 3DS and assume it worked on the other spec. They had to test it for each spec. Unfortunately I'm not sure if the "new 3DS" requires a brand new dev kit, or a firmware update for the same kit, but if its the former you can imagine the extra cost involved. Either way, you're talking about the same inputs and the same outputs, Both are handhelds with one only marginally faster than the other. That's completely different than a home console and a handheld console, both of which having different inputs (stylus for handheld, two extra shoulder buttons for home), both outputting at two completely different resolutions, both made to be viewed at completely different distances. They would literally need to be treated as two entirely separate platforms. The same code more or less runs on both platforms? So what? You have practically the same code running on PS4 and X1. The game code is not the problem. Its the engine that's the problem.

Yes, we do have games that are currently developed for 8 different platforms. It's not very common because its very very expensive to do. It happens very rarely because puting in any one change to the code would literally take a couple of hours of verification by a single developer (assuming they have one team working on all platforms) or they have multiple teams taking care of multiple platforms (which really drives up costs, the most expensive part of making video games is paying the people who do the work). Now you want to add more platforms to this, and drive up the costs even further. Again this is what I'm talking about. Just because some developers are willing to put out a game on 8 different platforms doesn't mean that the average publisher which is releasing games on 2-3 is willing to make the plunge to start developing on the equivalent of 5-6. That would literally double the cost of a lot of game development areas.

As for your last paragraph, that's literally fanboy level thinking. You simply do not know any better. A multispec platform doesn't have the same architecture, it has similar architecture. Take two processors, one is literally half the specs of the other in every way but "the same architecture" as the other. Do you think that the "half processor" can process the same command at half the rate? The answer is not necessarily. It may be the case for 95% of the commands you give it, but the half processor might take 2-3 times as one would expect to process the other 5% of commands. It depends on if that processor running at half the specs introduces bottlenecks not seen in the "full processor" because it has half the resources available to compute it. If that turns out to be the case, guess what? You need to develop a processor-specifc work around. These are the type of things I have literally lost sleep over trying to fix in some sweaty office at 3 in the morning hours before a deadline. Therse are real problems in the console video game world.

There is no reason to think that developing for these hypothetical NX specs would be any easier than developing for PS4 and X1. None at all. Both the PS4 and X1, have practically the same architecture as well. In fact the differences between the PS4 and X1 in terms of architecture would be a pretty similar comparison between the supposed different specs of the NX, except with the NX you'd have diffferent inputs and outputs per spec, which would make it more complicated to develop for a multispec NX. Again, best case scenario it would be like developing for PS4 and X1. Also about those developer kits. Do you think these things are made of hopes and dreams? Making a dev kit that can handle the different inputs, display at the different outputs, and "scale" down to different modes that replicate the actual hardware exactly isn't exactly a small feat, and it would be incredibly expensive to make.

I will keep stating and re-stating this. It is not that simple. At all. Everyone that thinks this is feasible glosses over major, major roadblocks as if they are arbitrary. They aren't, I assure you.

All Nintendo talk about unifying hardware and software teams, same architecture, same OS, same platform, they even mentiond example of iOS and Android, is beacuse there will be no need like before to develop game separate for handheld and separate for home console, they will develop one game in same time for both consoles.

If Nintendo will benefit from that, there is no reason why 3rd party couldn't too.

 

Nintendo reorganized its R&D divisions and integrated the handheld device and home console development teams into one division under Mr. Takeda. Previously, our handheld video game devices and home video game consoles had to be developed separately as the technological requirements of each system, whether it was battery-powered or connected to a power supply, differed greatly, leading to completely different architectures and, hence, divergent methods of software development. However, because of vast technological advances, it became possible to achieve a fair degree of architectural integration..

For example, currently it requires a huge amount of effort to port Wii software to Nintendo 3DS because not only their resolutions but also the methods of software development are entirely different. The same thing happens when we try to port Nintendo 3DS software to Wii U. If the transition of software from platform to platform can be made simpler, this will help solve the problem of game shortages in the launch periods of new platforms. Also, as technological advances took place at such a dramatic rate, and we were forced to choose the best technologies for video games under cost restrictions, each time we developed a new platform, we always ended up developing a system that was completely different from its predecessor. The only exception was when we went from Nintendo GameCube to Wii. Though the controller changed completely, the actual computer and graphics chips were developed very smoothly as they were very similar to those of Nintendo GameCube, but all the other systems required ground-up effort. However, I think that we no longer need this kind of effort under the current circumstances. In this perspective, while we are only going to be able to start this with the next system, it will become important for us to accurately take advantage of what we have done with the Wii U architecture. It of course does not mean that we are going to use exactly the same architecture as Wii U, but we are going to create a system that can absorb the Wii U architecture adequately. When this happens, home consoles and handheld devices will no longer be completely different, and they will become like brothers in a family of systems.

Still, I am not sure if the form factor (the size and configuration of the hardware) will be integrated. In contrast, the number of form factors might increase. Currently, we can only provide two form factors because if we had three or four different architectures, we would face serious shortages of software on every platform. To cite a specific case, Apple is able to release smart devices with various form factors one after another because there is one way of programming adopted by all platforms. Apple has a common platform called iOS. Another example is Android. Though there are various models, Android does not face software shortages because there is one common way of programming on the Android platform that works with various models. The point is, Nintendo platforms should be like those two examples. Whether we will ultimately need just one device will be determined by what consumers demand in the future, and that is not something we know at the moment. However, we are hoping to change and correct the situation in which we develop games for different platforms individually and sometimes disappoint consumers with game shortages as we attempt to move from one platform to another, and we believe that we will be able to deliver tangible results in the future. 

What we are aiming at is to integrate the architecture to form a common basis for software development so that we can make software assets more transferrable, and operating systems and their build-in applications more portable, regardless of form factor or performance of each platform. They will also work to avoid software lineup shortages or software development delays which tend to happen just after the launch of new hardware.

That's talking about making games easier to port from handheld to home console and vice versa. However, at the end of the day you are still making two distinct builds have to test for two distinct platforms. This is a solid idea. It makes it a far easier decision for Developers to support both the NX Home and NX Handheld instead of one or the other. It will lead to more games on both platforms.

However, let's be clear, this is far and away from what others have been saying. This is describing two distinct platforms. Not one platform in two form factors. Sure it'll be easier than say it is to make a Wii game run on the 3DS, or a PS3 game run on the Vita, but it's still not as easy as "Make one game that plays on two different form factors of the same console". You still have to make device-specific optimizations. You still need to test on every platform. Let's use an example Nintendo discusses. iOS - not every piece of software available for the iPhone is available for the iPad and vice versa. There are reasons for that. Even though they share very similar operating systems, and the same development API, they are two distinct platforms. You need to modify your iPhone app to work properly on the iPad. You don't just make an iPhone app and it "just work" on the iPad. Not even close. Even if you're just making an iPhone app. If you want your app to work well on the iPhone 4, 4S, 5, 5S, 6, and 6S you need to optimize the app for each device and test it on the 4, 4S, 5, 5S, 6, and 6S. You need to have all of those devices to deploy your app to. You can't just make an app for the iPhone 5 and assume it will work just as well on the 4 and 6.

The same will go for the next Nintendo platform. Even if the development environment is more unified, even if the different devices share similar architecture, they will be treated as two separate devices, have separate software, and developers will not be forced to develop for both simultaneously.

Thank you for providing further concrete evidence to support what I've been saying all along. I appreciate it.



Around the Network
forethought14 said:
zorg1000 said:

Here is a quote from Miyamoto from last year, I would post the whole article but my phone sucks at copying links.

"What I can say is, certainly, within Nintendo the fact that our development environment for our home console is different from the development environment for our portable system is certainly an area of stress or challenge for the development teams. So as we move forward, we're going to look at what we can do to unify the two development environments.

So, particularly with digital downloads now and the idea that you're downloading the right to play a game, that opens up the ability to have multiple platform digital downloads where you can download on one and download on another. Certainly from a development standpoint there is some challenge to it, because if you have two devices that have different specs and you're being told to design in a way that the game runs on both devices, then that can be challenging for the developer—but if you have a more unified development environment and you're able to make one game that runs on both systems instead of having to make a game for each system, that's an area of opportunity for us."

As for ur second paragraph, what makes u think Nintendo has any intention of competing with Sony/Microsoft in terms of power? If anything, the last decade or so should be a pretty good indicator that they care very little about going head to head in that aspect. It wouldn't be playing just handheld games at a higher resolution, it would simply be playing games.

The majority of Nintendo's IP play perfectly fine on either form factor. Mario platformers, Mario Kart, Smash Bros, Metroid, Zelda, Xenoblade, Kirby, Yoshi, Animal Crossing, Paper Mario, Luigi's Mansion, Donkey Kong, Mario & Luigi, Fire Emblem, Mario Party, etc. none of these series works on one but not the other.

You do realize that Miyamoto is referencing downloadable titles, right? I extremely doubt Nintendo is gonna do away with physical even with NX, so with Miyamoto referencing downloadable titles, I'd imagine stuff like smaller digital-only content, virtual console, some retail games, etc. Full retail games? He doesn't mention that, unless Nintendo plans to make NX fully digital, which is highly unlikely. 

I didn't say Nintendo needs to have hardware at the level of Sony or Microsoft hardware, you're putting those words in my mouth. And yes, it would be playing Handheld games in higher resolution because there are certain things consoles have that handhelds don't have: spec advantage. You will not see technologically ambitious games from Nintendo anymore because they must take the lowest denominator into consideration: the Handheld. Games cannot be very complex because the Handheld wouldn't be able to handle them, all games will need to be at the level of the Handheld with only minor tweaks for the Console. Mario Kart 8 for example at the same level of fidelity and complexity would not be possible on whatever Handheld they come up with next, neither will the next Zelda. You will also not see increases in complexity of their current games, Splatoon 2 for example wont see an increase in online players or gralhical fidelity, especially considering that Nintendo is conservative with hardware specs. Considering Nintendo's philosophy regarding the differences between handhelds and consoles, I doubt Nintendo will merely want Handheld games in higher resolution for the Console. I'm not expecting a powerful Console, but it needs to be powerful enough to even bother purchasing. With no notable difference between the Console and Handheld, people will just choose the cheaper option, there will be no point to the Console.

There are differences among those IPs on consoles vs the Handheld versions. Mario Kart for starters has more racers, and the complexity of that game is much higher. Plus, there are features in Mario Kart 8 that aren't in 7 for technological reasons. Super Mario 3D World's levels are much larger in scale than 3D Lands, and the ability to play 4 player multiplayer at once, on the TV is only possible due to being on a stronger system. Smash Bros. Wii U has not only larger stages, but also the characters can be more complex, more action can occur on the screen, and the CPU/AI is more intelligent because Wii U can handle it. Mario Maker wouldn't be friendly on the smaller screen of a Handheld. Xenoblade needed extra cores of CPU power to run at all, and at a severely reduced fidelity. It's these power differences that allow these major gameplay mechanics to even be possible. During 3D Worlds development for example, Motokura stated that with the increase in resolution, they were able to express different and new ideas that they never thought of before. Miyamoto also commented on 4K with Pikmin, and stated that being able to see more details would make the game more fun. It's increases in power that bring about new ideas and larger scope-concepts. If theres no notable difference in hardware power, these ambitions and new ideas wont come about. Until handhelds reach the ability to match consoles at a reasonable price, the differences will always be noticeable, and should be noted. 


If the portable is near/equal to the Wii U, I think will be fine for Nintendo if the console is basically just the same games in 1080P. If some devs want to go beyond that and add other effects, that would be their perrogative. 

Nintendo's consoles have dug their own grave, only the Wii has been truly a success in the past 15-20 years and that was again largely off the back of cheap casual games not big epic titles. Nintendo can't afford to keep their best franchises away from where 80% of their existing audience is, right now they're selling games like Splatoon and Mario Maker to a niche 10 million group of people that like playing Nintendo games on their TV, and that's an insanely stupid business model which they probably are very keen to change. 

I don't think Nintendo really has any huge interest in making today's "epic" games anyway, I don't think they're looking at Uncharted and saying "wow, we should be making games like that". 3D Zelda is really their only hugely epic franchise from a budget/scale POV and that looks great on existing Wii U tech as is. 

And no, I was never talking about two completely distinct architectures. 



Soundwave said:
JustBeingReal said: Snip:

It would have to work in a handheld. I mean it's easy to say "well they can get PS4 level performance in a console now" ... but, lol, I don't really see that taking Nintendo anywhere in and of itself. The handheld can't just be an after thought here, business wise it is actually the lead hardware variant if they do unify without a doubt. 

If they can use all HBM, great, but I don't think they can put that much HBM into a handheld. But I'm wondering if they could use a small pool of it integrated onto the GPU. By the way, the 3DS even uses embedded RAM, so Nintendo is obsessed with fast RAM caches. 

I think Mullins/Beema tech might be more likely the basis of Nintendo's processor tech than Carrizo if you have to account for the portable. It can't just be a thing of "lets just slap something together for the handheld and call it a day". 

Truth be told I don't think the NX console will do very well. The PS4 is going to dominate the general console market (as it exists now) likely well into 2019, and just making a system equal to or slightly better than the PS4 by fall 2016 isn't going to impress anyone but Nintendo fans. In other words, the handheld has to carry the mail here, that's why it's vital to the NX concept that it be at least decent hardware wise IMO. 

Tegra X1 is weak in the CPU department, ok, but is there a reason why AMD couldn't give them a portable GPU as powerful as that but with a better CPU? 

I just don't think you can have a console have PS4 quality graphics and then the portable is some thing that barely runs PS3/360 engines at 640x480, that's not really any different from where Nintendo is at today and would not help change their current business situation. 

It's all in the math and pixel count dude. 1080p is 1920X1080= 2,073,600 pixels, 480p is 640X480=307,200, 1080p is 6.75X more demanding to render than 480p.

That's with the same level of effects, physics, everything being the same comparing one resolution to another. If the NX console had the same level of hardware as PS4 then the NX handheld could run the exact same level of visuals and every other feature just at native 480p if it was packing 1/6.75 of the hardware of the console.

When I say Carrizo cut down, it's just an easier way of saying an APU that uses a smaller amount of Excavator CPU cores and GPU tech than what Carrizo has, clocked low enough to fit within X amount of a wattage budget. The CPU core type could be Puma or Puma+, that would still outstrip Jaguar since Puma is a new architecture, with better performance per watt.

As for the HBM comment, HBM can be used in whatever quantities Nintendo needs, dependent on cost they have outlined or factory output. From a power consumption perspective 1 watt can allow for 35GB/s of bandwidth, it's a flexible technology that would work for any platform holder's needs in their next platform, very efficient indeed, about 3X better than GDDR5 is currently.

 

Personally I think NX has a great chance of doing well, particularly if it is designed to put Nintendo's games at the forefront and break down any barriers between the handheld and console audiences that Nintendo covers.

If Nintendo focused on making a platform that caters to the 3rd party development community it could very well be Nintendo's chance to gain business back for not only the console market, but also the handheld. Even 3DS lacks the major releases like Witcher 3, Batman games, etc, if NX console can run good versions of those games and a handheld version can play those games too, even at 480p with lower graphical details it could mean big business for both Nintendo and their 3rd party partners.

 

As far as your last comment goes, you're not looking at things in the right way.

From a technical perspective it's the added pixels that cause the major demand on hardware, sure extra gameplay features add to that demand, but they're not the biggest performance hog on hardware, rendering extra pixels is.



Soundwave said:


If the portable is near/equal to the Wii U, I think will be fine for Nintendo if the console is basically just the same games in 1080P. If some devs want to go beyond that and add other effects, that would be their perrogative. 

Nintendo's consoles have dug their own grave, only the Wii has been truly a success in the past 15-20 years and that was again largely off the back of cheap casual games not big epic titles. Nintendo can't afford to keep their best franchises away from where 80% of their existing audience is, right now they're selling games like Splatoon and Mario Maker to a niche 10 million group of people that like playing Nintendo games on their TV, and that's an insanely stupid business model which they probably are very keen to change. 

I don't think Nintendo really has any huge interest in making today's "epic" games anyway, I don't think they're looking at Uncharted and saying "wow, we should be making games like that". 3D Zelda is really their only hugely epic franchise from a budget/scale POV and that looks great on existing Wii U tech as is. 

And no, I was never talking about two completely distinct architectures. 

I honestly don't think Nintendo will be okay with low-level physics, AI, etc on a Console. Those things won't change with resolution.

So you think they should just give up? The point of a Console according to them, is to offer an experience that can only be experience on their console. They screwed up on Wii U, they've screwed up some of their consoles; therefore they should just give up? I doubt they'll give up trying to " innovate"

I'm not saying they should make an Uncharted; however a difference in power does open up new possibilities, as we've seen with Wii U. I doubt they'd want to break away from that.



potato_hamster said:
Miyamotoo said:

All Nintendo talk about unifying hardware and software teams, same architecture, same OS, same platform, they even mentiond example of iOS and Android, is beacuse there will be no need like before to develop game separate for handheld and separate for home console, they will develop one game in same time for both consoles.

If Nintendo will benefit from that, there is no reason why 3rd party couldn't too.

 

Nintendo reorganized its R&D divisions and integrated the handheld device and home console development teams into one division under Mr. Takeda. Previously, our handheld video game devices and home video game consoles had to be developed separately as the technological requirements of each system, whether it was battery-powered or connected to a power supply, differed greatly, leading to completely different architectures and, hence, divergent methods of software development. However, because of vast technological advances, it became possible to achieve a fair degree of architectural integration..

For example, currently it requires a huge amount of effort to port Wii software to Nintendo 3DS because not only their resolutions but also the methods of software development are entirely different. The same thing happens when we try to port Nintendo 3DS software to Wii U. If the transition of software from platform to platform can be made simpler, this will help solve the problem of game shortages in the launch periods of new platforms. Also, as technological advances took place at such a dramatic rate, and we were forced to choose the best technologies for video games under cost restrictions, each time we developed a new platform, we always ended up developing a system that was completely different from its predecessor. The only exception was when we went from Nintendo GameCube to Wii. Though the controller changed completely, the actual computer and graphics chips were developed very smoothly as they were very similar to those of Nintendo GameCube, but all the other systems required ground-up effort. However, I think that we no longer need this kind of effort under the current circumstances. In this perspective, while we are only going to be able to start this with the next system, it will become important for us to accurately take advantage of what we have done with the Wii U architecture. It of course does not mean that we are going to use exactly the same architecture as Wii U, but we are going to create a system that can absorb the Wii U architecture adequately. When this happens, home consoles and handheld devices will no longer be completely different, and they will become like brothers in a family of systems.

Still, I am not sure if the form factor (the size and configuration of the hardware) will be integrated. In contrast, the number of form factors might increase. Currently, we can only provide two form factors because if we had three or four different architectures, we would face serious shortages of software on every platform. To cite a specific case, Apple is able to release smart devices with various form factors one after another because there is one way of programming adopted by all platforms. Apple has a common platform called iOS. Another example is Android. Though there are various models, Android does not face software shortages because there is one common way of programming on the Android platform that works with various models. The point is, Nintendo platforms should be like those two examples. Whether we will ultimately need just one device will be determined by what consumers demand in the future, and that is not something we know at the moment. However, we are hoping to change and correct the situation in which we develop games for different platforms individually and sometimes disappoint consumers with game shortages as we attempt to move from one platform to another, and we believe that we will be able to deliver tangible results in the future. 

What we are aiming at is to integrate the architecture to form a common basis for software development so that we can make software assets more transferrable, and operating systems and their build-in applications more portable, regardless of form factor or performance of each platform. They will also work to avoid software lineup shortages or software development delays which tend to happen just after the launch of new hardware.

That's talking about making games easier to port from handheld to home console and vice versa. However, at the end of the day you are still making two distinct builds have to test for two distinct platforms. This is a solid idea. It makes it a far easier decision for Developers to support both the NX Home and NX Handheld instead of one or the other. It will lead to more games on both platforms.

However, let's be clear, this is far and away from what others have been saying. This is describing two distinct platforms. Not one platform in two form factors. Sure it'll be easier than say it is to make a Wii game run on the 3DS, or a PS3 game run on the Vita, but it's still not as easy as "Make one game that plays on two different form factors of the same console". You still have to make device-specific optimizations. You still need to test on every platform. Let's use an example Nintendo discusses. iOS - not every piece of software available for the iPhone is available for the iPad and vice versa. There are reasons for that. Even though they share very similar operating systems, and the same development API, they are two distinct platforms. You need to modify your iPhone app to work properly on the iPad. You don't just make an iPhone app and it "just work" on the iPad. Not even close. Even if you're just making an iPhone app. If you want your app to work well on the iPhone 4, 4S, 5, 5S, 6, and 6S you need to optimize the app for each device and test it on the 4, 4S, 5, 5S, 6, and 6S. You need to have all of those devices to deploy your app to. You can't just make an app for the iPhone 5 and assume it will work just as well on the 4 and 6.

The same will go for the next Nintendo platform. Even if the development environment is more unified, even if the different devices share similar architecture, they will be treated as two separate devices, have separate software, and developers will not be forced to develop for both simultaneously.

Thank you for providing further concrete evidence to support what I've been saying all along. I appreciate it.

Sorry, but I don't see this describing two distinct platforms.

Of Course that 100% of same software will not work on home console and handheld, but I expect 80%.

All this means that developing games for NX home console and handheld console will be much easier and faster than developing games for instance for Xbox 360 and PS3 or 3DS and Vita, or even PS4 and XboxOne.



Around the Network
forethought14 said:
Soundwave said:


If the portable is near/equal to the Wii U, I think will be fine for Nintendo if the console is basically just the same games in 1080P. If some devs want to go beyond that and add other effects, that would be their perrogative. 

Nintendo's consoles have dug their own grave, only the Wii has been truly a success in the past 15-20 years and that was again largely off the back of cheap casual games not big epic titles. Nintendo can't afford to keep their best franchises away from where 80% of their existing audience is, right now they're selling games like Splatoon and Mario Maker to a niche 10 million group of people that like playing Nintendo games on their TV, and that's an insanely stupid business model which they probably are very keen to change. 

I don't think Nintendo really has any huge interest in making today's "epic" games anyway, I don't think they're looking at Uncharted and saying "wow, we should be making games like that". 3D Zelda is really their only hugely epic franchise from a budget/scale POV and that looks great on existing Wii U tech as is. 

And no, I was never talking about two completely distinct architectures. 

I honestly don't think Nintendo will be okay with low-level physics, AI, etc on a Console. Those things won't change with resolution.

So you think they should just give up? The point of a Console according to them, is to offer an experience that can only be experience on their console. They screwed up on Wii U, they've screwed up some of their consoles; therefore they should just give up? I doubt they'll give up trying to " innovate"

I'm not saying they should make an Uncharted; however a difference in power does open up new possibilities, as we've seen with Wii U. I doubt they'd want to break away from that.


They can still make the occassional wacky/"innovative" game that can only function with a console controller. A lot of people may have thought the Wii was giving up too, Nintendo abandoning competing with Sony/MS to recycle the GameCube for a motion controller. 

They can't continue on the way they are now. Approximately 300 GFLOPS is plenty for physics and AI even on a portable platform particularily for the games Nintendo makes. 

The fact is their last remaining strong hold, the portable market is also under attack. They can't afford to have their software library split between two platforms any longer, this is like an army that's suffering heavy losses trying to hold two fronts. 

Beyond that it's not "giving up" to acknowledge you can't support two high-end platforms at once. That's just accepting reality. Supporting the Wii/DS is one thing when development costs of those platforms was low, but they can't even hack it with a PS3-level console and a PS2-level handheld today. This is supposed to magically get better with a PS4-level console and a Vita++ level handheld? 

Yeah no. Nintendo also says a lot of things, like how they'll never make smartphone games. Sony wouldn't be able to support two platforms like that at once either, the PS4 is like almost 3 years old and it feels like just now is Sony finally "warming up" with their development teams. There's no chance Nintendo could juggle that while also having to make several high end handheld games per year on top of that. 

And yes, the one thing people don't talk about right now is that quite frankly it's terrible business to have your highest end/most expensive games only available to 20% of your fanbase. 80% of Nintendo hardware buyers this gen won't be playing Splatoon, Mario 3D World, Legend of Zelda U, Mario Maker, or Mario Kart 8. That just isn't smart business any way you spin it and it has to change. 



JustBeingReal said:
Soundwave said:
JustBeingReal said: Snip:

It would have to work in a handheld. I mean it's easy to say "well they can get PS4 level performance in a console now" ... but, lol, I don't really see that taking Nintendo anywhere in and of itself. The handheld can't just be an after thought here, business wise it is actually the lead hardware variant if they do unify without a doubt. 

If they can use all HBM, great, but I don't think they can put that much HBM into a handheld. But I'm wondering if they could use a small pool of it integrated onto the GPU. By the way, the 3DS even uses embedded RAM, so Nintendo is obsessed with fast RAM caches. 

I think Mullins/Beema tech might be more likely the basis of Nintendo's processor tech than Carrizo if you have to account for the portable. It can't just be a thing of "lets just slap something together for the handheld and call it a day". 

Truth be told I don't think the NX console will do very well. The PS4 is going to dominate the general console market (as it exists now) likely well into 2019, and just making a system equal to or slightly better than the PS4 by fall 2016 isn't going to impress anyone but Nintendo fans. In other words, the handheld has to carry the mail here, that's why it's vital to the NX concept that it be at least decent hardware wise IMO. 

Tegra X1 is weak in the CPU department, ok, but is there a reason why AMD couldn't give them a portable GPU as powerful as that but with a better CPU? 

I just don't think you can have a console have PS4 quality graphics and then the portable is some thing that barely runs PS3/360 engines at 640x480, that's not really any different from where Nintendo is at today and would not help change their current business situation. 

It's all in the math and pixel count dude. 1080p is 1920X1080= 2,073,600 pixels, 480p is 640X480=307,200, 1080p is 6.75X more demanding to render than 480p.

That's with the same level of effects, physics, everything being the same comparing one resolution to another. If the NX console had the same level of hardware as PS4 then the NX handheld could run the exact same level of visuals and every other feature just at native 480p if it was packing 1/6.75 of the hardware of the console.

When I say Carrizo cut down, it's just an easier way of saying an APU that uses a smaller amount of Excavator CPU cores and GPU tech than what Carrizo has, clocked low enough to fit within X amount of a wattage budget. The CPU core type could be Puma or Puma+, that would still outstrip Jaguar since Puma is a new architecture, with better performance per watt.

As for the HBM comment, HBM can be used in whatever quantities Nintendo needs, dependent on cost they have outlined or factory output. From a power consumption perspective 1 watt can allow for 35GB/s of bandwidth, it's a flexible technology that would work for any platform holder's needs in their next platform, very efficient indeed, about 3X better than GDDR5 is currently.

 

Personally I think NX has a great chance of doing well, particularly if it is designed to put Nintendo's games at the forefront and break down any barriers between the handheld and console audiences that Nintendo covers.

If Nintendo focused on making a platform that caters to the 3rd party development community it could very well be Nintendo's chance to gain business back for not only the console market, but also the handheld. Even 3DS lacks the major releases like Witcher 3, Batman games, etc, if NX console can run good versions of those games and a handheld version can play those games too, even at 480p with lower graphical details it could mean big business for both Nintendo and their 3rd party partners.

 

As far as your last comment goes, you're not looking at things in the right way.

From a technical perspective it's the added pixels that cause the major demand on hardware, sure extra gameplay features add to that demand, but they're not the biggest performance hog on hardware, rendering extra pixels is.


Is the Carizzo really all that special? I think the chip Nintendo would want AMD to start with is not the Carizzo, but the Wii U chip itself. That's a 40nm part, I'm sure today they can probably look at that chip and improve its efficiency even further, then you make a new chip based on that design. Replace the IBM CPU, get it on a 14nm process, remove the embedded RAM on the GPU for likely more power efficient HBM2. 

The Wii U chip gets about 11 GFLOPS/watt (if we believe the 350 GFLOP number that's used for it) at 40nm, if it was somehow possible to shrink that to 14nm, that would probably be in the range of 22-23 GFLOPS/watt which is fairly comparable to a Carizzo, no? Considering it's a 4-5 year old chip, that's not bad, I'm sure they could today improve it's power efficiency even more without just die shrinks too. 

As for pixels, I think for the portable maybe 960x540 resolution for really demanding 3D games, which would correspond 1:4 to 1920x1080 for the console. For lower end games, I think even the portable could just run things like Kirby's Rainbow Curse, Yoshi's Wooly World, Star Fox Zero, at 1280x720 though. 

1 watt per 35GB of bandwidth sounds ok, but wouldn't that be too power hungry for a portable? The design of the handheld is going to have to be the trick here I think, because the console is relatively easy to figure out because you can just plug it into the wall. 

*If* they can get PS4 level engines (with some effects stripped down) to run on a portable at a reduced resolution, yes I agree, they would probably get a shit-ton of Japanese third party support at least, and probably an OK amount of Western support too. I wonder if that's doable though. 



Miyamotoo said:
potato_hamster said:
Miyamotoo said:
 

All Nintendo talk about unifying hardware and software teams, same architecture, same OS, same platform, they even mentiond example of iOS and Android, is beacuse there will be no need like before to develop game separate for handheld and separate for home console, they will develop one game in same time for both consoles.

If Nintendo will benefit from that, there is no reason why 3rd party couldn't too.

 

Nintendo reorganized its R&D divisions and integrated the handheld device and home console development teams into one division under Mr. Takeda. Previously, our handheld video game devices and home video game consoles had to be developed separately as the technological requirements of each system, whether it was battery-powered or connected to a power supply, differed greatly, leading to completely different architectures and, hence, divergent methods of software development. However, because of vast technological advances, it became possible to achieve a fair degree of architectural integration..

For example, currently it requires a huge amount of effort to port Wii software to Nintendo 3DS because not only their resolutions but also the methods of software development are entirely different. The same thing happens when we try to port Nintendo 3DS software to Wii U. If the transition of software from platform to platform can be made simpler, this will help solve the problem of game shortages in the launch periods of new platforms. Also, as technological advances took place at such a dramatic rate, and we were forced to choose the best technologies for video games under cost restrictions, each time we developed a new platform, we always ended up developing a system that was completely different from its predecessor. The only exception was when we went from Nintendo GameCube to Wii. Though the controller changed completely, the actual computer and graphics chips were developed very smoothly as they were very similar to those of Nintendo GameCube, but all the other systems required ground-up effort. However, I think that we no longer need this kind of effort under the current circumstances. In this perspective, while we are only going to be able to start this with the next system, it will become important for us to accurately take advantage of what we have done with the Wii U architecture. It of course does not mean that we are going to use exactly the same architecture as Wii U, but we are going to create a system that can absorb the Wii U architecture adequately. When this happens, home consoles and handheld devices will no longer be completely different, and they will become like brothers in a family of systems.

Still, I am not sure if the form factor (the size and configuration of the hardware) will be integrated. In contrast, the number of form factors might increase. Currently, we can only provide two form factors because if we had three or four different architectures, we would face serious shortages of software on every platform. To cite a specific case, Apple is able to release smart devices with various form factors one after another because there is one way of programming adopted by all platforms. Apple has a common platform called iOS. Another example is Android. Though there are various models, Android does not face software shortages because there is one common way of programming on the Android platform that works with various models. The point is, Nintendo platforms should be like those two examples. Whether we will ultimately need just one device will be determined by what consumers demand in the future, and that is not something we know at the moment. However, we are hoping to change and correct the situation in which we develop games for different platforms individually and sometimes disappoint consumers with game shortages as we attempt to move from one platform to another, and we believe that we will be able to deliver tangible results in the future. 

What we are aiming at is to integrate the architecture to form a common basis for software development so that we can make software assets more transferrable, and operating systems and their build-in applications more portable, regardless of form factor or performance of each platform. They will also work to avoid software lineup shortages or software development delays which tend to happen just after the launch of new hardware.

That's talking about making games easier to port from handheld to home console and vice versa. However, at the end of the day you are still making two distinct builds have to test for two distinct platforms. This is a solid idea. It makes it a far easier decision for Developers to support both the NX Home and NX Handheld instead of one or the other. It will lead to more games on both platforms.

However, let's be clear, this is far and away from what others have been saying. This is describing two distinct platforms. Not one platform in two form factors. Sure it'll be easier than say it is to make a Wii game run on the 3DS, or a PS3 game run on the Vita, but it's still not as easy as "Make one game that plays on two different form factors of the same console". You still have to make device-specific optimizations. You still need to test on every platform. Let's use an example Nintendo discusses. iOS - not every piece of software available for the iPhone is available for the iPad and vice versa. There are reasons for that. Even though they share very similar operating systems, and the same development API, they are two distinct platforms. You need to modify your iPhone app to work properly on the iPad. You don't just make an iPhone app and it "just work" on the iPad. Not even close. Even if you're just making an iPhone app. If you want your app to work well on the iPhone 4, 4S, 5, 5S, 6, and 6S you need to optimize the app for each device and test it on the 4, 4S, 5, 5S, 6, and 6S. You need to have all of those devices to deploy your app to. You can't just make an app for the iPhone 5 and assume it will work just as well on the 4 and 6.

The same will go for the next Nintendo platform. Even if the development environment is more unified, even if the different devices share similar architecture, they will be treated as two separate devices, have separate software, and developers will not be forced to develop for both simultaneously.

Thank you for providing further concrete evidence to support what I've been saying all along. I appreciate it.

Sorry, but I don't see this describing two distinct platforms.

Of Course that 100% of same software will not work on home console and handheld, but I expect 80%.

All this means that developing games for NX home console and handheld console will be much easier and faster than developing games for instance for Xbox 360 and PS3 or 3DS and Vita, or even PS4 and XboxOne.


80% in common would be less in common than a PS4 game has with its X1 counterpart. And again, when developing for those platforms they are treated as two separarate entities, because they are. You know not of what you speak. There is no reason to expect it will be easier to develop a game for both NX home and NX handheld than it will be to develop a game for PS4 and X1.

For example do you know much extra work it takes to turn an iPhone game into an iPad game? Do you think that's trivial or complex?



Soundwave said:
JustBeingReal said:
Soundwave said:
JustBeingReal said: Snip:

It would have to work in a handheld. I mean it's easy to say "well they can get PS4 level performance in a console now" ... but, lol, I don't really see that taking Nintendo anywhere in and of itself. The handheld can't just be an after thought here, business wise it is actually the lead hardware variant if they do unify without a doubt. 

If they can use all HBM, great, but I don't think they can put that much HBM into a handheld. But I'm wondering if they could use a small pool of it integrated onto the GPU. By the way, the 3DS even uses embedded RAM, so Nintendo is obsessed with fast RAM caches. 

I think Mullins/Beema tech might be more likely the basis of Nintendo's processor tech than Carrizo if you have to account for the portable. It can't just be a thing of "lets just slap something together for the handheld and call it a day". 

Truth be told I don't think the NX console will do very well. The PS4 is going to dominate the general console market (as it exists now) likely well into 2019, and just making a system equal to or slightly better than the PS4 by fall 2016 isn't going to impress anyone but Nintendo fans. In other words, the handheld has to carry the mail here, that's why it's vital to the NX concept that it be at least decent hardware wise IMO. 

Tegra X1 is weak in the CPU department, ok, but is there a reason why AMD couldn't give them a portable GPU as powerful as that but with a better CPU? 

I just don't think you can have a console have PS4 quality graphics and then the portable is some thing that barely runs PS3/360 engines at 640x480, that's not really any different from where Nintendo is at today and would not help change their current business situation. 

It's all in the math and pixel count dude. 1080p is 1920X1080= 2,073,600 pixels, 480p is 640X480=307,200, 1080p is 6.75X more demanding to render than 480p.

That's with the same level of effects, physics, everything being the same comparing one resolution to another. If the NX console had the same level of hardware as PS4 then the NX handheld could run the exact same level of visuals and every other feature just at native 480p if it was packing 1/6.75 of the hardware of the console.

When I say Carrizo cut down, it's just an easier way of saying an APU that uses a smaller amount of Excavator CPU cores and GPU tech than what Carrizo has, clocked low enough to fit within X amount of a wattage budget. The CPU core type could be Puma or Puma+, that would still outstrip Jaguar since Puma is a new architecture, with better performance per watt.

As for the HBM comment, HBM can be used in whatever quantities Nintendo needs, dependent on cost they have outlined or factory output. From a power consumption perspective 1 watt can allow for 35GB/s of bandwidth, it's a flexible technology that would work for any platform holder's needs in their next platform, very efficient indeed, about 3X better than GDDR5 is currently.

 

Personally I think NX has a great chance of doing well, particularly if it is designed to put Nintendo's games at the forefront and break down any barriers between the handheld and console audiences that Nintendo covers.

If Nintendo focused on making a platform that caters to the 3rd party development community it could very well be Nintendo's chance to gain business back for not only the console market, but also the handheld. Even 3DS lacks the major releases like Witcher 3, Batman games, etc, if NX console can run good versions of those games and a handheld version can play those games too, even at 480p with lower graphical details it could mean big business for both Nintendo and their 3rd party partners.

 

As far as your last comment goes, you're not looking at things in the right way.

From a technical perspective it's the added pixels that cause the major demand on hardware, sure extra gameplay features add to that demand, but they're not the biggest performance hog on hardware, rendering extra pixels is.


Is the Carizzo really all that special? I think the chip Nintendo would want AMD to start with is not the Carizzo, but the Wii U chip itself. That's a 40nm part, I'm sure today they can probably look at that chip and improve its efficiency even further, then you make a new chip based on that design. Replace the IBM CPU, get it on a 14nm process, remove the embedded RAM on the GPU for likely more power efficient HBM2. 

The Wii U chip gets about 11 GFLOPS/watt (if we believe the 350 GFLOP number that's used for it) at 40nm, if it was somehow possible to shrink that to 14nm, that would probably be in the range of 22-23 GFLOPS/watt which is fairly comparable to a Carizzo, no? Considering it's a 4-5 year old chip, that's not bad, I'm sure they could today improve it's power efficiency even more without just die shrinks too. 

As for pixels, I think for the portable maybe 960x540 resolution for really demanding 3D games, which would correspond 1:4 to 1920x1080 for the console. For lower end games, I think even the portable could just run things like Kirby's Rainbow Curse, Yoshi's Wooly World, Star Fox Zero, at 1280x720 though. 

1 watt per 35GB of bandwidth sounds ok, but wouldn't that be too power hungry for a portable? The design of the handheld is going to have to be the trick here I think, because the console is relatively easy to figure out because you can just plug it into the wall. 

*If* they can get PS4 level engines (with some effects stripped down) to run on a portable at a reduced resolution, yes I agree, they would probably get a shit-ton of Japanese third party support at least, and probably an OK amount of Western support too. I wonder if that's doable though. 


Remember I just say Carrizo because it represents AMD's latest GCN core, along with using the Excavator line of CPU, all in the same SOC package, I think Beema and Mullins use different varients of Puma, one's Puma and the other Puma+, not sure about the level of GCN core in either of those.

AMD doesn't make IBM PowerPC CPU technology, it's all their own tech that they can provide Nintendo with, not IBM stuff, a single SOC makes way more sense than separate parts for a console, just because it's much cheaper and faster for manufacturing.

Whether AMD uses Puma, Puma+ or Excavator CPU technology it's all more capable than both Wii U's PowerPC and AMD's own Jaguar CPU tech.

The GPU in Wii U is an older architecture than the 7000 series tech in PS4 and XB1, the later stuff is more efficient on even the 28nm node it's built on.

 

AMD's current GCN core offers up 23.4GFlops per watt, so more than 2X more efficient than the GPU tech in Wii U, power consumption tests don't really state specifically how much energy the GPU alone uses, but the Carrizo System on Chip is packing a Quad Core Excavator CPU, at 2.1GHz, along with a 819GFlop GPU, as I said building a SOC with an 8 Core Excavator CPU and a 1638GFlop GPU would only require 70 watts for the SOC.

AMD's own tech is readily available, if Nintendo wrote an emulator they could emulate Wii U and 3DS.

 

As for the whole resolution on handheld thing, any higher resolution really isn't needed, it just adds cost and power demands to the system, for a handheld the more efficient it is the better. Going higher than 480p on a tiny handheld screen, when you can keep all of the gameplay features, most of the visual punch, enough of everything overall to make a good approximation of what the home console games look like is all you need to do.

This handheld spec I'm suggest can handle PS4 level visuals, with some slight concessions at 480p, things you won't even notice will be cut from the handheld version of games. That spec can definitely handle any games the Wii U had.

For reference 720p is 1280X720=921,600, as I said before 640X480=307,200, that's exactly 1/3rd of the pixels of 720p.

Our 819GFlop Carrizo APU runs on 35 watts, at 5 watts it's outputting 117GFlops (not counting the Excavator CPU performance or even talking about the better IPC gains over PowerPC), if Wii U has a 350GFlop GPU, then 1/3rd of it's performance is 116.6666666666667GFlops, the cutdown Carrizo is slightly more capable in the GPU department, plus the SOC would be HSA, allow way better multicore performance between the CPU and GPU.

Wii U level games would be no problem running everything the same, except for the resolution, 3DS games would be dominated by this system.

 

Re: Bandwidth HBM could be running slower, 35GB/s is at 1 watt, 25GB/s would only be 0.71watts, 1watt isn't too power hungry though, I mean the SOC would only be 5 watts, so memory and SOC would be 6 watts together. That's a 117GFlop GPU and probably 2 Excavator CPUs clocked at 1.2Ghz, in a handheld, very good performance IMO.

Bare in mind AMD Zen or the K12 CPU could be on the cards for Nintendo, that's a 14nm part, an even newer GCN part could be available too them and a newer HBM chipset, all with better performance.

 

Nintendo needs to think about what the development community outside of Nintendo wants, considering Nintendo are a member of the Khronos foundation I would think some version of Vulkan could very well be used in NX and perhaps they'll get help with their dev tools from AMD as a part of any collaboration on developing NX's processing tech.

The main area I think this will help Nintendo is in their own software sales, they'll push more 3DS and Wii U games if NX can run everything, which I don't see as unlikely, it would just require Nintendo to build an emulator for those games to run on NX. Personally I think there's no reason why Nintendo can't have both the handheld and console ready to go on the same day.

As I've shown AMD's tech is scalable, cutdown parts can definitely be made to fulfill the handheld's requirements, reducing power consumption, it all fits with what's required for a unified platform, hell even the whole API issue can easily be solved by Nintendo's involvement with Khronos and DeNA (from what I understand) can sort out the issues Nintendo have had with their online.

There's even positive rumors about 3rd party being happy with what they've been told about Nintendo's next platform. This all hinges on AMD providing everything in the processor department and them learning where they've gone wrong in the past.

 

The only area that I haven't heard a thing about is in Nintendo's own development camp. If Nintendo wants to gain back 3rd party then they need to make western focused games that will appeal to the western 3rd party market, of course any major shift in software focus would remain under wraps until NX gets officially announced, so this is something we probably won't find out about until E3 2016 happens, unless leaks begin to happen at some point this year or before E3 next year.

The hardware and API stuff seems like it could be covered, based on educated guesses and some logic, but it's still all just speculation for now.



potato_hamster said:
Miyamotoo said:
potato_hamster said:
Miyamotoo said:
 


80% in common would be less in common than a PS4 game has with its X1 counterpart. And again, when developing for those platforms they are treated as two separarate entities, because they are. You know not of what you speak. There is no reason to expect it will be easier to develop a game for both NX home and NX handheld than it will be to develop a game for PS4 and X1.

For example do you know much extra work it takes to turn an iPhone game into an iPad game? Do you think that's trivial or complex?


The major differences between PS4 and XB1 are API based, one's a version of DirectX/3D and the other's a custom variation of a LibGCM API, similar to OpenGL.

Hardware wise the major difference is in memory type, the CPU and GPU are of the same architecture, but there are also odd additions that one has, which the other doesn't.

CPU and GPU architecture in PS4 and XB1 are identical, quantities of GPU tech are grander in PS4 and there are clock speed difference.

 

The NX being invisioned by most people basically has the same API running on handheld and console, also the architecture would be identical, with the handheld only packing a smaller amount of tech or different clock speeds for parts to allow it to run on less energy. Developers could build the console version of a game, made to run at 1080p 30FPS, but the rendering pipeline could have options built in it to simply turn off that 1080p option and run the game at 480p, same goes for disabling AA, AF or any major performance hog.

Platform specific optimizations would stretch to reducing settings, in much the same way you'd turn off settings on a PC game because you're rig can't handle the higher resolution or GFX version of a game.

The difference between NX console and handheld is power, not architecture or API based.