I've already said this to you before, but I'll repeat it again, because it has to be pressed firmly into the mind of anyone make these kinds of comparisons. Each company that makes a processor measures GFlops in their own way, therefore it makes absolutely no sense to compare the A9X to anything AMD, comparisons cannot be made in that way.

Also each company measures their Transistor's size differently.

So yeah from those measurements it's basically a pointless comparison. Nintendo aren't going to use that chip, not going to happen, because they have a relationship with AMD, they've had a relationship with ATI since before AMD bought them, that isn't going to change and AMD can provide a single SOC, with all of their technology needs met as far their Processor requirements go.

Also matching PS4's specs actually does a lot, it's not just a matter of graphics, it's also a matter of what's possible from a gameplay perspective, physics in games like Horizon or Uncharted 4 are actually way more important and improve the experience overall, comparisons betweeen PS4 and some hypothetical NX console are useful, because we can see what PS4's level of tech can allow game creators to accomplish. A Zelda game with the kind of physics and massive world of Horizon, along with a tonne of dungeons and so on would be what fans of that series have been wanting for years and it would look crip with horizon's native resolution. Obviously to get something like that to hit 60FPS would be what most of the vocal community of fans would want, but that would require 2X the performance of PS4, which consequently is actually within reach, but to do that would make it more difficult for Nintendo to fit the same kind of experience into a handheld, so we can see that actual comparisons with PS4's level of tech are actually very useful.

Nintendo offering a console and handheld that share the same library, with a very similar level of experience between the two is actually going to set Nintendo apart, that's something which I think the community could get on board with.

Right now no one is providing the kinds of games you can play on your TV at home on the move, with a native, latency free experience on the handheld. This could actually be Nintendo's opportunity to return to their former glory or at least lay the groundwork for a future where they can become competitive once again. With a family of systems that they can evolve to meet the capabilities of the competition, capable of running multiplats on the move, along with Nintendo's own great core experiences and potentially new IP that are more catered to the west it could really be the thing Nintendo needs to get them back in the race.

Nintendo never really lost, they just stopped competing. They stopped trying to appeal to fans of theirs which they created back in the day. Gamers who matured and still wanted to game, but Nintendo just wanted to carry on making games for the same age range, thinking that the only other kinds of games that would work for them, from a business perspective could be the other extreme of titles like horror games (partnering with Ubisoft on games like ZombiU or making things like Eternal Darkness or violent stuff).

If Nintendo comes out and says oh we have this new system that can play games just as well the competiton, gets all of the multiplats, but if you want you can also buy a handheld that will play all of that on the move too and if you want to own both, having a bigger experience on a big TV at home, but also take it with you and own a collection to works on both systems then that could be huge and it would be something unique to Nintendo.

School kids could be playing COD on their NX Handheld, then carry on playing with their friends at home on the big screen.

As for the last part, the handheld would only need to be the size of a normal tablet. Cost wise I doubt it would even need to cost that much, the console would be comparable to PS4's price in late 2016, assuming NX has the same size HDD, Bluray and so on, the handheld would use a cut down version of the console's processor, so would cost a fraction of that price.

Maybe $200 at launch for the handheld and $300 or $350 for the console.

I never said they'd use that exact chip, I said they would want something from AMD that has similar type of performance at a similar power envelope. 

35 watts for 800 GFLOPS give of take isn't even that great to be honest considering that it's probably less than 2.5x the Wii U and the Wii U is a 40nm chip from 2010/11 at 31-32 watts. 

They should be able to study these new mobile chips, especially looking at their own Mullins/Beema tech and push the envelope further than that. 

And yes, like I said I understand they utilize GFLOPS in different ways, but I still maintain I would not be surprised at all if an A9X allowed to operate and scaled up to a massive 35 watt power envelope would probably beat that Carizzo. I wouldn't be surprised if it was fairly equal to the XBox One.  

Nintendo did lose badly with the GameCube. Not only did Sony kick their ass, they got beat by a newcomer that had no market presence in Japan and the XBox still sold more than the GCN even though the GCN had a full extra year on the market. 

I'd go with $250 for the handheld. You can't just fart out a low cost handheld and cry "unified platform" if there's too much of a gap in form factors. What will end up happening is many devs will end up just optimizing for the handheld spec because it'll likely be the top selling version by far, and then the console people will be stuck with portable games on the TV. So in that case you want some real grunt power to the handheld. 

$350 is too much for the console too, seeing as how PS4 will be $300 by next fall ($50 cut this year, $50 next year sounds about right). 

No because the wii u is much stronger than soundwave thinks.

That's why Nintendo will keep failing unless they change their approach. If the console is barely faster than Wii U, it won't have a shot at 3rd party games. In that case, the console has automatically failed before it has even come out. 

By end of 2016, PS4 and Xbox One will have dropped in price. We should see Xbox One at $299 and PS4 at $349, at most, and even lower prices for PS4 in Japan. No matter how you slice it, from a consumer's perspective Nintendo WILL be competing against Xbox One and PS4 unless they make a console that's priced at $199 and below. Also, because the NX is a mid-cyclel console, it will inevitably have to compete with even more price cuts on Xbox One/PS4 in the years 2017-2019 before the next generatino of systems. 

You have already been told that AMD does not manufacture ARM SoCs and has none in the pipeline for 2016. Therefore, your entire theory is flat out wrong. It's also suicidal to not have an x86 SoC powering Nintendo's next console because it means all the successors to the NX will be unnecessarily crippled/held back by the weak CPU/GPU horsepower of ARM SoCs compared to X86. The smartest strategy is to have both x86 AMD APU + ARM SOC for broad interoperability. You have already been shown how weak the A9 chip in iPhone 6S is compared to even the anemic Surface Pro 3, but Apple has the fastest ARM SoC in the market out today, something AMD can't even dream of matching in 2016 in ARM land. Since Nintendo cannot source Apple's SoCs, it's a foreone conclusion that if AMD won the NX contract, the chip inside is x86, and a possible co-support for a 3rd party ARM SoC. 

Also, 2-3X the performance of the Wii U is not going to accomplish much. If you understand anything about graphics you will know that more advanced graphics require expontentially more powerful hardware. 

That means for Nintendo to address one of the biggest complains from 3rd parties - lack of bare minimum GPU horsepower - they will need to increase horsepower by at least 4-5X to be relevant. 

You think they are not competing with Sony but for market participants/consumers, they are. When someone goes to buy a console for their kid/themselves, guess what? The Wii U is priced not much cheaper than Xbox One and PS4. Their prices are very similar, especiallly once we consider the lack of sufficient storage space on the Wii U. Hence, if Nintendo were to release the NX, it will compete with/for Xbox One/PS4 customers whether Nintendo likes it or not. 

We have already seen that unless Nintendo magically figures out the next popular fad (Wii motion controls), their 1st party support cannot maintain the sales sufficient for their business to be profitable. We started seeing issues with N64 and later this becames far worse with GameCube and then the Wii U.

You keep repeating how Nintendo should just give up 3rd party support and making a console at least as powerful as Xbox One but honestly it comes off more as to what YOU would want not what's BEST for Nintendo's NX. You also keep laughing at any notion of the NX ending up as powerful as the Xbox One or possibly coming in roughly between Xbox One and PS4's power. Again, sounds like you actually WANT the NX to fail rather than discussing what Nintendo can and should do to get themselves back to having a greater 3rd party support and a shot at capturing some of the core audience it had during N64/GameCube days thats since migrated to PlayStation/Xbox brands. 

A part of the reason Nintendo couldn't include a more powerful APU (CPU+GPU) in the Wii U has nothing to do with their technical inability to do so (i.e., you imply that Nintendo would never put a 70-95W TDP APU because they've never done it). This has everything to do with their strategy of trying to make $ on the hardware. Based on estimates around Wii U's launch, the system at that time cost $180 to manufacture, with roughly $50 of that going to the controller.  

http://techland.time.com/2012/04/09/how-much-for-that-wii-u-in-the-window/

http://kotaku.com/5900153/rumor-the-wii-u-has-180-worth-of-parts-wont-be-sold-under-300

I haven't found more credible sources so I could very well be wrong. However, even if we were to add manufacturing/labor, marketing, supply chain/distribution and logistics costs, Nintendo likely sold the Wii U at a significant profit margin compared to what MS/Sony did with their launch PS4/XB1 consoles. 

If Nintendo continues to try to sell us $180-200 worth of hardware for $350 with the NX, gamers aren't going to fall for this crap because this has failed them already with the Wii U. If they sell a console for $199-249 only, given their historic profit margin, that would imply really crappy budget parts too. Both of these strategies sound like failures out of the gate. 

If you say the NX is going to be primarily a portable console, then it's also going to fail since we've seen what happens with the PS Vita when you try to shove semi-decent portable hardware into a well-built portable console - but guess what the casuals are gaming on smartphones and tablets. Also, if they try to make a console that's both a home + portable, it's going to be a jack of all trades, master of none. 

If Nintendo does some combo of a home console NX + portable NX for say $400-500, that's another story entirely but if their portablel NX is supposed to also be a home console, it's impossible to produce a portable console that's much more powerful than the Wii U so as I alluded to earlier that would NOT solve any of the major issues with Wii U currently has already with gamers and 3rd party developers.

Nintendo themselves don't have the capacity to pump out 100s of 1st party games either to keep the system affloat.

What were those files bitrates at? Because it could be nothing, if the bitrate was low. I could equal the same amount of complexeity, if I loaded up 4 1080p into a timeline, and play same time. I do that when I'm needing to do split screen work.

Ya that can only be an x86 APU, not an ARM processor since AMD makes no such products. That's exactly what many here are trying to tell you but you keep discussing scaling up A9X made by a completely different firm - Apple. Do you think if AMD could produce a chip that's a scaled version of A9/A9X they wouldn't have done so by now? AMD has no ARM SOC with A9/A9X level of performance scheduled to launch in 2016 so what is this magic ARM SOC chip AMD could sell to Nintendo? 

No 2.5X greater than 800GFlops takes us near PS4's GPU rating. The Wii U is nowhere close to that. I bet the Wii U's maximum single precision output is around 350-400 GFLOPS. The Wii U seems to have a maximum memory bandwidth of just 12.8GB/sec.

A very close GPU family that resembles the Wii U is HD4650 series with 320 stream processors and 16GB/sec memory bandwidth. Nintendo likely used the more modern HD5xxx derivative though but still that's miles behind HD7790 level in Xbox One and HD7850+ level in PS4. It's no wonder developers couldn't easily port 3rd party games to the Wii U because the GPU performance level in it could already be purchased September 2008 more or less. 

But Apple will not sell A9X to Nintendo and AMD makes no such products and has no such products in the pipeline in 2016. So no, there is no mythical ARM SOC Nintendo could use that could beat Carizzo. 

In this very thread, you have already been proven wrong on this point. Even if we literally doubled the perfomrance of A9, that barely gets us to beat Surface Pro 3, which itself gets pounded by the best Intel APUs, and those are destroyed by HD7790 - the GPU in the Xbox One. 

Go read the responses already as it's all laid out for you. Stop ignoring facts. Unless you are an Apple shareholder, this A9X and NX rumors are baseless.

Or Nintendo integrates an x86 APU and an 3rd party ARM SOC into the home console - let's call it NX Home. The NX Portable would be a separate device. Everyone who would have the NX portable, would be able to play all NX Portable games on the NX Home. What that means is if you buy an NX Portable game, you get 2 games in one. That's the point of a uniform eco-system. This way the home console can be made powerful enough for 3rd party XB1/PS4/PC games and have interoperability with the NX portable. 

Trying to cram a very powerful ARM SOC into the NX portable that would also be able to function as a solid home console would be prohibitevely expensive. In fact, from a technological point of view, there will not be any ARM SOC powerful enough to even match the Xbox One in 2016. 

They could make it as powerful at least as powerful as the XB1 and sell for less than $350. How? Ship it with a standard controller. Ship it with no optical drive (digital distribution games only). Ship it with minimal profits on hardware like Sony/MS are doing. It's common sense that if Xbox One would sell for $299 by 2016, Nintendo can likely aquire similar hardware for $299. But since it'll be nearly 3 years since the chips in PS4/XB1 came out, it should be possible to acquire even greater level of perfomrance for $100-110 for a 2016 AMD APU. Alternatively, AMD is far more desperate right now to win business than it was during 2013 when their finances were in better shape. That should allow Nintendo to negotiate an even better deal than what Sony/MS got in 2013 because AMD literally needs every revenue stream it can get right now. 

Nintendo is in the best possible position to introduce a great console since XB1/PS4 are going to be 3-years-old, but it's just about what they want out of their console and who are they targetting. 

You have also ignored a point raised earlier how many an ARM SOC only corners Nintendo into future options for NX2, NX3, etc. because to make the next versions of their consoles BC, they would need an ARM SOC design or if they switch to x86, they'd need a secondary ARM SOC which adds cost. 

3rd parties would have a lot of trouble porting x86 code games if the NX is solely an ARM SOC design like the A9X.

Bingo. Some people are trying to pigeonhole the NX as one console - either a handheld or a home console or a hybrid but it can be another option you and I both discussed -- NX Home console and NX Portable console. Together, they start to form the NX eco-system.

The home console is the most powerful so it can play NX portable games, smartphone, tablet Nintendo games. The NX portable is a complementary device. Among the entire NX game eco-system, you get cross-buy and cross-play. If you buy a Nintendo game on an Android or their NX Portable, it's playable from your account on the NX home console. 

This is actually what the rumors have:

• With NX, Nintendo wants to create a platform where they can develop software for multiple devices with ease, from the home console to the portable to smartphone and tablets. They'll have a big catalogue available for multiple platforms, with cross buy, cross saves and cross play, similar to what Microsoft is doing with Xbox One and PC thanks to Windows 10 (that allows to play many Xbox games on PC, which is also getting a bunch of ports from Xbone). Each platform will still have exclusive games
• They'll show the portable in spring 2016, will be out for the end of the year or spring 2017 max. Specs will be higher than PS Vita but nothing mindblowing, screen resolution should be 540p, considering 720p if costs go down.
• The home will be out 6-12 months after the portable, creating the "NX system" that will allow Nintendo to better use their resources in games development
• They approached third parties during E3. Capcom, Square-Enix, Ubisoft and Electronic Arts already have devkits.

Your entire thread is about the A9X and it being what Nintendo will use in NX, it's the topic you keep bringing up all of the time, either when you make a new thread or when you comment about NX, so actually that is what you've been saying all of the time.

FYI Carrizo basically offers a CPU with very close levels of performance to PS4 and XB1, also most curret gen 3rd party games are 1080p, with the levels of effects we've been seeing. In some cases XB1 is achieving 1080p with a 1.31TFlop GPU and pretty lacking memory system, an 819GFlop GPU, with a more modern feature set, more efficient, modern architecture would easily allow for that level of games at 900p.

Nintendo can have a SOC with a far superior CPU to both competing platforms, along with a GPU with less compromises than XB1's for a lower level of power consumption, basically we're talking about at least 50% less power consumption, for similar levels of performance compared to PS4 and XB1, very attractive IMO for Nintendo and they could easily have a box that's a little over half the size of PS4.

It's a pointless comparison, an A9X style chip isn't what Nintendo will use, you can maintain whetever you like, doesn't make it true. The visuals produced by that Apple chip isn't anywhere near to achieving Wii U levels of performance in it's power range and it's certainly not comparable to Carrizo, even if it was scaled up.

It's most definitely not achieving XB1 levels of performance in a 35 watt package, Carrizo seems more efficient.

Nintendo didn't really try, they didn't produce the kind of games to appeal to a western audience, didn't have titles with the same level of diversity as Sony did or even does now. Sony had presence in Japan, with their overall tech image, they came out with games that no one had seen before and they targeted a portion of the market that had a whole in it, with a lack of content in that area.

Hell Gamecube missed the mark when it comes to lacking DVD and Nintendo refused to make that standard on their platform, like I said they didn't really compete, they just let Sony provide for the masses and focused on their niche and over time that niche has shrunken. They got lucky with the Wii and managed to expand through a gimmick, but it was a one trick pony for most of the 70M+ plus people that would never have bought into a games console if it hadn't had such a feature that was unique to it, at such a cheap price.

Who said anything about just farting out the handheld? The whole platform in this instance is built around the same core architecture, with the same core capabilities, it's just that the more powerful system allows for some effects features that are out of reach of the lower power consumption device. Developers wouldn't just optimize for the handheld, because they'd be building their games to take advantage of the whole platform.

It's a unified platform because the whole thing has been designed to be that. So what if the handheld sells more copies? The API and architecture make it so that you're building your games for the whole and the whole idea is that you're giving the platform userbase freedom to take their games anywhere across either system, you think developers making Apps for Apple products or Android cares that one platform has more than the other? No they just make the code accessible to all, then they gain whatever sales are available across all platforms.

The handheld can still handle the same basic level of experience the home console can and there aren't any roadblocks making it hard to put games on both, it's easy. The whole point here is that the hardware is just a vessel to accomodate the games, if more people choose the weaker system because cost is substantially cheaper it's not prohibiting the console from having a better experience for those people that want to have that.

The handheld would have plenty of grunt to achieve current gen level gameplay and graphics, just at 480p, with a few reductions in effects, but it's still the same core game.

As for cost, I did say it could be $300 for the console, tbh this level of tech would be more mature and less expensive next year, so it could even be less expensive than that for the home console and if the core processing tech is cheaper in the console, then the handheld would be cheaper to make, since it's a piece of silicon that has the same tech, but with reduced quantities of that.

If the SOC in NX console costs $100, then the one in the handheld would be less than $20.

$200 for the handheld and $300 for the console, with a package of both at $450 seems about right IMO.

You have to remember that Sony isn't reducing the price of PS4 because they don't really have to in order to keep getting huge sales, Nintendo on the other hand are trying to appeal more to the market which they've lost, costs need to be appealing and the level of hardware involved in this hypothetical NX wouldn't have a cost beyond that.

While I agree with almost everything you're saying, it stops at x86. Unless they can find an affordable way to get the NX BC with the Wii U and maybe even the 3DS eshop while also adopting x86 hardware, I don't see it happening. It's not just a unified platform Nintendo are going for, but a legacy platform. They want to be able to build from what they started with the Wii U, especially when it comes to the library. In order for that to happen, they absolutely need to be able to launch the NX with the full eshop library, so they don't have to spend rescources and time porting SMB3 for a 4th time. I genuinely think, if having to choose, Nintendo will choose weaker hardware over absolutely no digital BC with at least the Wii U eshop.

This isn't an issue of backwards compatability, but of legacy. I think Nintendo is shooting for a steam-like platform where games from every platform they've had is playable. From the NES to the Wii U. That's why they've been adding DS and Wii games. They're not going to send all that to die, especially after they had to do that with the Wii VC. They want to absorb the Wii U's architecture. While I hope they can get the both worlds, using the x86 while also absorbing the Wii U to keep BC with the eshop, I don't see any way that is feasable.

Also, the gamepad is staying. I don't know how anyone can think it isn't. Same with the two screen set up of the DS and 3DS for the handheld. That functionality will be imperitive for the NX, and we already have patents backing this up. Do I think the gamepad will go through a massive revision, making it much smaller, much more traditionally shaped, and much cheaper? Absolutely. But it's definitely not being dropped for a Pro Controller.

...Or something else.

I don't think a $20 processor gets you PS4/XB1 graphics, not even scaled down at 480p resolution. RAM would have to be $30 at least probably more. Unless Nintendo is using a shit screen again that's $50 at least for your touch panel. Sensors/NFC/WiFi is another cost. NAND Flash. And you're going to need a huge battery to power the thing too. 

I have doubts they can sell that at $200 without taking a loss. Like I said the mentality here is too much of "well just design the console first and then we'll slap together whatever for the handheld". For a unified platform, particularily where the handheld is more likely to be 4:1 the best seller, you can't have that approach. 

I think they have to build the portable chip first, then it's easy to scale that up, but that's not workable IMO using desktop PC components or even laptop components that have the benefit of being cooled internally by a huge casing and a fan. 

I don't see Nintendo magically eating $50 of the hardware cost if you buy both either. What's in it for them to take $50 less on the hardware (that likely is already being sold at margin/loss) just because you want to buy both?