Why Nintendo's NX Console Needs To Be x86-Based (Op-Ed)

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
"At least with the Genesis you could get a converter and run Master System games (I've done it, it works). "

You can thank the onboard Z-80 for that---Genesis used it for Genesis games as a sound processor, whereas it was the main cpu in the older SMS. In a sense, thats what Nintendo tried to do with the SNES, put in a backwards compatible 16-bit chip, they just didnt pull it off as well as Sega did. Might have made more sense to put a 68000 in there, like you said, and put a 6502 alongside it, similar to how Sega did their backwards compatibility. One think I always remember SNES devs complaining about was what a dog the 6502-native compatible WDC 65c816 was compared to the Genesis' 68K. Which was a shame, because everything else about the SNES hardware was far more capable than Genesis.

Again, probably came down to a matter of cost.
 
I don't think x86 is important at all; I think the best thing that Nintendo could do would be to open up development and encourage their consoles for indie developers. After all, Nintendo's consoles have always had a reputation for the "cuter" games such as Mario titles, and they're somewhat more affordable, which makes them well suited to indie developers.

Currently indie developers are looking more towards Android powered devices, but as with Android phones this is a fragmented market with numerous hurdles; if Nintendo could offer a single, clear platform for non-AAA titles then it could work well in their favour, and ought to attract more indie development of local multiplayer titles.

Even better if they could apply the same principles to a new handheld, and focus on having them interact with each other. I didn't so much mind the concept of using a tablet as a controller for the WiiU, except that it was just so inferior to an actual DS. It would be much better if they just made the DS the controller for the new console and sold it in packages with and without the controller (the latter if you already have one). Making the games portable like Mii profiles would be a good step too, so my friend could just bring the new Mario Kart on their DS, and everyone could then play it through my console.
 
Nintendo is a master of franchises; that's where their strength really lies. And that is indeed their weapon - their characters. I feel Nintendo hasn't utilized some of it's franchises very well at all in recent years - look at Starfox! What an amazing series .. first 2 games were simply awesome! If they put their heart to it, Starfox has the potential to be something totally out of this world - literally. I could go on.

For Nintendo, architecture shouldn't really matter. They will go for whatever will provide them maximum ease with compatibility.

It's the games that matter in the end and Nintendo has always been very strong in game design and game mechanics. Too bad, we haven't seen a lot of activity from them on console front lately. I wouldn't mind some of that old Nintendo magic coming back. In fact, I very much miss it! Nintendo seems lost. C'mon Nintendo ... only YOU can find yourself !!!
 
The Wii U actually benefited from a number of high profile game ports when it was first released. One of the things that turned me off initially about the Wii U was titles like Mass Effect 3, Sniper Elite 2, Tekken Tag Tournament, the EA Sports franchises like Madden and FIFA. There were a lot of titles that got ported over from XBox 360 and from PS3, but not a lot of original titles.

The Wii U exclusives have trickled out, and they've been pretty spectacular. It's because of Super Smash Brothers and Mario Kart that the Wii U is finally finding some traction and Nintendo is making money.

Betting the farm on going with x86 architecture and hoping your "me too" console will reach critical mass when there is nothing to differentiate itself from the competition sounds like a risky business plan.

Plenty of companies have gone the ARM route with consoles like the GameStick, the Ouya, and the MOJO. You couldn't get an easier platform for which to develop games, yet none of these systems has been a success. Why did they fail? No good games. A good phone game is mildly amusing for three minutes, a console game has to be more captivating than casual.

Nintendo has always been about providing a differentiated and accessible gaming experience that put fun above technology. Video gaming isn't drag racing; Nintendo doesn't need the most horsepower to win.

Nintendo simply needs to answer the question, "Do you know what would be really fun?" with a gaming system that provides the answer in a unique way. Light gun (Duck Hunt) games, party games (Mario Party), and motion control games (Wii Sports) have all grabbed the attention of new generations of gamers. Nintendo simply needs to find its next "big" thing.
 
What does CPU architecture and producing games have in common? Performance is dependent on main CPU frequency, number of cores and IPC. Games represent a piece of software that utilize hardware infrastructure including CPU, hence CPU is only a fraction of whole solution. There is no discussion as to why x86 set of instructions matter more than that of flavors of ARM.

It is clear that speculation in the article only touches CPU in rather limited discussion of software development. However, for a game system you would need to specify heat dissipation requirements along with what graphics subsystem you can have at best. And currently heat is produced most heavily by x86 CPU offerings.

ARM ecosystem would offer you good performance for minimal heat envelope. Nintendo is also know for minimalistic graphics and slow hardware, so it actually fits in with their long term strategy.
 


True! The only problem is that ARM core means you will get games ported from tablets and mobilephones... Not bad, but those games are / will be playable with a phone or a tablet. Why would you like to have those games in your gaming console. (in addition to normal Nitendo stuff of course)
x86 would means more games ported from other "big" x86 consoles and PC world (in addition to normal Nitendo stuff) If customer chose between those alternatives, the easier choice is/will be x86. Will the Nitendo go to that direction... It is hard to say, but it would allow very cheap and very small console, so it has merits, but there alreasy are guite many, not so successful android consoles that are just substitutes to the phone and/or tablets.
 


True! The only problem is that ARM core means you will get games ported from tablets and mobilephones... Not bad, but those games are / will be playable with a phone or a tablet. Why would you like to have those games in your gaming console. (in addition to normal Nitendo stuff of course)
x86 would means more games ported from other "big" x86 consoles and PC world (in addition to normal Nitendo stuff) If customer chose between those alternatives, the easier choice is/will be x86. Will the Nitendo go to that direction... It is hard to say, but it would allow very cheap and very small console, so it has merits, but there alreasy are guite many, not so successful android consoles that are just substitutes to the phone and/or tablets.

It is true that vast majority of ARM devices can and do run Android as their operating system. Android game development can target phones, tablets, TVs that run android and even small android "sticks". All in all Android ecosystem is beneficial. ARM platform is not Android and have many factors included such as built-in hardware and various sub-systems that you might use per manufacturer, per family and per category. The same is true for x86 integration. CPU instructions can be similar and will contain superset of x86, however it means a little due to the fact that operating system (I suspect you're referring to Windows when talking about x86 platform for games) needs to be different to maximize hardware performance. Operating system will be able to provide drivers for graphics sub-system and controllers sub-system at least. It is also true that for x86 you have many choices for operating system from DOS implementations to Linux and even Windows(embedded or not). Operating system will dictate then real-time requirements and what drivers would need to be written.

The question is more about operating system and drivers rather than hardware. x86 is just "another instruction set" in world of CPU nowadays. That is the reason for fast ARM adoption.
 
So, I'm not getting the x86 argument. It uses a lot of power, and it isn't any faster than the Power Architecture. Is it a speed issue?
 


The misconception is that an x86 console will be easiest for 3rd Party developers to create new games. The Wii U saw a big influx of 3rd Party game ports when it first launched; it didn't take off, however, until several Wii U exclusives finally shipped.

My impression at the launch of the Wii U was that it was more of an upgraded Wii than a whole new platform. Nintendo wanted to take advantage of the tablet craze with the new controller as well as finally offer Hi Def graphics. They thought players would be happy continuing to play Wii games while springing for the occasional Hi Def Wii U game. Without new games, Wii owners had no reason to upgrade and folks without a Wii never felt like they were missing out on anything particularly great.

I can't speak for most gamers, but I generally only buy consoles when there is a AAA must-have exclusive title that I can't play anywhere else. If Nintendo had lined up titles like Mario Kart 8 and Super Smash Bros. as launch titles, the Wii U would still be $350 and the whole world would have one.

Games sell game systems. Good original games sell a lot of game systems.
 
Status
Not open for further replies.