Nintendo announce: Nintendo NX

Status
Not open for further replies.
I think to help win back the hearts and minds of gamers, help maintain the loyalty of those who bought the wiiu which will be getting its life cut short, and help stop the customer bleeding to Sony and Microsoft, and even help gain new customers, Nintendo really really should consider taking a loss on hardware this generation. Now or never to make their new platform as enticing as possible or risk fade into a Sega.
 
Last edited:
We don't want a Nintendo console with fewer flops than PS4, years after PS4. It would just be not quite almost enough, and late to the party to boot.

This is a machine that will have Mario, Yoshi, and Kirby games targeting primarily kids, why the need for big graphical power? Nintendo didn't exactly go to town in Star Fox graphics, recently..

Why would we want 20-30W TDP in a stationary console? It didn't grant Wuu any measure of success at all, if anything it just held the system back. Wuu is ~50% the volume of PS4 and only a fifth the power dissipation, yet even less than a fifth in actual computing performance, of course. Once you go below a certain level of computing power you cease to really scale down in physical size, at least as long as you want an optical drive inside the casing as well, and I think we all assume Nintendo would want that. If you get too small, you start running into Ouya territory where your box is so small and light the cables attached to it will manhandle it, and the whole product just feels dumb and cheap.

Cost and size.. acceptance in Japanese homes.. Nintendo efficiency over performance tendency.. Patents already suggesting no optical drive.. The future trend is the small digital, box.

Do they? I'd assume they simply want more performance and the fewer cores that they can achieve that with the better. Putting die size/cost aside (as it's Nintendo's problem from devs pov) I'd assume devs would prefer 4 high clocked Excavator cores that 8 low clocked Puma cores.

Good point.

Wow, that's very nice. Can we more precisely extrapolate gains that 10nm finfet will enable when the time for PS5 deployment arrives?

Is precision a special achievement for you?.. It's very simple to extrapolate from Carrizo an APU in the X1/PS4 performance envelope in the sub-40 watt range in 2016. 60% TDP improvement from 16nmFinFet+ is already known..
 
I realize there is one weak point Nintendo could leverage to get back into the console race and win people over mid cycle against competitors with a large mindshare and large software library and userbase lead. These consoles are not being sold at much of a loss.
If Nintendo is willing to take a $100 loss on BOM unlike PS4 or XO, the noticeably superior performance across identical 3rd party titles, could entice people to buy their system. Developers are willing to leverage the performance benefits of a small segment of the PC community the top 3% of pc gamers, so it'd be likely that they'd be willing to leverage the performance of a new Nintendo console.

I don't think Microsoft would want to attack Nintendo's high BOM by dropping its price to hurt nintendo, as they'd like their profits, especially early when the Nintendo user base is small. Sony probably is even in a less likely position to go into a pricewar vs Nintendo and eat into profits.
 
Last edited:
There is too much arrogance in Nintendo and too vast a difference in software and online capabilities for Nintendo to overtake existing XB / PS4 during this generation. They can not win over the gamers.
 
There is too much arrogance in Nintendo and too vast a difference in software and online capabilities for Nintendo to overtake existing XB / PS4 during this generation. They can not win over the gamers.
Yeah overtaking PS4/XO is impossible, I was merely suggesting they could help prevent their obsolescence and survive in the console space by selling the console at a loss for once.
 
Last edited:
If the NX has parity with PS4/Xbone and something like the wiimote that provides genuine novelty to gamers and non gamers then they could do great numbers. Based on the WiiU it'll have neither but I'll not count them out yet!

You could argue that by going all in on VR, Sony's going to grab the novelty market share too (using novelty as term for initial attention grab btw, as oppose to something I think will be short lived)
 
This is a machine that will have Mario, Yoshi, and Kirby games targeting primarily kids, why the need for big graphical power?
Then why launch a new console at all? The Wii U seems already perfect for that!
Oh wait I know why. It's because the Wii U was a commercial failure in every possible metric.

Cost and size.. acceptance in Japanese homes..
The PS4 consumes 4x more power than the Wii U but it's been outselling the Nintendo console for over a year in Japan.
Perhaps it's time we stop excusing Nintendo for simply being cheap on their hardware.
 
And so it begins.

Nintendo started distributing the software version of the "NX" development kit to qualified publishers and developers this week. We had a brief chat with a senior developer at a major game publishing company based in the U.S., and according to him, Nintendo NX is going to have very impressive hardware. Based on the development kit, the sheer processing capabilities of the hardware (which still hasn't been finalized) are going to be "incredibly powerful" and quite possibly faster than whatever Sony and Microsoft have in store. Specifically, one software demo included with the kit crunches so many polygons that it's currently impossible to run at 60fps using a current-generation Intel (we're assuming a Core i7 Skylake) CPU and a nearly top-of-the-line graphics card (no specifics provided, but they probably used a single graphics card).

Hi, found this thread through neogaf. I am a mid level-ish environment artist/designer at Ubisoft, and our office did get the SDK this week. I wasn't technically supposed to see it but was working late Friday and managed to sneak a peek of others playing with the demos. Can't comment on the fps specifically, but both demos I saw seemed to be running fine on our (relatively decked out) machine. One of the demos particularly looked gorgeous, and the impression around the office is that people are pleasantly surprised with this kind of focus on graphics from Nintendo. I'm not working on anything Nintendo-related and can't confirm anything else, but I can agree with the processing power statements. It should be an exciting little machine and from what I see, fun to develop for.
 
This is a machine that will have Mario, Yoshi, and Kirby games targeting primarily kids, why the need for big graphical power? Nintendo didn't exactly go to town in Star Fox graphics, recently..
That's precisely my point. Hardly anyone wants a shit console that can only run kiddy games.

People have cellphones and tablets for that kind of gaming these days,

Cost and size.. acceptance in Japanese homes..
First off, if Nintendo is so blinded by what the Japanese market deems acceptable that the only design on the table is the size of two pocketbooks, then they've fucking lost already. Japan always was a minority player in the console market, and these days that market is shrinking. Most of the world is not Japan!

Second, PS4 is very close to the original PS2 (and PS for that matter) in size. PS3 slim too by the way. All of these consoles sold plenty in Japan and were accepted just fine.

Nintendo efficiency over performance tendency..
Wuu is not particularly efficient though; it's just bad that's all, that's why it doesn't draw a lot of power.

Patents already suggesting no optical drive.. The future trend is the small digital, box.
The future isn't here yet. Most of the world does not have an internet infrastructure that can easily support a digital-only delivery system. Many still subsist with single-digit megabit per second internet connections, downloading just one modern gaming title could take a day or more. Also, data caps; the scourge of our existence. Just one or two large games could wipe out your cap for the entire month.
 
All Nintendo needs to do is combine their Console and Handheld market into a single customer base, they actually can easily overtake XB1 even with 3DS and Wii U sales numbers (nearly 60m 4 and a half years in)

Considering a single platform with multiple form factors could share the marketing budget of both Wii U and 3DS, their combined libraries and the handheld's 3rd party support. Then actually name these products NX home and NX portable, or something distinct, drop the discs and use 32GB gamecards with the option to move to 64GB cards, while targeting $199 price point for the handheld (n3DS sells at this price right now) with the option to sell the console for $250 (cheaper than the Wii U currently) They SHOULD be able to exceed Wii U and 3DS LTD numbers fairly easily, we also are hearing from insiders that there is a 3rd device for this platform, so 60m in 5 years actually looks like the minimum to me.

Notice the position I've outlined above, it doesn't target 3rd party AAAs and it doesn't mention meeting/exceeding PS4 specs. Nintendo consoles biggest competition has always been their own handhelds which have always outsold their consoles during the same generation they are a part of, even the Wii lost to the DS. Combining these platforms into one really is Nintendo's best choice to turn around its biggest problems: Droughts, Marketing (budget is no longer split), success of individual devices only matter in the sum of the platforms numbers, unique 3rd party support (handheld 3rd party support is pretty good) and lastly Gimmicks (you simply don't have to ride out a gimmick from 1 device if your platform doesn't rely on it, which multiple form factors should avoid)
 
Lol, "more poweful than the most powerful PC's available today"... how I've missed these pre-console launch rumours :p

We don't actually know anything about the machine yet, just a couple of apparent first-hand experiences from people claiming to be developers. I guess we need to question whether these sources are considered accurate and believable. Apparently some are corroborating, but that's also quite easy for people (with nothing better to do) to fake.

If the second source is believable, then the "relatively decked out" PC they're running the test on is likely to be of a higher spec than the Nintendo box since framerate was said to be of a decent quality. They also only mention "one" of the demos being anything impressive.

What would a relatively decked out machine be? A 960? If so, then it's likely put Nintendo's console somewhere around the same power as the PS4.

The first source seems to suggest that a near top-of-the-line GPU and CPU aren't running the demo brilliantly well. I agree that it sounds very unlikely that Nintendo will release an uber powerful console that outperforms a quality PC, but we shouldn't rule out the possibility based on our on personal biases. We can only say that it's "unlikely", based on costs and historical evidence.
 
Lol, "more poweful than the most powerful PC's available today"... how I've missed these pre-console launch rumours :p

Hum? I didn't see that sentence anywhere.
In fact, what they wrote was pretty down to earth. The first claimed that one demo seemed to be pushing too many polygons for the test system they were using (for which he didn't know the specs) and the other said all the demos were running okay it a relatively high-specced machine.

Needless to say, they're talking about the Software Development Kit, meaning those demos are probably an idea of what the final hardware will be capable of, which may or may not be close to the end product.
For example, there were lots of talks about the first Wii U System Development Kits coming with a Radeon 4850 (which ended up being 4-8 times faster than the console itself).
 
Last edited by a moderator:
The first source seems to suggest that a near top-of-the-line GPU and CPU aren't running the demo brilliantly well. I agree that it sounds very unlikely that Nintendo will release an uber powerful console that outperforms a quality PC, but we shouldn't rule out the possibility based on our on personal biases. We can only say that it's "unlikely", based on costs and historical evidence.

The demo specifically mentions a "crunching polygons" faster then a current top CPU/GPU. It's certainly possible that Nintendo (AMD?) is adding some sort of hardware enhancement to make specific use cases faster than what a traditional console or PC can do in the same scenarios. That doesn't mean the overall performance will be comparable. (Maybe it's some sort of new fancy and efficient tessellator,)
 
The demo specifically mentions a "crunching polygons" faster then a current top CPU/GPU. It's certainly possible that Nintendo (AMD?) is adding some sort of hardware enhancement to make specific use cases faster than what a traditional console or PC can do in the same scenarios.

I wouldn't be surprised if the Arctic Islands come with a focus on enhanced geometry performance, since that's been bottleneck-ish for GCN.
For example, Tonga has better geometry performance than Hawaii.
 
I wouldn't be surprised if the Arctic Islands come with a focus on enhanced geometry performance, since that's been bottleneck-ish for GCN.
For example, Tonga has better geometry performance than Hawaii.
Exactly my thoughts, it might just be GCN 2.0 is much better with polygons. It could also point to a high clock rate for the GPU since GCN is 2 poly per hertz. If Nintendo is going with 14nm/16nm and they are using GCN 2.0 which is twice as efficient in power / performance over current GCN parts, it might just be a 1GHz clock speed and a nice new tessellation engine. it might still only have 8-16CUs and give a performance of 1tflops - 2tflops
 
Hum? I didn't see that sentence anywhere.
In fact, what they wrote was pretty down to earth. The first claimed that one demo seemed to be pushing too many polygons for the test system they were using (for which he didn't know the specs) and the other said all the demos were running okay it a relatively high-specced machine.

The fact that they're talking about polygons at all raises doubt as to the accuracy of the source. Since when was the number of polygons being pushed an accurate measure for graphics quality or system performance? In 2001 maybe, but this is 2015. Likely "polygons" was just a proxy for general graphics and the suggestion was that a near top end PC wasn't enough to handle the demo at the presumably target 60fps.
 
Status
Not open for further replies.
Back
Top