Nintendo announce: Nintendo NX

[MrFox]

If we count the first year sales to consumers, plus filling the channels, plus assembly line stock, 20 million sounds okay. They will still need those parts for a few months into the next year anyway if the sales are under expectation. The alternative of under-supply is a much worse situation. I always wondered what the 7th gen final sales figures would have looked like if all three consoles had plenty of stock in the first year, like we saw with current gen.

Or... it's a memory supplier and nintendo ordered 20 million memory chip for 5 million consoles.

 

[BRiT]

Or for 2.5 million consoles if memory is split evenly between console and handheld controller tablet.

 

[liolio]

I would not be too bullish wrt to the 10nm transition and its impact on lesser lithography availability or cost, no matter what Apple does or doesn't.
I see no reason for Nintendo to use anything but the widely available 28nm process.
Especially if NIntendo to include a handheld in its console they need something cheap, 28nm is cheap and plenty of progress have been done when it comes to perfs per watts and perfs per mm2 by multiple IP providers.
As for the rumor of late they sound a lot like an off beat viral campaign, I say off beat because 'lots of polygon /geometry" sounds a lot like a PR argument that missed the Mantle/Metal/Vulcan/etc. bandwagon but was released into the wild anyway... I guess there are no bad advertisements.

 

[McHuj]

I'll be disappointed if Nintendo doesn't go with a Finfet process for their next console. For the home console, they'd have to go with a large die (250+ mm) to be competitive with the other consoles. For the portable, even if the level of performance is similar with what they would build on 28 or 14nm, the power consumption savings are worth it.

 

[cheapchips]

And 14/16nm + hmb + no blurray means it could be an ultra dinky console. Which would be nice.

 

[Grall]

Which would be nice.

Ultra dinky things give off the impression it is a toy, and/or weak. There's literally no inherent ownership value in ultra dinkyness.

 

[cheapchips]

Ultra dinky things give off the impression it is a toy, and/or weak. There's literally no inherent ownership value in ultra dinkyness.

Er, Ok. It would be ultra *insert machismo marketing version of very small* here.

 

[dobwal]

Well no, actually they don't. Because who will make the launch lineup for NX if their dev teams (which really aren't THAT many) are locked up supporting the ailing Wuu?

Wuu is a millstone around Nintendo's neck, and they know this. I'd be hugely surprised if there'll be any software support at all for Wuu into '17, much less all the way up to the holiday season.

I personally feel its not a good ideal to abandon your less than stellar current platform for newer hardware without considering those thats supporting your current system. At the very least I would focus on supporting both with every new title to make sure the new console has a healthy launch library while rewarding those who have supported me all along.

 

[Entropy]

Ultra dinky things give off the impression it is a toy, and/or weak. There's literally no inherent ownership value in ultra dinkyness.

Well, you already heard one vote in favour, and I have to say that the prospect of buying another big black plastic box that needs to live on an open shelf for heat dissipation reasons has very little appeal. I belong to the group who think both MS and Sony did well in this generation moving to less power hungry and quieter devices. If Nintedo also ditches a whirring optical drive, they have the potential to produce something pretty nice from a living room point of view.
That said, if they are making a dedicated stationary consoly it has to offer reasonable oompf to justify its existance at all, otherwise they could just have commissioned a good mobile solution from Mediatek for instance, and ditched the archaic stationary concept completely. So I guess anywhere from 40-150W could be reasonable.

 

[Entropy]

I'll be disappointed if Nintendo doesn't go with a Finfet process for their next console. For the home console, they'd have to go with a large die (250+ mm) to be competitive with the other consoles. For the portable, even if the level of performance is similar with what they would build on 28 or 14nm, the power consumption savings are worth it.

A 250+mm2 16nmFF device would be competitive with the R9 Fury! While it would give me no small amount of evil satisfaction to see the glass jaws of forum console warriors shatter as they hit the floor, I can't see how that would be justified inside a board room.

 

[DieH@rd]

A 250+mm2 16nmFF device would be competitive with the R9 Fury!

Not if its paired with slightly upgraded PowerPC CPU from WiiU and 4-8 GB of DDR3.

Although it would indeed be funny if Nintendo went with Zen+R400@16nmFF+HBM2+hybrid spinning/SSD 1TB drive. Forum meltdowns would be legendary.

 

[Grall]

I have to say that the prospect of buying another big black plastic box that needs to live on an open shelf for heat dissipation reasons has very little appeal.

I didn't say 'big black plastic box' now did I? There are more degrees of differentiation than just 'ultra dinkyness' and 'big black box'.

PS4 is a very good middle-of-the-road kind of device, IMO. It's not so tiny that it becomes wimpy, but still diminutive enough to fit into a quite small space, especially if turned on its side.

 

[Grall]

Although it would indeed be funny if Nintendo went with Zen+R400@16nmFF+HBM2+hybrid spinning/SSD 1TB drive. Forum meltdowns would be legendary.

I hope my neighbors hear the thump after I collapse of an excitement-induced heart attack and call an ambulance for me.

Unfortunately, I don't think I'll ever get to experience that though; the last console that really knocked people to the floor with its over the top hardware crazyness was PS2. PS3 came close, but it married the crazy Cell with a middling and already aging GPU, so it doesn't manage to steal the crown IMO...

 

[McHuj]

A 250+mm2 16nmFF device would be competitive with the R9 Fury! While it would give me no small amount of evil satisfaction to see the glass jaws of forum console warriors shatter as they hit the floor, I can't see how that would be justified inside a board room.

Sorry, I wasn't clear. I meant a 250+ mm die on 28nm for the home console. If clocked sufficiently high it could be competitive with current gen, but I doubt Nintendo would go for the power consumption.

 

[DieH@rd]

Sorry, I wasn't clear. I meant a 250+ mm die on 28nm for the home console. If clocked sufficiently high it could be competitive with current gen, but I doubt Nintendo would go for the power consumption.

PS4 has 348mm2 28nm APU that does not have any on-board cache memory
Xbone has a bit bigger chip [same CPU, smaller GPU, and ~1.5 billion of transistors for 32MB of ESRAM]

The largest die space that is possible at 28nm HKMG is ~600mm2 [FuryX is 596mm2 [as is 175W N9 Nano]].

 

[dobwal]

Sorry, I wasn't clear. I meant a 250+ mm die on 28nm for the home console. If clocked sufficiently high it could be competitive with current gen, but I doubt Nintendo would go for the power consumption.

I doubt Nintendo wants to invest in a chip competitive with current gen. I think what ultimately dictates the prowess of Nintendo's hardware is its first party software. And I think Nintendo's isn't willing to accommodate the increased level of development costs to take advantage of the hardware. Either Nintendo would have to bolster the size of dev houses or consolidate groups which either leads to greater costs or lesser output.

Nintendo wants cheap development and a wide offering of first party titles which has led Nintendo to offer things like motion control, stereoscopy and other tech because they didn't come with a permanent increase in software dev costs.

 

[Entropy]

I doubt Nintendo wants to invest in a chip competitive with current gen. I think what ultimately dictates the prowess of Nintendo's hardware is its first party software. And I think Nintendo's isn't willing to accommodate the increased level of development costs to take advantage of the hardware. Either Nintendo would have to bolster the size of dev houses or consolidate groups which either leads to greater costs or lesser output.

Nintendo wants cheap development and a wide offering of first party titles which has led Nintendo to offer things like motion control, stereoscopy and other tech because they didn't come with a permanent increase in software dev costs.

I don't think the first paragraph holds water. Why would it be all that more expensive to develop a game for 48CU part than for a GPU with 8CUs (in todays AMD GPU parlance)? The total cost of game production, promotion and distribution would seem to have very little to do with shading power and bandwidths.

The second paragraph makes logical sense, but seems like crossing the bridge for water - why wouldn't it simply be that they want to be able to offer new gaming experiences to their customers like they say, hoping of course that they will resonate with new and hopefully large groups of people? I find it difficult to see it as a pattern motivated by a wish to save on game development costs. This part of your argument mostly hinges on the first paragraph being true, and I don't think it is.

 

[Entropy]

I didn't say 'big black plastic box' now did I? There are more degrees of differentiation than just 'ultra dinkyness' and 'big black box'.

PS4 is a very good middle-of-the-road kind of device, IMO. It's not so tiny that it becomes wimpy, but still diminutive enough to fit into a quite small space, especially if turned on its side.

The PS4 is a better black box than its predecessor, I'll give you that. :smile2:
And if Nintendo produces a stationary console, I guess it will fall somewhere between the WiiU and the PS4 in size for pure heat dissipation reasons, I don't think an AppleTV like device is particularly likely. Like I said, if that was an option at all, why not go fully mobile? It would be a gaming device after all, not a TV programming accessory, so being tethered to a TV set doesn't make any sense if wired connections aren't necessary for power reasons anyway.

At this point in time we simply don't know appreciably more than we did back at the DeNA reveal.

 

[Deleted member 13524]

It will really be a shame if/when Nintendo doesn't take advantage of technologies that became available in 2015 like 14/16nm and HBM to make a more powerful (or at least equivalent) console than the PS4.

Nintendo wants cheap development and a wide offering of first party titles which has led Nintendo to offer things like motion control, stereoscopy and other tech because they didn't come with a permanent increase in software dev costs.

Nonsense. Mario Kart 8, Mario Bros U and Pikmin 3 would look phenomenal at 1080p with 8xAA, 16xAF, Ambient Occlusion and other post processing that are either free or super cheap to implement.
Nintendo didn't put terrible hardware in the Wii U in order to spare their first-party developers from ballooning costs. Nintendo games strive on gameplay above all, and that wouldn't be affected by having decent hardware.

 

[Shifty Geezer]

Nintendo didn't put terrible hardware in the Wii U in order to spare their first-party developers from ballooning costs. Nintendo games strive on gameplay above all, and that wouldn't be affected by having decent hardware.

Actually, they've said they like weaker hardware because it keeps costs down. A AAA game on PS4 that maxes it out costs way more more than a AAA game on PS2. The cost to create all those realistic assets with all their variety is just way more. Of course, you don't have to max the hardware out and Nintendo could still stick with simple graphics to fit a budget, letting 3rd parties go expensive with variety, but it has been N.'s reasoning none-the-less, to have hardware with lower expectations from users so they needn't spend so much making games.