The wiiU's CPU is not faster than this in any scenario. This is still on par with the PS4 base model CPU.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Nintendo Switch Dev Kit Stats Leaked? Cortex A57, 4GB RAM, 32GB Storage, Multi-Touch.
- Thread starter Robert at Zeboyd Games
- Start date
- Status
The wiiU's CPU is not faster than this in any scenario. This is still on par with the PS4 base model CPU.
Thanks for the explanation. Why should we assume that because there is a fan, the machine should be more powerful than it is? Maybe it just needs the fan to stay cool.
As I said, I'm afraid you guys are searching another reason to dissapoint yourselves, but whatever.
This will still be a great device with amazing graphics for a handheld. We just got too excited about the specs and didn't think about all of the other factors (heat, battery life, pricing).
And we don't know the fan is actually used in portable mode, do we? If the chip is downclocked this much in docked mode to preserve battery, it doesn't make sense to use the fan. I'm betting it won't use the fan much in portable mode, and the rumors we've been hearing of an audible fan may be due to devs might not have access to the dock, so those Switch units may be running in the "docked mode", which requires the fan running.
And we don't know the fan is actually used in portable mode, do we? If the chip is downclocked this much in docked mode to preserve battery, it doesn't make sense to use the fan. I'm betting it won't use the fan much in portable mode, and the rumors we've been hearing of an audible fan may be due to devs might not have access to the dock, so those Switch units may be running in the "docked mode", which requires the fan running.
Xdrive05
Member
Truth.
1080p Zelda should be a given now, by the way. Considering the clock boost they don't have a good reason to miss that mark. If Zelda has to run perfectly playable undocked, then docked should hit full HD.
Of course there will also probably be games where they go 540p/900p, reductions that somewhat scale together similarly.
James Scott
Banned
A lot of Wii U games look really good but they run at 720p, so sure.
If it was just a portable then it would've just been 720p at all times so this is an improvement.
Well, other apparatus have more power on display and run fine without a fan. That is why it is so strange that the Switch would need an active fan while in undocked mode. Because of this, some people theorise that the system could have more CUDA cores than the standard X1 has, explaining the need for a fan (since it would, in that case, necessarily draw more power) and in that scenario also adding more graphical power to the system.
Skittzo0413
Member
You could very well be right but just going by what similarly specced devices do (Pixel C) and that the Vita has apparently the same power draw with less surface area, it adds up to be very confusing that Nintendo would spend the space and money for an internal fan. Especially when a fan inside of a portable is a very high risk point of mechanical failure for the device.
I'm not an expert here but based on everything we know about this, the Pixel C and the Vita, there should be absolutely no need for active cooling in either portable or docked mode.
If a hypothetical VR dock does exist then that could be a reason for one.
Just spit-balling. There have been complaints from some that the Gear for example gets too hot. Computation and display in the same package inches from the face is hard to keep comfortable.
LordOfChaos
Member
I imagine the power can be applied how the developer sees fit, a 1080p requirement docked on hardware of this level would be wasteful. I'm imagining a lot of 900p games with moderate visual boost, resolution apart.
MisterR
Member
See, there you go. Keep low expectations and you can end up happy. I actually agree with you on the portable part. I expected Vita level for the 3DS replacement.
For the jump from 720p to 1080p, you need 2,25 times the power (that many times the pixels are rendered), so the difference in power very neatly fits a jump from 720p to 1080p. I would guess that that is what the power differential is for.
That clock rate when undocked does seem too low to warrant a fan, specially when you consider that the old TK1 Shield Tablet didn't have one.
There's the possibility of the fan only kicking in when docked, to prevent the unit from feeling too warm when you remove it from the dock. There's also the possibility of some sort of separate framebuffer cache to speed up rendering.
There's the possibility of the fan only kicking in when docked, to prevent the unit from feeling too warm when you remove it from the dock. There's also the possibility of some sort of separate framebuffer cache to speed up rendering.
As far as we know, Switch has no anti-exploding tech. Yet again, it's a generation behind.
I mean I have a $2k PC for a reason. If you're looking for Nintendo be your only system and provide everything the competition does then yeah you will probably be disappointed. But Nintendo has been a supplementary gaming company at best for years now for the majority of people. The Switch is at least their first real admittance of that fact.
Nintendo delivers games you don't find elsewhere for the most part. That goes for their 3rd party support as well, at least in the handheld 'verse. You don't need photorealism for great games, and while more power and visual effects are great regardless of rendering style it doesn't change the fact that art direction and design is still king. That's the reason why Wind Waker or Super Mario Galaxy still looks so good to this day, and look downright amazing when run in HD on emulators.
The Switch isn't without its improvements over the Wii U and it is a handheld. So having that kind of power and fidelity on the go is a pretty huge deal. Then at home you can continue that experience in a console like environment with the minimum increase in image quality to match it. Sounds like a perfect device for me as I've been a PC/3DS person for the tail end of last gen and the entirety of this current one so far. But that might not be enough for others. We'll have to see how on point Nintendo's marketing is and how the market responds to it.
Skittzo0413
Member
I'm gonna post Thraktor's writeup here:
This also doesn't even go into the possibility of a fan in the Switch AND one in the dock, as Laura Dale reported a few weeks ago. The patent doesn't support that report of hers, but that is something that could have been easily added to the final hardware since the patent was filed in June.
I haven't had time to read through every response here, so I'm probably repeating what others have already said, but here are my thoughts on the matter, anyway:
CPU Clock
This isn't really surprising, given (as predicted) CPU clocks stay the same between portable and docked mode to make sure games don't suddenly become CPU limited when running in portable mode.
The overall performance really depends on the core configuration. An octo-core A72 setup at 1GHz would be pretty damn close to PS4's 1.6GHZ 8-core Jaguar CPU. I don't necessarily expect that, but a 4x A72 + 4x A53 @ 1GHz should certainly be able to provide "good enough" performance for ports, and wouldn't be at all unreasonable to expect.
Memory Clock
This is also pretty much as expected as 1.6GHz is pretty much the standard LPDDR4 clock speed (which I guess confirms LPDDR4, not that there was a huge amount of doubt). Clocking down in portable mode is sensible, as lower resolution means smaller framebuffers means less bandwidth needed, so they can squeeze out a bit of extra battery life by cutting it down.
Again, though, the clock speed is only one factor. There are two other things that can come into play here. The second factor, obviously enough, is the bus width of the memory. Basically, you're either looking at a 64 bit bus, for 25.6GB/s, or a 128 bit bus, for 51.2GB/s of bandwidth. The third is any embedded memory pools or cache that are on-die with the CPU and GPU. Nintendo hasn't shied away from large embedded memory pools or cache before (just look at the Wii U's CPU, its GPU, the 3DS SoC, the n3DS SoC, etc., etc.), so it would be quite out of character for them to avoid such customisations this time around. Nvidia's GPU architectures from Maxwell onwards use tile-based rendering, which allows them to use on-die caches to reduce main memory bandwidth consumption, which ties in quite well with Nintendo's habits in this regard. Something like a 4MB L3 victim cache (similar to what Apple uses on their A-series SoCs) could potentially reduce bandwidth requirements by quite a lot, although it's extremely difficult to quantify the precise benefit.
GPU Clock
This is where things get a lot more interesting. To start off, the relationship between the two clock speeds is pretty much as expected. With a target of 1080p in docked mode and 720p in undocked mode, there's a 2.25x difference in pixels to be rendered, so a 2.5x difference in clock speeds would give developers a roughly equivalent amount of GPU performance per pixel in both modes.
Once more, though, and perhaps most importantly in this case, any interpretation of the clock speeds themselves is entirely dependent on the configuration of the GPU, namely the number of SMs (also ROPs, front-end blocks, etc, but we'll assume that they're kept in sensible ratios).
Case 1: 2 SMs - Docked: 384 GF FP32 / 768 GF FP16 - Portable: 153.6 GF FP32 / 307.2 GF FP16
I had generally been assuming that 2 SMs was the most likely configuration (as, I believe, had most people), simply on the basis of allowing for the smallest possible SoC which could meet Nintendo's performance goals. I'm not quite so sure now, for a number of reasons.
Firstly, if Nintendo were to use these clocks with a 2 SM configuration (assuming 20nm), then why bother with active cooling? The Pixel C runs a passively cooled TX1, and although people will be quick to point out that Pixel C throttles its GPU clocks while running for a prolonged time due to heat output, there are a few things to be aware of with Pixel C. Firstly, there's a quad-core A57 CPU cluster at 1.9GHz running alongside it, which on 20nm will consume a whopping 7.39W when fully clocked. Switch's CPU might be expected to only consume around 1.5W, by comparison. Secondly, although I haven't been able to find any decent analysis of Pixel C's GPU throttling, the mentions of it I have found indicate that, although it does throttle, the drop in performance is relatively small, and as it's clocked about 100MHz above Switch to begin with it may only be throttling down to a 750MHz clock or so even under prolonged workloads. There is of course the fact that Pixel C has an aluminium body to allow for easier thermal dissipation, but it likely would have been cheaper (and mechanically much simpler) for Nintendo to adopt the same approach, rather than active cooling.
Alternatively, we can think of it a different way. If Switch has active cooling, then why clock so low? Again assuming 20nm, we know that a full 1GHz clock shouldn't be a problem for active cooling, even with a very small quiet fan, given the Shield TV (which, again, uses a much more power-hungry CPU than Switch). Furthermore, if they wanted a 2.5x ratio between the two clock speeds, that would give a 400MHz clock in portable mode. We know that the TX1, with 2 SMs on 20nm, consumes 1.51W (GPU only) when clocked at about 500MHz. Even assuming that that's a favourable demo for the TX1, at 20% lower clock speed I would be surprised if a 400MHz 2 SM GPU would consume any more than 1.5W. That's obviously well within the bounds for passive cooling, but even being very conservative with battery consumption it shouldn't be an issue. The savings from going from 400MHz to 300MHz would perhaps only increase battery life by about 5-10% tops, which makes it puzzling why they'd turn down the extra performance.
Finally, the recently published Switch patent application actually explicitly talks about running the fan at a lower RPM while in portable mode, and doesn't even mention the possibility of turning it off while running in portable mode. A 2 SM 20nm Maxwell GPU at ~300MHz shouldn't require a fan at all, and although it's possible that they've changed their mind since filing the patent in June, it begs the question of why they would even consider running the fan in portable mode if their target performance was anywhere near this.
Case 2: 3 SMs - Docked: 576 GF FP32 / 1,152 GF FP16 - Portable: 230.4 GF FP32 / 460.8 GF FP16
This is a bit closer to the performance level we've been led to expect, and it does make a little bit of sense from the perspective of giving a little bit over TX1 performance at lower power consumption. (It also matches reports of overclocked TX1s in early dev kits, as you'd need to clock a bit over the standard 1GHz to reach docked performance here.) Active cooling while docked makes sense for a 3 SM GPU at 768MHz, although wouldn't be needed in portable mode. It still leaves the question of why not use 1GHz/400MHz clocks, as even with 3 SMs they should be able to get by with passive cooling at 400MHz, and battery consumption shouldn't be that much of an issue.
Case 3: 4 SMs - Docked: 768 GF FP32 / 1,536 GF FP16 - Portable: 307.2 GF FP32 / 614.4 GF FP16
This would be on the upper limit of what's been expected, performance wise, and the clock speeds start to make more sense at this point, as portable power consumption for the GPU would be around the 2W mark, so further clock increases may start to effect battery life a bit too much (not that 400-500MHz would be impossible from that point of view, though). Active cooling would be necessary in docked mode, but still shouldn't be needed in portable mode (except perhaps if they go with a beefier CPU config than expected).
Case 4: More than 4 SMs
I'd consider this pretty unlikely, but just from the point of view of "what would you have to do to actually need active cooling in portable mode at these clocks", something like 6 SMs would probably do it (1.15 TF FP32/2.3 TF FP16 docked, 460 GF FP32/920 GF FP16 portable), but I wouldn't count on that. For one, it's well beyond the performance levels that reliable-so-far journalists have told us to expect, but it would also require a much larger die than would be typical for a portable device like this (still much smaller than PS4/XBO SoCs, but that's a very different situation).
TLR
Each of these numbers are only a single variable in the equation, and we need to know things like CPU configuration, memory bus width, embedded memory pools, number of GPU SMs, etc. to actually fill out the rest of those equations to get the relevant info. Even on the worst end of the spectrum, we're still getting by far the most ambitious portable that Nintendo's ever released, which also doubles as a home console that's noticeably higher performing than Wii U, which is fine by me.
This also doesn't even go into the possibility of a fan in the Switch AND one in the dock, as Laura Dale reported a few weeks ago. The patent doesn't support that report of hers, but that is something that could have been easily added to the final hardware since the patent was filed in June.
Quoting myself for the new page.
I think the issue for me is that I was okay with Wii and Wii U each being a generation behind as consoles. But Switch now feels a bit like 2 generations behind. Maybe even 2.5 generations behind if you include the PS4P and Scorpio.
As a console, Switch is, relative to its time of release, both Nintendo's weakest console to date as well as arguably their strongest handheld. And that's a bit disappointing to those of us who have preferred their console offerings.
So far, the device doesn't offer a much different console experience than the Wii U, but is still going to cost a few hundred bucks to gain admission to Nintendo's titles. It feels a bit lack luster for Wii U owners who planned on using this primarily as a console. Basically, this is just Wii U 2 with a slight resolution bump.
Hopefully, there's more cool hardware features which allow for new experiences to be revealed on the 12th.
As a console, Switch is, relative to its time of release, both Nintendo's weakest console to date as well as arguably their strongest handheld. And that's a bit disappointing to those of us who have preferred their console offerings.
So far, the device doesn't offer a much different console experience than the Wii U, but is still going to cost a few hundred bucks to gain admission to Nintendo's titles. It feels a bit lack luster for Wii U owners who planned on using this primarily as a console. Basically, this is just Wii U 2 with a slight resolution bump.
Hopefully, there's more cool hardware features which allow for new experiences to be revealed on the 12th.
LordOfChaos
Member
It's a moderate help I guess. Having the same low level API accross major OSs (except you, macOS! *stares*) and consoles would be great. But that's largely the work of engine makers and middleware to port, most games without bespoke engines aren't dabbling in low level API stuff all day, the engine and middleware does a lot for them.
I had no doubts it would be more powerful than Espresso.
I'm not seeing how it's on-par with the PS4, though. Jaguar isn't very powerful, but from what I've seen an A57 benches pretty close clock-to-clock with them. Hence the request for benchmarks.
Where was this confirmed?
You're talking like its some kind of challenge to put Maxwell onto 16nm. Also no idea why you'd refer to increasing core count as a experiment.. Its just customising a chip to your liking, which is what Nintendo does every time. Considering 16nm doesn't really give you less die area compared to 20nm I don't even see it as necessary for 3SM's.
I don't see any need for a exotic RAM solution in the setup you're talking about though. You already have a chip that keeps frame buffer access on chip (akin to XBox One) so there's no need for a on chip frame buffer. Then you already have at least 26GB/s main memory bandwidth, which given the amount of performance and RAM is more than enough (its more bandwidth per flop than XBox One and with half the memory to feed from).
With the system you're talking about (393Gflop Tegra Maxwell with 4GB RAM) it would be quite the waste of resources IMO spending extra on exotic embedded memory setups. With that system the most exotic I think they should go is maybe a small amount of cache (like 3-4MB shared between CPU/GPU) to keep performance very predictable, but that shouldn't take up the kind of space that would make extra SM's impossible.
Of course the fan is going to be there to keep it cool. But the point is you don't need a fan to keep a 300Mhz 2SM Tegra cool, or 4 x 1Ghz A57 CPU's. Its questionable that you'd even need a fan for docked mode (at 700Mhz+).
Pretty much confirms what I had been suspecting after reading some replies from people who know a thing or two. This article opens up as many questions as it answers. Anything's still possible. It's time to chill again.
January, where art thou?
Skittzo0413
Member
At this point I'm betting that they won't release any specs at the January event. We'll have to wait until April probably.
Kikorin
Member
Not that I care so much about power in Switch, but I'm wrong or N already overclocked 3ds during lifetime to improve performance? Maybe at start they only need this level of performance and they'll overclock in future if needed.
Also, why we are assuming this is a standard chip and not customized? Nvidia didn't said worked for hundred engineering years in this?
Also, why we are assuming this is a standard chip and not customized? Nvidia didn't said worked for hundred engineering years in this?
They did not. The new 3DS has a new CPU with two extra cores and 3x clock speed. One of the cores is reserved for the face tracking, however. They never upped the clock of the original 3DS.
th3sicknness
Member
It is possible that the customization could just be the clocking and battery management to hit what Nintendo considered acceptable battery life.
That's not how GAF rolls ;p
Everyone is assuming it's custom, and it's still unclear how powerful it actually is. The fan makes no sense for a standard x1 at these clock speeds.
th3sicknness
Member
In related news this is a positive?
Khronos updated the list of conformant products with Vulkan, OpenGL 4.5 and OpenGL ES to include the upcoming Nintendo Switch console.
http://wccftech.com/nintendo-switch-supports-vulkan/
Khronos updated the list of conformant products with Vulkan, OpenGL 4.5 and OpenGL ES to include the upcoming Nintendo Switch console.
http://wccftech.com/nintendo-switch-supports-vulkan/
Milly Osworth
Member
As someone that doesn't care about handheld gaming, the latest news really sucks. It will be a handheld, with a kinda console mode.
Give me a $/400 box to play Nintendo games on on my tv already
Give me a $/400 box to play Nintendo games on on my tv already
In a handheld battery management always comes into the design to a large degree. But if you limit things this aggressively for battery longevity you also get the added benefit of a cool handheld which doesn't need a cooling fan. Also you don't need to customise anything to drop clock speeds.
Skittzo0413
Member
Clock speed is just a software customization though. Nvidia strongly implied that this chip is custom hardware, not just a TX1 instructed to have lower maximum clock speeds.
Hahaha that one got me, well done!
Even digital foundry said there are additional custom features to x1.
blu
Wants the largest console games publisher to avoid Nintendo's platforms.
This is quite important, actually, as APIs and toolchains normally constitute 50% of the porting effort, and nintendo are largely removing that burden from devs.
KIMISHIMA, DROP THE BOMB!
I laughed
I'm curious about how the device handles the mode change. On Jimmy Fallon they hid that part but there seemed to be at least a message or loading popping up. I wonder if it takes longer depending of the game etc
I don't really believe it'll be as seamless as in the reveal video, though that would have been nice.
I don't really believe it'll be as seamless as in the reveal video, though that would have been nice.
Right now, the worst scenery looks like:
NGC -> Wii
Wiiu -> Switch
Looks really bad, but then I remember what Nvidia says: it's easy to port this generation games to switch
If that machine is really the worst situation, it make that clain a big and fat lie.
More than that, one machine like that can't run Skyrim Remaster. So, I belive that is more about the switch that we know right now.
NGC -> Wii
Wiiu -> Switch
Looks really bad, but then I remember what Nvidia says: it's easy to port this generation games to switch
If that machine is really the worst situation, it make that clain a big and fat lie.
More than that, one machine like that can't run Skyrim Remaster. So, I belive that is more about the switch that we know right now.
In the jimmy falllon video it looked liked it automatically paused the game to me and it looked seamless enough.
The Abominable Snowman
Member
Really?
Was there something in the Skyrim screens shown in the promo that looks better than S:SE or something? Bethesda is crafting completely new assets for the Switch version?
Skyrim SE in itself couldn't run on something with the specs of Wii U. I mean, it's not the best-looking game this year but still, it's far from being a low-key game.
amstradcpc
Member
What people waited from Nvidia?. They tricked MS,after that Sony, and now is Nintendo's turn.
What?
Smurfman256
Member
Nvidia oversold the GPU in both the original Xbox and PS3.
The graphics we see in January and the games announced will be more meaningful to the average gamer than detailed spec info.
- Status