• 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.

Digital Foundry: Nintendo Switch CPU and GPU clock speeds revealed

Status
Not open for further replies.

LordKano

Member
Would adding a third SM cost more than increasing the size of the battery? I think this is the tradeoff that may have been made there, as higher clock speeds for 2 SMs requires a larger battery.



Someone send out the Matt Signal!

I'm on it !
The+Duck+Signal+Banner+PNG.png
 
I feel like something doesnt add up - especially if this is getting Skyrim Remastered...i dont see Bethesda being interested in this if it were as weak the article suggest.
Could it be getting The Ledgendary version of vanila Skyrim? Basically the Legendary Version that's on the X360...?
 
In the Maxwell generation, 1 SM is 128 CUDA cores

Oh. Well then if we take the dev kit leak to be true, which both Emily Rogers and Digital Foundry seemingly confirm in this report, then we know the SoC has 2 Super Marios (256 cores). So why are speculating it might have more?
 

defferoo

Member
Well, I guess it will be cheap at least?

lower clocks don't necessarily mean cheaper, unless these are crap tegra parts that nobody else wanted, but given their custom nature, i'd expect they aren't.

The weird thing is running it at a lower than standard clock speed when docked. They could have easily kept it at 1000 MHz since they have active cooling in there. Why unnecessarily limit the docked version? I can understand CPU, but the GPU? Why wouldn't you let the devs have as much power as possible when it is docked?
 
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).

TL:DR

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.

Another great piece of analysis Thraktor, this should be added to the OP.
 

saskuatch

Member
I'm actually thinking these specs are quite optimistic. With all the crap they have crammed into the controllers and if the console comes with a dock included. Based on Nintendo's margins at 299 I would expect something along the lines of 50-100 tf docked in order to fit in a smaller less expensive battery.
 

diablos991

Can’t stump the diablos
Did people expect a portable PS4?

While it's depressing how weak this machine is it's relatively strong when compared to other Nintendo consoles.

This will be a pain for third parties and likely drive them away from direct ports.
 
Because it'll still look identical to Wii U games, but in 1080p, and because there's no logical reason for Nintendo have done this. Even with Maxwell, it could have easily hit 500MHz when portable without a running fan. And it's not like it makes the unit any cheaper either, since there's not really any savings from just lowering the clock speed. At this point, a 3DS successor at a lower price would have been better for pretty much everyone.

Lets get this clear first. We dont know for certain what the Switch is and what it is capable of. We are going off what people are telling us is fact based on something that isnt even out yet.

And what is wrong with getting Wii U games at 1080 p? This is an improvement over the Wii U with games coded specifically for it. For all we know we are going to get better games that look better than Wii u games on the Switch when it is developed for.

The reason Nintendo did this is because it was getting too hard / expensive supporting 2 platforms on 2 different architectures and OS's. They also most likely saw that their home console division was going downhill in addition to their portable division. This was the best way to consolidate resources across 1 device and put out the best product possible.

Lets be clear we dont fully know what is possible with this platform because Nintendo is waiting till January to answer any questions.
 
Crow eating the thread. Nintendo fans never learn.

Props to big n for releasing a 2017 console weaker than your avg smart phone

Some people still in denial too
 

KAL2006

Banned
Its the current clocks than don't add up, the current profile would draw a very, very low amount of power that should have very little trouble passively cooling even in docked mode and without a fan. Its a tiny TDP as currently laid out.

That's why it makes no sense. You could just have had a metal backplate and a simple fan in the dock for this kind of use if you had some reason to worry about minor heat dissipation issues. Putting the fan *in* the unit and it seemingly running in portable with these clocks is a big ?????.

The patent states it runs the fan in portable mode. But it's just a patent. Perhaps fans don't turn on in portable mode and only turns on in docked mode. As for just allowing fans to be on the dock instead of the hardware itself. Perhaps it's cheaper to just have the system to have a fan instead of dealing with the design and manufacturing process of having a dock with a fan that Works with the Switch system.
 
This is damage control & marketing just like calling the DS a "third pillar" was.



It's interesting. I'm now wondering if the 3rd pillar stuff could be an explanation for that low clock and basically would be a safe exist in a year or two, in case Switch is a failure, to come up with a real handheld that'd be compatible with all the stuff that was done on Switch and even backward compatible.
 

RootCause

Member
At least these specs basically confirm GameCube games will be native resolution and not be rendered in HD like Dolphin. For me that's dissapointment I may as well just play GameCube games on Dolphin
VC tittles have always been presented in their original state. Don't see why it would've been any different now.
 
The patent states it runs the fan in portable mode. But it's just a patent. Perhaps fans don't turn on in portable mode and only turns on in docked mode. As for just allowing fans to be on the dock instead of the hardware itself. Perhaps it's cheaper to just have the system to have a fan instead of dealing with the design and manufacturing process of having a dock with a fan that Works with the Switch system.

The problem with this last statement is that a fan in a portable device is a moving part, which is almost always the point of critical failure of a device. If Nintendo could have avoided placing a moving fan in the device- even if it cost a bit more to manufacture- then they certainly would have, as a fan in a portable will wind up costing them more in the long run due to warranty repairs/replacements.
 
Crow eating the thread. Nintendo fans never learn.

Props to big n for releasing a 2017 console weaker than your avg smart phone

Some people still in denial too



Right. Which average smartphone please ? Bring up facts, right now. Or maybe you're shitposting with console war crap ?
 

Vhalyar

Member
I have not turned on any of my consoles in almost nine months now, I honestly don't miss them. Specs are sad, but I'll gladly take a more portable Wii U.
 
Does anyone actually think that the Wii U was under powered for Nintendos own games? (the actually only important titles on the system, let's be honest). The Wii U has some of the most beautiful, gorgeous video games on any console. Marjority of it (all the high action ones) run at 60fps, they have amazing art and they are all play great.

Now we will get that in a handheld/console hybrid, but with more power. And, you won't have to buy TWO systems any more to play Nintendo games, that's a huge plus.
 

Xellos

Member
So regarding the possibility of more SMs:

I believe according to graphs posted here, you get higher performance per watt with more SMs than you do by simply increasing the clock speed. Assuming Nintendo is targeting something like 5 hours of battery life they would be better off using a lower clock speed regardless of the amount of SMs.

So when we look at this from a design perspective, the two cost related variables here are:

  • Die cost/SM cost
  • Battery cost

And two power related variables here which directly affect the cost variables above:

  • # of SMs/CUDA cores
  • Clock of SMs/CUDA cores

It may be possible that Nintendo opted to increase the number of CUDA cores to reach their target performance rather than increase the clock speeds, as increasing the clock speed would require a larger and more expensive battery. Does anyone know how much money a larger battery would cost a console maker relative to an additional SM on the die?

The most likely scenario is that Nintendo went with 2 SMs, lower clocks, and a smaller battery to reach the lowest costs possible. Whether Switch is using 2 SM or 3 SM, we're looking at 150 GFLOPS or 230 GFLOPS (portable). I doubt Nintendo puts much value on 80 GFLOPS, especially if they're getting the same CPU/RAM either way. If a game can run well on 230 GFLOPS GPU, it can almost certainly be scaled down to run on 150 GFLOPS. No sense spending more money if that's the case.
 
Does anyone actually think that the Wii U was under powered for Nintendos own games? (the actually only important titles on the system, let's be honest). The Wii U has some of the most beautiful, gorgeous video games on any console. Marjority of it (all the high action ones) run at 60fps, they have amazing art and they are all play great.

Now we will get that in a handheld, but with more power. And, you won't have to buy TWO systems any more to play Nintendo games, that's a huge plus.

Does Zelda run at 60fps all the time, while also providing a pop-up free open world?
 

kIdMuScLe

Member
What's your opinion on the fact that there is at least one fan then? Thraktor's writeup did a good job of explaining why active cooling should not be necessary if the specs are as Digital Foundry is saying, in portable mode or console mode.

Malo doesn't know anything as he has been constantly been proven wrong
 
These specs don't mean much without Nintendo hardware to compare it to.

How does the Switch, both docked and un-docked, stack up to the Wii U? Better, worse, or same?

To the 3DS? How much better? Leaps and bounds?

That's all that matters.
 

Zedark

Member
Crow eating the thread. Nintendo fans never learn.

Props to big n for releasing a 2017 console weaker than your avg smart phone

Some people still in denial too

People pointing out some odd inconsistencies that arise from the DF article and speculating what that could mean for the system is apparently equivalent to being in denial. You heard it here first guys!
 
These specs don't mean much without Nintendo hardware to compare it to.

How does the Switch, both docked and in-docked, stack up to the Wii U? Better, worse, or same?

To the 3DS? How much better? Leaps and bounds?

That's all that matters.

To 3DS, SIGNIFICANTLY more powerful.
To Wii U, a good bump.
 

KAL2006

Banned
The problem with this last statement is that a fan in a portable device is a moving part, which is almost always the point of critical failure of a device. If Nintendo could have avoided placing a moving fan in the device- even if it cost a bit more to manufacture- then they certainly would have, as a fan in a portable will wind up costing them more in the long run due to warranty repairs/replacements.

Surface Pro and many laptops have no issues having a fan. Although it is a moving part of the part only moves when the system is stationary and not when portable it's probably easier to include the fan in the system rather than the dock.
 
The most likely scenario is that Nintendo went with 2 SMs, lower clocks, and a smaller battery to reach the lowest costs possible. Whether Switch is using 2 SM or 3 SM, we're looking at 150 GFLOPS or 230 GFLOPS (portable). I doubt Nintendo puts much value on 80 GFLOPS, especially if they're getting the same CPU/RAM either way. If a game can run well on 230 GFLOPS GPU, it can almost certainly be scaled down to run on 150 GFLOPS. No sense spending more money if that's the case.

That's definitely a good point and a distinct possibility. It still doesn't explain the fan though, because going by the clock speeds and specs assumed by DF a fan shouldn't be necessary even in docked mode.

Yet adding a fan to a portable device greatly increases its chance of mechanical failure, thus greatly increasing their warranty related costs.
 

Seik

Banned
So, is this thing less powerful than a WiiU? Has Nintendo learned nothing?

Equals or a bit better than Wii U in portable mode.

2.5X stronger than a Wii U on the dock/TV.

If they're downclocking I would see why they even went with an X1 instead of a cheaper chip.

I wonder how much more Tegra X2 are.

Since these are Pascal based, wouldn't those be more powerful and even more power efficient? I guess the choice of the X1 was purely based on price.
 

Instro

Member
Does anyone actually think that the Wii U was under powered for Nintendos own games? (the actually only important titles on the system, let's be honest). The Wii U has some of the most beautiful, gorgeous video games on any console. Marjority of it (all the high action ones) run at 60fps, they have amazing art and they are all play great.

Now we will get that in a handheld/console hybrid, but with more power. And, you won't have to buy TWO systems any more to play Nintendo games, that's a huge plus.

No they don't. And yes many WiiU games looked like they were being held back. Particularly stuff like Xenoblade and Zelda.
 
Status
Not open for further replies.
Top Bottom