• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Ventura Beat: Nintendo Switch graphics are based on Nvidia's Maxwell Architecture

How reliable are VentureBeat? I've always thought of them as just a typical tech blog, and there's not really a lot in this article that sets them apart from the likes of Engadget and Gizmodo.

Well the author is not a random blogger for sure, Dean Takahashi is journalist since 1988 and he wrote several famous books, especially about the creation of the Xbox.
 
Yup, the article is directly comparing, so they imply FP32. I think the better interpretation, though, is that the author has no clue what he is talking about, and we should disregard the article until someone who actually understands what his sources tell him corroborates it.
This is my approach. Honestly, I'll just wait until the NDA has lifted and someone asks Nvidia directly (because we know Nintendo won't answer this question).
 
I have no interest in the Switch before or after this article.

I don't know why people are getting so hung up on specs. The bottom line for people buying this console is Nintendo games. That's the draw. Not high specs. Everyone should already know Nintendo bowed out of the arms race with the other two consoles. If someone was basing their purchase on specs and power, then they would just buy a gaming rig, PS4 pro or wait for Scorpio.

I have no doubts that this console has more than enough power to achieve what Nintendo aims to achieve with their games. And don't forget how talented Nintendo are at making great looking games. I don't think there is anything to worry about.
 
This place would melt down if the custom Tegra in Switch used elements from the Pascal-based Parker SoC and Nvidia's next-next gen Volta-based Xavier SoC.

wEGnnFq.jpg
 
Well, them having limited tech knowledge and being right on their source isn't incompatible. The problem is people makes it a huge issue, as if Maxwell to Pascal would've made it a different device.

I'm more questioning the journalistic integrity and reach of aforementioned sites. Not so
much their understanding of things.

Were this, say, the WSJ, I most likely would not be asking these things, because they have a track record of being on point.
 
This place would melt down if the custom Tegra in Switch used elements from the Pascal-based Parker SoC and Nvidia's next-next gen Volta-based Xavier SoC.



Now THIS would be a big change. Not because of the architecture change but because of the 512 cuda cores. That's twice more than Maxwell X1 and Pascal X2.
 
This is only worthwhile quote from that article:

“I don’t see Nintendo’s strategy as a risk,” said Peddie. “Too many pundits and fan boys and investors make a serious mistake when they try to compare and contrast Nintendo with Sony and Microsoft. Nintendo has a niche in the affordable, accessible product, and performance is never a leading criteria for them. It is gameplay and immersion. They are never a technology pioneer. Trying to compare Nintendo to Sony is like comparing a Volkswagen to a Corvette. It’s a facetious and fallacious analogy and a discredit to fans who love Nintendo.”
 
Anyone can summarize this thread? Really no time to read through 20 pages rn.

Meltdown over a dubious article.

Slightly longer explanation: the article says two sources confirm that Switch uses maxwell architecture. Meltdown ensues, but there are some glazig issues with the article. First, the writer shows he does not understand how pascal and maxwell architectures work, and Secondly it is unclear whether he is referring to the same dev kit sources that we already know about (though it figures he doesn't since his understanding of the source material indicates to me he won't understand that the chip in dev kits does not necessary end up in the final product).
 
On the bright side if you think of it as a 3DS successor instead of a Wii U successor then it's a mind blowing leap!

If you thought of it as a WiiU successor it would be a 6-7x leap according to this article (which is massive).

Except the article is complete rubbish.
 
We're going to get more home gaming on the go than proper portable games though.

That isn't true at all, Nintendo will focus most of their development at this platform, there are a lot of teams who made 3DS games that will now make Switch games, and not all of them will be Xenoblade... Xenoblade 3 is going to be so good...
 
That's really not surprising at all if true. A Pascal-based Tegra doesn't even exist yet. Nintendo using truly cutting edge technology goes against everything they've done in the last 20 years.
 
I think we can all agree GAF (including myself) is not tech savy. 99% doesn't know what Maxwell, Pascal, TFLOPS and all those terms actually mean.

This is my take away from every tech-related thread on GAF. 99% of the thread has no idea what the information they're given even means (myself included) and yet some still feel compelled to be chicken little.
 
This means dick-all.


It's a Nintendo console. Y'all were never getting ports regardless. And even with it being Maxwell based, you're talking a substantial jump for Nintendo hardware, which will show in the exclusives.

Ubisoft's Chicken-fucking-littlez.
 
That's really not surprising at all if true. A Pascal-based Tegra doesn't even exist yet. Nintendo using truly cutting edge technology goes against everything they've done in the last 20 years.
We could argue that the 3D screen on the 3DS is pretty cutting edge buy hey, you probably didn't meant it that way.
 
I tend to believe this and it does not really matter to me. I'm still gonna buy it because of Nintendo softwares.

I always know that Nvidia will going to have a better GPU on their next Tegra... because they are Nvidia. Nintendo will choose technology that suits them, rather than aiming at the latest and greatest components.
 
I only care about this because it means worse battery life. :(

Even if it is maxwell architecture, that would still be a premature conclusion. If they have maxwell architecture with a die shrink to 16nm the efficiency will be on par with Pascal (that shrink is the reason Pascal is typically more efficient).
 
This means dick-all.


It's a Nintendo console. Y'all were never getting ports regardless. And even with it being Maxwell based, you're talking a substantial jump for Nintendo hardware, which will show in the exclusives.

Ubisoft's Chicken-fucking-littlez.

Quoting myself from earlier

...except we've had several posts from a very reliable insider like Matt stating that yes, Switch can get ports from PS4 and Xbox One on a technical level? It seems this time the real matter will be if games generate enough ROI to be considered a profitable venue in the long run.
 
Following the 1TF claim from the article it'd also be big compared to Wii U with a x8 jump in flops.

That would be a record-breaking flop per watt performance on a chip under 7W. I don't think nVidia has gotten anywhere near that in sustained performance, only on paper. Even if they were talking FP16, I think its safe to say 300-400GFLOP/s FP32 is an absolute ceiling. And it's still incredible.
 
We could argue that the 3D screen on the 3DS is pretty cutting edge buy hey, you probably didn't meant it that way.

I guess!

But it's also an example that going that route didn't do them much good. Had to drop the price massively soon after release, and eventually they even released a 2D alternative.
 
Where does the 5-8 hour battery life report (from Laura's sources iirc) come into this? We know for sure (from a Ubisoft source, iirc) that the current X1-based devkits get 3 hour battery.
 
Not in handheld mode since its got a 720p screen so it will downgrade it.

Unless they can do 1080p in handheld mode too downsampled making it look even better

That's likely how it will be. Rendered at native 1080p (or 900p) and then downsampled in 720p on the screen.
 
This means dick-all.


It's a Nintendo console. Y'all were never getting ports regardless. And even with it being Maxwell based, you're talking a substantial jump for Nintendo hardware, which will show in the exclusives.

Ubisoft's Chicken-fucking-littlez.

Ports are announced.
 
Where does the 5-8 hour battery life report (from Laura's sources iirc) come into this? We know for sure (from a Ubisoft source, iirc) that the current X1-based devkits get 3 hour battery.

Well, dev kits tend to take more power because it needs to run a dev environment in addition. Also, the dev kits are not necessarily representative of the final product, so it is possible the final product is more efficient than the dev kit.
 
I guess!

But it's also an example that going that route didn't do them much good. Had to drop the price massively soon after release, and eventually they even released a 2D alternative.

It's hard to separate anything from the fact that the launch 3DS (and Wii U) software were completely unappealing to the masses. Most people forget that the original Nintendo DS was suffering low sales at launch until Nintendo just started releasing hit after hit in the following year.
 
That's likely how it will be. Rendered at native 1080p (or 900p) and then downsampled in 720p on the screen.
That's a waste on the battery. Expect native 720p on the handheld with 720/900/1080p docked depending on the processing overhead. Probably also 4k video when docked.
 
Where does the 5-8 hour battery life report (from Laura's sources iirc) come into this? We know for sure (from a Ubisoft source, iirc) that the current X1-based devkits get 3 hour battery.
I thought that came from NateDrake? Anyway wasn't that more they were aiming for 5-8. Companies always aim higher especially if they are currently working on the spec's for the handheld side. Downclocking the handheld as low as it can feasibly go to increase battery performance would be part of the finalizing.
 
I have a few immediate thoughts after reading through the article:


  • Firstly, it's worth noting the difference between Maxwell and Pascal is almost entirely down to the manufacturing process. Maxwell was made on 28nm (and in the case of the TX1, 20nm) whereas Pascal is made on 16nm. The actual architectural difference between the two is minimal, and aside from improved color buffer compression, largely irrelevant for a device like the Switch.
  • Despite that, the article never makes any mention of the manufacturing process. I find that extremely strange, as it's obviously the defining difference between the two sets of GPUs.
  • In fact, the article gets the difference between the two completely the wrong way around, saying "Nintendo’s box is relatively small, and so it has to fit into the heat profile of a portable device, rather than a set-top box. That’s another reason that explains the older Maxwell technology, as opposed to the Pascal’s state-of-the-art tech." Pascal is literally a more power efficient version of Maxwell, so the incentive would be the other way around.
  • The author says "we expect the Nintendo Switch to be more than 1 teraflop in performance", which is notably higher than even those of us who were expecting Pascal were considering (I literally posted earlier today with a 500-750 Gflop estimate). If this is a Maxwell chip, then that would mean at least 4 SMs (512 "CUDA cores") at 1GHz, as they're not going to be able to push much past that on 28/20nm. This is a much larger GPU than most people would have been expecting.
I see a few different scenarios here:


  1. The Switch SoC uses Maxwell at 20nm, and simply has a much larger GPU than anticipated to account for the performance.
  2. Nintendo looked at the feature-set planned for Pascal when design started, realised that the new features were largely irrelevant, and decided that they would save time and just use a straight-forward die shrink of Maxwell to 16nm instead. That would technically be a Maxwell GPU, but would be almost completely indistinguishable from Pascal in terms of performance.
  3. The sources are wrong about Maxwell, the 1 Tflop performance, or both.
Basically, if you're to take the article as being accurate, then the only worthwhile takeaway is this quote:



A Maxwell Tflop is identical to a Pascal Tflop, and it's largely irrelevant to us whether they achieved that by using a larger Maxwell GPU on 20nm/28nm at a lower clock or a smaller Pascal GPU on 16nm at a higher clock.

Great summary, guess we'll have to wait and see which scenario it is.
 
If this is lacking power, then it would still be pinicle. The only thing I really want is apps, battery, and no load screens.
 
That's a waste on the battery. Expect native 720p on the handheld with 720/900/1080p docked depending on the processing overhead. Probably also 4k video when docked.

That would mean the docked image resolution would be 720p upscaled ? No way it will be what. That would be worse than Wii U on a much more powerful machine.
 

The Graphics idiots on GAF already are "OUT IF CONFIRMED RAAAH!" forgetting that hte article states "it can't handle 4K graphics".... like the PS4 can LOOOOOL! (The original not the pro, but if you had expected 4.8 teraflops in the Switch you were an idiot anyway.). It's about 1 teraflop and most people know this already, and with being Nvidia rather than AMD, you can't directly compare them. Wait until it's out.. rather than go "OMG cartoony graphics... argh so kiddie!" in here... like the Article who states that all Nintendo's graphics are "cartoony" due to them having an ART STYLE rather than just realism(tm).
 
Top Bottom