Support NeoGAF

[DekuLink]

http://venturebeat.com/2016/12/14/nintendo-switch-specs-less-powerful-than-playstation-4/

The semi-custom Nvidia Tegra processor in the machine is still powerful enough to play typical Nintendo cartoon-style games (like the Mario series)

Always found quotes like this a bit silly, I mean of course Nintendos games will run, they make their games around their own hardware, and as far as other developers goes, it's not like cartoony games are always less demanding, they just tend to need less powerful hardware than realistic games to look good, like with Wind Waker.

If Switch is significantly less powerful than ps4/xbone, a cartoony game made to take advantage of those consoles to the fullest won't run on Switch just because it's cartoony

 

[disap.ed]

So not only do these idiots somehow said that Pascal would be hotter and larger than previous chips (except the opposite is true), they also said the Switch is going to be at least 1TF anyway?

Seriously, this is idiotic. I think we can safely throw this rumor into the trash pile.

Honestly, I think we should trust Nate over this "report".

 

[Risk Breaker]

I can't say I'd be surprised, this is Nintendo after all. But a rebranded Shield would be shameless from nVidia and their hundreds of man years of engineering.

Again, not surprised especially since I saw Zelda on the actual NS looking ROUGH, but I also don't believe everything I read on the Internet, so I have no reason to trust this rumor over any other NS rumor. I'll wait for facts.

I'm not familiarized with hardware arquitecture, but considering Switch is more a portable than a home console, I didn't expect it to be very powerful.

And it has UE4 support, that was pretty surprising for me.

And we all want Switch to be cheap, right?

My 3 year old phone had UE4 support 2 years ago my man.

 

[EVH]

Dont know whats more ridiculous, the way this article is written or people expecting PS4 and PS4K power in a portable. Sure it needs to last for 8h battery with that power and be less than 200$ amirite GAF?

 

[Sesuadra]

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.

I tweeted this to dean and hope to get an answer, maybe he'll say something about it.

 

[Nanashrew]

What was rough about the Switch version of Zelda? It was on a terrible projector under heavy studio lighting so everything looks washed out and terrible. Frame rate was the only thing we could tell and it was much better compared to the Wii U version.

I tweeted this to dean and hope to get an answer, maybe he'll say something about it.

He or someone at Venture Beat are already hastily and stealthy editing the article from what it was. Likely based from Thraktor's post. I think Dean might have been spammed with Thraktor's post at this point lol.

 

[PingPongvonLaserstein]

And we all want Switch to be cheap, right?

Yes. Cheap, paper thin, at least as powerful as a PS4, 10-15 hours of battery life, and for the love of all things holy, no bezel.

Is that so much to expect?

 

[Risk Breaker]

What was rough about the Switch version of Zelda? It was on a terrible projector under heavy studio lighting so everything looks washed out and terrible. Frame rate was the only thing we could tell and it was much better compared to the Wii U version.

LODs, the terrain in general, I'm on mobile so I can't be concrete right now, but the previous trailers I found amazing-looking, and the NS footage at Fallon was just... meh. Still really excited about it, don't get me wrong, just want it to look like that trailer with the frog jumping out of the water, or better.

 

[Sesuadra]

What was rough about the Switch version of Zelda? It was on a terrible projector under heavy studio lighting so everything looks washed out and terrible. Frame rate was the only thing we could tell and it was much better compared to the Wii U version.



He or someone at Venture Beat are already hastily and stealthy editing the article from what it was. Likely based from Thraktor's post. I think Dean might have been spammed with Thraktor's post at this post lol.

what? really? if so I just...I don't know lol.

Yes. Cheap, paper thin, at least as powerful as a PS4, 10-15 hours of battery life, and for the love of all things holy, no bezel.

Is that so much to expect?

don't forget no plastic. only lightweight aircraft aluminium or titan.

 

[Herb Alpert]

Always found quotes like this a bit silly, I mean of course Nintendos games will run, they make their games around their own hardware, and as far as other developers goes, it's not like cartoony games are always less demanding, they just tend to need less powerful hardware than realistic games to look good, like with Wind Waker.

If Switch is significantly less powerful than ps4/xbone, a cartoony game made to take advantage of those consoles to the fullest won't run on Switch just because it's cartoony

Yeah that's what baffles me. How a gaming "journalist" can write something like that, this makes absolutely no sense. "nintendo games are cartoony kiddy games, so of course thay are less demanding than the XboxPs games which are mature and edgy and therefore more demanding"
I could expect that from some dumb posters here but from someone who should know his stuff this really stupid.

Honestly, I think we should trust Nate over this "report".

There's a post from nate in this thread where he says he's not so sure now. Acording to his sources during summer they were just seriously considering pascal. They may have chosen maxwell in the end.

Again, this doesn't mean "lol another WiiU" at all. I think a lot of people were eagerly waiting for "bad" switch news, cause overall the tone was very positive these lasts weeks. A lot of people were nowhere to be seen in these positive threads and now they just reappear. They must feel some relief.

 

[DekuLink]

I hope most of you guys are not serious.


Pascal vs Maxwell is not about the graphical power.

It's more about power consumption. This article is.. questionable.

Well if there is a power consumption difference only, that still means it can't be clocked as high in portable mode due to battery life and heat, thus giving a more power effecient chip the possibillity to be clocked higher.

 

[KZObsessed]

Well, at least it'll be cheap. I'm still more interested in Switch than I ever was for Wii and Wii U.

 

[Hermii]

So it's a 1 tflop maxwell chip that can only do cartoon graphics. Makes sense.

 

[Instro]

I hope most of you guys are not serious.


Pascal vs Maxwell is not about the graphical power.

It's more about power consumption. This article is.. questionable.

Confused as to why people keep making this statement. It is in the sense that a smaller node allows improves efficiency and requires less cooling, thereby allowing them to increase clockspeeds both in portable and docked mode.

 

[oti]

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.

Thank you for this post. Very interesting context for us who don't understand much about the tech. Very appreciated.

 

[AquaWateria]

The question is why would Nintendo chose Maxwell instead of Pascal especially if they're all about efficiency.

I mean is Maxwell cheaper than Pascal?

 

[ShadowFox08]

So if this article is true (and OMG how badly written it is) the switch is more powerful than a lot of people here expected. I mean, the consensus was clearly below XboxOne.

But why all the morons saying "lol nintendo being cheap", "lol every nintendo new console", "they're doomed lmfao"?

Even if the maxwell chip ends up being true, don't expect everything else to be accurate as well. Half of what he is saying sounds like conjecture, including the 1TFLOP expectation. BTW the 1 TFLOP for Maxwell, is when its fp16.
http://www.anandtech.com/show/8811/nvidia-tegra-x1-preview/2

Realistically speaking it can't be 1TFLOP all the time. It's going to average out with the 512 GLFOPS at fp16. So something around +750 might be possible. Nvidia and Nintendo says its a custom chip, so it could be better or worse. I'm not expecting better.
http://www.anandtech.com/show/8811/nvidia-tegra-x1-preview/2

 

[Jonneh3003]

So why are people saying "I'm out"? What are you expecting?

 

[Mariolee]

The question is why would Nintendo chose Maxwell instead of Pascal especially if they're all about efficiency.

I mean is Maxwell cheaper than Pascal?

Yes?

 

[Herb Alpert]

So it's a 1 tflop maxwell chip that can only do cartoon graphics. Makes sense.

LOL
I just read the article again, OMG the stealth editing! The guy seems so amateurish...

 

[PingPongvonLaserstein]

don't forget no plastic. only lightweight aircraft aluminium or titan.

Of course. I thought that was a given.

 

[Schnozberry]

The editing after the fact has changed the entire tone of the article. It doesnt even seem like it was written by the same person.

 

[SinCityAssassin]

I'd be mad too if I can't get The Pascal Graphics. Where else can I get the Pascal Graphics.

 

[Nanashrew]

LOL
I just read the article again, OMG the stealth editing! The guy seems so amateurish...

That's the weirdest thing. He's a tech journalist and has been one for 25 years now according to his profile, and he's even worked for WSJ and LA Times.

But this article is written so terribly written and amateurish. It's bad.

 

[dtm808]

We could just wait less than a month and see what this system can do. If it runs games well, looks good on the handheld and TV, and devs like it, then i do not see a problem with the underlying hardware.

 

[KHlover]

So people are upset because the Switch will use a slightly older chip, BUT according to this rumor still is stronger than even the most generous guesses expected on GAF?

Sometimes I genuinely don't understand this place.

 

[Jonneh3003]

So people are upset because the Switch will use a slightly older chip, BUT according to this rumor still is stronger than even the most generous guesses expected on GAF?

Sometimes I genuinely don't understand this place.

People just like to have something to be angry about

 

[guit3457]

So why are people saying "I'm out"? What are you expecting?

They were never in

 

[Rubixcuba]

So people are upset because the Switch will use a slightly older chip, BUT according to this rumor still is stronger than even the most generous guesses expected on GAF?

Sometimes I genuinely don't understand this place.

I've only just seen the news, is it anything more than just a good old Nintnedo is doooooooomed?

 

[dark10x]

This thread, man...

 

[mrklaw]

Pascal uses less shaders at higher frequencies to beat Maxwell. The GTX 980 uses 2048 shaders (16SM) @ just over 1.1ghz to hit 4.6tflops. GTX 1060 uses 1280 shaders @ ~1.7ghz for 4.3 tflops. It does all this on less power (120watts vs 165watts) otherwise the architecture is pretty much identical from a flop to flop perspective.

so are we approaching semantics now? Whats the practical difference between a maxwell @16nm and a pascal @16nm?

I suppose what matters is... *is* it at 16nm, or is it a fat old 28nm chip? Thats the big question for me. I thought there were issues getting older processes die shrunk and it'd actually be cheaper to take the newer chip - aren't Xbox one S and PS4 Pro using a later architecture for their shrinks?

 

[The_Ayertosco]

So people are upset because the Switch will use a slightly older chip, BUT according to this rumor still is stronger than even the most generous guesses expected on GAF?

Sometimes I genuinely don't understand this place.

Don't you know? Unfortunately, Maxwell has only Cartoon-style based shaders, where Pascal use the new Mature & Realistic shaders.

 

[Jonneh3003]

This thread, man...

When someone from Digital Foundry says this I dont think there's much reason to worry.

 

[TheMagician]

So why are people saying "I'm out"? What are you expecting?

They were never in.

Nintendo threads on Neogaf get bombarded with trolls but the mods never stop it.

 

[KingSnake]

This is one of the strangest articles I've read in a while. I read it after being edited and it seems like the author learned that there is a Pascal based Tegra only after he published this.

WTF at the paragraph about Nvidia 1060?

There is no clear mention about the fab process. (which is the biggest issue with Maxwell from my point of view, using 20nm is in no way future proofing your new console)

Also says that it will have above 1 Tflops, which is more than expected anyhow, be it Pascal or Maxwell.

This article is really a strange mix.

 

[ShadowFox08]

So people are upset because the Switch will use a slightly older chip, BUT according to this rumor still is stronger than even the most generous guesses expected on GAF?

Sometimes I genuinely don't understand this place.

The 1 TFLOP expectation thing is BS, because X1 Maxwell is only at 1 TFLOP when its fp16, which won't be most of the time(its 512 GFLOPS at fp32). Expect over 700 at the most, if the custom chip is exactly using maxwell x1.


Anyway, that is interesting to note though about the nvidia shield comparision to switch.. Why would Nvidia want to rerelease the shield on a nintendo console with nearly the same specs with slight modifications(3 to 4GB ram for one)? Didn't the shield sell for like $200 at launch too?

 

[Gamer @ Heart]

You guys are asking alot in hoping gaffers would READ the article AND actually take a minute to process and think about what it means.

1 TF would be insane and I'm shocked if they had multiple sources comfirming that to them.

 

[mrklaw]

also there is the Nintendo conservative factor. Maybe they weren't willing to take a last minute decision or risk waiting for Pascal and wanted something locked down much earlier, which would lead them to sitting with Maxwell. Might also have led them to accept 28nm and design the rest of the system around that SDP.

Middle ground is a 20nm maxwell solution which exists and possibly earlier enough for Nintendo to go that route.

 

[mrklaw]

This is over of the strangest articles I've read in a while. I read it after being edited and it seems like the author learned that there is a Pascal based Tegra only after he published this.

WTF at the paragraph about Nvidia 1060?

There is no clear mention about the fab process. (which is the biggest issue with Maxwell from my point of view, using 20nm is in no way future proofing your new chicks)

Also says that it will have above 1 Troops, which is more than expected anyhow, be it Pascal or Maxwell.

This article is really a strange mix.

Nintendo + strangely worded article + rumours = 20+ page thread on Gaf. Its a potent combination.

 

[Sesuadra]

This thread, man...

you need a hug? :/ It's kind of horrible sometimes in such threads..

 

[MatrixMan.EXE]

Dont know whats more ridiculous, the way this article is written or people expecting PS4 and PS4K power in a portable.

The people expecting an Xbox One/PS4 power tier device, even after we saw what the thing looked like. The argument as to why Switch would be getting certain games was always, "but it's compatible with Unreal Engine 4". Yeah, so are iPhones. Doesn't mean they can run high-end UE4 games.

That said, for a hybrid device, it'll certainly pack a punch. Even having something in between Wii U and Xbox One is good enough for a device like this.

 

[Lilalaunebaer]

Lol, this article and parts of the thread.

Maxwell, shitty in comparison to Pascal. It´s only 1TF and can only do those cartoony mario nintendo games. No way, PS4/XB1 games can run on this thing. It´s almost as strong as the XB1.

Article being edited on the fly. Yeah, i´m gonna go ahead and say it´s bull.

 

[Gamer @ Heart]

also there is the Nintendo conservative factor. Maybe they weren't willing to take a last minute decision or risk waiting for Pascal and wanted something locked down much earlier, which would lead them to sitting with Maxwell. Might also have led them to accept 28nm and design the rest of the system around that SDP.

Middle ground is a 20nm maxwell solution which exists and possibly earlier enough for Nintendo to go that route.

If they were willing to skip the holiday season for the sake of a smoother hardware production and software release schedule, I can't imagine they were against waiting for the right chip that meets their strict handheld demands.

 

[SinCityAssassin]

The argument as to why Switch would be getting certain games was always, "but it's compatible with Unreal Engine 4". Yeah, so are iPhones. Doesn't mean they can run high-end UE4 games.

Because they aren't dedicated gaming systems built for that.

 

[GarciaHotspur]

People expecting Switch to be as powerful as PS4 are insane ?

 

[TheGreatMightyPoo]

 

[SinCityAssassin]

People expecting Switch to be as powerful as PS4 are insane ?

If they still believe supermetaldave at this point, they might as well be.

 

[KingSnake]

Taking the article at face value, Nintendo Switch is a portable console that has over 1 Tflops in processing power. Think about that.

 

[ggx2ac]

Why are we still getting people in here that think that other people think the Switch is as powerful as a PS4?

 

[Gamer @ Heart]

People expecting Switch to be as powerful as PS4 are insane ?

No one who actually followed any of this or knows even a fraction about the hardware market.

The way this article framed this 'report' implies the author may not be one of those.