• Hey everyone, staff have documented a list of banned content and subject matter that we feel are not consistent with site values, and don't make sense to host discussion of on Famiboards. This list (and the relevant reasoning per item) is viewable here.
  • Do you have audio editing experience and want to help out with the Famiboards Discussion Club Podcast? If so, we're looking for help and would love to have you on the team! Just let us know in the Podcast Thread if you are interested!

StarTopic Future Nintendo Hardware & Technology Speculation & Discussion |ST| (Read the staff posts before commenting!)

nanite is, in general, heavy. not as heavy as VSM or lumen, but there's a cost. that cost scales with how many polygons your asset is and your resolution. don't throw polygons at your model just because you can with nanite. those "sculpt in zbrush, import directly to UE5" claims are still unrealistic. nanite falls back to mesh shaders once your triangle takes up 4 pixels, so you're going to hardware acceleration then
Ahhh I see, thank you. Yeah, I like the concept but it sounds like the technology needs to be shored up a bit to really offload all that LOD work.
 
Then he goes on to say the PS3 was half as powerful as the target Sony showed off?? Am I remembering his words correctly? That's not good if also true for Switch 2! I have no clue the event he's referencing about the PS3, was that for devs or consumers? Could he be right, and something similar could happen for NG?
You are misunderstanding his point, his point is that tech demos on simulated hardware doesn’t give him confidence that the demo is illustrating what is possible, citing the PS3 scenario with MGS4 where the developer demo ran much better and looked better than the actual hardware was capable of.

He’s not saying this as a way to say that Drake isn’t capable, he isn’t confirming that. Simply that simulated hardware is giving him a pause and making him go with a side eye.

My two cents with this is that this doesn’t make sense for Nintendo because Nintendo gains nothing by doing this. Intentionally lying or crafting something that isn’t possible to show developers for them to end up not being able to do that on the hardware isn’t helping anything?? Developers have no issues dropping Nintendo’s platform.

And second, the demo he’s referring to was very deep in the console war trenches against the 360, and it was showcased to the wider public, this is a private demo that is meant only for developers to know about. Sony really wanted to set the tone that their console was powerful and could do these amazing things and the power of the CELL or whatever.


Nintendo is different. They aren’t marketing this and they only have a goal of showing it to devs of the features and capabilities, along with Epic and Nvidia for all the demos of the system at the point in time: faster load times, DLSS, Ray Tracing, Nanite/Mesh Shaders in work, etc.


The simulated target specs are what Nintendo is aiming for though.
 
You are misunderstanding his point, his point is that tech demos on simulated hardware doesn’t give him confidence that the demo is illustrating what is possible, citing the PS3 scenario with MGS4 where the developer demo ran much better and looked better than the actual hardware was capable of.

He’s not saying this as a way to say that Drake isn’t capable, he isn’t confirming that. Simply that simulated hardware is giving him a pause and making him go with a side eye.

My two cents with this is that this doesn’t make sense for Nintendo because Nintendo gains nothing by doing this. Intentionally lying or crafting something that isn’t possible to show developers for them to end up not being able to do that on the hardware isn’t helping anything?? Developers have no issues dropping Nintendo’s platform.

And second, the demo he’s referring to was very deep in the console war trenches against the 360, and it was showcased to the wider public, this is a private demo that is meant only for developers to know about. Sony really wanted to set the tone that their console was powerful and could do these amazing things and the power of the CELL or whatever.


Nintendo is different. They aren’t marketing this and they only have a goal of showing it to devs of the features and capabilities, along with Epic and Nvidia for all the demos of the system at the point in time: faster load times, DLSS, Ray Tracing, Nanite/Mesh Shaders in work, etc.


The simulated target specs are what Nintendo is aiming for though.
Yeah i think the difference is that for Sony nothing happens if they deceive third party devs during a tech demo, Nintendo probably can't get away with stuff like that though because third party devs routinely ignore Nintendo consoles while they are always desperate to release everything they have on a playstation system. Nintendo have to be very careful to get even the tiniest amount of third party support to their consoles and thus can't use a tech demo to troll like Sony did with PS3. If Nintendo uses Matrix Awakens to show the power of their next system and then presents the real deal that is much worse in practice, all that will lead to is all those devs dropping the console like a hot potato and continue to make Playstation/Xbox games like they´ve been doing the last 20 years.
 
No, this doesn't follow, because of the additional limit of 20 attachments. Which is another hard gameplay limit that I didn't even mention. So, effectively, there can be a maximum of 10 objects that aren't single dragon parts that don't despawn at extremely close range. You can't fell a forest and just have the logs sit there.

You could argue, "well, Nintendo doesn't want that", except Nintendo have programmed it so that it remembers that a tree has been chopped until the next blood moon. You can fell the forest. It will stay felled for a decent amount of time. It's just that, any product of that chopping, the log, the fruits, the useful consequence of the chopping, disappear at close range. There is the intention to remember the player's interactions, but an inability to follow through fully. All they can do is set flags: Chopped, not chopped.

Nintendo can't make the world more persistent with the Switch's hardware. And we know that making a more persistent Hyrule is something they've wanted to do for a very long time because when talking about Ocarina of Time's 64DD expansion, Miyamoto was focusing on ideas like, "if you drop a weapon, and then go exploring for like fifty hours and then come back, the weapon is still there. If you chop a sign, it stays chopped."
Even the free-form physical interactions in open-world Zelda are merely the manifestation of ambitions that are decades-old. Refer to Miyamoto's delight that a chopped sign can fall in the water and float there in Ocarina of Time.
Aonuma, in at least one interview and one statement that I can recall, has acknowledged that fans want to be able to change the world of these games, but has so far only been able to deliver that in limited, scripted form.

The Hudson construction signs, for example, are a small example of this. Does it improve gameplay that the signs are fixed at the angle you left them in? No, of course not, but it adds to the tactility of the world. Which is clearly a focus for the Zelda team at this moment in time. But Tears of the Kingdom is a game that is so obviously held back by its hardware that it just continually blows my mind that people keep bringing it up as an example of the opposite.

And this is just persistence. Another consequence of the "we can't store anything in our RAM" is that you also can't really do things at a distance. Throwing a bomb in a cluster of trees in Breath of the Wild fells the trees. Firing a canon from 200 metres away into a cluster of the trees does nothing.

You talk about the old Zeldas, but Link to the Past and Ocarina of Time both featured the draining and filling of large bodies of water. This is something that isn't currently possible under the "we're doing shit for real this time" paradigm of open-world Zelda. Probably not even on a PS5. But it's a clear goal to work towards. Fluid mechanics that you can manipulate in real time will be a mind-blowing gamechanger whenever we eventually see it, people aren't ready for it. But also, from this, you can also infer that traditional Zelda wasn't constrained by the same limitations. Once you made the world seamless, all you could do is make it prettier. But that is the difference between open-world Zelda and, say, God of War. And I say the former has much greater room for growth, and stands much more to gain from technological advancement, and to suggest that Tears of the Kingdom, of all games, wasn't severely limited by hardware is just absurd.

Some fair points but then you have to also take into account decisions like these are not purely based on hardware limitations but also limitations of time / design and budget. It's not always a case of things not being present due to technical limits but rather limits of development scope in general. You can somewhat discern this simply by looking at ToTK and BoTW and how much more complex one is over the other despite them running on the same hardware. A lot of things ToTK does you might have said previously were not in BoTW due to technical limitations whereas it was not the case. It was limitations of design and time.

But also as to your point, the distance for despawning / number of items was quite substantially increased in ToTK from BoTW which does tell us that this was a limit they were looking to increase and probably would have been set to a higher bound if they had more wiggle room. Would they like to further increase it? Probably. But also that does not mean they want these things to be limitless or to completely do away with the process as the game has to be built around certain parameters from a design standpoint
 
I'm reading a lot about Digital Foundry being optimistic, but I don't know, I interpreted their words quite the opposite!
(well Rich's words at least, not John's).

Rich would hardly entertain the idea that the reports are accurate. First saying that the demo might be downgraded to a similar degree that Witcher 3 on Switch was, which in my opinion is not consistent with the reports, unless developers were only shown the Matrix demo running on a small screen to simulate portable mode AND were shown it from far away, in which case devs could have mistaken the graphics for looking better than they were (as I think Rich suggests was the case).

Then he goes on to say the PS3 was half as powerful as the target Sony showed off?? Am I remembering his words correctly? That's not good if also true for Switch 2! I have no clue the event he's referencing about the PS3, was that for devs or consumers? Could he be right, and something similar could happen for NG?
From everything I've read over pages and pages of this thread, it seems to me that most believe it's too late for such a large spec change, in which case, why did Rich say that? Is he making a big mistake or what?

I don't know but either Rich is right, and the situation could be way worse than expected, or Rich loses a bit of credibility after those comments, at least in my eyes.
I don't think Witcher 3 on Switch was playable in docked mode, the blur was just toooo much. And if he's right that history could repeat and Nintendo is overestimating themselves like Sony did, then that's just as bad. So is he right that these are possible?

Edit: I fully expect that the Matrix demo was downgraded compared to the PS5 version, but "comparable" doesn't mean identical, and Rich seems to be combatting the concept that they were even comparable. At least that's how I interpreted his words, especially when he brings up the PS3 thing as well.
Nintendo literally can't afford to do that, their consoles already aren't the main development platform since forever. Deceiving everyone and their mother at Gamescom with an unrealistic, impossible to achieve demostration? That doesn't make any sense, it's going to seriously hurt Switch 2's third party support.
 
Nintendo is different. They aren’t marketing this and they only have a goal of showing it to devs of the features and capabilities, along with Epic and Nvidia for all the demos of the system at the point in time: faster load times, DLSS, Ray Tracing, Nanite/Mesh Shaders in work, etc.
Indeed, when Nintendo announces Drake Redacted the trailer will be effectively like the Switch launch trailer. It will show off and perhaps announce a few new Nintendo titles and highlight a couple of third party titles that previously weren't on the Switch.

I could see an EA games title like Madden or the new 2K NFL title being featured the way NBA 2K was featured in the original trailer.
 
Yes, except for console exclusive deals (as always)

Big 3rd partners and AAA games "impossible" for the actual hardware is in the works for this new system (both ports and new releases)
Will them launch day & date with the console release? — I don't know yet (but the possibility remains)
I hope Nintendo manages their launch window lineup well. Wii U and 3DS has problems with 3rd party games not selling when they were just dumped.at launch.

Games should be spaced out and given time to breathe and give users a chance to jump in. Most likely launch day users will all be playing the big Nintendo game so major 3rd party releases outside of a few support games may not be a good idea as they may not get the attention especially with a smaller initial installed base
 
Broke: There is no Nintendo Direct announcement because they're probably announcing it tomorrow.

Woke: NG is being announced this month. :cool:
 
F51W6dubYAA26L8


Looks like we going to have to wait one more day for the NX2 Reveal
 
Broke: There is no Nintendo Direct announcement because they're probably announcing it tomorrow.

Woke: NG is being announced this month. :cool:
Furukawa tomorrow: ''This september we replace the Nintendo direct with our announcement that the NG will release March 2024, please understand''.
 
Yeah i think the difference is that for Sony nothing happens if they deceive third party devs during a tech demo, Nintendo probably can't get away with stuff like that though because third party devs routinely ignore Nintendo consoles while they are always desperate to release everything they have on a playstation system. Nintendo have to be very careful to get even the tiniest amount of third party support to their consoles and thus can't use a tech demo to troll like Sony did with PS3. If Nintendo uses Matrix Awakens to show the power of their next system and then presents the real deal that is much worse in practice, all that will lead to is all those devs dropping the console like a hot potato and continue to make Playstation/Xbox games like they´ve been doing the last 20 years.
Frankly I do not see the point of lying to developers / partners about the power of the thing. Lying to consumers on the other hand.
 
I hope Nintendo manages their launch window lineup well. Wii U and 3DS has problems with 3rd party games not selling when they were just dumped.at launch.

Games should be spaced out and given time to breathe and give users a chance to jump in. Most likely launch day users will all be playing the big Nintendo game so major 3rd party releases outside of a few support games may not be a good idea as they may not get the attention especially with a smaller initial installed base

This is a reason why I lean towards a Spring release being better than a November release. If you do a Spring release like the Switch originally was, you can continue to launch game after game every month leading up to the holidays. The problem with November is that you really only have a few weeks to get big games out before the end of the year. Yes, there are exceptions, but December isn't normally a big video game release month. Once you get to mid-December, there is basically a 6 week window where nothing big gets released.

Even if Nintendo opts for a September release, you can still get a few months to build up momentum with the new system and not have that lull period right away.
 
Because PC Gaming is x86 land and Nvidia doesn't has the rights to create an x86 processor and thus can't create an power-efficient x86 CPU + Nvidia GPU with access to DLSS, RTX, etc.
Nintendo, by virtue of being a console manufacturer and not being burdened by legacy applications, is able to pick whatever ISA they want for their console and thus, for Nintendo, it's viable to choose Nvidia Arm + Nvidia GPU SoC. But for others PC handheld, they need to use x86 processors because that's what is supported on Windows PC Land and that means they're locked to AMD or Intel.
Thats an interesting insight, thanks for pointing that out. I wasn't aware that the rights to producing an x86 processor lie with Intel and AMD who probably have very little interest in sharing these rights.

I never expected to learn so much from this forum, there is an absolut wealth of knowledge on offer. Best thing reddit ever did was bring Fami to my attention.
 
The original post was "ToTK proves Switch hardware isn’t holding back gameplay or grand designs or huge worlds. "
I would say from a gameplay perspective, as in say, controling your character and its interaction with the world there is not much new compared to say other soul games, its just that when you add its massive world that things get considerably more complex. But i dont think there's any breakthrough in gameplay department. But thats my opionion.
 
Hey, I know you, and others, really expect Nintendo to treat this new hardware exactly like Sony treats the ps5, but I’ve been around long enough to know that the smart money is not expecting Nintendo to follow the same path as Sony/Microsoft. They don’t have the same motivations or goals. At all.

I see your point... and as a matter of fact, I don't believe Nintendo is going to pull a Sony at all myself, either.

On the other hand though, they will likely target another Switch-like lifecycle i.e. at least 5-6 years of decent first/second party support. Plus, they're certainly aware that third parties exist too, and making "easy" ports in the current market scenario is a very welcomed feature.

To cut a long story short, I'm not expecting Nintendo (nor third parties, which will keep giving Sony/MS their priorities) to have any significant policy shift regarding their output; we'll obviously get the usual 4-5 "AAA-like" big guns per gen but yes, I really can't see them becoming a AAA factory with basically most of their teams devoted to this very specific kind of game. But if they play it smart (and I expect they do) we might end up gaining a bit more third party support, which probably won't hurt.

I'm always going to push back against this idea.

Hardware will always limit software. A more powerful Switch would have definitely allowed for a smaller amount of constraints, but I think in the end we just have to accept any given platform... is what it is. Not to mention it's not like there's a plethora of games comparable to TotK in scope/features running on similarly specced hardware.
 
0
The original post was "ToTK proves Switch hardware isn’t holding back gameplay or grand designs or huge worlds. "
Fromsoftware usually don't release their games on Nintendo plattforms. So i am unsure if they will release Elden Ring on Switch 2, their close ties to Sony may mean that Sony asks them to not release Elden ring on Switch 2 due to Sony being terrified of Switch 2 as a viable competitor to PS5 and will want to do what they can do kneecap third party support on the Switch 2.
 
I would say from a gameplay perspective, as in say, controling your character and its interaction with the world there is not much new compared to say other soul games, its just that when you add its massive world that things get considerably more complex. But i dont think there's any breakthrough in gameplay department. But thats my opionion.
That's fine but the whole discussion was over whether Elden Ring was not ported due to hardware constraints or for some other reasons ("graphics" only was claimed).
 
0
Fromsoftware usually don't release their games on Nintendo plattforms. So i am unsure if they will release Elden Ring on Switch 2, their close ties to Sony may mean that Sony asks them to not release Elden ring on Switch 2 due to Sony being terrified of Switch 2 as a viable competitor to PS5 and will want to do what they can do kneecap third party support on the Switch 2.
Yes, that makes sense, happy for it to go on the platform closest to theirs in terms of target market (Xbox) but they're "terrified" of a future platforms will start "kneecapping".
 
I thought the ceiling was 2 and 3.45 TFLOPS mode each? Did something change?
There is some wiggle room, but I'm also noticing some "FLOP drift", partially because some folks are talking in terms of "mixed-precision TFLOPS" which is confusing. Let's talk "limits" for a second.

On the handheld side, Nintendo is limited by battery and heat. At around 550MHz (1.6 "standard" TFLOPS), each small jump in performance causes a larger and larger fall in battery life, and a bigger and bigger increase in heat. Nintendo can make any compromise they want, it's just that the trade-offs get really bad at that point.

In docked mode, heat still matters, but there is also memory bandwidth. There is a point where adding 20% more GPU performance doesn't add you 20% more game performance, just because the GPU can't access memory fast enough to actually use the power it has. Our best guess is that line is around 1000MHz, or 3.0 TFLOPS.

Again, Nintendo can make any compromise they want, but past that point the returns start getting smaller while the costs - a bigger heat sink, a heavier device, fewer chips that meet the quality standard, increased risk of hardware failure - rise quickly.

I would be very surprised if Nintendo goes past those lines. Not because I don't expect Nintendo to delivery high performance, at this point that's exactly what I expect. And not because I expect Nintendo to "cheap out." It's because, as a performance engineer, I would be surprised that Nintendo believed that power was going to be usable.

This thing has 8 CPU cores. It has some form of modern storage. It has a new cartridge format. It has some unknown amount of memory. All of these things eat power, all of these things make heat. All of these things have different "sweet spots" where efficiency is good and where efficiency falls off a cliff. All of these things influence the performance of the others. Want HDR on your screen? That costs power. New Joy-Con features? Power.

When we talk about limits, what we really mean is "costs go up." That's not a hard limit - as @LiC pointed out to me recently, marginal gains are still gains! But it's not just "how much does it cost" but "what other things do we have to spend on." I find it hard to imagine that Nintendo has half a watt to spare and decides "Instead of a 25% more CPU performance, what I really need is 5% more GPU, heavily limited by memory bandwidth."
 
Yes, that makes sense, happy for it to go on the platform closest to theirs in terms of target market (Xbox) but they're "terrified" of a future platforms will start "kneecapping".
Fromsoftware has released 3 games on Nintendo home consoles: Lost Kingdoms I, Lost Kingdoms II on Gamecube and Dark souls remastered on the Switch, so their total lack of history of Nintendo support means that its unlikely that they would release Elden ring on the Switch 2. They don't seem to find the Nintendo ecosystem as worth giving the slightest breadcrumbs to.

Ironically as always it will be easier for Nintendo to get western third party studios to support Switch 2, Japanese third party studios like Fromsoftware will probably continue their Playstation/Xbox only strategies in the future.
 
0
Thats an interesting insight, thanks for pointing that out. I wasn't aware that the rights to producing an x86 processor lie with Intel and AMD who probably have very little interest in sharing these rights.

I never expected to learn so much from this forum, there is an absolut wealth of knowledge on offer. Best thing reddit ever did was bring Fami to my attention.
If you want to know more and why the PC market shaped to be like what it is now, this Anandtech article from more than a decade ago is an excellent read:


TLDR: Nvidia and Intel fought regarding Nvidja rights to make Intel motherboards chipsets and about Nvidia having the capability to create an x86 processor (Nvidia does hold some x86 patents from their acquisitions). Ultimately, it was settled that Nvidia couldn't create an x86 processor nor x86 emulator but, in exchange, Nvidia got a welthy sum of money (Quite a lot of money for the size Nvidia was back then).

It was one of the reasons Nvidia kept investing heavily into Arm ecosystem and why Nvidia can't create an Nvidia x86 CPU + GPU SoC/APU.

Another problem is that PC games are made with x86 in mind and nothing else. And Windows on Arm is still in a terrible state, with the Windows x86 application emulator being very meh.

However, Nvidia is rumored to be working on a Arm SoC for a Microsoft Surface product scheduled for 2025. So we might get to see a heavier push and investment into Windows on Arm and, with Nvidia expertise and GPU IP, perhaps PC Games finally being playable on Arm devices. Nvidia did a demo back then that might be a taste of what a ISA agnostic PC Gaming future might shape to be:
 
Tech wizards of Fami: can someone ELI5 how Botw is supposedly running at 4k60fps with no/reduced load times? Knowing what we know is this technically possible? There’s gotta be sacrifices or compromises somewhere right? Maybe input lag is higher since it’s using DLSS? Or maybe the tech demo file for botw only had Hyrule field and wasn’t running the full game which would explain why it ran so well?

Excuse my ignorance😭 I’ve been keeping it for almost 2 years now and didn’t expect the Switch 2 to supposedly be THIS powerful. I want to believe and get hype but I can’t help but feel a little hesitant. I know DLSS is magic so I was expecting a resolution bump but I didn’t know increased framerate and reduced load times were on the table too
 
There is some wiggle room, but I'm also noticing some "FLOP drift", partially because some folks are talking in terms of "mixed-precision TFLOPS" which is confusing. Let's talk "limits" for a second.

On the handheld side, Nintendo is limited by battery and heat. At around 550MHz (1.6 "standard" TFLOPS), each small jump in performance causes a larger and larger fall in battery life, and a bigger and bigger increase in heat. Nintendo can make any compromise they want, it's just that the trade-offs get really bad at that point.

In docked mode, heat still matters, but there is also memory bandwidth. There is a point where adding 20% more GPU performance doesn't add you 20% more game performance, just because the GPU can't access memory fast enough to actually use the power it has. Our best guess is that line is around 1000MHz, or 3.0 TFLOPS.

Again, Nintendo can make any compromise they want, but past that point the returns start getting smaller while the costs - a bigger heat sink, a heavier device, fewer chips that meet the quality standard, increased risk of hardware failure - rise quickly.

I would be very surprised if Nintendo goes past those lines. Not because I don't expect Nintendo to delivery high performance, at this point that's exactly what I expect. And not because I expect Nintendo to "cheap out." It's because, as a performance engineer, I would be surprised that Nintendo believed that power was going to be usable.

This thing has 8 CPU cores. It has some form of modern storage. It has a new cartridge format. It has some unknown amount of memory. All of these things eat power, all of these things make heat. All of these things have different "sweet spots" where efficiency is good and where efficiency falls off a cliff. All of these things influence the performance of the others. Want HDR on your screen? That costs power. New Joy-Con features? Power.

When we talk about limits, what we really mean is "costs go up." That's not a hard limit - as @LiC pointed out to me recently, marginal gains are still gains! But it's not just "how much does it cost" but "what other things do we have to spend on." I find it hard to imagine that Nintendo has half a watt to spare and decides "Instead of a 25% more CPU performance, what I really need is 5% more GPU, heavily limited by memory bandwidth."
also a more powerful console mean more expensive games to make, around a budget of $200/400 milions dolar with the next Legend of Zelda/3D Mario for example, with longer development cycle too, Tears of the Kingdom took 6 years of development, imagine how expensive and time consuming the next Legend of Zelda will be , can we really expect Nintendo to willing spent a bugdet equivalent to huge Hollywood movies, Avengers Endgame on it games?
 
Knowing what we know is this technically possible?
Yes.


There’s gotta be sacrifices or compromises somewhere right?
No.


reduced load times were on the table
Just with the information we know for certain, it has a far, far better CPU and a dedicated decompression block, so this was a given even if storage speed doesn't considerably increase (given this demo it appears that is also the case). It's not surprising storage speed would increase too, since UFS storage is actually more readily available and will be for longer compared to the eMMC they're currently using.
 
Tech wizards of Fami: can someone ELI5 how Botw is supposedly running at 4k60fps with no/reduced load times? Knowing what we know is this technically possible? There’s gotta be sacrifices or compromises somewhere right? Maybe input lag is higher since it’s using DLSS? Or maybe the tech demo file for botw only had Hyrule field and wasn’t running the full game which would explain why it ran so well?

Excuse my ignorance😭 I’ve been keeping it for almost 2 years now and didn’t expect the Switch 2 to supposedly be THIS powerful. I want to believe and get hype but I can’t help but feel a little hesitant. I know DLSS is magic so I was expecting a resolution bump but I didn’t know increased framerate and reduced load times were on the table too
Thraktor has you covered.
 
Tech wizards of Fami: can someone ELI5 how Botw is supposedly running at 4k60fps with no/reduced load times? Knowing what we know is this technically possible? There’s gotta be sacrifices or compromises somewhere right? Maybe input lag is higher since it’s using DLSS? Or maybe the tech demo file for botw only had Hyrule field and wasn’t running the full game which would explain why it ran so well?

Excuse my ignorance😭 I’ve been keeping it for almost 2 years now and didn’t expect the Switch 2 to supposedly be THIS powerful. I want to believe and get hype but I can’t help but feel a little hesitant. I know DLSS is magic so I was expecting a resolution bump but I didn’t know increased framerate and reduced load times were on the table too
BotW is a Wii U game at heart. very low end compared to modern hardware and architectures. simple as for the most part. it was optimized for 3 PPC cores and a pissant AMD gpu. 7 modern ARM cores, a buttload of gpu power and much faster storage makes do. throw in DLSS and you can go higher to 4K/60
 
also a more powerful console mean more expensive games to make, around a budget of $200/400 milions dolar with the next Legend of Zelda/3D Mario for example, with longer development cycle too, Tears of the Kingdom took 6 years of development, imagine how expensive and time consuming the next Legend of Zelda will be , can we really expect Nintendo to willing spent a bugdet equivalent to huge Hollywood movies, Avengers Endgame on it games?
It seems that Nintendo is aware of this and has been hiring more people the last few years, due to the need to increase dev team sizes for the next console.
 
0
Yes that happened. It was an extreme situation that is not going to be repeated here

This and PS3 was already $600 at launch, they weren't going to launch it with two HDMI's, two enthernet cables, and the specs it had unless it was going to be like $800+
 
There is some wiggle room, but I'm also noticing some "FLOP drift", partially because some folks are talking in terms of "mixed-precision TFLOPS" which is confusing. Let's talk "limits" for a second.

On the handheld side, Nintendo is limited by battery and heat. At around 550MHz (1.6 "standard" TFLOPS), each small jump in performance causes a larger and larger fall in battery life, and a bigger and bigger increase in heat. Nintendo can make any compromise they want, it's just that the trade-offs get really bad at that point.

In docked mode, heat still matters, but there is also memory bandwidth. There is a point where adding 20% more GPU performance doesn't add you 20% more game performance, just because the GPU can't access memory fast enough to actually use the power it has. Our best guess is that line is around 1000MHz, or 3.0 TFLOPS.

Again, Nintendo can make any compromise they want, but past that point the returns start getting smaller while the costs - a bigger heat sink, a heavier device, fewer chips that meet the quality standard, increased risk of hardware failure - rise quickly.

I would be very surprised if Nintendo goes past those lines. Not because I don't expect Nintendo to delivery high performance, at this point that's exactly what I expect. And not because I expect Nintendo to "cheap out." It's because, as a performance engineer, I would be surprised that Nintendo believed that power was going to be usable.

This thing has 8 CPU cores. It has some form of modern storage. It has a new cartridge format. It has some unknown amount of memory. All of these things eat power, all of these things make heat. All of these things have different "sweet spots" where efficiency is good and where efficiency falls off a cliff. All of these things influence the performance of the others. Want HDR on your screen? That costs power. New Joy-Con features? Power.

When we talk about limits, what we really mean is "costs go up." That's not a hard limit - as @LiC pointed out to me recently, marginal gains are still gains! But it's not just "how much does it cost" but "what other things do we have to spend on." I find it hard to imagine that Nintendo has half a watt to spare and decides "Instead of a 25% more CPU performance, what I really need is 5% more GPU, heavily limited by memory bandwidth."
Assuming we're approaching the same battery life than the original Erista model, I guess that's where the ceiling ("max") values are coming from? 2 and 3.45 TFLOPS are the values I heard about the most, at least.
also a more powerful console mean more expensive games to make, around a budget of $200/400 milions dolar with the next Legend of Zelda/3D Mario for example, with longer development cycle too, Tears of the Kingdom took 6 years of development, imagine how expensive and time consuming the next Legend of Zelda will be , can we really expect Nintendo to willing spent a bugdet equivalent to huge Hollywood movies, Avengers Endgame on it games?
Marginal changes in clocks aren't really going to move the needle on that, this is already going to be a console capable of running all GOW: R, Horizon: FW and TLOU2 with several improvements and even 60 FPS modes without much trouble. Their budgets are ballooning either way, unless they essentially made this thing just to keep getting 3rd parties while their 1st party offerings stay in low budgets (very unlikely).
 
Tech wizards of Fami: can someone ELI5 how Botw is supposedly running at 4k60fps with no/reduced load times? Knowing what we know is this technically possible? There’s gotta be sacrifices or compromises somewhere right? Maybe input lag is higher since it’s using DLSS? Or maybe the tech demo file for botw only had Hyrule field and wasn’t running the full game which would explain why it ran so well?

Excuse my ignorance😭 I’ve been keeping it for almost 2 years now and didn’t expect the Switch 2 to supposedly be THIS powerful. I want to believe and get hype but I can’t help but feel a little hesitant. I know DLSS is magic so I was expecting a resolution bump but I didn’t know increased framerate and reduced load times were on the table too
BOTW was a game made for 2010 hardware, being ported (as a tech demo) by the developers to 2021 hardware. These types of leaps are absolutely not surprising.

Load times are heavily dependent on storage format and CPU capability. We know what the CPU on the Switch 2 is and it's miles ahead of what the Switch 1 has, plus double the cores, so it's not shocking to see load times reduced by that much. We also know Switch 2 has a dedicated file decompression engine which helps offload some of that loading from the CPU. We have no idea what storage its using but emmc was already pretty fast and heavily limited by the Switch's CPU.

4k could be possible even without DLSS. Again BOTW was built for the Wii U, runs at 900p on Switch via a hastily ported version (the developers' own words that the port from Wii U to Switch was really fast), and we're again looking at a massive jump in GPU power. 60fps definitely takes a lot of retooling since the game's systems are so complex but again for the actual developers of the game this shouldn't be that impossible of an ask with the hardware we know of.
 
This is a reason why I lean towards a Spring release being better than a November release. If you do a Spring release like the Switch originally was, you can continue to launch game after game every month leading up to the holidays. The problem with November is that you really only have a few weeks to get big games out before the end of the year. Yes, there are exceptions, but December isn't normally a big video game release month. Once you get to mid-December, there is basically a 6 week window where nothing big gets released.

Even if Nintendo opts for a September release, you can still get a few months to build up momentum with the new system and not have that lull period right away.

This has been my take for quite some time as well. Launching in November guarantees a shortage of units. Demand is always high at launch regardless of when it launches and then the holiday season increases demand further. Unless they have a stock pile of ten million units for launch, which means they will have to warehouse them for months, there is no way to keep up with demand when launching in November. If its H2, September is probably the best month but even that will likely see sustained demand that surpasses supply. Launching in H1 gives them the best chance of meeting consumer demand. The hardcore Nintendo gamer will be there day one and burn through the initial five million units. Then create a consistent lineup of software releases that sustains high demand all the way to the holiday season where they can sell upwards of ten million units.

Third parties will also benefit from allowing a userbase to grow and establish itself earlier in the year. Nintendo is going to launch with at least one strong first party titles, probably 3D Mario. So at launch the SNG will have a near one to one attach ratio for the hardware and 3D Mario, making it tougher for third party titles to get noticed. Nintendo hardware almost always have worse attach ratios for third party with a few exceptions like Sonic games, so having lots of third party titles right at launch will likely result in diluted sales for most of those games. I know Wii U was a flop, but it did sell 3.5 million units at launch and had over twenty launch titles from third parties, and they all did very poorly. Nintendo needs a larger userbase built up than Sony or Microsoft in order for third party developers to see good sales figures for their games.
 
Regarding Apple Event today, Apple now uses pre-recorded videos for their presentation instead of the live presentations.And there are many other companies doing the same transitions. I’m curious of Nintendo’s choice for Switch successor conference.
 
Regarding Apple Event today, Apple now uses pre-recorded videos for their presentation instead of the live presentations.And there are many other companies doing the same transitions. I’m curious of Nintendo’s choice for Switch successor conference.
They should have a pre-recorded presentation in Japanese but still have a live translator for some reason. That was so funny when the guy was struggling to keep up in the 2017 presentation.
 
They should have a pre-recorded presentation in Japanese but still have a live translator for some reason. That was so funny when the guy was struggling to keep up in the 2017 presentation.
You should've seen the RGG summit a few months back. They had 1 translator for like 5 guys on stage and another translator for the 1 girl. It was unwatchable live, it was a disaster lol
 
This is a reason why I lean towards a Spring release being better than a November release. If you do a Spring release like the Switch originally was, you can continue to launch game after game every month leading up to the holidays. The problem with November is that you really only have a few weeks to get big games out before the end of the year. Yes, there are exceptions, but December isn't normally a big video game release month. Once you get to mid-December, there is basically a 6 week window where nothing big gets released.

Even if Nintendo opts for a September release, you can still get a few months to build up momentum with the new system and not have that lull period right away.
Yeah, I'm increasingly considering that Nintendo might look at the Switch's first year as a genuine strategy worth replicating, even though it was dictated by circumstance (as they had originally planned to launch in November 2016). One thing I think is very important to Nintendo is to continue the cadence of major releases, which they've maintained for 6 and a half years now, right up through and beyond the launch of new hardware. They'll still probably want a bit more than a Switch-esque "one big game" launch, but rather than a launch blowout, I think they'll still want to let things be spread out in the following months of the year. That kind of release scheduling would also fit with Nintendo wanting third party games to also be relatively spaced out into the second half of the year, not clustered around the launch date.
 
They should have a pre-recorded presentation in Japanese but still have a live translator for some reason. That was so funny when the guy was struggling to keep up in the 2017 presentation.
I remember when Pokémon did a presentation when Pokémon Go Plus+ and Sleep was revealed. It was live translation with the translater seeing the products for the first time, and he was so befuddled by the name. "Pokémon Go Plus? Plus?"
 
Last edited:
so apparently gta 6 is gonna have a file size of 750gb and over 400 hours of gameplay AND cost $150 as previously reported.

journalism nowadays is sensational for sensations sake. gotta get them clicks any way any how
 
Please read this staff post before posting.

Furthermore, according to this follow-up post, all off-topic chat will be moderated.
Last edited:


Back
Top Bottom