• 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.
  • Furukawa Speaks! We discuss the announcement of the Nintendo Switch Successor and our June Direct Predictions on the new episode of the Famiboards Discussion Club! Check it out here!

StarTopic Future Nintendo Hardware & Technology Speculation & Discussion |ST| (New Staff Post, Please read)

But ultimately what they needed, and what they need, is TIME. Time to have more than one draft, time to optimise, time to implement. They've spoken to trying to improve quality while maintaining release cadence and that makes me uneasy, because I don't think those are possible at the same time. Something has to give, no matter how much next generation hardware smooths things over, some things just can't be fixed with MORE technology.
they're most likely not building games and engine at the same time and haven't been since Sword and Shield. this iteration of their tools date back to Pokemon Legends, so they're probably not concurrent

also it seems like the engine's name is Trinity according to the credits of the DLC. I'm far from that point though
 
I'm giving anyone whom has multiple monitors AND throwing shade to Dual screens a big side eye. The DS is a legend (I mean look at them sales). It should be treated as such.
DS is huge, but that doesn't mean every specific thing about it was the result of its success and should be continued forever. Dual screens felt like an outgrowth of their particular situation at the time. Let's have a touch screen! OK, but if we do that then if it flips like the GBASP then that's not going to be very stable when people start poking or drawing. OK, so if we put the touch screen on the base, what do we do with the part that flips up? How about a cheaper secondary screen that can be counted on to not be covered by people's hands?

Functionally, DS's two screens was just one vertical screen with a gap in the middle, and that's certainly not something I'd want to see standard. No particular aspect ratio or screen layout gives much inherent advantage over another. They can all show as much of a play area as a developer wants, or spend as much area as they want showing maps or stats or anything. It's only a hassle when taking something heavily designed around one type of screen(s) and making it work on another.
 
another year? LOL

this motherfucker better have ai + vr + ar + 2 screens + phone + web + pixels + sound pixels + back compat + 12 hr batt + split screen + friends + free online + zelda remake collection + mario remake collection + new smash + new mk + $299

people literally dying out here before switch2 wtf
 
Here's something cool:

FSR 3 Frame Interpolation has been modded into games (which is why the HUD elements are wonky) and has had promising results from the Lenovo Legion Go handheld:





I think it would be cool if developers could implement custom versions of FSR 3 for Docked Mode on the new Switch (which sounds slightly less powerful than the Legion Go) to interpolate games that run at 35-45fps to 60fps. I think it would be awesome, especially for slightly higher end games.
 
They had the tools. The artists. The assets.

They weren't and aren't destined for a technological trainwreck. What we got still feels like a functioning alpha build, like a working vertical slice that had to be pulled and stretched over the entire game.

We probably won't see what Paldea could have been for ten, fifteen years, if even then, and isn't that sad?

I still believe the Unreal Engine thing and I think not having to literally create their own engine and build it out should help immensely, having time to do other things.

But ultimately what they needed, and what they need, is TIME. Time to have more than one draft, time to optimise, time to implement. They've spoken to trying to improve quality while maintaining release cadence and that makes me uneasy, because I don't think those are possible at the same time. Something has to give, no matter how much next generation hardware smooths things over, some things just can't be fixed with MORE technology.
They desperately need to work with Monolith Soft to get their shit together.
 
Having an engine that can deliver more sophisticated and complex games doesn’t mean they have the talent for it, or that they can even manage that.

They could deliver a game that looks really good, but still performs pretty poorly. And they could customize it, but they’d be hacking and slicing to their specific needs.

Hell, we don’t really know what goes on with GF’s engine these days, but Unity or Unreal aren’t exactly perfect either. They are helpful though but not perfect.

GF’s engine could, in theory, be an engine that is actually completely fine for what they need it for, and switching to those aforementioned engines wouldn’t offer anything better but learning the ways of the new engine if they haven’t already.

I know people bash the games but let’s say for the second that the engine the game runs on has all the features that UE or Unity has, they just still can’t deliver a game that looks like other UE or Unity games due to other reasons
Like I said, I don't have detailed knowledge about their engine, I just assumed it's full of legacy/ leftovers from previous platforms/ technical debt. Maybe this assumption is wrong.

If that is the case, sometimes it's better to start over. Maintaining an in house engine is hard, and not a lot of studios can do it well.

My point wasn't that we need a Pokémon with Nanite, it was that it's hard to make quality with bad tools. I know using a third party engine is no magic fix either, but might be a step in the right direction.
 
0
I hope we get the option to stream games to the Dock while using the device in handheld so we can get a proper mario maker experience + ds emulation
The anti-Wii U. It would be so awesome. Games like Nintendo Land could be ported and played by more than the 7 people who bought a Wii U! It'd never happen but I hope they return to the 2 screen set up some day with video streaming.
 
I'm hoping for Wii U-like functionality with a dongle that lets you cast wirelessly to the the TV. I also hope for some sort of camera implementation, perhaps for even more advanced AR games.

I sincerely doubt all the R&D they put into the 3DS AR capabilities are just going to disappear forever. GameBoy Camera, DSi, 3DS, Wii U Gamepad Camera -- It was actually sort of odd the Switch didn't have a camera. I'm curious what sorts of fun, quirky, 3DS-like experiences might be possible with the addition of a camera on the back.
I hope we get the option to stream games to the Dock while using the device in handheld so we can get a proper mario maker experience + ds emulation

Glad to see someone else share this sentiment! When I brought it up I felt like it kinda got shot down.
 
Glad to see someone else share this sentiment! When I brought it up I felt like it kinda got shot down.
It's both totally extremely incredibly awesome and desirable, especially for anyone who's actually used a Wii U and understands the benefits, and technologically totally infeasible at the moment. But we can dream!
 
For some gimmicks that might be cool there's an easy "flowchart":

Could gameplay benefit from it?
No: Nintendo won't care for it
Yes:
Is it done without much engineering work, or won't be a hassle for consumers to setup/use?
No: Nintendo won't care for it
Yes:
Can it be done cheap enough to not make the consumer price too high or the production cost too much?
No: Nintendo won't care for it
Yes:
Could be something Nintendo might do, but not necessarily with the next upcoming system.

Like, i'm sure there's gonna be a day and a system from Nintendo that goes in hard on AR and VR. But that might be console-generations off until this stuff isn't either a hassle to set up or to use.
For example.
 
I hope we get the option to stream games to the Dock while using the device in handheld so we can get a proper mario maker experience + ds emulation

I thought about the same idea. However, outside of DS emulation, it'd have limited use and it'd limit performance to handheld clocks.
More likely variant would be grip for a phone and streaming from the console.
 
I thought about the same idea. However, outside of DS emulation, it'd have limited use and it'd limit performance to handheld clocks.
More likely variant would be grip for a phone and streaming from the console.
If a future Nintendo device used the technique from the patent, you could put the upper screen in the dock and use the other screen as a „gamepad“. But since Switch 2 probably won‘t have dualscreens it most likely isn‘t gonna be used.
 
It's both totally extremely incredibly awesome and desirable, especially for anyone who's actually used a Wii U and understands the benefits, and technologically totally infeasible at the moment. But we can dream!
I'm sorry, but what is infeasible about a Switch 2 device coming equipped with whatever was in the Wii U Gamepad and also a "dock" that has power, a chip, and HDMI out so it serves as both a stream stick and dock at the same time? If the Wii U could do it super cheap in 2012, why can't it happen in 2024?

The dock wouldn't need to have a disc drive so it wouldn't have to be unwieldy like the console part of the Wii U.
 
I'm sorry, but what is infeasible about a Switch 2 device coming equipped with whatever was in the Wii U Gamepad and also a "dock" that has power, a chip, and HDMI out so it serves as both a stream stick and dock at the same time? If the Wii U could do it super cheap in 2012, why can't it happen in 2024?

The dock wouldn't need to have a disc drive so it wouldn't have to be unwieldy like the console part of the Wii U.
...That is neither how the Wii U functioned nor an accurate description of its market position.

The Wii U didn't do it cheaply, the Wii U didn't stream from the GamePad to the television, the Wii U certainly didn't have to wirelessly stream the signal in HD. Wii U used what was essentially a draft standard of a cutting edge wireless protocol to achieve even 480p60 in 2012. The video production was done in the console on Wii U, the part with wall power, not on a power constrained handheld, and it was not done particularly well given the range and reliability. I love my Wii U, I really do, but the technology it explored being used for 4K from a handheld at a mass market price simply does not demonstrably exist; it is utterly, and with a great degree of objectivity, infeasible.

I'll also note that this has been explored in much better detail throughout this thread when this idea is brought up, I can heartily suggest one reads back a bit. This thread has been going on for so long and gone over so many topics, there's bound to be something worth reading.
 
https://www.reddit.com/r/GamingLeaksAndRumours/s/zo8cheAYCE


Next gen xbox might come on about 2 years, I wonder if this means switch 2 might be 1 generation behind the other guys in a few years.
1. I don't buy it. There are announced games for the Xbox Series that won't come out until 2026. Plus, all of the major 8th-gen systems have lasted a while. Hell, the PS4 is still getting games next year. At least 2028, if not longer.
2. So just like the Switch has been for the past 4 years, and that certainly hasn't stopped it.
 
https://www.reddit.com/r/GamingLeaksAndRumours/s/zo8cheAYCE


Next gen xbox might come on about 2 years, I wonder if this means switch 2 might be 1 generation behind the other guys in a few years.
It MAY come out soon-ish, but... So what, honestly? Very few pieces of software have even caught up to meet Xbox Series X where it is, and as diminishing returns continue to pile up, and the diversity of hardware in use increases, game scalability naturally increases both due to market forces and technical reasons.

Xbox's hardware strategy isn't even particularly generational and isn't meant to be. Xbox's idea of their own hardware is more like a gradient of capability, and they haven't actually released much first party software that's exclusively on the current generation. It would also stand alone in the tenth generation of consoles for several years, with little reason to develop for it.

Even now, we see Gen9 games make their way onto Nintendo Switch. Is it so hard to believe a game designed to do reconstructed 8K on the Xbox HEX2 or whatever couldn't eek out a reconstructed 1080p on NG Switch? That's a 16x difference in pixels, but 1080p is FINE!

Games are more scalable, generations are less relevant, T239 is already ahead of current gen architecturally. It just isn't worth worrying about.
 
I'm sorry, but what is infeasible about a Switch 2 device coming equipped with whatever was in the Wii U Gamepad and also a "dock" that has power, a chip, and HDMI out so it serves as both a stream stick and dock at the same time? If the Wii U could do it super cheap in 2012, why can't it happen in 2024?

The dock wouldn't need to have a disc drive so it wouldn't have to be unwieldy like the console part of the Wii U.

The CPU/GPU/RAM/etc. on the Wii U didn't have to run off battery power. All of that was in the console itself, which meant the gamepad could use a very simple, low power chip to achieve a somewhat reasonable battery life. If a Switch 2 were to stream to the TV, it would have to run both its internal display and the TV display from the SoC in the part you're holding in your hands, which means it would be running off battery power. To get usable battery life out of it, it would have to use the same clocks as handheld mode, which means it'll look worse than the normal TV mode (even before compression from streaming), and have worse battery life than handheld mode usually has, as it has to continuously stream data to the dock.

Cost is another factor here. The Wii U wasn't a cheap console to manufacture, and although there were a variety of reasons for that, the custom streaming hardware required to achieve their latency requirements didn't help.
 
1. I don't buy it. There are announced games for the Xbox Series that won't come out until 2026. Plus, all of the major 8th-gen systems have lasted a while. Hell, the PS4 is still getting games next year. At least 2028, if not longer.
2. So just like the Switch has been for the past 4 years, and that certainly hasn't stopped it.
With Xbox's hardware strategy how it is, their 2026 console would probably be more like a Series X Pro with a new brand and a couple rare exclusives.
 
With Xbox's hardware strategy how it is, their 2026 console would probably be more like a Series X Pro with a new brand and a couple rare exclusives.
That isn't necessarily true considering Microsoft is pivoting to ARM chips and the FTC leak came out that they're R&Ding an ARM based xbox console. Could be something completely different.
 
MS should focus on their current console atm, at the rate it's going it might end up below the Xbox One which is....not good

Also Monolith doesn't have an infinite numbers of employees christ. People expect them to work on everything under the sun at times..
 
That isn't necessarily true considering Microsoft is pivoting to ARM chips and the FTC leak came out that they're R&Ding an ARM based xbox console. Could be something completely different.

They were considering an ARM based console, but it didn't suggest there was any serious R&D work on it (just a cut-off date for deciding between x86 and ARM). If they've brought the console forward, then I'd say it's almost certainly x86, as it's the lower risk option.
 
That isn't necessarily true considering Microsoft is pivoting to ARM chips and the FTC leak came out that they're R&Ding an ARM based xbox console. Could be something completely different.
They're pivoting to arm in the server space. There's not too much benefit from going to ARM in a console and more to lose
 
https://www.tomshardware.com/pc-com...-and-nvidias-approach-that-uses-interpolation

I thought this was interesting.

So what's happening is compared to AMD, or Nvidia's approach where they interpolate frames (using two individual frames, and insert one additional frame between the two frames), Intel is looking to Extrapolate frames (using one frame to generate a new frame)

Is it possible if typical DLSS 3 FG isn't feasible on Drake, could a custom version that extrapolates frames rather than interpolate frames be feasible?

EDIT: Intel's reasoning is reduced latency, which appears to be an issue for AMD, and Nvidia's solutions right now. Given the precious render time requirements, and limitations, this could have some benefits.
 
https://www.tomshardware.com/pc-com...-and-nvidias-approach-that-uses-interpolation

I thought this was interesting.

So what's happening is compared to AMD, or Nvidia's approach where they interpolate frames (using two individual frames, and insert one additional frame between the two frames), Intel is looking to Extrapolate frames (using one frame to generate a new frame)

Is it possible if typical DLSS 3 FG isn't feasible on Drake, could a custom version that extrapolates frames rather than interpolate frames be feasible?

EDIT: Intel's reasoning is reduced latency, which appears to be an issue for AMD, and Nvidia's solutions right now. Given the precious render time requirements, and limitations, this could have some benefits.



Link to the publication; https://poiw.github.io/ExtraSS/

There's a video demo, gifs and explanation on there.

They test and compare to other methods on an RTX 3090

Paper was already posted by dakhil ;

Moreover, I think some media outlets are too quick to call this method as an existing solution by intel v.s. something that's in research with academia being involved and intel being involved. As it may not even come into existence.
 
the general idea for these solutions, I think, is that frame times are so low that perception of artifacts is minimal while perceived smoothness is maximized. so the question would best be, is 60fps from 30fps a low enough frame time for that?
 
https://www.tomshardware.com/pc-com...-and-nvidias-approach-that-uses-interpolation

I thought this was interesting.

So what's happening is compared to AMD, or Nvidia's approach where they interpolate frames (using two individual frames, and insert one additional frame between the two frames), Intel is looking to Extrapolate frames (using one frame to generate a new frame)

Is it possible if typical DLSS 3 FG isn't feasible on Drake, could a custom version that extrapolates frames rather than interpolate frames be feasible?

EDIT: Intel's reasoning is reduced latency, which appears to be an issue for AMD, and Nvidia's solutions right now. Given the precious render time requirements, and limitations, this could have some benefits.
Interesting. I‘ve thought myself that extrapolating frames based on the two previous frames (or better 3, so two real frames and one generated one) the game logic (which runs at 60fps) and also checking the controller at 60hz. Even though it probably won‘t be feasible for every Switch 2 game (CPU heavy games).
 
Whether these Xbox rumors (I have severe doubts) and PS5 Pro rumors (I am much more willing to believe those) are true, it is very interesting that these companies seem to be moving towards hardware-accelerated ray tracing and machine learning while sticking to AMD, a company that's always been about raster performance first and foremost. Maybe this could see them take a shift towards GPU specialization in the same way that Nvidia and Intel have with GeForce RTX and Arc? It'll also be very interesting to see Nintendo as the one being ahead of the curve on these technologies. As someone who's less than a year older than the Wii, a Nintendo this forward-thinking is definitely an exciting prospect.
 
Whether these Xbox rumors (I have severe doubts) and PS5 Pro rumors (I am much more willing to believe those) are true, it is very interesting that these companies seem to be moving towards hardware-accelerated ray tracing and machine learning while sticking to AMD, a company that's always been about raster performance first and foremost. Maybe this could see them take a shift towards GPU specialization in the same way that Nvidia and Intel have with GeForce RTX and Arc? It'll also be very interesting to see Nintendo as the one being ahead of the curve on these technologies. As someone who's less than a year older than the Wii, a Nintendo this forward-thinking is definitely an exciting prospect.
What do you mean Wii babies are adults now.

Why. Why would you do that.

I agree re: exciting about Nintendo being on the cutting edge again, but they sort of were in 2017, it was the best thing on the shelf at the time, and they were cuttings edge in wireless technology on Wii U. Though being not merely "impressive for what it is" but "significant architectural advantages" is a new era for Nintendo. They haven't been in that position since N64.
 
I'm sorry, but what is infeasible about a Switch 2 device coming equipped with whatever was in the Wii U Gamepad and also a "dock" that has power, a chip, and HDMI out so it serves as both a stream stick and dock at the same time? If the Wii U could do it super cheap in 2012, why can't it happen in 2024?

The dock wouldn't need to have a disc drive so it wouldn't have to be unwieldy like the console part of the Wii U.

You'd need to split the current "tablet" device into wireless screen and "compute" module that would be docked to TV. That probably wouldn't be impossible, but it'd add a lot complexity for pretty much useless gimmick, that can be fully replaced by a cheap device that can act as a wireless external display.
 
To add my two cents to the Wii U discussion: Nintendo heavily customized Wii U to support its purpose of extreme low latency wireless casting, to the point of even customizing the GPU to work better with their streaming solution. There's a reason why Wii U still is much better than some current streaming solutions like PS Portal. And this amount of work towards such feature isn't worth for a Switch like device neither makes sense. There's difference between phones and PCs supporting home streaming and a specialized console.

I'd honestly suggest people to read Rodrigo Copetti work into analyzing Wii U and the amount of work Nintendo needed to do support their low-latency streaming solution.
 
I’m not as high as everyone else. Nintendo scares me every time they release new hardware. For me personally it’s all about software. Obviously hardware and architecture plays a big part in what software your system can run. I really hope the next system works out.
 
0
To add my two cents to the Wii U discussion: Nintendo heavily customized Wii U to support its purpose of extreme low latency wireless casting, to the point of even customizing the GPU to work better with their streaming solution. There's a reason why Wii U still is much better than some current streaming solutions like PS Portal. And this amount of work towards such feature isn't worth for a Switch like device neither makes sense. There's difference between phones and PCs supporting home streaming and a specialized console.

I'd honestly suggest people to read Rodrigo Copetti work into analyzing Wii U and the amount of work Nintendo needed to do support their low-latency streaming solution.

Heck, just the first page shows how much went into making the Gamepad work.

You absolutely could imagine it being done in reverse on a Switch, with Gamepad hardware (plus power management) integrated into the dock (maybe using Wifi 6E) and a preference for sending a 480p stream to the dock to limit bandwidth and performance hit on the mobile clocks.

I just have significant trouble imaging the end result performing anywhere near good enough to justify the extra cost. The BOM on the Gamepad was around $80, so stripping out the controller and display should bring that down a fair amount, even if the end result is $15 extra cost. . . Not enough people cared about the asymmetric gimmick to drive Wii U sales.

Heck, if Nintendo wanted to give it a go again, they would be better to make a few games like Pac-man VS and Nintendoland free on Switch online and just require folks have two Switches. One in the dock and one as the primary controller using the existing local WAN functions built into the system.
 
Yeah I used to be on the "stream from handheld mode" train until I realized that you'd be extremely limited by performance in handheld mode.
Limited by battery
Limited by low clocks and low resolutions

It just wouldn't work great, and there isn't a cheap easy solution to that problem. Streaming a final handheld image to an external display (one screen) is one thing but essentially splitting the handheld performance in two (for dual screen gaming) is rough... like 2 player Hyrule warriors on the gamepad rough... maybe even worse honestly...
 
Last edited:
Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
Last edited:
So what's happening is compared to AMD, or Nvidia's approach where they interpolate frames (using two individual frames, and insert one additional frame between the two frames), Intel is looking to Extrapolate frames (using one frame to generate a new frame)

Is it possible if typical DLSS 3 FG isn't feasible on Drake, could a custom version that extrapolates frames rather than interpolate frames be feasible?
I don't know that any of us here are in a good place to know how the real world requirements for such a thing would compare to FSR or DLSS frame generation. Though if so I'd be interested to hear.

Back when we first started hearing about DLSS3 and it was unclear how it worked, I was thinking about how DLSS upscaled by combining data from a new frame with data from previous frames, and wondering what the result would be if it just had the data from previous frames. Given how DLSS works I imagine that would take the same amount of time as doing the DLSS pass on any frame. But I also imagine it could look much worse than the frames with actual new data.
Heck, if Nintendo wanted to give it a go again, they would be better to make a few games like Pac-man VS and Nintendoland free on Switch online and just require folks have two Switches. One in the dock and one as the primary controller using the existing local WAN functions built into the system.
Right. Especially with the new generation coming up it increases the likelihood of a single person/household having access to multiple devices. Have your Mario Maker 3 for Switch 2 docked and interact with a dummy program on Switch 1.
 
I mean.. define “insider knowledge”. Is it contacts within Nintendo or a professional 3rd party developer?

Is it the research pro’s like (sry if I get names wrong) Oldpuck, Luigiblood, and many others?

I “MAY” have been told (along with many other shareholders) H1 too based on just stock research and other villainy in those waters. But H1 “MAY” not happen and in which case a whole lot of Nintendo shareholders will have been wrong and early. It’s not a surprise that Nintendos stock breaks out of a descending channel the moment gamescom happened. The rich shareholders know something. They always do much to my anger.

I wonder if I worded that as poorly as possible to avoid getting flamed and banned. :LOL:
Why do you keep putting May in all caps? What a strange fella 😂👀
 
I “MAY” have been told (along with many other shareholders) H1 too based on just stock research and other villainy in those waters. But H1 “MAY” not happen and in which case a whole lot of Nintendo shareholders will have been wrong and early. It’s not a surprise that Nintendos stock breaks out of a descending channel the moment gamescom happened. The rich shareholders know something. They always do much to my anger.
Unless they announce it in January, a May release sounds way too early. Assuming Nate is right about a March announcement, that'd only leave 2 months from starting MP to release, which does not sound like enough time to build up sufficient stock, and a May reveal seems too late, since they'd want some good news for shareholders such as yourself coming into the new fiscal year in April. To my eyes, this is probably just shareholders letting their confirmation biases do the talking.
 

Heck, just the first page shows how much went into making the Gamepad work.

You absolutely could imagine it being done in reverse on a Switch, with Gamepad hardware (plus power management) integrated into the dock (maybe using Wifi 6E) and a preference for sending a 480p stream to the dock to limit bandwidth and performance hit on the mobile clocks.

I just have significant trouble imaging the end result performing anywhere near good enough to justify the extra cost. The BOM on the Gamepad was around $80, so stripping out the controller and display should bring that down a fair amount, even if the end result is $15 extra cost. . . Not enough people cared about the asymmetric gimmick to drive Wii U sales.

Heck, if Nintendo wanted to give it a go again, they would be better to make a few games like Pac-man VS and Nintendoland free on Switch online and just require folks have two Switches. One in the dock and one as the primary controller using the existing local WAN functions built into the system.

That whole link is a goldmine is technical goodness. I love reading about this stuff even if I don't fully understand all aspects of it.
 
I don't know if this has been posted ITT before but:

Seems like the steam deck isn't powerful enough to run the new avatar game due to RAM limitations, not GPU nor CPU.
The game uses a lot of video memory and system memory without even taking frame generation into account (which is currently unavailable on steam deck on linux anyways).
 
Please read this new, consolidated staff post before posting.

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


Back
Top Bottom