• 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.

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

I'm sure that's true for the CPU but I have no idea about the GPU side. Regardless, it would be the same concept of a BC test platform.
you don't need BC for the cpu in the first place. a translation layer would work for the SDK too. NVN and NVN2 both run natively in windows, so devs can test the translation layer on a dev pc like they would their native games games
 
Except in Japan, the Switch is loosing momentum, falling behind even Xbox in most markets in hardware sales.

Just so we are not inventing narratives no this is not true lol.

This year the Switch is leading the Xbox Series and the PS5 in units in the US. I was third place only for Oct. In Sept it was second. We will have to see what stock looks like but the odds it is not the best selling hardware this year in the US seem low.

In the UK the three consoles are in a deadlock. Switch is third but "barely".

In the rest of Europe neither PS5 nor Xbox Series have had stock to be close to the Switch this year. Xbox is not beating the Switch in most markets. Plainly untrue.

Lets not misrepresent sales here. Neither the PS4 or the Xbox One ever moved more than 6 million units in the US in a calender year. This will be the 3rd time the Switch does it. The platform is still extremely healthy hardware units wise.

You want to argue that it is a good time to launch hardware. By all means. But lets not get it twisted. By combined metrics (hardware/software/subscription and digital) the Switch is basically going into/coming out of its mature phase in the product life cycle. There is no actual business danger by conventional metrics if they did not launch new hardware for the next 24 months.

Do I think they should in the next 12? Yes. Is it critical? The metrics say no.
 
You might be interested with this link as well:

I've seen. it's more of a port from x86 to ARM and is using the same libraries as the windows version.



something I missed was Mediatek had their own event showing off the Dimensity 9200 and it's ray tracing capabilities. Arena Breakout was playable there (which is where the previous video came from)





both of these events imply to me that both ARM and Qualcomm are designing a simpler RT core that's more akin to AMD's implementation than Nvidia's: focus on ray triangle intersection and leave bvh traversal to the compute units. it might result in a smaller area for RT functions and would force simpler workloads for power limited hardware. shadows, mirror reflections, and AO are prominent in all the demos between the two, with Qualcomm touting a demo running below 5W and Mediatek showing a proper game running on proper hardware (phone body et al).

it begs the question what this could mean for Drake, who's SoC would have to contend with similar power constraints. don't think Drake would be limited to just mirror reflections and shadows, but with a low power budget, it might be a case of tradeoffs of which RT effects devs choose to go with. it's gonna be much better than phones can provide, but how much better?
 
If Nintendo truly wanted a pro or mid gen refresh...they had their chance when Nvidia shrank the Tegra x1 logan into the 16nm Tegra x1 mariko. Nintendo could had fully utilized the Mariko to it's fullest capabilities and added more ram and called it the switch pro. Instead they clocked mariko to match the original x1 used in the first switch just to extended battery life.

At the end of the day splintering your fan base between pro only games and switch games just doesn't make sense.

A mid gen refresh option in 2019?

Naw, Nintendo had already realized the success of the Switch and were already talking about an unusually long lifecycle for Switch compared to usual consoles (6 years in the market and then release a “successor”)

If Nintendo was looking at 8-9 years instead, the mid gen model would come 4-5 years in, not 2.

But more importantly, around 2019 Nvidia was already selling them the future of tensor cores and DLSS that would now be in every Tegra variant going forward.

Nintendo would want to use Drake as their stepping stone into DLSS development. Being able to do that slowly with an upgrade model while the family of systems is still strong is the ideal option. Again, they didn’t want to make the same mistakes they made going from Wii—> Wii U where they were caught flat footed trying to move to HD development for a new system well after the previous system had significantly declined.

DLSS is their new HD. They want to develop for it WITH the current Switch family and library while engagement is high. Not wait and leave it for the successor to carry the load.

I still believe they were waiting for the shortages to be over, it would had been suicide to release a new console during the chip shortage. They would had been in the same boat as sony and microsoft struggling to make the drake console and losing out on sales due to shortages.

Shortages/supply issues have never…ever…stopped new tech from launching. It’s not an isssue.

It never stopped the new Xboxes from launching, the new PlayStations from launching, the OLED Switch from launching, the new iPhone from launching, new TVs from launching, new cars from launching, new tablets from launching, the Steam Deck from launching etc etc etc.

Whatever prevented the Drake model from releasing in 2021 or 2022 had NOTHING to do with being worried about having enough supply . (I’m betting it was them not liking the yield of the 2019/2020 SoC they were working with and decided to upgrade a bit which pushed it to where we are now.)
 
Last edited:
Lanching in May 2023 would mean launching with prolly the most awaited Nintendo game, and still riding on the previous launches of what's likely the most exciting Pokemon game in years (not counting Arceus), and two evergreens, an old and a new one, still getting active support.

To me that feels like the most ideal timeframe.

Had the new model launched November 2022 alongside Pokémon, it would have sold out.

Had the new model launched in March 2022 alongside Kirby, it would have sold out.

Had the new model launched in October of 2021 alongside Metroid Dread, it would have sold out.

Had the new model launched in February of 2021 alongside 3D Mario Bowsers Fury, it would have sold out.

Launching in May 2023 isn’t the “perfect time” in terms of software or Drake sales success.

The decisions by Nintendo when to release this thing is entirely hardware development based, not software window timings.
 
Not sure if this has been posted yet but I found a patent regarding DLSS for the Switch.


So in the diagram, looks like Nintendo will be using a based 540p to 1080p conversion. Hopefully, this means a 1080p screen portable.

US20220335570A1-20221020-D00002.png
 
Not sure if this has been posted yet but I found a patent regarding DLSS for the Switch.


So in the diagram, looks like Nintendo will be using a based 540p to 1080p conversion. Hopefully, this means a 1080p screen portable.

US20220335570A1-20221020-D00002.png
perhaps that resolution is simply an example of the capacities.



Hey, it would be great if this technology was applied in the next console to the so-called "impossible ports" because even lowering the resolution of (for example) Plague Tale 2 to 520p to make it work well, it would look good after going through that dlss
 
Last edited:
Had the new model launched November 2022 alongside Pokémon, it would have sold out.

Had the new model launched in March 2022 alongside Kirby, it would have sold out.

Had the new model launched in October of 2021 alongside Metroid Dread, it would have sold out.

Had the new model launched in February of 2021 alongside 3D Mario Bowsers Fury, it would have sold out.

Launching in May 2023 isn’t the “perfect time” in terms of software or Drake sales success.

The decisions by Nintendo when to release this thing is entirely hardware development based, not software window timings.

My post was more aimed at the thoughts about it releasing in late 2023 or even 2024, you know.

Also, going with that opinion, they could've launched in 2017 without Zelda, of course the initial batch would've sold out. But having BotW at launched helped.
And having TotK for the Drake launch on top of a new FE, a still new Pokemon and Splatoon 3 will help Drake in the long run.
 
A mid gen refresh option in 2019?

Naw, Nintendo had already realized the success of the Switch and were already talking about an unusually long lifecycle for Switch compared to usual consoles (6 years in the market and then release a “successor”)

If Nintendo was looking at 8-9 years instead, the mid gen model would come 4-5 years in, not 2.

But more importantly, around 2019 Nvidia was already selling them the future of tensor cores and DLSS that would now be in every Tegra variant going forward.

Nintendo would want to use Drake as their stepping stone into DLSS development. Being able to do that slowly with an upgrade model while the family of systems is still strong is the ideal option. Again, they didn’t want to make the same mistakes they made going from Wii—> Wii U where they were caught flat footed trying to move to HD development for a new system well after the previous system had significantly declined.

DLSS is their new HD. They want to develop for it WITH the current Switch family and library while engagement is high. Not wait and leave it for the successor to carry the load.
The impact of DLSS and HD on game development just aren't comparable in the slightest. DLSS is, for the most part, just an extra step in the rendering pipeline. The only new feature on Drake that could have anywhere near the level of impact of HD development (at least in a way that wouldn't make the games completely unworkable on the current Switch) is RT, and even that broadly tends to make development easier, not harder, and having to support non-RT hardware saps a lot of the benefit.

Also the Wii declined in large part because of the move to HD development, which is representative of why softer generational transitions are quickly becoming the norm.
 
Holy hell, what a find! It's arguably the most noteworthy piece of information we have had since the Nvidia leak.

And yes, the text of the patent in the PDF mentions that the 540p input and 1080p output resolutions are just examples.
Yeah that's kind of what it is. Still hoping for a 1080p screen though.

Anyway, this statement confirms DLSS although it is just an "example":

In certain example embodiments, the techniques herein may advantageously take advantage of NVIDIA's tensor cores (or other similar hardware). A tensor core may be a hardware unit that multiplies two 16×16 FP16 matrices (or other sized matrices depending on the nature of the hardware), and then adds a third FP16 matrix to the result by using fused multiply—add operations, and obtains an FP16 result. In certain example embodiments, a tensor core (or other processing hardware) can be used to multiply two 16×16 INT8 matrices (or other sized matrices depending on the nature of the hardware), and then add a third INT32 matrix to the result by using fused multiply-add operations and obtain an INT32 result which can then be converted to INT8 by dividing by the appropriate normalization amount (e.g., which may be calculated during a training process, such as described in connection with FIG. 9). Such conversions may be accomplished using, for example, a low processing cost integer right shift. Such hardware acceleration for the processing discussed herein (e.g., in the context the separable block transforms) may be advantageous.

A 1080p screen would probably do wonder in BC rendering dock mode power in a portable setting (ala ReverseNX).
 
Not sure if this has been posted yet but I found a patent regarding DLSS for the Switch.


So in the diagram, looks like Nintendo will be using a based 540p to 1080p conversion. Hopefully, this means a 1080p screen portable.

US20220335570A1-20221020-D00002.png
Interestingly they mention "NVIDIA's tensor cores" but not DLSS directly.
 
The publication date shows its recency but isn't that the same as the old NERD patent?

EDIT: yea this looks like a resubmission with changes

Old submissions
 
The patent also confirms that 'DLSS' is updateable via patch (network).

Fig 9. shows a training method, maybe it's DLSS 1.0?

I'm an absolute layman in this, but even i don't think nVidia would want to keep supporting DLSS 1.0 in any form. Basing it on DLSS 2.0 is the only working option.

Maybe it retains some features from 1.0.
 
100+ Monsters planned for release, mostly included in Rise, World and Generations.
Battle Pass system to replace expansions. You will be able to buy the battle pass once the season ends, the only thing you can miss is event quests, whic hcan come back later as well.
Will have an in-game item shop where you will be able to buy microtransactions/cosmetics.
Will have a system in which you will be able to call NPC clones of real players for solo plays, similar to NPC quests in Ryse.
New story will be about a human main villain trying to awake Lethalis.
Lethalis is the flagship monster, a black/gold wyvern with massive 4 leg/wings which is the natural rival/wyvern version of Fatalis.
Hunter Arts will be introduced, which will allow players to infuse their characters with an element, which will give his hunters special abilities like faster movements, explosive attacks, eletric body etc.
Planned to last until the end of PS5/XSX generation. Also coming to PC and next Nintendo console.

Leak from 4chan


Is this believable?
 
Not sure if this has been posted yet but I found a patent regarding DLSS for the Switch.


So in the diagram, looks like Nintendo will be using a based 540p to 1080p conversion. Hopefully, this means a 1080p screen portable.

US20220335570A1-20221020-D00002.png
perhaps that resolution is simply an example of the capacities.



Hey, it would be great if this technology was applied in the next console to the so-called "impossible ports" because even lowering the resolution of (for example) Plague Tale 2 to 520p to make it work well, it would look good after going through that dlss
Holy hell, what a find! It's arguably the most noteworthy piece of information we have had since the Nvidia leak.

And yes, the text of the patent in the PDF mentions that the 540p input and 1080p output resolutions are just examples.
I wouldn’t really read into it like that, it’s a filing they’ve been trying to patent for a while that seems like it is a method that they can also use on other hardware in the chance that they need to move off of NVIDIA. It doesn’t allow them to be limited to that.



Here’s also a thing, it’s resolution is simply an example case, 540-1080.

At step 200, a 540p source image 205 is rendered by game engine 110. In certain example embodiments, and as discussed herein, the source image may come from other sources, such as real cameras, movies, television shows, broadcast television, or the like. For example, the techniques herein may be used to transform a source 540p signal that is received for a television program (for example a live broadcast of a sporting event) into a 1080p signal that is then output for display to the user. Further, while a 540p is discussed in connection with the example in FIG. 2 (and elsewhere herein), the techniques may be applied to images of other sizes. It will be appreciated that the details of the neural network 112 (e.g., the coefficients or L and R) employed as part of the upconversion process will change should the details of the source and/or converted image change (e.g., should the resolution of such images be adjusted). For example, a neural network for upconverting to 1080p from 540p will be different than one upconverting from 1080p to 1440p (e.g., 2560×1440). It will also be appreciated that while the example shown in FIGS. 3-7 relates to transforming a 540p image to a 1080p image, the techniques herein may be applied to other image sizes (e.g., 720p to 1080p; 480p to 1080p, 1080p to 1440p, 1080p to 4k/3840×2160, 720p to 4k, etc.).



System 900 includes a dataset preparation module 902 that is used for preparing images (e.g. 1080p images) that are provided from a training set database 906. The images are prepared and then used to train a neural network (e.g., to determine the coefficients of L & R, including each layer of sums of L & R transforms, that are discussed herein) via the neural network trainer module 904. The neural network trainer module 904generates one or more trained neural networks that are stored into database 908. Trained neural networks 908 may then be communicated via a network 912 (e.g., the Internet) or via physical media (like game cartridges) to various game devices 1, 2, 3, 4, 5, etc. (each of which may be an example of game device 100). In certain example embodiments, one or more trained neural networks may be delivered along with a game that is acquired by a user. For example, a user may download a game from an online store or the like and one of the components of the game may be a neural network for processing images produced by the game. Similarly, games that are provided on cartridges or other physical media may include one or more neural networks that can be used by the user to transform images produced by the game. In certain examples, multiple neural networks may be provided with the same instance of a game (e.g., an individual download or specific physical media instance) to allow for the game to output to different types of displays (e.g., 1080p in one instance, 1440p in another, 4k in another, etc.).
 
I've seen. it's more of a port from x86 to ARM and is using the same libraries as the windows version.



something I missed was Mediatek had their own event showing off the Dimensity 9200 and it's ray tracing capabilities. Arena Breakout was playable there (which is where the previous video came from)





both of these events imply to me that both ARM and Qualcomm are designing a simpler RT core that's more akin to AMD's implementation than Nvidia's: focus on ray triangle intersection and leave bvh traversal to the compute units. it might result in a smaller area for RT functions and would force simpler workloads for power limited hardware. shadows, mirror reflections, and AO are prominent in all the demos between the two, with Qualcomm touting a demo running below 5W and Mediatek showing a proper game running on proper hardware (phone body et al).

it begs the question what this could mean for Drake, who's SoC would have to contend with similar power constraints. don't think Drake would be limited to just mirror reflections and shadows, but with a low power budget, it might be a case of tradeoffs of which RT effects devs choose to go with. it's gonna be much better than phones can provide, but how much better?

Well, we say "similar power constraints", but they're prioritizing different things in their power budgets, so not sure the constraint is as similar as that.
Like, if Drake is still operating with the 7W (?) portable TDP rate we saw in Switch, that's still a higher thermal budget overall than most smartphones and there's going to be undoubtedly far less overhead for Drake, being that Nintendo will probably stick to a super lightweight micro-kernel system software, among other important differences.
 
0
I wouldn’t really read into it like that, it’s a filing they’ve been trying to patent for a while that seems like it is a method that they can also use on other hardware in the chance that they need to move off of NVIDIA. It doesn’t allow them to be limited to that.
I get it. But it's an unmistakable sign that Nintendo is dipping its toes in the machine learning supersampling and that only reinforces the probability to see Drake provide this service.

Finally a good reason to go on being excited.
 
0
The patent also confirms that 'DLSS' is updateable via patch (network).

Fig 9. shows a training method, maybe it's DLSS 1.0?
It's not DLSS.
It's something more like nvidia's AI upscaling on Shield tvs, can be applied to any video feed, while DLSS requires data from real time rendering. Just a continuation of NERD's old patent anyway.
 
Leak from 4chan


Is this believable?
As someone with easily 10k+ hours in Monhun total, 99.9% of absolutely not lol

"Paradise" was a name thrown around that indeed spurred from a Discord leak of sorts. But all these details are extremely against the current design philosophy of modern Monster Hunter - to the point where it feels so blatant, that whoever made these up were probably just trying to get some attention.

"100+ monsters", unless they literally and successfully ported every single monster from from Iceborne/Sunbreak, and then added a new game's worth of new ones, this is a hard no (unless they're counting small monsters...?)

"Battle Pass system", but there's just no feasible way to do this. It doesn't even make sense for MH, in the way it's progression is structured. The game itself IS the progression along the ranks, fighting stronger and stronger monsters to make better and better equipment.

Mispelled "Ryse"...

Human villains just aren't a thing. Sure, there may be some internal character drama here or there (like in 4U or World/IB), but the point of these games is that we're just dealing with nature. And nature is scary.

"Hunter Arts" were already a thing from Generations / Generations Ultimate. This point is the one that really convinced me that whoever wrote this has no idea what they're talking about. Besides, the actual mechanic itself makes no sense - "special abilities like faster movements, explosive attacks, eletric body etc." are just generic "powers" that really don't fit in with the combat style of Monster Hunter.

Then again, I will gladly eat my words if this ends up being true, knowing that Ryozo completely lost his mind when he pushed for this.

-

With that said, I really, REALLY wish MH6 or whatever it's called is a cross-play game, available to ALL platforms, including Drake. One can dream.
 
Not sure if this has been posted yet but I found a patent regarding DLSS for the Switch.


So in the diagram, looks like Nintendo will be using a based 540p to 1080p conversion. Hopefully, this means a 1080p screen portable.

US20220335570A1-20221020-D00002.png
This is a continuation of the one NERD filed back in 2020. Continuations happen frequently when the applicant wants slightly more or different legal coverage in the claims but it's for the same overall incention/concept.
 
As someone with easily 10k+ hours in Monhun total, 99.9% of absolutely not lol
This was proven fake back in July or so on Era. It was a random name found in discord.
The image is likely fake, but MHParadise may exist as a thing. But as a guess somewhere, not as something meant to be a game.
Any chance that Switch operates at 18W in docked mode instead of 15W like Switch does? 3W extra would potentially boost CPU and GPU clock speeds a bit for DLSS up to 4K/30fps
Mm, no not really. It’s not really good for them to just boost the CPU as it creates disparity which you wouldn’t want that. GPU and CPU will be limited because of their memory bandwidth, so it’s not a good idea to go above the slowest ship in the fleet.


As I’ve said before, it seems like the ceiling for them in docked is 3.2TF, and 1.6TF in portable mode, and that’s with not pushing the memory bandwidth constraints and having even lower relative memory bandwidth than the other ampere parts.


But this is just speculation.
 
Let’s hope Scarlett and Violet essentially being neutered by performance issues gets this thing out the gate.

Even though I think it’s moreso about what GameFreak didn’t do as opposed to the current Switch performance since Xenoblade 3 is good.
 
Let’s hope Scarlett and Violet essentially being neutered by performance issues gets this thing out the gate.

Even though I think it’s moreso about what GameFreak didn’t do as opposed to the current Switch performance since Xenoblade 3 is good.

Nintendo has had no issue with releasing poorly running games. I wouldn't imagine pokemon, the game that sells itself, would make them reconsider. The game looks way worse than even Xenoblade 2 though in render quality. The difference between the new pokemon and xenoblade 3 is like early gen vs late gen quality.
 
Mm, no not really. It’s not really good for them to just boost the CPU as it creates disparity which you wouldn’t want that. GPU and CPU will be limited because of their memory bandwidth, so it’s not a good idea to go above the slowest ship in the fleet.


As I’ve said before, it seems like the ceiling for them in docked is 3.2TF, and 1.6TF in portable mode, and that’s with not pushing the memory bandwidth constraints and having even lower relative memory bandwidth than the other ampere parts.


But this is just speculation.

Oh I forgot about the CPU disparities. How about feeding that extra 3W to the GPU then? That would be good for games and I don’t think bandwidth will hinder that much like it does on Switch
 
Let’s hope Scarlett and Violet essentially being neutered by performance issues gets this thing out the gate.

Even though I think it’s moreso about what GameFreak didn’t do as opposed to the current Switch performance since Xenoblade 3 is good.
Don't think it makes any difference.
Game Freak are technically a decade behind any Nintendo studio.
They should be embarrassed at the state of the visuals in their games.
 
Not to drag this discussion in here any further, but two points:

1. ScarVio's issues, or generally GF Pokemon game issues, are not rooted on what console they are on. It's a variety of specific things happening within and around Gamefreak.

Even if Drake was available right now, the games would at best only be able to keep their taget framerate a bit more often.

2. After watching a couple of reviews over the lunch break, i noticed that the most glaring problems aren't "performance" that much (though it is noticeable), but downright graphical/lighting glitches and errors.

I don't expect GF to adress performance much with patches, but the glichtes and errors i've seen are definitely issues where a dev usually does have patches in the pipeline. This is not a "well, couldn't make it better" but a "well, no time, we gotta ship it like this".
 
Whatever they end up doing, I’m definitely not looking forward to trying to apply another screen protector. Went through like 6 packs of them before I got a good fit on my OLED.
 
re: the patent.

This revision more heavily emphasizes the advantages of using this technique in real time rendering, and the specific advantages it provides. The technique as described would be an excellent tool for upscaling emulated games. It had been rumored that this patent was the product of work done by NERD on 3D All Stars, so that tracks.

Unlike DLSS it doesn't automatically anti-alias the resulting image, especially useful if you're underlying game already is anti-aliased. It uses a tiled approach to keep the working set entirely in cache in order to never touch DRAM is possible. It uses a series of AI models of varying levels of complexity, and dynamically detects the underlying games frame rate, and selects a model complexity to complete inside the frame budget. It doesn't require in-engine integration.

If I wanted to wildly speculate (and I do!), the best use for this technique in the context of Drake would be to provide high quality upscaling to BC Switch games which use an emulated GPU.
 
re: the patent.

This revision more heavily emphasizes the advantages of using this technique in real time rendering, and the specific advantages it provides. The technique as described would be an excellent tool for upscaling emulated games. It had been rumored that this patent was the product of work done by NERD on 3D All Stars, so that tracks.

Unlike DLSS it doesn't automatically anti-alias the resulting image, especially useful if you're underlying game already is anti-aliased. It uses a tiled approach to keep the working set entirely in cache in order to never touch DRAM is possible. It uses a series of AI models of varying levels of complexity, and dynamically detects the underlying games frame rate, and selects a model complexity to complete inside the frame budget. It doesn't require in-engine integration.

If I wanted to wildly speculate (and I do!), the best use for this technique in the context of Drake would be to provide high quality upscaling to BC Switch games which use an emulated GPU.
This theory makes a lot more sense to me than the “future proof for changing gpu vendor” theory.
 
A mid gen refresh option in 2019?

Naw, Nintendo had already realized the success of the Switch and were already talking about an unusually long lifecycle for Switch compared to usual consoles (6 years in the market and then release a “successor”)

If Nintendo was looking at 8-9 years instead, the mid gen model would come 4-5 years in, not 2.
Any hardware manufacturer who is a couple years in and making plans like they're assured to still be relevant at 8 or 9 is pretty full of themselves. A "Switch Pro" 2.5 years after launch wouldn't have been very different from PS4 Pro being 3 years after launch.
 
0
re: the patent.

This revision more heavily emphasizes the advantages of using this technique in real time rendering, and the specific advantages it provides. The technique as described would be an excellent tool for upscaling emulated games. It had been rumored that this patent was the product of work done by NERD on 3D All Stars, so that tracks.
Generally a continuation needs to have (legally speaking) the exact same disclosure as the originally filed application. The only differences should be in the claims. So I'm not sure I see them emphasizing anything different here.

Unless they decided to amend the disclosure after filing which would be unusual and confusing.


I do like your theory though.
 
0
Paradise is a fake name. And we know World 2 is a think thanks to Capcom leaks
World 2 would be the internals codename, but not representative of what CAC wants to market it as. They named World what it was to get away from numbering in marketing and not confuse people.

World 2 doesn’t stop the possibility of Paradise in marketing
Oh I forgot about the CPU disparities. How about feeding that extra 3W to the GPU then? That would be good for games and I don’t think bandwidth will hinder that much like it does on Switch
As I’ve mentioned, it likely isn’t a good idea you’d be pushing it more into the switch direction imo. If it had LPDDR5X, then yeah if they wanted to. But I’m aiming to keep everything within balance.

A balanced system is a better system.

Switch is unbalanced since it has a really good GPU, but the rest is bad.
 
They likely have a NVN-unique solution based on nVidia's standard DLSS.
The NVN solution is a means of interfacing with the DLSS library, not a separate upscaling technique like what's being described in this patent. NVN is also made by Nvidia, while this is a Nintendo patent.

And of course people have already pointed out that this is a continuation of a previous patent by Nintendo, and not related to DLSS.
 
Any chance that Switch operates at 18W in docked mode instead of 15W like Switch does? 3W extra would potentially boost CPU and GPU clock speeds a bit for DLSS up to 4K/30fps
I think it might even be likely for the docked power consumption to be much higher than the previous generation, I think 30W is possible. The Dock with LAN Port supports power outputs up to 39W, which no Switch system has used yet. I still believe the Dock with LAN Port is the dock that the new system will use; and I don't think it'll even change power brick. 39W output, 4K60 support, better ventilation than the previous dock, it has everything it really needs. It could perhaps even support fast charging with the same brick.

As for compatibility with the original Nintendo Switch Dock, I actually have my doubts. I think it's possible given the original dock's power constraints and ventilation that a Drake in one would either refuse to output to a TV (and just charge) or operate only at its handheld/tabletop mode profile. This isn't such a big deal given this dock can't output 4K anyway, but I don't see how else it'd work given the lower output compared to the Dock with LAN Port, and the worse ventilation. I suppose it could always use the same power requirements as the original Switch and not have any issues, but then I'd wonder why the Dock with LAN Port can output 39W at all- other than perhaps fast charging support.
 
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