• 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!)

Maybe i'm wrong, but wasn't it the VGC article that was published around the same time that had these claims?

No, you are remembering correctly:

Another VGC source claimed that Nintendo showcased Epic’s impressive The Matrix Awakens Unreal Engine 5 tech demo – originally released to showcase the power of PlayStation 5 and Xbox Series X in 2021 – running on target specs for its next console.

The demo is said to have been running using Nvidia’s DLSS upscaling technology, with advanced ray tracing enabled and visuals comparable to Sony‘s and Microsoft’s current-gen consoles (however, it should be noted this does not mean the Switch successor will sport raw power anywhere near that of PS5 or Xbox Series X, which aren’t portable devices).

 
Then around 2020 there was this idea floating around that it made financial sense for Nintendo to stop producing the regular Switch as it was not cost effective anymore, a console they're still producing in 2024
They haven’t produced the original switch since 2019.
 
One of these anniversaries Nintendo will actually celebrate correctly.
  1. 25th not only was a dud but the duddiest dud that ever dudded.
  2. 30th was better but mostly a dud (thanks for Super Mario Maker though).
  3. 35th was even better, enough not to be a dud but still kind of lacking for the world’s most recognizable character. It was mostly missing a brand new Mario platformer. Bowser’s Fury was good, tying it to a port of Super Mario 3D World made it seem like a side-attraction.
While I don’t think a Mario’s 40th anniversary is the reason for a 2025 Switch 2 launch, I hope Nintendo goes all out with Mario using all the whiz-bang stuff T239 can do.
Nintendo also did nothing to celebrate Donkey Kong 40th anniversary(not even a tweet to knowlege Donkey Kong 40th anniversary they did)
 
hot take: i dont think ps5 pro is coming this year.
like, why release a pro model now if there will be no/barely any software releases in 2024?
gta 6 is in 2025 as well, and i can totally understand them having new hardware ready before the game's release but like, what if gta 6 is very early 2025 like idk, march~april? then a ps5 pro releasing in november 2024 would have a very small timeframe to deal with scalpers and stock issues
 
they released a PSVR2 with barely no games, and a PSPortal while you already can play on any device with the remote play.

I don't think the lack of software will hold them from releasing a pro - not to mention we still don't know much about the lineup for this year.
 
hot take: i dont think ps5 pro is coming this year.
like, why release a pro model now if there will be no/barely any software releases in 2024?
gta 6 is in 2025 as well, and i can totally understand them having new hardware ready before the game's release but like, what if gta 6 is very early 2025 like idk, march~april? then a ps5 pro releasing in november 2024 would have a very small timeframe to deal with scalpers and stock issues
Surely the simple pleasure of owning a new high-tech box to put on a shelf is sufficient to base any hardware launch on!
 
Stop attacking my decision to buy a PS5 at launch. ;_;
I only got mine at launch because the ps4 pro sounded like a jet engine 😅 .
With the pro the improvements need to be really big to make me buy one at launch (and if the rumours that the cpu is pretty much the same I doubt it). But even if they do not have the software, nothing is stopping them from partnering with third party to have "updated versions" of the current games. Launching early may also help with stock issues when future games do indeed drop.
 
0
About CPU, the Steam Deck has a variable frequency @ 2.4 – 3.5 GHz, right? Can Switch 2 use the same strategy? Something like 1.2 – 2.1 GHz of variable frequency can give a good gain in performance or a fixed frequency is better?

And what CPU can be better at the end: AMD Zen 2 with 4-cores @ 2.4 – 3.5 GHz or a Cortex A78C with 8-cores @ 1.2 – 2.1 GHz?
 
I'm pretty sure the switch CPU already does variable frequency. It's quite common nowadays specially on a mobile device where you can save battery
 
About CPU, the Steam Deck has a variable frequency @ 2.4 – 3.5 GHz, right? Can Switch 2 use the same strategy? Something like 1.2 – 2.1 GHz of variable frequency can give a good gain in performance or a fixed frequency is better?

And what CPU can be better at the end: AMD Zen 2 with 4-cores @ 2.4 – 3.5 GHz or a Cortex A78C with 8-cores @ 1.2 – 2.1 GHz?
First question: No. SteamDeck has variable CPU clocks because it's a PC and power budget is automatically handled. On a console, you want predictable and repeteable performance. Nintendo does already offers a variable CPU clock strategy for faster loading, which is a scenario mostly dictated by CPU and storage and thus it can be allowed for the GPU to sit back.

Second question: Both CPU are comparable performance per clock. It depends on the workload. There are workloads which can't be parallelized and benefit more from higher single core frequency. Others will benefit more from more cores available. In general though, you want to have more cores. And games specially have been trying more and more to be able to benefit and make usage of multiple cores. Avatar and Cyberpunk are good examples of games which benefit from having more cores available.
 
About CPU, the Steam Deck has a variable frequency @ 2.4 – 3.5 GHz, right? Can Switch 2 use the same strategy? Something like 1.2 – 2.1 GHz of variable frequency can give a good gain in performance or a fixed frequency is better?

And what CPU can be better at the end: AMD Zen 2 with 4-cores @ 2.4 – 3.5 GHz or a Cortex A78C with 8-cores @ 1.2 – 2.1 GHz?
First question: No. SteamDeck has variable CPU clocks because it's a PC and power budget is automatically handled. On a console, you want predictable and repeteable performance. Nintendo does already offers a variable CPU clock strategy for faster loading, which is a scenario mostly dictated by CPU and storage and thus it can be allowed for the GPU to sit back.

Second question: Both CPU are comparable performance per clock. It depends on the workload. There are workloads which can't be parallelized and benefit more from higher single core frequency. Others will benefit more from more cores available. In general though, you want to have more cores. And games specially have been trying more and more to be able to benefit and make usage of multiple cores. Avatar and Cyberpunk are good examples of games which benefit from having more cores available.
I would also like to add that theoretical performance is only part of the story. SD games run on a translation layer (Proton) which I assume mostly take up cpu cycles, while console games run natively. Console games also have hardware specific optimizations.

Edit: SD also lacks dedicated decompression, that takes load of the cpu.
 
Last edited:
This question isn't tangentially related to Switch 2 but it's not really about Nintendo, it's about Sony. With how Sony/Playstation's financials are going and the massively ballooning dev costs due to higher fidelity and bigger games, does releasing a PS5 Pro make that exact problem even worse for them? Is it worth it to Sony to release the PS5 Pro if it just means more dev cost investment and it's not gonna do much to grow their ecosystem, just replace current PS5 owners with Pro owners. Do they just cancel their plans for it and retreat to more cost cutting and efficient development?

In relation to Nintendo, is the poor economy, tech inflation, and struggles people are going through a factor in some of their plans to deal with how they price and market this system? Even if the tech is a generational leap, are we expecting their games to reflect that if it means massively ballooning dev costs?
Imho, the sustainability of current dev costs and the "need" for a Ps5 Pro are partially unrelated issues.
A new hardware variantion per se won't do anything to grow the overall volume of the market, it's a matter of acquiring more marketshares in the current market. More specifically I think a Pro would target both current Ps5 owners, as well as gamers who are already planning to buy a 9th gen system or a gaming PC.

As for Nintendo, I would expect more games to be priced at 70$ like ToTK, but overall I don't think it would be too different from now. Expecting games to keep their full price years after the release.
 
with it being march do y'all think we might hear something this march, like nintendo just mentioning they're developing new hardware, because with CDG coming we might hear a floodgate of information of which ports we might see from developers and also finally hear the true specs
(and hopefully one developer will just say nintendo are using 4NM)
 
Arm only mentioned two options for the Cortex-A78C: 6 Cortex-A78C cores or 8 Cortex-A78C cores.
A78C-blog-post-image5.JPG
But Dak, you don't understand, in his heart 🫶🏻 because Nintendo.
And Nintendo's love of cheapness defies reality!
 
with it being march do y'all think we might hear something this march, like nintendo just mentioning they're developing new hardware, because with CDG coming we might hear a floodgate of information of which ports we might see from developers and also finally hear the true specs
(and hopefully one developer will just say nintendo are using 4NM)
The closest we will get is a vague announcement at the next investors meeting.

Also no developer is going to know the node used. We won’t find out until a while after launch
 
with it being march do y'all think we might hear something this march, like nintendo just mentioning they're developing new hardware, because with CDG coming we might hear a floodgate of information of which ports we might see from developers and also finally hear the true specs
(and hopefully one developer will just say nintendo are using 4NM)
I doubt that lithography will be in the specs sheet that Nintendo give on their development kits. But maybe they can give other useful infos, like the quality of the ports their are making, how good ray tracing and DLSS are, or, if we get luck, the TFlops that machine can run.

Right now what everyone want, in realistic side, is a hybrid close as possible it can be from Series S but with some improvements for it be a Nvidia and have more RAM.

Maybe the developers can tell us if what we are expecting is realistic, without talk about node.
 
Last edited:
The closest we will get is a vague announcement at the next investors meeting.

Also no developer is going to know the node used. We won’t find out until a while after launch
Makes sense, but couldn't we speculate which node it'll be with the ports the developer are doing, because then we'll know how overclocked it'll be.
Like the only ports we know of as of right now is persona 3 reload, metaphor and the new monster hunter game that's releasing next year.

(correct me if i'm wrong because my knowledge wtih computing is minimal)
 
I doubt that lithography will be in the specs sheet that Nintendo give on their development kits. But maybe they can give other useful infos, like the quality of the ports their are making, how good is ray tracing and DLSS is, or, if we get luck, the TFlops that machine can run.

Right now what everyone want, in realistic side, is a hybrid close as possible as can be of Series S but with some improvidentes for it be a Nvidia and have more RAM.

Maybe the developers can tell us if what we are expecting is realistic, without talk about node.
If we're somehow able to get clock speeds we might be able to infer which node process seems more likely (considering there is in theory "optimal" clock speeds for a given node process)
 
i
I doubt that lithography will be in the specs sheet that Nintendo give on their development kits. But maybe they can give other useful infos, like the quality of the ports their are making, how good is ray tracing and DLSS is, or, if we get luck, the TFlops that machine can run.

Right now what everyone want, in realistic side, is a hybrid close as possible as can be of Series S but with some improvidentes for it be a Nvidia and have more RAM.

Maybe the developers can tell us if what we are expecting is realistic, without talk about node.
I'm curious about ram because it's likely the developers had something to do with it, because the switch was originally meant to have only 2 Gb of ram, but was increased to 4 because of capcom, so i'm expecting 12-16 gb ram. Because Capcom will put out a lot of ports like re4 remake and the new monster hunter game.
 
i

I'm curious about ram because it's likely the developers had something to do with it, because the switch was originally meant to have only 2 Gb of ram, but was increased to 4 because of capcom, so i'm expecting 12-16 gb ram. Because Capcom will put out a lot of ports like re4 remake and the new monster hunter game.
the gigaleak shown it was always planned to have 4GB. the capcom comment might have been feelers put out by Nintendo to gauge if their direction was correct
 
Gonna be rough when I have to deal with anti-aliasing for Switch 2 because I'm just not a fan of the way it looks. Give me the raw jaggies.
I have mixed feelings. All else being equal I prefer maximum detail, with jaggies intact, over anti-aliased-but-weirdly-soft. But temporal instability drives me nuts in motion.


TFB: We want independence
MS: Sure
TFB: We need your collaboration
MS: o_O
The way I read this is "we want to keep doing everything we've been doing, but we absolutely do not want to be sent to the Call of Duty mines"
 
The way I read this is "we want to keep doing everything we've been doing, but we absolutely do not want to be sent to the Call of Duty mines"
You make 16 skins, what do you get?
Another day older and deeper in debt
St. Peter, don't you call me 'cause I can't go
I owe my soul to the Microsoft Store
 
About CPU, the Steam Deck has a variable frequency @ 2.4 – 3.5 GHz, right? Can Switch 2 use the same strategy? Something like 1.2 – 2.1 GHz of variable frequency can give a good gain in performance or a fixed frequency is better?

And what CPU can be better at the end: AMD Zen 2 with 4-cores @ 2.4 – 3.5 GHz or a Cortex A78C with 8-cores @ 1.2 – 2.1 GHz?
if they will give 5W budget for CPU and 5W for GPU we will have 2.5GHz for CPU
 
@BDGAME This will be of your interest regarding the dynamic clocks question. It isn't something that Nintendo overlooked, but rather they received a request and decided against it:

Capcom asked Nintendo about implementing a feature to adjust the GPU clock corresponding on scenes (note: this is a development term), but Nintendo said that even if you lower the clock, the overall processing time will be longer and it won’t necessarily give an advantage in the end.

with it being march do y'all think we might hear something this march, like nintendo just mentioning they're developing new hardware, because with CDG coming we might hear a floodgate of information of which ports we might see from developers and also finally hear the true specs
(and hopefully one developer will just say nintendo are using 4NM)
No. Expect something to be said in May, at the Shareholders Meeting/FY25 guidance. If Nintendo doesn't say anything in May, it's not coming in FY25.

But March is out of question, even if we hear some whispers due to GDC. Nintendo will simply deny and remember that they have official communication channels and people shouldn't believe baseless rumors.
Makes sense, but couldn't we speculate which node it'll be with the ports the developer are doing, because then we'll know how overclocked it'll be.
No. A game port has nothing to do with the device performance nowadays, but how much money and manpower is being put into it. Even if Switch 2 is super low clocked and fabbed on Samsung 8nm, a lot of ports will happen because the device is expected to sell a lot and publishers will want to take advantage of the new revenue stream.

One way to try to guess the node, however, is when both clocks and battery size information are leaked. So we'll be able to infer, when comparing with the current and public energy x clock data we have of 8nm Orin, if it's fabbed on 8nm or 4N. But it would be just an educated guess, of course.
I'm curious about ram because it's likely the developers had something to do with it, because the switch was originally meant to have only 2 Gb of ram, but was increased to 4 because of capcom, so i'm expecting 12-16 gb ram. Because Capcom will put out a lot of ports like re4 remake and the new monster hunter game.
The RAM story is something that has been increased with the passage of the years. What happened is that when Nintendo was asking third-parties what they wanted in the next device, Capcom and others Third-parties repeatedly asked for a increase in the RAM amount:
The first thing that came up in the exchange of opinions was the capacity of the main memory. The numbers initially presented by Nintendo were sufficient compared to other hardware at the time, but Capcom dared to say, "This is not enough!" The reason for this was that the development of a new engine, the RE ENGINE, was underway within the company at the time, and we thought that memory was absolutely necessary to meet the specifications required by the engine.

Mr. Mitsuyoshi also said, "I remember very well what was said about memory in the first voice." Regarding memory, it seems that many points were made by people other than Capcom. However, if you ask a programmer how much memory they want, they will say "I want infinitely." It seems that it was Yamayama who wanted to put a lot of Nintendo on it, but it was difficult to balance the cost and it took time to adjust. Mr. Ijuin said that he was relieved that Capcom's request was finally installed as it was (however, the actual memory capacity was not mentioned, and only commented that "it was raised considerably").

As you can see, the memory wasn't doubled neither it was 2GB. It increased considerably from what Nintendo considered to be a good approach on amount x price per requests of developers. But that wasn't something that was changed mid-development of Switch, but before. As per Gigaleak, we know that since ever the first specification draft of Switch, it already included 4GB of RAM memory.

Mind you, this isn't something unusual, but quite the way consoles are developed. Manufacturers will look at request from 3P just as much as general tech direction and evolution and start from there. For example, here are two quotes from Iwata on how they gathered feedback from 3P and answered accordingly:

Iwata:
But it was pretty hard preparing a two-gigabyte ROM in time for release of the Nintendo 3DS system. But the Licensing department that works with outside companies strongly requested it hearing feedback from those companies. Some at Nintendo were saying, "Do you even need all this storage space?" and "How are you going to fill it all with data?"

Not only is there the draw of being able to see things in 3D, you've added a number of other features that may become a topic to be discussed in the public, and this title put itself very naturally into the flow of the Street Fighter revival project. I can tell that the Capcom team is very adamant about implementing the wireless features. The tech staff of Nintendo had to match your enthusiasm. It was a very challenging but interesting job

I had heard reports that our Licensing department, which is in charge of contacting software developers, has received many comments to the effect that some games couldn't be made for the Nintendo DS system because of graphics limitations, so we wanted to do something about that.

And I think we did that with the Nintendo 3DS system. Based on the developers' demands, we also prepared a ROM with lots of memory space so they can pack in the necessary graphics data.

The way I read this is "we want to keep doing everything we've been doing, but we absolutely do not want to be sent to the Call of Duty mines"
Toys for Blob were already working on Call of Duty. In fact, their last work in doing assistance for Call of Duty Warzone Mobile. But I do agree this is a chance they saw with the new leadership to be independent by doing a deal with Microsoft. Not too dissimilar to what Bungie did to also separate themselves from Microsoft Game Studios.
 
Implicitly clock speeds will give us a very good idea though.
Yes, although not with 100% certainty, especially since there's a nonzero (although not particularly high) chance Nintendo and Nvidia could use TSMC's N6 process node for fabricating Drake, which is much better than Samsung's 8N process node, but not as good as TSMC's 4N process node.
 
Thanks for digging deeper than I was able to. To follow up what you discovered, I found that ADATA exhibited their microSD Express cards (256GB/512GB) at the CES 2024:

P6Vh1f5.png
c1jt0aZ.png


At the CP+ 2024, the SD Association booth displayed three manufacturers’ microSD Express cards—GTS, Phison, and SanDisk. See the lower right corner of the following photo (source):

63a28_1223_67bdceac998e00b9d4edd5c8ba862e82.jpg


GTS is a small Japanese OEM. On their website, two microSD Express cards (256GB/512GB) with their test results are shown:

gts04-1.jpg

gts04-3.jpg


SanDisk also exhibited their SD and microSD Express cards at the CP+ 2024, and even did some speed tests live (source):

007l.jpg

008l.jpg


It seems that the SD Association and some card manufacturers are giving the microSD Express format a renewed push. I suspect that the proliferation of handheld gaming PCs caught their attention. The recent introduction of 3-in-1 UHS-I/UHS-II/Express controller IC and card socket also gives the Express format a fighting chance in the professional camera market.

The speed of the ADATA card is interesting, as it's only claiming 600MB/s read speeds. I'd wager this is a limitation of using the existing 28nm Phison SDE controller, and they have to throttle down to get it to work in a microSD form factor without melting.

The host controller which supports both UHS-II and SD Express is interesting, but I doubt it will change much in the camera market, as almost every manufacturer has already adopted CFexpress. Panasonic might adopt it, as they have a history with SD, having developed the cards in the first place, and I don't believe that they've used CFexpress at all yet, but all the other manufacturers are pretty well committed to CFe by this point. My guess is that the renewed push is more focussed on the microSD form factor, either driven directly by Nintendo, or just generally the number of devices (handheld gaming PCs, drones, etc.) that are limited by current microSD performance but are unlikely to adopt the physically larger CFe format.

Incidentally, speaking of CFexpress, I missed that the first CFe 4.0 Type A cards were announced by Lexar, with up to 1TB of capacity and read speeds of 1.8GB/s. No word on price, but likely very expensive due to the small market and professional customers. Incidentally, they also recently launched a 1TB M.2 2230 SSD which you can get for $80, and is likely using identical hardware to their CFe 4.0 cards, but without the plastic case.
 
I do have one question though in regards to the questionability of 8 nm, that I don't think has been brought up (or if it has, I missed it, so apologies)

Wouldn't it just be a practical problem for the foundry to continue having to manufacture on an increasingly old process as time goes on? Don't companies like TSMC, Samsung, etc. want to just keep moving to the newest processes? Working on an older ones probably crowds out the better, newer ones I'd imagine. And if Switch 2 would have the same success as Switch, Samsung or whoever, would be making 8 nm chips all the way to 2032, when even the PS5 and XBSX/S moved to newer processes all the way back in 2020?

Not all semiconductor products need to be on a leading edge or near-leading edge process.

Take TSMC's Q4 2023 earnings report: https://www.anandtech.com/show/21239/tsmc-q4-2023-earnings-3nm-revenue-share-jumps-5nm-overtakes-7nm

About 67% of their revenue comes from N7, N5, and N3 class nodes. However, they still have billions of dollars in revenue from 16nm, 28nm, 40/45nm and 65nm, and even have customers utilizing processes as old as 0.25 um (250nm).

Things like microcontrollers, smart appliances, automotive chips, interposers, etc. utilize these older node families because they don't require the transistor density, performance, or power consumption that advanced nodes provide. While these older processes may be in some cases more expensive per transistor than newer nodes, design and validation costs for chips on advanced processes has exponentially increased, so overall it makes economic sense for these sorts of semiconductor products to remain on older nodes.

Building a fab costs billions of dollars, and upgrading an old foundry to have the capability of manufacturing advanced process technologies ranges in the tens of millions to billions. Once a fab site has become net profitable, there's few reasons to upgrade it: perhaps long term, customers may switch to a newer processes, or demand for advanced processes so far exceeds supply that you risk losing customers to a competitor foundry because they can't book capacity on your leading edge nodes.

TSMC has actually pissed quite a few customers off recently with their announcement that they will not be building any more 65 or 40nm capacity and encouraging customers on or considering these processes to move to 28nm. While these older fabs are net profitable for TSMC, the profit per wafer is lower than newer processes. (As a small aside, 28/20nm is TSMC's last planar transistor node, with all newer processes using FinFETs and costing more in design and validation than planar nodes.) So while there is an incentive to push customers toward newer processes due to their higher profitability for the foundry, foundries don't only maintain advanced nodes in their portfolio.

As another example, Samsung's advanced processes include their 3GAP/E, 4LPP/E, and 5 and 7nm class nodes. Yet Samsung uses their 8nm node for their Elpis and Pascal NVME controllers (found in their 980 Pro and 990 Pro SSDs respectively) and use 12nm for their latest DDR5 Memory ICs.
 
Last edited:
I thought they were already under contract with Nvidia before the Tegra X1 and Shield TV came out. Hasn't Nintendo been involved in any last-minute changes to the SoC, particularly in terms of security?
I know they were in contact with Nvidia back in the days of what eventually became the Nintendo 3DS, but I have to admit I have no idea if their contractual commitment to the Switch predates the Shield TV's conception, although I have to admit I'd be a bit surprised if it did, as the Shield TV was unveiled at CES 2013, just a few weeks after the Wii U's release, and long before its failure was a foregone conclusion.
If Nintendo says "yes, we could have gone to 5nm, but instead we've worked closely with Nvidia to build a massive power-saving architecture that allows us to get 5nm performance at the price of 8nm", I can't see how anyone could see that as a bad thing!
That's why I don't understand people's premature concern. We know for a fact that more RAM is better. We know that more storage space is better. We know for a fact that 1080p is better than 720 and that OLED is better than LCD. On the other hand, we don't know anything about the node that will be used, and even if we did, we wouldn't know anything about its optimisation, the engineering involved, or how it actually works from a customised perspective.
The idea that "I didn't think they'd wait this long" makes no sense, because these products are supposed to last more than 5 years.
And this is true for games consoles too. The PS4 and Xbox Series came out 7 years after the PS3 and Xbox One. The idea of making the Switch 1's potential 8-year lifecycle seem like some kind of mind-boggling, insane surprise seems far-fetched, to say the least.
Given how bespoke the T239 is, Samsung 8nm would seem to be a massive fucking oversight in the design. A big "Oops. My bad!"
As others have pointed out, the TX1, although Maxwell, was 20nm. So I don't see why the T239 would necessarily be 8nm, albeit Orin. Nvidia and Nintendo may have excellent commercial and industrial reasons for wanting to work with Samsung, but I don't understand why working with Samsung would necessarily mean opting for 8nm, as if it were impossible to do anything else with Samsung.
Given that the recent job offers at Retro are aimed at conceptual artists, they're probably wrapping up development of Prime 4 and moving on to their next project.
Honestly, I'd be very surprised if the groundwork for Prime 5 hadn't already begun. They've put so much effort and resources into rebuilding the Retro teams with profiles suited to Metroid, that the prospect of capitalising on those resources by making full use of the Switch 2's capabilities feels like an absolute no-brainer. It's also one of the reasons why, while I still fundamentally think there's a ton of sense in Prime 4 being crossgen, I don't think it's entirely impossible for it to come out before the Switch 2 and for the Switch 2's Prime to be Prime 5. Plus, as much as I endlessly love DK Country, I get the feeling Nintendo really intends to bring DK home. The fact that Miyamoto is taking time out from a Nintendo Direct to talk about DK's design is far from an insignificant detail, in my opinion.
OMG, I know why Nintendo delayed the Switch 2 until next year, it's because next year is Mario's 40th anniversary and what better way to launch the Switch 2 with a bunch of Mario games.
They're re-releasing Super Mario 3D All Stars, in a deluxe version with no sell-by date this time, and including Galaxy 2 and 3D Land.
 
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