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

[Paul_Subsonic]

I think battery will be def bigger than in Switch 1, so its possible, and i think if Switch 2 will be relased in 2024, it will be for sure on TSMC 4N

It's not just about battery though.

A higher clocked chip will require more power, which means less battery life but also, and maybe more importantly, higher temperatures.
And you've gotta find a way to cool down those higher temperatures.

This is even more a problem considering frequency doesn't scale linearily to power : what this means is that 2 GHz won't consume 2x the power over 1 GHz, but much more than 2x the power.

For Switch to accomodate a 2 GHz A78c it would have to be super bulky, like a Steam Deck if not worse. And we all know Nintendo is NEVER making a handheld as bulky as a Steam Deck.

 

[kvetcha]

It's not just about battery though.

A higher clocked chip will require more power, which means less battery life but also, and maybe more importantly, higher temperatures.
And you've gotta find a way to cool down those higher temperatures.

This is even more a problem considering frequency doesn't scale linearily to power : what this means is that 2 GHz won't consume 2x the power over 1 GHz, but much more than 2x the power.

For Switch to accomodate a 2 GHz A78c it would have to be super bulky, like a Steam Deck if not worse. And we all know Nintendo is NEVER making a handheld as bulky as a Steam Deck.

Power consumption is also closely tied to manufacturing node, so depending on what they end up going with they may have some play with higher clock speeds. Do we have power consumption estimates for A78c at 2ghz?

The Steam Deck is running a 2.5-3.4ghz quad-core x64 processor. I wouldn't be at all surprised if it was significantly hotter and more power hungry than an 8-core A78C.

 

[Hairway to Steven]

Pretty much the same. A lot of the AAA current gen games will skip the Switch 2. The console performance will be low enough that ports to Switch 2 will take some real effort, and many of these franchises just don't sell on Nintendo consoles the way they do on other consoles. Plenty of PS4 ports will come over because the hardware performance will make it low handing fruit.

Hopefully, then, there will be plenty of exclusives to make up for it.

 

[Bosintang]

Nate's new podcast dropped today. He and MVG were both in agreement that it seems like Metroid Prime 4 will launch this holiday.

He then said that he doesn't expect Nintendo to talk about new hardware until April 2024, with the next console then releasing in November 2024.

When goalposts constantly move, they eventually end up in the right place.

But yeah, finally he's probably going to be right any something hardware released. 2024 is most likely the year, and most likely always was.

 

[Paxie]

When goalposts constantly move, they eventually end up in the right place.

But yeah, finally he's probably going to be right any something hardware released. 2024 is most likely the year, and most likely always was.

This isn't even a leak, he's reiterated that this is personal speculation ever since he started talking about that specific release window

 

[Look over there]

My naive simple ballparking has 8 A78's needing 3-3.5 watts for 2 ghz on an N5 family node. Per ARM's own claims; an A78 needs 1 watt to hit 3 ghz on a '5 nm node' (assume TSMC). Going down from 3 ghz to 2 ghz should be in the ballpark of (2/3)^2, or 4/9th of a watt per core. Multiply by 8 to get ~3 5/9. You can cheat a bit and assume the OS to not necessarily run at that clock to get you down to below 3.5. And if it were a refined variant instead of base N5, you could probably get down closer to 3 watts.
The catch here is that I'm not sure on how high of a power budget we're getting for CPU. OG Switch was like 2 watts, or close to it. But IIRC, that was also the long term max after thermal throttling? Not necessarily a coincidence.

Regarding Steam Deck's CPU power draw: oh, definitely more than a couple watts. I distinctly recall seeing in some Steam Deck videos the CPU power draw hitting 10 watts at times.
I also have the feeling that a mobile Zen 2 CCX has a floor of somewhere in the ~1.25 watt range when not idle.
From here:

~2.5-3 watts is as low as it gets for the 4800H here. The 4800H is 8 cores/2 CCX's, so halving that is in the ~1.25-1.5 watt range.

... waaaait a second, I've done handwavey napkin math crap about this before, here.
...did I end up being right about a high of ~10 watts?

 

[kvetcha]

My naive simple ballparking has 8 A78's needing 3-3.5 watts for 2 ghz on an N5 family node. Per ARM's own claims; an A78 needs 1 watt to hit 3 ghz on a '5 nm node' (assume TSMC). Going down from 3 ghz to 2 ghz should be in the ballpark of (2/3)^2, or 4/9th of a watt per core. Multiply by 8 to get ~3 5/9. You can cheat a bit and assume the OS to not necessarily run at that clock to get you down to below 3.5. And if it were a refined variant instead of base N5, you could probably get down closer to 3 watts.
The catch here is that I'm not sure on how high of a power budget we're getting for CPU. OG Switch was like 2 watts, or close to it. But IIRC, that was also the long term max after thermal throttling? Not necessarily a coincidence.

Regarding Steam Deck's CPU power draw: oh, definitely more than a couple watts. I distinctly recall seeing in some Steam Deck videos the CPU power draw hitting 10 watts at times.
I also have the feeling that a mobile Zen 2 CCX has a floor of somewhere in the ~1.25 watt range when not idle.
From here:

~2.5-3 watts is as low as it gets for the 4800H here. The 4800H is 8 cores/2 CCX's, so halving that is in the ~1.25-1.5 watt range.

... waaaait a second, I've done handwavey napkin math crap about this before, here.
...did I end up being right about a high of ~10 watts?

Thanks for this. I thought I remembered there being a conversation about it at some point.

 

[Deleted member 887]

My naive simple ballparking has 8 A78's needing 3-3.5 watts for 2 ghz on an N5 family node. Per ARM's own claims; an A78 needs 1 watt to hit 3 ghz on a '5 nm node' (assume TSMC). Going down from 3 ghz to 2 ghz should be in the ballpark of (2/3)^2, or 4/9th of a watt per core. Multiply by 8 to get ~3 5/9.

Looking at Geekerwan's power curves for phone SOCs (which can hide a lot of fixed SOC costs, so should be taken with a grain of salt), the curve from 2GHz to 3GHz for A78 is pretty brutal. ARM provides curves but with no data points on them but since the A78 maxes out at 3.3, we can assume that's the far right side of those curves.

At ~2GHz, ARM says that A78 on 5nm FinFET is 50% TDP of A77 on 7nm FF. And they say that A78 is only 4% more efficient ISO process. Nvidia seems to indicate about 6W of TDP for 2GHz 8 core cluster on Orin, which is SEC 8nm, and known to be considerably more power hungry than TSMC 7nm

All of this to say - I think that there is a lot more power savings to be found below 3GHz.

 

[Look over there]

Ooh, yea, ~2 ghz seems like the dream target, balancing still-reasonable-perf/watt, power cost that shouldn't be too far out of reach from the OG, and being able to roughly maintain the same ratio of single thread strength as OG Switch:PS4.

...but to be clear for the rest of the folks reading this thread: I treat it as a possibility, but not necessarily likely. I hope for it, but don't necessarily expect it.
My expectations are more in the ~2-2.5 watt range. Clocks are of course node-dependent, but I'm sitting on a range of low to mid 1 ghz.
(ie I'm implicitly ruling out Samsung 8LPP; I still, to this day, cannot believe that one can power 12 SMs on that node and simultaneously have a battery life that satisfies Nintendo)

 

[Paul_Subsonic]

Oh wow. Didn't expect Steam Deck to be THAT power hungry.

So yeah, claiming that 2 GHz would mean a machine as bulky as a Steam Deck... that was pretty bad take.
But even then, it's clear 2 GHz is very unlikely.

 

[Look over there]

The Steam Deck as a whole is in the 20 to 25 watt range (wikipedia entry says 25). By default, the CPU+GPU power limit is 15 watts (but this should be user adjustable like any other PC that isn't locked down by an OEM). Then there's everything else that adds up to those 5-10 watts (RAM, display, fan, etc.)

 

[Procheno]

I'm still sticking by the idea that Metroid Prime 4 is 2024. Ain't no way is that game going to be Nintendo's "big holiday game" along with a bunch of DLC and Pokémon expansions. I also don't believe that they'd release Prime in the same year as MP4. Doesn't make sense

 

[Dragoncaine]

I'm still sticking by the idea that Metroid Prime 4 is 2024. Ain't no way is that game going to be Nintendo's "big holiday game" along with a bunch of DLC and Pokémon expansions. I also don't believe that they'd release Prime in the same year as MP4. Doesn't make sense

To me, the only reason to shadowdrop Metroid Prime Remastered is to plant a 9-to-13-month buffer between MPR and MP4. It's going to either be tail end of this year or Q1 of 2024, imo.

 

[Machrider]

So, we're all expecting Switch 2's (Japanese) 3rd party situation to be the same as it was on Switch?

Not that I mind it too much. Switch has had great Japanese support, but also more than a few painful glaring omissions (although I guess the same can be said in reverse too, since Nintendo's exclusives aren't on other platforms).

I would expect it to be better. I think Capcom will be all aboard this time rather than waiting years with MH Rise. So I'd expect Resident Evil early in the product cycle, Street Fighter, and Monster Hunter too.

 

[ILikeFeet]

Oh wow. Didn't expect Steam Deck to be THAT power hungry.

So yeah, claiming that 2 GHz would mean a machine as bulky as a Steam Deck... that was pretty bad take.
But even then, it's clear 2 GHz is very unlikely.

2GHz isn't unlikely on something like 4N

 

[JoshuaJSlone]

To me, the only reason to shadowdrop Metroid Prime Remastered is to plant a 9-to-13-month buffer between MPR and MP4. It's going to either be tail end of this year or Q1 of 2024, imo.

When Metroid Prime Trilogy kept being rumored and rumored and not showing up, it kind of became conventional wisdom that it was sitting on a shelf until closer to MP4, to help whet interest for that. It didn't turn out to be Trilogy after all but the same principle applies. MPR seems to have been ready for some time and for whatever reason they chose early this year to launch it.

 

[Hairway to Steven]

I would expect it to be better. I think Capcom will be all aboard this time rather than waiting years with MH Rise. So I'd expect Resident Evil early in the product cycle, Street Fighter, and Monster Hunter too.

Definitely. Same with S-E and Koei-Tecmo, I imagine.

I do wonder about Namco and Sega/Atlus.

 

[mjayer]

I think there's a high probability they are going to have MP4 and a new 3d Mario be their real drake show pieces. They can both launch closely to the Switch 2 without much concern about competing with each other as they are hitting seemly very different demographics.

 

[Dekuman]

We recorded an episode but I'm not totally happy with it. Listening back to it again & it'll either go live tomorrow morning or we'll rerecord and I'll post it on Sunday.

Just listened to the episode, it was good and always nice to hear you guys check in on your predictions from the previous week. Don't be too harsh on yourself. Thanks for posting it!

One thing i'd say about the summer direct is if they miss the June spot, then it's september. July/Aug seems to be dead in terms of marketing and i assume most people are on vacation at that time. With nothing announced for June or August , there may be at the very least a mini-direct where they shadow drop a June release and announce a few things covering up to Sept or October.

 

[Twilight]

I would expect it to be better. I think Capcom will be all aboard this time rather than waiting years with MH Rise. So I'd expect Resident Evil early in the product cycle, Street Fighter, and Monster Hunter too.

Tbf Capcom didn't wait years with Rise, the game was just in development for a good amount of time, between being the first MH game built on the RE Engine, and that to an extent development started before the Switch was out, plus covid.

 

[Pokemaniac]

Recent comments about multithreaded performance in the DF direct made me think of this. The complexities of multithreaded performance are hard for non-coders to understand. It's not only hard to extract the maximum theoretical performance out of a multithreaded system, but they are notoriously crash prone and insecure as well.

Technologies like Rust are starting to change that, but they require big overhauls of code to use, and are just now starting to trickle into video games. It'll be a while before reliably decent multithreading comes to games, which is why single core perf remains so important.

Of course the reason that multithreading perf is getting pushed at all is because we're hitting areas where pushing single core perf any further isn't possible on the scales that it once was. As huge as the jump for single core perf was with Zen 2, I'm dubious we'll be able to see a similar jump in the 10th gen consoles.

I hesitate to mention this, because someone is probably going to doom based on it, but one of the few portability disadvantages the Switch actually has is that some broken multithreaded code can work fine anyway on x86 because it has a memory model that's unusually lenient in terms of ordering memory accesses. I seriously doubt this is going to be (or has been) a significant factor in it a game will be ported or not (especially because all the consoles were PowerPC 10 years ago), but it is a thing. I think in theory, this lets ARM go faster in cases where order doesn't matter, as well.

Because 1,2 and 3 comes before 4 and playing games out of order ruinens the fun i think. I will say that I have not played MP 1-3 and don’t know how tightly they fit together storywise. For me it Wild be like playing Mass Effect 3 before 1 and 2.

The original Metroid Prime Trilogy has a highly self-contained story which is unlikely to connect to Prime 4.

 

[Erathix]

I'm still sticking by the idea that Metroid Prime 4 is 2024. Ain't no way is that game going to be Nintendo's "big holiday game" along with a bunch of DLC and Pokémon expansions. I also don't believe that they'd release Prime in the same year as MP4. Doesn't make sense

To be fair, dropping Metroid Fusion and Metroid Prime in the exact same year didn't exactly make sense either when they easily could've released a Metroid game once every year like clockwork back in the day with the releases it was having.

Dropping Prime Remastered in the same year, at the very beginning of it, is perfect for Prime 4 releasing later in the year. The general public, gamers who may have never experienced Prime before, are trying Prime for the first time through Remastered. To them it is like a whole new game, it's polished, it's stunning to look at, it teaches them the fundamentals of Prime and gets them invested in this part of the franchise after most likely playing Dread on Switch.

They will finish it within the week of release and by then, if they have become fans, they will be itching for more of it. Now that the general public knows what Prime is all about and how good it is, what do you do? You announce Prime 4 later on for a Holiday 2023 release now that you have people's attention firmly on Prime and that it has gone successfully. Prime 4 riding on the success and word of mouth of Prime Remastered will only benefit its sales potential to a wider audience outside of the hardcore Metroid fanbase.

I'm just entirely against the 2024 idea, both with this game releasing as well as Switch 2's release, it would be a mistake and the logical choice currently given we know nothing about the second half of 2023, is for the console to release for Christmas. There is no better time to maximise sales potential than a Switch 2 under the tree of a metric tonne of homes after people hear about it and what it will be capable of.

Have a new 3D Mario, perhaps that DK game that's been rumoured in the works for a while now, have a game like Prime 4 as well in the launch which will absolutely be a graphical showpiece (this was in development for Switch but honestly given what we know and how far back Drake development has been, who is to say that their goalpost never shifted and they pushed their technical aspirations further by having a Switch version and a Drake version with all of the bells and whistles, Switch version given what Remastered looks like will most likely be great and look fine but the Drake version? I imagine a game like Prime 4 to work with Ray Tracing in some capacity and to lock 4K, 60 FPS behind the Drake version would be more enticing as a selling point.)

I just don't think in my opinion that the first half of 2024 makes any sense, it made sense for May this year because of TOTK but outside of that factor I didn't think it would make sense at all to be released that far from a bigger shopping period than Christmas. Now that May is out of the question, it falls on Holiday 2023 to be when this is coming otherwise we'll be waiting until the end of next year for another opportunity at a release as I don't think there's any real way they would simply drop it in the first quarter of 2024 unless they seriously leave the second half of this year hung and dry and push games out to be in 2024, which we certainly do not know they will do yet and that is to be determined by how future directs this year go.

I'm all in on this year and unless some leak comes out that completely blows it apart, the next direct is going to reveal everything and determine whether or not it's coming Holiday 2023, same with Prime 4. And I'm adamant they will do Prime 4's reveal exactly how they did Dread's and release it in a similar window too.

 

[Erathix]

The original Metroid Prime Trilogy has a highly self-contained story which is unlikely to connect to Prime 4.

Not to mention Prime 1 itself is largely self-contained, if you don't bother to seek out and watch the secret ending reveal of Dark Samus, the game essentially completes its story and if you were none the wiser, you'd honestly never know. It's only when we get to Echoes do you need to start doing the whole trilogy for it to flow better.

I imagine they did this with Prime 1 because it is the most self-contained entry. It's the Super Metroid of 3D Metroid, a beautiful planet to explore, not as difficult or heart in mouth as Echoes gets rep for in some parts, but gets the balance just right, has a story that ends in a climatic showdown with Metroid Prime itself and resolves Phazon within the first game if you completely were unaware of where the story goes after and didn't bother to see Dark Samus rising up out of the remains. You don't need to play 2 or 3 to play 4 if you solely played 1. 1 immerses you into the world of the Prime side of the story and how it works mechanically and then from there you can play 4 which is most likely going to be far closer to 1 than it is to 2 or 3 because with the power of hindsight, Retro knows that 1 was the more acclaimed game of the trilogy and rightly so (no knock to Echoes or Corruption, Echoes was amazing in its own right but has its detractors and Corruption turned some off with its Halo-esque story beats and the way it handled exploration and how much easier they made the game after Echoes.)

It's easy to see where Nintendo's thinking lies with this release strategy for Prime.

 

[MP!]

Only way I could see mp2 and 3 coming out is if they were bundled together

 

[Dekuman]

Prime 1 is positioning to reintroduce a new audience to Prime 4. I strongly believe Prime 4 is coming in the next 12 months. and of those 12 months, holiday 2023 makes the most sense.

 

[LiC]

Launching late 24 makes sense given their latest declarations.
But next year is going to be very long for Switch...

What declarations? The boilerplate PR-speak answers from the investor Q&A that amount to "next year, we will release games and try to get more people to buy a Switch?" You can't get any information about unannounced plans from earnings releases, period.

 

[NintenDuvo]

What declarations? The boilerplate PR-speak answers from the investor Q&A that amount to "next year, we will release games and try to get more people to buy a Switch?" You can't get any information about unannounced plans from earnings releases, period.

Thats correct if there is going to be any talk of the future its going to be at the Fiscal briefing that should be in May

 

[carbonaxiom]

It's unwise to use prior precedents as the basis for any predictions about Metroid Prime 4 in light of the shadow drop of Metroid Prime Remastered, which was a completely unprecedented move. I do still think Metroid Prime Remastered was released now because some specific milestone in the development of Metroid Prime 4 was reached, but what that milestone is and how long before it actually releases is anyone's guess.

If Nintendo is so willing to shadow drop the first game, then they could easily do the same for remasters of Prime 2 and 3 on the way to Prime 4's release, and they could all be used in the latter's advertising if so. This implies that a Prime 3 Remastered could potentially release only a few months before Prime 4, which sounds like a bizarre thing to do, especially if it also costs $40 alongside Prime 4's $60 or even $70.

But Metroid has always been a franchise whose reputation far surpasses its sales. It's possible that Nintendo might be willing to take the risky strategy of releasing so many games in quick succession in return for providing a much needed introduction to the franchise for the Switch's enormous user base. Nintendo's newly created Metroid portal website, which lists all Metroid games currently available on the Switch, points to this possibility. For example, Metroid Fusion, despite having already been announced for Nintendo Switch Online, has yet to be added to this website, presumably because they're waiting for its release to do so. Remasters of Prime 2 and 3 might be in a similar wait, just without announcements.

Of course, in this scenario, there would need to be some way to clearly differentiate Metroid Prime 4 from the previous games in the series. What better way to do that than as a visually impressive release title for the new Switch.

 

[Salah Kalajo]

DLSS isn't the panacea people make it out to be, especially if the ps5 game is already rendering at a resolution like 1080p

please correct me f i am wrong, i know that GPU power is not the only difference between PS5 and the Switch Successor , but if the subject is rendering Capabilities , my understanding is even if PS5 Render at 1080P the Switch Successor can still render at 720P and Upscale to 1080P (or More) using DLSS . i am not saying the Switch 2 games will look better than PS5 , but i think this is still an Advantage to the Switch 2 to close the Gap with PS5/ XBOX Series ...

 

[Salah Kalajo]

To add to this, if a game does take full advantage of the PS5, then any somewhat significant gap would be enough of a reason for the developer to skip the Switch 2.

i think any game that take full advantage of the PS5 will be an Exclusive Game anyway , the game that can't run On Series S duo to technical limitations will be even harder to be released on Switch 2.

 

[ShadowFox08]

My naive simple ballparking has 8 A78's needing 3-3.5 watts for 2 ghz on an N5 family node. Per ARM's own claims; an A78 needs 1 watt to hit 3 ghz on a '5 nm node' (assume TSMC). Going down from 3 ghz to 2 ghz should be in the ballpark of (2/3)^2, or 4/9th of a watt per core. Multiply by 8 to get ~3 5/9. You can cheat a bit and assume the OS to not necessarily run at that clock to get you down to below 3.5. And if it were a refined variant instead of base N5, you could probably get down closer to 3 watts.
The catch here is that I'm not sure on how high of a power budget we're getting for CPU. OG Switch was like 2 watts, or close to it. But IIRC, that was also the long term max after thermal throttling? Not necessarily a coincidence.

Regarding Steam Deck's CPU power draw: oh, definitely more than a couple watts. I distinctly recall seeing in some Steam Deck videos the CPU power draw hitting 10 watts at times.
I also have the feeling that a mobile Zen 2 CCX has a floor of somewhere in the ~1.25 watt range when not idle.
From here:

~2.5-3 watts is as low as it gets for the 4800H here. The 4800H is 8 cores/2 CCX's, so halving that is in the ~1.25-1.5 watt range.

... waaaait a second, I've done handwavey napkin math crap about this before, here.
...did I end up being right about a high of ~10 watts?

Well at least we can say only 7 cores will be used max for gaming, while one core could be 1 Ghz or less for OS tasks.

I hesitate to mention this, because someone is probably going to doom based on it, but one of the few portability disadvantages the Switch actually has is that some broken multithreaded code can work fine anyway on x86 because it has a memory model that's unusually lenient in terms of ordering memory accesses. I seriously doubt this is going to be (or has been) a significant factor in it a game will be ported or not (especially because all the consoles were PowerPC 10 years ago), but it is a thing. I think in theory, this lets ARM go faster in cases where order doesn't matter, as well.

The original Metroid Prime Trilogy has a highly self-contained story which is unlikely to connect to Prime 4.

Sylux

 

[fwd-bwd]

What declarations? The boilerplate PR-speak answers from the investor Q&A that amount to "next year, we will release games and try to get more people to buy a Switch?" You can't get any information about unannounced plans from earnings releases, period.

Thats correct if there is going to be any talk of the future its going to be at the Fiscal briefing that should be in May

I'm still trying to wrap my head around this bizarro earnings call. As being pointed out earlier, from fiscal Q1 to Q3, 46% of Nintendo's gross profit came from their hardware business. And yet Furukawa, President of a Nikkei 225 corporation, publicly admitted that the management has no handle on the outlook of their hardware business (half of their profit!), declaring the situation "uncharted territory" twice on the call.

Their solution to the declining hardware sales is to [checks notes] maintain engagements and develop more titles? Aren't those what they are supposed to do irregardless?

There is no way that large institutional investors, such as the Master Trust Bank of Japan (17% stake) or Public Investment Fund of Saudi Arabia (6%), were so sanguine about it and not asking questions behind the scenes. It strains credulity that Nintendo is just gonna let Switch adrift in its seventh year without announcing any new hardware beyond a few Special Editions. Unless the circumstances are severely adverse—say, launch titles in development hell—revealing Switch 2 this year seems more sensible to me.

 

[Pokemaniac]

Sylux

I'm not going to say anything more on the subject, because this is veering way off topic for this thread, but he never actually appears in person in the original Prime Trilogy, just his ship in that one scene explicitly framed as a sequel hook. He's from Prime Hunters and makes a cameo in Federation Force, but neither of those are games you should play for their story. He's functionally a blank slate.

 

[Paul_Subsonic]

please correct me f i am wrong, i know that GPU power is not the only difference between PS5 and the Switch Successor , but if the subject is rendering Capabilities , my understanding is even if PS5 Render at 1080P the Switch Successor can still render at 720P and Upscale to 1080P (or More) using DLSS . i am not saying the Switch 2 games will look better than PS5 , but i think this is still an Advantage to the Switch 2 to close the Gap with PS5/ XBOX Series ...

1)
DLSS upscale is not as good as native resolution. So 720p upscaled to 1080p Switch will look worse than 1080p PS5.
2) The PS5 can also do upscaling; just not with DLSS, which requires specialized hardware. The PS5 can use FSR, the AMD equivalent of DLSS, which granted doesn't look as good as DLSS; but there is a possibility that many future PS5 titles will render at less than 1080p then upscale with FSR.

 

[Hermii]

1)
DLSS upscale is not as good as native resolution. So 720p upscaled to 1080p Switch will look worse than 1080p PS5.
2) The PS5 can also do upscaling; just not with DLSS, which requires specialized hardware. The PS5 can use FSR, the AMD equivalent of DLSS, which granted doesn't look as good as DLSS; but there is a possibility that many PS5 titles render at less than 1080p then upscale with FSR.

THE major advantage of dlss compared to fsr2 (other than slightly better IQ), is that it's using dedicated hardware to reduce runtime.

 

[ILikeFeet]

please correct me f i am wrong, i know that GPU power is not the only difference between PS5 and the Switch Successor , but if the subject is rendering Capabilities , my understanding is even if PS5 Render at 1080P the Switch Successor can still render at 720P and Upscale to 1080P (or More) using DLSS . i am not saying the Switch 2 games will look better than PS5 , but i think this is still an Advantage to the Switch 2 to close the Gap with PS5/ XBOX Series ...

This assumes a whole lot about the rendering load that's not the resolution. And that makes a lot of dlss talk not have much use without establishing what we are rendering first

 

[Paul_Subsonic]

THE major advantage of dlss compared to fsr2 (other than slightly better IQ), is that it's using dedicated hardware to reduce runtime.

That's indeed a very big advantage for a low power device like a Switch, but for something like a PS5 the cost of FSR is minor.
In the context of the discussion this means that comparing DLSS Switch with Native PS5/XS isn't really fair, because PS5/XS will use FSR.

 

[SNP]

When goalposts constantly move, they eventually end up in the right place.

But yeah, finally he's probably going to be right any something hardware released. 2024 is most likely the year, and most likely always was.

This is everything wrong with "rumors" spreading, don't want to hear from an actual source and just read some headline, then blame the original source when they say nothing as such... Or sometimes, just bad at reading comprehension.

 

[Concernt]

When goalposts constantly move, they eventually end up in the right place.

But yeah, finally he's probably going to be right any something hardware released. 2024 is most likely the year, and most likely always was.

Nah.

 

[YolkFolk]

If Prime 4 comes out before Switch 2 then it’ll be a long time before Switch 2 has a 3D Zelda or 3D Metroid game launching on it.

 

[Nintendojitsu]

To be fair, dropping Metroid Fusion and Metroid Prime in the exact same year didn't exactly make sense either when they easily could've released a Metroid game once every year like clockwork back in the day with the releases it was having.

Dropping Prime Remastered in the same year, at the very beginning of it, is perfect for Prime 4 releasing later in the year. The general public, gamers who may have never experienced Prime before, are trying Prime for the first time through Remastered. To them it is like a whole new game, it's polished, it's stunning to look at, it teaches them the fundamentals of Prime and gets them invested in this part of the franchise after most likely playing Dread on Switch.

They will finish it within the week of release and by then, if they have become fans, they will be itching for more of it. Now that the general public knows what Prime is all about and how good it is, what do you do? You announce Prime 4 later on for a Holiday 2023 release now that you have people's attention firmly on Prime and that it has gone successfully. Prime 4 riding on the success and word of mouth of Prime Remastered will only benefit its sales potential to a wider audience outside of the hardcore Metroid fanbase.

I'm just entirely against the 2024 idea, both with this game releasing as well as Switch 2's release, it would be a mistake and the logical choice currently given we know nothing about the second half of 2023, is for the console to release for Christmas. There is no better time to maximise sales potential than a Switch 2 under the tree of a metric tonne of homes after people hear about it and what it will be capable of.

Have a new 3D Mario, perhaps that DK game that's been rumoured in the works for a while now, have a game like Prime 4 as well in the launch which will absolutely be a graphical showpiece (this was in development for Switch but honestly given what we know and how far back Drake development has been, who is to say that their goalpost never shifted and they pushed their technical aspirations further by having a Switch version and a Drake version with all of the bells and whistles, Switch version given what Remastered looks like will most likely be great and look fine but the Drake version? I imagine a game like Prime 4 to work with Ray Tracing in some capacity and to lock 4K, 60 FPS behind the Drake version would be more enticing as a selling point.)

I just don't think in my opinion that the first half of 2024 makes any sense, it made sense for May this year because of TOTK but outside of that factor I didn't think it would make sense at all to be released that far from a bigger shopping period than Christmas. Now that May is out of the question, it falls on Holiday 2023 to be when this is coming otherwise we'll be waiting until the end of next year for another opportunity at a release as I don't think there's any real way they would simply drop it in the first quarter of 2024 unless they seriously leave the second half of this year hung and dry and push games out to be in 2024, which we certainly do not know they will do yet and that is to be determined by how future directs this year go.

I'm all in on this year and unless some leak comes out that completely blows it apart, the next direct is going to reveal everything and determine whether or not it's coming Holiday 2023, same with Prime 4. And I'm adamant they will do Prime 4's reveal exactly how they did Dread's and release it in a similar window too.

If Holiday 2023 is in the cards for the next Switch, I wouldn't expect concrete evidence of it until after July.

 

[MisterSpo]

If Prime 4 comes out before Switch 2 then it’ll be a long time before Switch 2 has a 3D Zelda or 3D Metroid game launching on it.

Yeah, brand new games in those series could be a ways off; but I suspect Nintendo will want - and will perhaps be putting things in place - to make sure that the next games don't spend 6 years plus in development.

The obvious options first, to get each brand on the successor, will be further remasters; Prime 2 and 3 are options for 3D Metroid and it wouldn't surprise me if work has already begun. Zelda has a bunch of options: Ocarina, Majora, Wind Waker, and Twilight Princess as 3D remasters; Adventure of Link, the Oracle games, and Minish Cap feel like the likeliest 2D remakes to me.

Both series also have the option of new '2D' entries: Metroid 6 in 2025 is a distinct possibility and, regardless of whether it makes 2025 or not, I think Metroid 6 is extremely likely in the works. Link's Awakening (2019) is actually one of the strongest performing Zelda titles with over 6 million sold; a new game in a similar style would be a good way of getting Zelda out on successor hardware without relying solely on spin-offs and remasters, and if it can sell similarly to Link's Awakening, it's a decent sized hit for Nintendo if they can get it out sometime in that 2024 to 2026 period when the next open world Zelda is extremely unlikely to be ready.

(Edit - just to be clear, I get what you're saying and why you specify 3D; but from Nintendo's perspective, getting new titles out in those franchises through the 2D style is a sensible and completely doable option for them - it gives those brands a presence that's a bit more exciting and noteworthy than re-releases; though I suspect a bells and whistles Ocarina remake would be a huge hit in the next few years on the Zelda front)

 

[Concernt]

If Prime 4 comes out before Switch 2 then it’ll be a long time before Switch 2 has a 3D Zelda or 3D Metroid game launching on it.

Yes and no. It might not have exclusives for quite some time after launch but those games launching briefly before Switch 2 probably just means they'll get Switch 2 enhancements on launch day.

See: every Xbox published title 2019-2020.

I think the dooming about a lack of meaningful Switch 2 excluives is misinformed. Switch has a 10 year lifespan, at least, and it's exceeded unit sales expectations so might have even longer. Much of that back half will be games launching on Switch 2 and Switch 1, like Xbox games in 2021, with the only exclusives being games that simply can't run on Switch 1, which I think is going to very very few until 2028-ish.

 

[Deleted member 887]

I am one of those people playing Metroid Prime for the first time. Can we maybe keep complete plot breakdowns of the Metroid franchise out of the hardware thread, for my sake?

 

[Goodtwin]

Plenty of the big hitters that are popular with the mass market are out there. Mario Kart 9 will be ready to go by then. Splatoon 3 and Xenoblade 3 are getting DLC this year, but those teams have likely started working on their sequels shortly after those titles released. Fire Emblem Engage development had completed quite some time ago and the game was held for strategic reasons, meaning Intelligent Systems will have been working on their next game. Seeing some big hitters release late in Switch's life shouldn't worry us, Nintendo has been releasing titles strategically for years now and will have a strong lineup ready for whenever Switch 2 comes out.

Switch hasn't been this successful because of any specific title. The hybrid concept itself remains an attractive proposition to consumers. Playing AAA games on the go remains attractive. Nintendo would release with no 3D Mario or Zelda at launch and its not going to flop. There needs to be compelling software certainly, but is there any reason to believe they couldn't release with Mario Kart 9 and Luigis Mansion 4 along with some quality third part ports? The merits of the hardware itself stand up just fine on its own.

I am one of those people playing Metroid Prime for the first time. Can we maybe keep complete plot breakdowns of the Metroid franchise out of the hardware thread, for my sake?

Certainly, but you will have to go digging for that plot if you want one. Without digging into the lore, you will find the plot to be about as complicated as your typical Mario game. Prime is one of my favorite games ever, but its not because of its story.

 

[YolkFolk]

Yes and no. It might not have exclusives for quite some time after launch but those games launching briefly before Switch 2 probably just means they'll get Switch 2 enhancements on launch day.

See: every Xbox published title 2019-2020.

I think the dooming about a lack of meaningful Switch 2 excluives is misinformed. Switch has a 10 year lifespan, at least, and it's exceeded unit sales expectations so might have even longer. Much of that back half will be games launching on Switch 2 and Switch 1, like Xbox games in 2021, with the only exclusives being games that simply can't run on Switch 1, which I think is going to very very few until 2028-ish.

We just don’t know have much cross gen software there will be. For sure from the indies, but it’ll be next to nothing for third parties. With Nintendo, I wouldn’t even be surprised if 3D Mario was a day 1 Switch 2 exclusive.

 

[LiC]

I'm still trying to wrap my head around this bizarro earnings call. As being pointed out earlier, from fiscal Q1 to Q3, 46% of Nintendo's gross profit came from their hardware business. And yet Furukawa, President of a Nikkei 225 corporation, publicly admitted that the management has no handle on the outlook of their hardware business (half of their profit!), declaring the situation "uncharted territory" twice on the call.

Their solution to the declining hardware sales is to [checks notes] maintain engagements and develop more titles? Aren't those what they are supposed to do irregardless?

There is no way that large institutional investors, such as the Master Trust Bank of Japan (17% stake) or Public Investment Fund of Saudi Arabia (6%), were so sanguine about it and not asking questions behind the scenes. It strains credulity that Nintendo is just gonna let Switch adrift in its seventh year without announcing any new hardware beyond a few Special Editions. Unless the circumstances are severely adverse—say, launch titles in development hell—revealing Switch 2 this year seems more sensible to me.

It's the same answer for their business strategy of every past year, copy-pasted into this year. It's not actually meant to be a solution to declining hardware sales, it's just restating a boilerplate mission statement, as many Q&A answers are, for the Switch platform as a whole.

To me it doesn't seem strange, it's just the natural conclusion of six years where Nintendo maintained the same strategy for every year, at the point where momentum is starting to decline, but before they're ready to announce their next system, since they still have major game releases (TotK) which they want to focus on for the current hardware. Of course they don't have "one weird trick" to revitalize sales, and the obvious answer is new hardware, but that's still unannounced plans which Nintendo won't talk about before they're ready, no matter how much investors raise eyebrows at them.

As someone who already thought hardware was coming this year, "we'll keep releasing games and conveying the appeal of the hardware" sounds like exactly what Nintendo would do within the year for the runup to new hardware, because a continuous release schedule and no lulls/droughts/coasts is an integral part of their strategy. But really it's just the strategy for every year, and it's what they'll say again next year if hardware isn't launching until then.

 

[HankyPanky]

just thinking aloud but if no Switch 2 this year I believe the rumoured new 2D Mario could easily serve as the 'big' holiday title. makes sense we'll see something Mario related this year and it's obvious they'll save the next 3D title for the new system. likewise I can't see any merit of Prime 4 releasing this year if it's only coming to the OG Switch. surely MP4 has to be cross-gen at this stage.

 

[MisterSpo]

Personally I'm expecting the biggest system sellers from EPD - stuff like 3D Mario, Splatoon, Animal Crossing, Mario Kart, new Zelda (whenever it arrives) - to be exclusive, especially because it gives EPD room to play around with the hardware. Smash I would also expect to be exclusive to new hardware, but the development arrangements for that aren't really clear yet (is it another Namco gig, or is it brought to EPD?). At first, big sellers like Pokemon and Mario Party could be cross gen; Pokemon especially depends on what Game Freak want to do, and Party might not necessarily need new hardware. 2D Mario is the one major EPD seller I could see coming to Switch later this year; unless new hardware arrives. I'd also expect some odds and ends from EPD4 to make it to Switch.

As for everybody else - I'd guess Next Level, Monolith Soft, Intelligent Systems, HAL, and (assuming Prime 4 is wrapping up) Retro are all working on stuff that's either cross gen or on new hardware. Mercury Steam and EPD7 are presumably also targeting new hardware with the next Metroid, though perhaps early development began on Switch.

 

[Foltzie]

I'm still trying to wrap my head around this bizarro earnings call. As being pointed out earlier, from fiscal Q1 to Q3, 46% of Nintendo's gross profit came from their hardware business. And yet Furukawa, President of a Nikkei 225 corporation, publicly admitted that the management has no handle on the outlook of their hardware business (half of their profit!), declaring the situation "uncharted territory" twice on the call.

Their solution to the declining hardware sales is to [checks notes] maintain engagements and develop more titles? Aren't those what they are supposed to do irregardless?

There is no way that large institutional investors, such as the Master Trust Bank of Japan (17% stake) or Public Investment Fund of Saudi Arabia (6%), were so sanguine about it and not asking questions behind the scenes. It strains credulity that Nintendo is just gonna let Switch adrift in its seventh year without announcing any new hardware beyond a few Special Editions. Unless the circumstances are severely adverse—say, launch titles in development hell—revealing Switch 2 this year seems more sensible to me.

46% of Nintendo's gross profit or revenue? Im having trouble finding a source attributing 46% of profit to hardware sales.

I would presume it was never Nintendo's plan to go seven years between major hardware revisions, however, we dont know all that has been happening in the background.

Some reports indicating that the Mariko chip may have been considered for additional profiles, but instead Nintendo opting to keep the experience the same for all Switch owners and take the extra battery life. I can't speak to why Parker and Xavier were skipped (though the CPU in both seems like a bad fit for Switch), so Orin (Nano) based it is. . . Perhaps Orin is delayed a bit due to Covid, who knows.

In either event, Nintendo can't ship hardware that isn't ready and they might opt to hold back hardware that is ready in order to make sure there is software available. Nintendo simply no longer has two major units moving at one time so their hardware cycle will be more cyclical than before. Their major shareholders would be aware of this, and should allow some understanding. Additionally, Nintendo offers a dividend, so long as that isn't cut major shareholders are likely to remain relatively passive in their interaction with Nintendo.