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

[bloopland33]

 

[chocolate_supra]

I've made this mistake.
Fyi ezgif deletes the stuff you make after a few minutes. You have to save the gif you made and post it to a place like imgur.

 

[Dark Cloud]

This is the big one from Nate. He got the info he needed for actually NEW updated information

 

[ILikeFeet]

As someone who knows shit about hardware, i just hope whatever this is will allow nintendo to be liberal with antialiasing lmao

Watch NERD get a patent about upscaling without anti aliasing

 

[Sol]

This is the big one from Nate. He got the info he needed for actually NEW updated information

Hopefully. But it seems like he could just talk about the details of the NVN2 leak with MVG (ie, the 12 SM Ampere GPU) without discussing any new target dates or updated production/release windows.

Either way I'm excited for more talk about this device. This time last year we were getting blown up with a new Bloomberg article every few weeks and the hype was through the roof. A year later here we are and it just seems so quiet.

 

[karmitt]

True, but he also said he didn't want to piecemeal his info. I think this will be discussing what he could source about the system and he finds relevant to discuss, and he won't be doing a definitive edition soon after.

Could be wrong, of course. We'll see in a week.

That’s what I thought as well. But I don’t want to have too high expectations here - I’m already happy enough that last episode he confirmed that nothing’s changed with his thoughts on the release window, and that MVG wasn’t disagreeing.

 

[bloopland33]

I've made this mistake.
Fyi ezgif deletes the stuff you make after a few minutes. You have to save the gif you made and post it to a place like imgur.

Good to know! I was a little amazed to think they were hosting it like that before I'd even downloaded it...welp, there goes that. It was a gif of Leia saying the classic line "Help us, Nate, you're our only hope."

 

[chocolate_supra]

This time last year we were getting blown up with a new Bloomberg article every few weeks and the hype was through the roof. A year later here we are and it just seems so quiet.

The calm before the storm baybee

Good to know! I was a little amazed to think they were hosting it like that before I'd even downloaded it...welp, there goes that. It was a gif of Leia saying the classic line "Help us, Nate, you're our only hope."

Your effort is recognized and appreciated.

 

[mariodk18]

I miss Mochizuki

 

[bloopland33]

Your effort is recognized and appreciated.

Hahaha thank you

 

[Alovon11]

What does it all mean!?

One moment Nate says there are exclusives ready for Fall 2022/March 2023
Then he says "Not until 2024 at the latest"

Then he likes my reply saying that Fall 2022/March 2023 is the most likely release window.

 

[bloopland33]

Then he says "Not until 2024 at the latest"

Already forgotten the exact quote and am on mobile but the quote was the next gen successor won’t come until at least 2024, right?

If Drake is positioned as a revision, then he’s not contradicting himself. I took that line to mean “yeah, you won’t see a next gen successor before 2024 - and likewise you won’t see a revision(Drake) in 2024+.” So hypothetically Drake comes out FY2023 as we all hope and then his comment the next gen hardware window doesn’t even begin until 2024+.

I don’t think he actually meant that he expects next gen in 2024 or even 2025. Just that, after Drake and once we enter 2024, what lies ahead will be next gen hardware.

I don’t think that came out as clearly in text as it was in my head, I hope you follow what I’m trying to get at

 

[LiC]

Already forgotten the exact quote and am on mobile but the quote was the next gen successor won’t come until at least 2024, right?

If Drake is positioned as a revision, then he’s not contradicting himself. I took that line to mean “yeah, you won’t see a next gen successor before 2024 - and likewise you won’t see a revision(Drake) in 2024+.” So hypothetically Drake comes out FY2023 as we all hope and then his comment the next gen hardware window doesn’t even begin until 2024+.

I don’t think he actually meant that he expects next gen in 2024 or even 2025. Just that, after Drake and once we enter 2024, what lies ahead will be next gen hardware.

I don’t think that came out as clearly in text as it was in my head, I hope you follow what I’m trying to get at

Makes sense to me!

 

[brainchild]

I think people are too focused on Nanite. it's amazing, but if you don't have the workforce to support such high resolution models, then I don't think you'll be making meshes that would be so crippling to current hardware. besides, there's a fallback for unsupported devices like mobile

I'd be more worried about Lumen. if more and more devs jump on the lumen train, then scaling down will be more work.

for example, here's what the Matrix city looks like at night when lumen is off

With all due respect, I completely disagree with this take.

First, you do not need a big team to create models with high poly counts and high resolutions. What you need are comprehensive resources and access to powerful procedural, generative, and/or machine learning tools. Houdini, for instance, allows you to create cities on the scale of that Matrix demo, even if you're only one person. I should know as I have already done it. If you use open street map data, your turnaround time will be even faster.

In fact, the Matrix demo artists used Houdini to generate the city in the demo:



And the presenter above explicitly mentions its benefits for smaller teams:

There are even tools that convert 2D photos into high-quality 3D models, which I also use. Converting such assets to Nanite is trivial and allows indies to create game worlds on a scale comparable to AAA productions.

Also, the Lumen comparison is flawed since the lighting was not replaced with any alternative solutions. Simply turning Lumen off as if any game would actually ship like that is not a fair comparison. RTXGI with sufficient probe sampling will produce better results (mainly due to reduced visual latency) and yet yield much better performance. Lumen is simply not production-ready and other solutions have already been proven to work just fine. The problem is that there aren't that many examples of Nanite assets being illuminated by GI solutions other than Lumen. I can assure you, however, RTXGI looks great on Nanite assets. In fact, procedurally produced Nanite assets + DirectX ray-traced reflections + RTXGI = Indie cheat code for scalable, photorealistic AAA graphics.

EDIT:

Also, NVIDIA Omniverse is great for indies. I'm excited to use the new UE5 connector, but I want to wait until I can migrate my UE4 project first.

 

[ReddDreadtheLead]

@nate

@NateDrake can you do me 1 favor, a single request, and this is for your next podcast.

Can you ask MVG to clarify what he means by “100% is not possible for BC unless they have the hardware inside it”?

And by that I mean, is he saying that unless they pull a “Wii inside the Wii U” scenario, then it cannot have BC in any fashion?

Or, is he saying “100% (as in every single title) will not run” and expecting absolutely every single title to run out the box is unrealistic, but that a situation like the PS5 (which has like 99% compatibility with the previous titles and needed a patch to run on the PS5) is a more realistic stance to have with respect to expectation?


The only thing, possibly ad verbatim.

Or, “are you saying 100% BC like the Series consoles is not possible, or are you saying a PS5 scenario who isn’t 100% compatible, but 99% compatible is what’s possible, but the 100% isn’t and need a patch for that 1%?”

The sole request.

@NateDrake in case you forgot

(Or if it’s even possible to slot in)

 

[Alovon11]

@nate

@NateDrake in case you forgot

(Or if it’s even possible to slot in)

Also on that note @NateDrake , maybe clarify what you meant by the 2024 statement?

Based on everything we can tell Drake/T239/The FY23 system is a generational leap hardware wise over OG Switch, so it would be very weird/impossible to get even more performance in 2024/2025 for a "True Next Gen"

 

[brainchild]

I don't see why Lumen can't run on drake when all the necessary precautions are taken. The Matrix is intended to be a heavy hitting demo from the start, so it's not the best barometer.

And Nvidia already has a scalable RT GI tool.

Clearly, you already know about a sufficient alternative so I'm not sure why you'd be worried about devs jumping on the Lumen train. If they don't make their games scalable enough, that's on them.

 

[mariodk18]

Haven’t we already determined that Drake is overkill for a simple “revision”? How much of a leap would a 2025 device provide anyway over Drake?

 

[Alovon11]

Clearly, you already know about a sufficient alternative so I'm not sure why you'd be worried about devs jumping on the Lumen train. If they don't make their games scalable enough, that's on them.

Yeah, and considering RTXGI is a plugin in UE4/5, it shouldn't be too hard to swap Lumen out for it considering they likely work off the same data right?

I'm personally curious as to how the City demo would run with RTXGI and DLSS on a NVIDIA card versus Lumen and UE-TSR

 

[brainchild]

Yeah, and considering RTXGI is a plugin in UE4/5, it shouldn't be too hard to swap Lumen out for it considering they likely work off the same data right?

Different approaches to achieve the same effect. Lumen uses screen tracing and signed distanced fields, RTXGI uses probe grids that have a hard limit on grid size due to the use of irradiance textures (the GPU limit on texture resolutions is the problem here), so the placement of the lighting captures for either solution is going to be different, unfortunately.

I'm personally curious as to how the City demo would run with RTXGI and DLSS on a NVIDIA card versus Lumen and UE-TSR

RTX 3080 and up shouldn't even break a sweat if Lumen was swapped with RTXGI and DLSS was used.

 

[LiC]

Haven’t we already determined that Drake is overkill for a simple “revision”? How much of a leap would a 2025 device provide anyway over Drake?

It's not really right to say we've determined anything about Drake when we have only partial specs, which could be subject to change, and extrapolations from them. Taking the things speculated about in this thread as a foregone conclusion is asking for trouble.

This isn't me arguing that it's a "simple revision" or underpowered or whatever. We know the new system exists. The features the system has and the games it enables are what's important, as well as when we can expect to see it. And that's what we'll hopefully get more info on before too long.

 

[ReddDreadtheLead]

How much of a leap would a 2025 device provide anyway over Drake?

Probably the difference between PS4 and XB1X or PS4 Pro. Yes, I didn’t use the XB1 here as the bump would be bigger (4x increase vs 2-3x increase)

Assuming Drake releases this year or next year (early or late 2023) of course.

There’s not exactly a lot that can make a generational leap in difference that makes one a Pro and another a 2/sequel system in terms of paper performance or that it matters much graphically on making a noticeable difference like 360 titles vs PS4 titles. Maybe the “pro” levels of difference.

I think if you asked about the difference between a 2019-2020 device to a late 26-late 27 device then you’d get a really nice bump up to offer a significant enough difference.

But a bump like switch to Switch 2 is quite a big one and probably won’t be replicated again for a quite some time or if the hardware generations are shorter. Longer generations give sufficient time of difference between a console and its successor, and can give enough time for new technology to arrive. But there comes a point where they would have to ask if it’s feasible to continue to stretch the generation even longer than the previous just to make enough of a difference.


Notice though, I don’t use definitives.


That said, paper performance increases are one thing, its feature set that would start to matter at that point a lot more for the consoles. So while the XBox Series Y released in 2027 could be only 3x the paper TFLOPs than the Series X (so 36TFLOPs) for example, it matter more on what RDNA 5 or 6 have feature wise that improve the performance beyond what paper numbers give you. RDNA5 or 6 in the Series X could probably do Blast Processing that the RDNA2 in the consoles couldn’t. Or can handle so much more ray tracing that it makes it look real. Or hell, can maybe walk your talk. I’m being facetious here at the end to show a point, that feature wise can put it much farther than simply having more paper specs.


And smart engineering.

 

[12cheeper]

Already forgotten the exact quote and am on mobile but the quote was the next gen successor won’t come until at least 2024, right?

If Drake is positioned as a revision, then he’s not contradicting himself. I took that line to mean “yeah, you won’t see a next gen successor before 2024 - and likewise you won’t see a revision(Drake) in 2024+.” So hypothetically Drake comes out FY2023 as we all hope and then his comment the next gen hardware window doesn’t even begin until 2024+.

I don’t think he actually meant that he expects next gen in 2024 or even 2025. Just that, after Drake and once we enter 2024, what lies ahead will be next gen hardware.

I don’t think that came out as clearly in text as it was in my head, I hope you follow what I’m trying to get at

here's what gets me about this whole thing:if we assume that drake is indeed a revision and even if it wasn't based on nvidias latest tegra chip and they had anything newer to actually work with by 2024 mvg has largely dismissed the idea of switch games being playable on a device that uses different architecture.....which drake does.so assuming that mvg is right, is the argument that this is an xbox one x/ps4 pro style revision....that doesn"t play any existing switch games?

this whole thing just feels like a really weird hill for nate to choose to die on.like,none of this makes a lick of sense.

 

[LiC]

here's what gets me about this whole thing:if we assume that drake is indeed a revision and even if it wasn't based on nvidias latest tegra chip and they had anything newer to actually work with by 2024 mvg has largely dismissed the idea of switch games being playable on a device that uses different architecture.....which drake does.so assuming that mvg is right, is the argument that this is an xbox one x/ps4 pro style revision....that doesn"t play any existing switch games?

this whole thing just feels like a really weird hill for nate to choose to die on.like,none of this makes a lick of sense.

"Positioned as a revision" doesn't really mean anything with regards to the hardware internals. It can use the latest Tegra and still be treated as a revision by Nintendo if they think that's the best way to market it. It does definitely imply it will have backward compatibility, though.

All this revision/2024 stuff is a mixture of speculation, getting hung up on definitions around hardware vs. marketing, and general wheel spinning from the lack of new info. Again, I hope we hear updated info about the device that we already know exists, and can set aside this very circular discussion.

 

[Kenka]

Lots of good inputs Brainchild, thank you!
These lines in particular:

In fact, procedurally produced Nanite assets + DirectX ray-traced reflections + RTXGI = Indie cheat code for scalable, photorealistic AAA graphics.

EDIT:

Also, NVIDIA Omniverse is great for indies.

It basically encapsulates what we want Drake to be capable of. Plus fast data streaming (SSDs or similar).

 

[12cheeper]

"Positioned as a revision" doesn't really mean anything with regards to the hardware internals. It can use the latest Tegra and still be treated as a revision by Nintendo if they think that's the best way to market it. It does definitely imply it will have backward compatibility, though.

All this revision/2024 stuff is a mixture of speculation, getting hung up on definitions around hardware vs. marketing, and general wheel spinning from the lack of new info. Again, I hope we hear updated info about the device that we already know exists, and can set aside this very circular discussion.

.....my point is that how do you market this thing as an xbox one x style revision that runs existing games better(as nate has suggested) if according to mvgs past comments it flat out couldn't play existing switch games on it anyway?

 

[brainchild]

@Alovon11

After thinking on it a bit more, I probably seem very biased in favor of RTXGI, so there are a few things I'd like to share about Lumen that I think provide a better experience for developers working on their projects:

Lumen pros over RTXGI:

• More inclusive in terms of light sources and lighting contributions (real skydomes can be captured and not just HDRI maps that represent skylight)
• More simplified workflow (there is less fiddling to get the lighting to look accurate and most of the default settings are sufficient)

If a dev doesn't want to have to fuss about with lots of parameters and settings in order for the global illumination solution to look as desired, then Lumen is going to be the better option for that dev.

That being said, personally, I will sacrifice ease of workflow when the performance difference is so drastic. Yes, you will have to make sure your DDGI volumes are "just right" in your scenes with RTXGI, but the time it takes to do that is not that big of a deal to me and I think it's totally worth the trade-off in productivity.

Also, a lot of the drawbacks with RTXGI in Unreal are due to Unreal issues and not inherently the technology itself, so those drawbacks will be mitigated as the plugin compatibility improves over time.

 

[chocolate_supra]

.....my point is that how do you market this thing as an xbox one x style revision that runs existing games better(as nate has suggested) if according to mvgs past comments it flat out couldn't play existing switch games on it anyway?

Pretty sure MVG's comments were not that it flat-out couldn't play existing Switch games, it was that it wouldn't play Switch games very well natively without some sort of emulation layer running for the GPU. Which, IIRC, is what they did with Mario Galaxy in 3D All-Stars.

 

[Alovon11]

@Alovon11

After thinking on it a bit more, I probably seem very biased in favor of RTXGI, so there are a few things I'd like to share about Lumen that I think provide a better experience for developers working on their projects:

Lumen pros over RTXGI:

• More inclusive in terms of light sources and lighting contributions (real skydomes can be captured and not just HDRI maps that represent skylight)
• More simplified workflow (there is less fiddling to get the lighting to look accurate and most of the default settings are sufficient)

If a dev doesn't want to have to fuss about with lots of parameters and settings in order for the global illumination solution to look as desired, then Lumen is going to be the better option for that dev.

That being said, personally, I will sacrifice ease of workflow when the performance difference is so drastic. Yes, you will have to make sure your DDGI volumes are "just right" in your scenes with RTXGI, but the time it takes to do that is not that big of a deal to me and I think it's totally worth the trade-off in productivity.

Also, a lot of the drawbacks with RTXGI in Unreal are due to Unreal issues and not inherently the technology itself, so those drawbacks will be mitigated as the plugin compatibility improves over time.

Not to mention that RTXGI being an open resource that can be implemented into any engine theoretically means that it has far more flexibility than Lumen beyond what it has in unreal alone.

There is a reason I feel Nitnendo may just slot in RTXGI into their Drake-Enhanced versions of games as they also could get direct assistance from NVIDIA to make it run even better in NVN2.etc

 

[Crusters]

here's what gets me about this whole thing:if we assume that drake is indeed a revision and even if it wasn't based on nvidias latest tegra chip and they had anything newer to actually work with by 2024 mvg has largely dismissed the idea of switch games being playable on a device that uses different architecture.....which drake does.so assuming that mvg is right, is the argument that this is an xbox one x/ps4 pro style revision....that doesn"t play any existing switch games?

this whole thing just feels like a really weird hill for nate to choose to die on.like,none of this makes a lick of sense.

I'm not sure dismissed is the right word, Nintendo didn't use the ptx for future compatability, that means the compiled Maxwell shaders aren't going to be automatically recognized, so the easy way is out, true. They need to be recompiled, there are solutions for that. Shoot, maybe they can be dynamically recompiled with the extra grunt.

 

[ShadowFox08]

Also on that note @NateDrake , maybe clarify what you meant by the 2024 statement?

Based on everything we can tell Drake/T239/The FY23 system is a generational leap hardware wise over OG Switch, so it would be very weird/impossible to get even more performance in 2024/2025 for a "True Next Gen"

Maybe he meant no next gen first party exclusives until 2024. Gets treated like a pro model until then. Better performance for the same games. Where switch 2 can play switch games via backwards compatibility is another thing all together .

Sony and MS weren't that much different with their new consoles when they launched too

 

[Adulfzen]

one thing is certain in my mind, nintendo won't do a complete change regarding the internals for a revision with drake if the power bump is as moderate as Nate was implying.

So the FY 2022 Switch is either drake with the specs that have been discussed since the leak which will be a generation leap in all but name (called Switch 4k or something).

Or it's just yet again another mariko update that doesnt have DLSS besides the clock speeds being higher (why it wouldn't be possible on the other mariko skus aside from the lite would be a mystery).

 

[ReddDreadtheLead]

Pretty sure MVG's comments were not that it flat-out couldn't play existing Switch games, it was that it wouldn't play Switch games very well natively without some sort of emulation layer running for the GPU. Which, IIRC, is what they did with Mario Galaxy in 3D All-Stars.

Considering the headroom they get from Drake from the looks of it, they could brute force this if they wanted to.

 

[Hermii]

I think Nate and MVG are mixing in their own theories (worth about as much as a random forum member) and believes with what they actually heard from sources, And it’s not always clear what’s what.

I think the stuff about “ampere can’t do bc”, exactly how it will be positioned, and definitely about a “real next gen” platform in 2024+ goes under the former category. It’s not gospel.

To me, it’s extremely illogical that Nintendo will release the best hardware Nvidia can offer them in 2022/2023 while simultaneously working on a “real next gen” platform for a couple of years after that. And don’t get me started about BC.

 

[12cheeper]

I'm not sure dismissed is the right word, Nintendo didn't use the ptx for future compatability, that means the compiled Maxwell shaders aren't going to be automatically recognized, so the easy way is out, true. They need to be recompiled, there are solutions for that. Shoot, maybe they can be dynamically recompiled with the extra grunt.

oh,i'm not saying it's impossible,microsoft already proved there are work arounds after all.even so between recompiling at least the most popular 100 or so switch games and incorporating brand new architecture this is an awful lot of work for a device that will supposedly only be on the market for a year.

 

[Thugstas]

I think Nate and MVG are mixing in their own theories (worth about as much as a random forum member) and believes with what they actually heard from sources, And it’s not always clear what’s what.

I think the stuff about “ampere can’t do bc”, exactly how it will be positioned, and definitely about a “real next gen” platform in 2024+ goes under the former category. It’s not gospel.

To me, it’s extremely illogical that Nintendo will release the best hardware Nvidia can offer them in 2022/2023 while simultaneously working on a “real next gen” platform for a couple of years after that. And don’t get me started about BC.

We will get more leak within the next months, I guess.

 

[Crusters]

I think Nate and MVG are mixing in their own theories (worth about as much as a random forum member) and believes with what they actually heard from sources, And it’s not always clear what’s what.

I think the stuff about “ampere can’t do bc”, exactly how it will be positioned, and definitely about a “real next gen” platform in 2024+ goes under the former category. It’s not gospel.

To me, it’s extremely illogical that Nintendo will release the best hardware Nvidia can offer them in 2022/2023 while simultaneously working on a “real next gen” platform for a couple of years after that. And don’t get me started about BC.

Thats pretty gospel, although the way you are interpreting it isn't the way it is. It can do bc, it just can't do the automatic bc Nvidia has provided.... Because Nintendo decided not to use it.

PTX is forward-compatible. Meaning PTX is supported to run on any GPU with compute capability higher than the compute capability assumed for generation of that PTX. For example, PTX code generated for compute capability 7.x is supported to run on compute capability 7.x or any higher revision (major or minor), including compute capability 8.x. Therefore although it is optional, it is recommended that all applications should include PTX of the kernels to ensure forward-compatibility. To read more about cubin and PTX compatibilities see Compilation with NVCC from the Programming Guide.

This whole optional forward compatability thing, Nintendo opted out of it. They likely had good reasons, but I don't know what they are, and it's more fun to just assume they didn't. So when switch games were compiled, this wasn't part of it, so when the architecture changed, portions of what was compiled on Maxwell won't be recognizable by ampere.

That doesn't mean ampere can't do bc, it just means the native out of the box solution wasn't used.

I honestly don't think Nintendo would have did what they done did, if they didn't already have a plan for this.

 

[mariodk18]

I mean if the Steam Deck, which has a completely different architecture (arm vs x86) can do a pretty good job of emulating an entire Switch, then wouldn't it stand to reason that Drake could emulate the Switch GPU really well? Or (and I have no idea if this is possible) maybe they emulate just the necessary "precompiled shader" portion of Switch?

 

[ReddDreadtheLead]

I mean if the Steam Deck, which has a completely different architecture (arm vs x86) can do a pretty good job of emulating an entire Switch, then wouldn't it stand to reason that Drake could emulate the Switch GPU really well? Or (and I have no idea if this is possible) maybe they emulate just the necessary "precompiled shader" portion of Switch?

Let’s put it more to context so you can see something that’s crazy about what you just said, and it’s not because it’s wrong because it isn’t wrong, it’s right in a way:


The developers of the Nintendo switch emulator, the ones that do not have any official documentation of the hardware, the ones that do not have any ins and outs of NVidia architecture, are saying it is impossible for Nintendo to achieve backwards compatibility on their own hardware, the two multi billion dollar corporations where one of these two is industry leader in graphics. Let that sink in for a second….

I think that speaks more for itself , and answers part of your question.

They can find a way to achieve it, it isn’t easy but it’s never easy to do this. But the idea that freelancers believe it is completely impossible is… contentious at best.

 

[Lelouch0612]

Re: Drake positionning.

If Microsoft released the Xbox Series S as a premium revision of the Xbox One, it would have looked like this :

Holiday 2013 : Xbox One
Q3 2016 : Xbox One S, replacement SKU
Holiday 2017 : Xbox Series S, a premium revision
Holiday 2020 : Xbox Series X, Series S repositionned as the entry model

Which would translate to this for the Switch :

Q1 2017 : Switch
Q3 2019 : Switch v2, replacement SKU
October 2021 : Switch OLED
Q4 2022/Q1 2023 : Drake, premium revision, OLED becomes entry SKU.
2024/2025 : Switch 2, Drake becomes the entry SKU.

Focused on the main Switch rather than the Lite since we are talking about the next hybrid model.

So basically, it is something that could indeed work. It is probably even smarter than the current release strategy of releasing a Pro SKU that would have a very few years of support.

Under the pretense of launching a revision, you are starting the next generation early.

 

[Adulfzen]

Re: Drake positionning.

If Microsoft released the Xbox Series S as a premium revision of the Xbox One, it would have looked like this :

Holiday 2013 : Xbox One
Q3 2016 : Xbox One S, replacement SKU
Holiday 2017 : Xbox Series S, a premium revision
Holiday 2020 : Xbox Series X, Series S repositionned as the entry model

Which would translate to this for the Switch :

Q1 2017 : Switch
Q3 2019 : Switch v2, replacement SKU
October 2021 : Switch OLED
Q4 2022/Q1 2023 : Drake, premium revision, OLED becomes entry SKU.
2024/2025 : Switch 2, Drake becomes the entry SKU.

Focused on the main Switch rather than the Lite since we are talking about the next hybrid model.

So basically, it is something that could indeed work. It is probably even smarter than the current release strategy of releasing a Pro SKU that would have a very few years of support.

Under the pretense of launching a revision, you are starting the next generation early.

wouldn't it be way further than 2024/2025 for any hardware that could be called switch 2 if drake is considered the entry point in that example.

 

[Hermii]

Re: Drake positionning.

If Microsoft released the Xbox Series S as a premium revision of the Xbox One, it would have looked like this :

Holiday 2013 : Xbox One
Q3 2016 : Xbox One S, replacement SKU
Holiday 2017 : Xbox Series S, a premium revision
Holiday 2020 : Xbox Series X, Series S repositionned as the entry model

Which would translate to this for the Switch :

Q1 2017 : Switch
Q3 2019 : Switch v2, replacement SKU
October 2021 : Switch OLED
Q4 2022/Q1 2023 : Drake, premium revision, OLED becomes entry SKU.
2024/2025 : Switch 2, Drake becomes the entry SKU.

Focused on the main Switch rather than the Lite since we are talking about the next hybrid model.

So basically, it is something that could indeed work. It is probably even smarter than the current release strategy of releasing a Pro SKU that would have a very few years of support.

Under the pretense of launching a revision, you are starting the next generation early.

Under this scenario, Drake and Switch 2 has 100% library parity?

 

[Lelouch0612]

wouldn't it be way further than 2024/2025 for any hardware that could be called switch 2 if drake is considered the entry point in that example.

I would feel more like saying 2025/2026 yeah. PS4 Pro was 3 years after the Ps4 for instance.

Under this scenario, Drake and Switch 2 has 100% library parity?

Yes (at least until Switch 2 Drake). Not expecting a generational gap between the two but more something like XSS/XSX.

 

[Adulfzen]

I would feel more like saying 2025/2026 yeah. PS4 Pro was 3 years after the Ps4 for instance.

Yes (at least until Switch 2 Drake). Not expecting a generational gap between the two but more something like XSS/XSX.

tbh iirc it was already speculated drake could be refreshed later on by going from 8nm to 5nm later which could apply to the scenario you're talking about.

 

[NeonRed]

XONE - Switch
XONE S - Switch V2/Lite
XONE X - Switch Pro
Series X - Switch 2
Series S - Switch 2 Lite

Switch Pro is taking longer than ONE X because of covid. I think that we will see Pro early next year. After 2024 Switch Pro will have more and more exclusive games. Switch 2 is late 2026 or early 2027.

 

[Skittzo]

.....my point is that how do you market this thing as an xbox one x style revision that runs existing games better(as nate has suggested) if according to mvgs past comments it flat out couldn't play existing switch games on it anyway?

Because he never said that.

He and Nate were both adamant that it would absolutely have BC, it just wouldn't be a simple process of plug and play from Nintendo's perspective. And it probably would only wind up being 99% of games available via BC.

 

[karmitt]

My take on the real successor talk? When it does come it’ll have little to do with ‘power’

Perhaps we’ll see them using a die shrunk Drake in a few years, leveraging improvements in energy efficiency to deliver new functionality - maybe handheld could deliver a good enough resolution for VR at that point. Who knows.

I do think we should stop dwelling on ‘2024’ as it could have just been said without a ton of thought. If anybody really wants to know what Nate and MVG were thinking around that, just give ‘em a donation and ask

 

[ReddDreadtheLead]

Plot twist: the successor is the GameCube to Wii scenario

:0

Sell a product and after a few years shrink it, clock it higher, change and add more RAM, change the controller scheme and call it revolution baby!

 

[thegodsend]

I think the reality will be that especially Nintendo but also many third parties will want to release their games on Mariko and Drake. There's a huge active userbase on Switch that won't upgrade immediately.

What this means is that for most games, the Switch hardware will be the baseline for performance, using Drake's capabilities mainly for 4K, better framerate and such.

This is a scenario that just makes business sense for everyone - which is more important than anything else. And this is the scenario that if you want do compare to existing platforms, it makes Switch Drake similar to PS4 Pro.

At the same time, there will be a few exclusives if it makes sense for third parties to port games but it would rather be the exception.

Now what comes next? Nintendo is always working in hardware but they won't abandon something with a healthy lifecycle. So either Brake sales give the Switch platform and ecosystem an new breath of life and boost sales significantly, which means it would become the de facto standard to develop for in 2024/25 with an active userbase.

Or, another option is that Drake sells well but not comparable to a "new generation" - which would be a hardware that is branded differently and spots more unique features than "just" more horsepower.

In this case, Nintendo could introduce such a hardware in 2024/25, maybe even using Drake on a different production pipeline with higher clocks to sustain compatibility, gut changing branding and features of the platform to differentiate from a saturated Switch market as something new and exciting.

This is something Nate could only speculate on, as it heavily depends on the performance of Switch Drake and Nintendo's needs to act accordingly. And that explains his statements, with a potential successor in the distant future and his impression of Drake as a "PS4 Pro" approach to the Switch ecosystem.

You guys think too much about power as she defining factor for a new generation, when it's actually all about market saturation, growth potential and new product and service offerings for a publisher or company.

Also @NateDrake , I'm sorry if I now spoiled your next podcast

 

[Adulfzen]

think the next revision will be out of stock regardless so we don't even have to worry about it selling well or not because Nintendo won't be able to produce enough of them where it'd become an issue.