• Hey everyone, staff have documented a list of banned content and subject matter that we feel are not consistent with site values, and don't make sense to host discussion of on Famiboards. This list (and the relevant reasoning per item) is viewable here.
  • Furukawa Speaks! We discuss the announcement of the Nintendo Switch Successor and our June Direct Predictions on the new episode of the Famiboards Discussion Club! Check it out here!

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

Like I said before, even just the few details that are solidly rumored about Dane place it pretty far outside "performant SKU" territory. It won't be the largest generational leap ever, but it doesn't need to be. Several of the new hardware features specifically enable improved efficiency, which will let developers do more with less.

Also, what's the point in doing a "performant SKU" when software has reached a point where anything you'd want to do with such a system could be done with the successor instead? It's a lot less work for both Nintendo and third parties when there are fewer distinct performance targets that have to be dealt with, especially on a hybrid system like the Switch which already creates more work than usual on that front.

Then you believe we will get “full” Dane in the next 12-18 months?
 
Yes, at least I very much do.

Devkits have been out since late 2020 with a performance target for Dane.

Nate has mentioned software exclusives to Dane from third parties targeted to come out in Late 2022 or early 2023.

Yeah, that seems like the timeline the rumors have been converging on. If that doesn't happen, then that would suggest fairly exceptional circumstances.

I hope you guys are right!
 
0
GBC was a stop-gap until they could release the GBA. Do you really think they're releasing the "true next-gen Switch" in 2025?
I don't, honestly the GBC comparison is only because it's the closest thing in the console space to what Dane will be.

Imho, describing it as Galaxy S to S2/iPhone to iPhone 3G is more accurate as Nintendo will likely stick with Dane for at least 4 years after Launch, maybe a refresh of it for better battery life aka Mariko, but power wise sticking with it for 4 years minimum.

When we say Smartphone model, we don't mean every year, we just mean power-increasing updates that build the platform out, dropping old hardware over time as it's rendered obsolete.

They are all under the Switch family for this example, but when, say 4 years after Switch 2 is out, Switch 1 Family Units will hit true EoL software support wise.

And the cycle will begin again with a Switch 3 coming out (Assuming they do that model of release).

The key thing is the accelerated release of new hardware with the caveat of the increase gen-on-gen not being as big as traditional console generations due to the sped up release cycle.

But the cross gen period between each hardware level is extended (which we already see with the current gen), giving people time to upgrade and find reasons to while stuff is releasing for both, but allowing Nintendo to keep the old system around as the budget model.
 
I don't, honestly the GBC comparison is only because it's the closest thing in the console space to what Dane will be.

Imho, describing it as Galaxy S to S2/iPhone to iPhone 3G is more accurate as Nintendo will likely stick with Dane for at least 4 years after Launch, maybe a refresh of it for better battery life aka Mariko, but power wise sticking with it for 4 years minimum.

When we say Smartphone model, we don't mean every year, we just mean power-increasing updates that build the platform out, dropping old hardware over time as it's rendered obsolete.

They are all under the Switch family for this example, but when, say 4 years after Switch 2 is out, Switch 1 Family Units will hit true EoL software support wise.

And the cycle will begin again with a Switch 3 coming out (Assuming they do that model of release).

The key thing is the accelerated release of new hardware with the caveat of the increase gen-on-gen not being as big as traditional console generations due to the sped up release cycle.

But the cross gen period between each hardware level is extended (which we already see with the current gen), giving people time to upgrade and find reasons to while stuff is releasing for both, but allowing Nintendo to keep the old system around as the budget model.
I don't think it makes sense to call that the smartphone model. That's the Xbox model.
 
I don't think it makes sense to call that the smartphone model. That's the Xbox model.
The Xbox 360 and One had a fairly quick cutoff.

The key distinction is the Cross-Gen period for big First-Party releases.

Third Parties can choose to publish on both or Switch 2 alone depending on the game, but Nintendo will keep Switch 2 exclusives from themselves rare or nonexistent until 2+ years after the model comes out.

by 4 years (if they move to an iterative release model), when the Switch 3 or Switch 2S or whatever comes out, that is when First Party support for OG Switch will be over.
 
The Xbox 360 and One had a fairly quick cutoff.

The key distinction is the Cross-Gen period for big First-Party releases.

Third Parties can choose to publish on both or Switch 2 alone depending on the game, but Nintendo will keep Switch 2 exclusives from themselves rare or nonexistent until 2+ years after the model comes out.

by 4 years (if they move to an iterative release model), when the Switch 3 or Switch 2S or whatever comes out, that is when First Party support for OG Switch will be over.
I really doubt there's going to be any hard and fast rule about cross-gen for Nintendo. It will depend on expectations, context, and the demands of development.

I expect something like Mario Kart 9, for example, to just be exclusive to Dane since there's already a MK for Switch OG.
 
GBC was more or less the Game Boy specs with everything doubled more or less.
That's not what the new SoC we expect is. That was what Nintendo was exploring with the Mariko at one point. From one of the older Speculation threads, someone said they heard Nintendo was looking at doing a 2GHZ CPU configuraiton and maybe also doubled GPU clocks on Marioko but the power draw was too much so they axed it. That was allegedly the actual Switch Pro that could have happened

I'm not sure why GBC keeps coming up in relation to the Dane SoC when it's clearly a generational leap ahead.

It's iterative in the sense that the CPU/GPU configuration SoC will be very similar to the Switch and highly compatible and that's rarely happened with Nintendo tech, and actually the closest thing isn't the GBC but the Super Nintendo where the SNES CPU essentially had a 6502 Core for compatability but was 'super' specced as well. Outside of that, Nintendo's designs often involved proprietary or old tech. There was no viable Z80 successor chip for them to use in the GBC and even when they Switched to ARM CPUs, the SoCs were so customized that compatability was only kept by having the previous SoC in the next design. Not really in the same situation now with nvidia's SoCs and the A57 inside the Switch and A78 CPUs we expect will be in the next SoC.
 
Last edited:
GBC was more or less the Game Boy specs with everything doubled more or less.
That's not what the new SoC we expect is. That was what Nintendo was exploring with the Mariko at one point. From one of the older Speculation threads, someone said they heard Nintendo was looking at doing a 2GHZ CPU configuraiton and maybe also doubled GPU clocks on Marioko but the power draw was too much so they axed it. That was allegedly the actual Switch Pro that could have happened

I'm not sure why GBC keeps coming up in relation to the Dane SoC when it's clearly a generational leap ahead.

It's iterative in the sense that the CPU/GPU configuration SoC will be very similar to the Switch and highly compatible and that's rarely happened with Nintendo tech, and actually the closest thing isn't the GBC but the Super Nintendo where the SNES CPU essentially had a 6502 Core for compatability but was 'super' specced as well. Outside of that, Nintendo's designs often involved proprietary or old tech. There was no viable Z80 successor chip for them to use in the GBC and even when they Switched to ARM CPUs, the SoCs were so customized that compatability was only kept by having the previous SoC in the next design. Not really in the same situation now with nvidia's SoCs and the A57 inside the Switch and A78 CPUs we expect will be in the next SoC.
The GBC is a weird case where its unusually strong software support (which I'd argue has more to do with its unique context rather than any meaningful distinction from the DSi and New 3DS) has lead some people to view it as more than it really is: an overclocked Game Boy with a color screen. Even Nintendo doesn't consider it a distinct system from the Game Boy. Not only is Dane a very different beast from a technical standpoint, it's distinct enough that I have a hard time seeing it not getting a full generational cycle.
 
I think a possibly better way to frame this discussion would be to approach it from a software point of view. This is because- and it may seem obvious- Nintendo is a hardware-software integrated company. How does that change things? Well, I think there is a reason you don't see Nintendo support games for more than 3 years and that is due to the simple fact that they have to sell hardware. Therefore, we should look at where their development teams are at the moment and what their next title might be.

For example, what system do you all think Botw 3 or Odyssey 3 will release on? The next Animal Crossing? In terms of development timelines, the next Animal Crossing is more likely to release before the next 3D Zelda (after Botw 2) because they finished development on ACNH in 2020 with DLC wrapping up recently. AC is the type of game, like Mario Kart, to get one game per system. It would also benefit greatly from improved hardware (as seen by some of the loading and framerate hitches in busy islands). Speaking of Mario Kart, that is another series that could more easily do something new and innovative with the hardware of Dane (open world multiplayer racing? Diddy Kong Racing style solo adventure?). These are two franchises that I think would launch within the first 1-3 years of the next Switch's life.

Sure, we can argue about release timing, but the more interesting question is when Nintendo's marquee teams will begin using Dane as the baseline for development. Seeing as Botw 2 and 3D Mario are likely to release by the end of 2023, we likely won't see their successors until 2025 at the earliest. And given that Bowser's Fury has already shown that Nintendo is seriously bumping up against the limits of the Switch, I find it highly unlikely that Botw 3 or Odyssey will be made with it in mind.
 
This company made all of us pay to transfer all of our already purchased Virtual Console Titles to the Wii U... for what? Gamepad play? save states?
I don't think 4K upgrade costs are off the table
I'm more confident in my mario party example.
It's recent and not from the wiiu era...with the mess that was the different virtual consoles.
 
0
I think a possibly better way to frame this discussion would be to approach it from a software point of view. This is because- and it may seem obvious- Nintendo is a hardware-software integrated company. How does that change things? Well, I think there is a reason you don't see Nintendo support games for more than 3 years and that is due to the simple fact that they have to sell hardware. Therefore, we should look at where their development teams are at the moment and what their next title might be.

For example, what system do you all think Botw 3 or Odyssey 3 will release on? The next Animal Crossing? In terms of development timelines, the next Animal Crossing is more likely to release before the next 3D Zelda (after Botw 2) because they finished development on ACNH in 2020 with DLC wrapping up recently. AC is the type of game, like Mario Kart, to get one game per system. It would also benefit greatly from improved hardware (as seen by some of the loading and framerate hitches in busy islands). Speaking of Mario Kart, that is another series that could more easily do something new and innovative with the hardware of Dane (open world multiplayer racing? Diddy Kong Racing style solo adventure?). These are two franchises that I think would launch within the first 1-3 years of the next Switch's life.

Sure, we can argue about release timing, but the more interesting question is when Nintendo's marquee teams will begin using Dane as the baseline for development. Seeing as Botw 2 and 3D Mario are likely to release by the end of 2023, we likely won't see their successors until 2025 at the earliest. And given that Bowser's Fury has already shown that Nintendo is seriously bumping up against the limits of the Switch, I find it highly unlikely that Botw 3 or Odyssey will be made with it in mind.

I think just like with Xbox and PS5 we're going to see cross platform releases and targets for the first two years of a new Switch at the very least. Which also fits with the comments about where the Switch is in its life cycle. Personally I expect games to launch with different performance targets. Game Design isn't nearly as compromised by hardware performance as PR and GPU manufacturers would like you to believe. If anything, tools and pipelines are the biggest limiting factors in any creative business production from my experience... (That includes incompetent or toxic leadership)

Since we have no freaking clue what projects the different Devs at Nintendo are up to right now, speculation on any launch or hardware target is soothsaying at best.
 
I think just like with Xbox and PS5 we're going to see cross platform releases and targets for the first two years of a new Switch at the very least. Which also fits with the comments about where the Switch is in its life cycle. Personally I expect games to launch with different performance targets. Game Design isn't nearly as compromised by hardware performance as PR and GPU manufacturers would like you to believe. If anything, tools and pipelines are the biggest limiting factors in any creative business production from my experience... (That includes incompetent or toxic leadership)

Since we have no freaking clue what projects the different Devs at Nintendo are up to right now, speculation on any launch or hardware target is soothsaying at best.
But that was my point. I wasn't focused on talking about what might launch with the system. I'm looking at when the next 3D Mario will come out after the current one is released. Given then Botw 2 is due for late 2022 at the earliest, Botw 3 would come out 2025 in an absolute best case scenario.

Let's say that Nintendo goes the latest possible release date that many are expecting in here, which would be 2024. Zelda and 3D Mario still wouldn't be ready by then. How much good would it do for Nintendo to release Botw 3 on Switch and Switch 2 in 2025? The Switch will have been over 8 years old by that point and have received 1.5 new 3D Zeldas and 2 new 3D Marios. At a certain point, they have to release software to sell their hardware. And 2025 would still be 2 years after late 2023/early 2024, so it's not like there wouldn't be other cross gen releases before then.
 
0
The more I hear about this alleged iterative strategy, the more it sounds just like what Sony and Microsoft are doing right now. A new gen strategy but with a more robust and long cross gen period and a unified account system that carries over and blurs the line a bit compared to previous gens. Nothing more, nothing less. With Xbox it was blurrier because of their branding, similar UI and controller, with PS5 it was a bit clearer. But that's the spectrum.
 
Last edited:
  • Nintendo released a new console, the OLED Switch
  • PS4/XB1 sales stopped because they announced successors, if that hadn't happened before COVID they would've seen a huge resurgence in demand. And I fact the PS4 at least did, it's been sold out globally for like a year and a half, they had to cut production of it to make way for PS5 production but that didn't stop demand for PS4s.
  • If they wait 1-2 years there will still be a chip shortage, releasing this same chip only a couple years later does nothing for them except hurt their yearly revenue
  • There is no shortage at all for 8nm parts, that's one of the few nodes that's almost fully available

  • OLED is not new console, its revision of current model with exactly same chip like current models.
  • PS4/XB1 sales start slowing down even before announcement PS5/XsX, but they did get sales bump during Covid (IMO Sony stoping PS4 production was big mistake)
  • Fact is that you dont know how big exactly chip shortages will be in year or two, it could be better situation, also more time can easily mean more chip produced and prepared for Switch Pro/2 manufacture, so for instance if they want full Switch 2 launch they will not launch it only if they have small number of consoles available in 1st year, even Nate said "supply constraints make it hard to know exactly when it is possible to release, meaning could miss holiday season"
  • We dont know for sure if it will be 8nm, with chip Covid and chip shortages plans are changing constantly, and with fact that Switch is selling like crazy entering its 6. year on market, they dont need rush new console on market
 
I see comparison with GBC, we dont need to go back so much, IMO PS4 and PS5 are much better example, similar tech/architecture with similar dev tools,
thats why even 90%+ of Sony onw games are still cross gen for PS4/PS5.
If we talking about Switch 2 (IMO "Pro" dont exist any more and we talking about Switch 2 now), difference is that Nintendo will aim for even better compatibility between Switch and Switch 2 in any case (compared to PS4 and PS5) with huge difference that Nintendo will keep selling current Switch models for at least around 2 years after Switch 2 launch like low price point.
 
0
The more I hear about this alleged iterative strategy, the more it sounds just like what Sony and Microsoft are doing right now. A new gen strategy but with a more robust and long cross gen period and a unified account system that carries over and blurs the line a bit compared to previous gens. Nothing more, nothing less. With Xbox it was blurrier because of their branding, similar UI and controller, with PS5 it was a bit clearer. But that's the spectrum.
Exactly. All console manufacturers has embraced this strategy to some degree, Xbox most of all because of smart delivery, more robust bc, nearly identical controller/ os/ digital store.
 
0
The GBC comparisons I'm making are all about product positioning and messaging, not the actual tech specs or anything.
  • OLED is not new console, its revision of current model with exactly same chip like current models.
  • PS4/XB1 sales start slowing down even before announcement PS5/XsX, but they did get sales bump during Covid (IMO Sony stoping PS4 production was big mistake)
  • Fact is that you dont know how big exactly chip shortages will be in year or two, it could be better situation, also more time can easily mean more chip produced and prepared for Switch Pro/2 manufacture, so for instance if they want full Switch 2 launch they will not launch it only if they have small number of consoles available in 1st year, even Nate said "supply constraints make it hard to know exactly when it is possible to release, meaning could miss holiday season"
  • We dont know for sure if it will be 8nm, with chip Covid and chip shortages plans are changing constantly, and with fact that Switch is selling like crazy entering its 6. year on market, they dont need rush new console on market
  • In the context of chip shortages that makes a very minor difference. Most of the components in short supply are things other than the SoC. Things like screens, controllers, antennae, etc. The OLED uses a number of new components from these categories.
  • We agree here
  • They can't just spend time amassing inventory if that's what you're suggesting, that's extremely costly paying for inventory storage
  • The most reliable info we actually have at the moment is that it's on 8nm
 
The GBC comparisons I'm making are all about product positioning and messaging, not the actual tech specs or anything.

  • In the context of chip shortages that makes a very minor difference. Most of the components in short supply are things other than the SoC. Things like screens, controllers, antennae, etc. The OLED uses a number of new components from these categories.
  • We agree here
  • They can't just spend time amassing inventory if that's what you're suggesting, that's extremely costly paying for inventory storage
  • The most reliable info we actually have at the moment is that it's on 8nm

-Context of new console was about chip shortages, revision with existing chip dont change nothing about talk of launch new console that have chip shortages, and we know that chip shortages mainly refers to CPU/GPU and in some capacity RAM, not about screen or antenna
-I dont talk about sitting on produced and finished consoles, Nintendo would rather want to secure better production and to gureantee some solid number of production instead have console launch with small number in time when they dont need to have new console launch because Switch is still selling like crazy
-We don't have any new info at moment, last one was few months ago, plenty things could change in meantime

Talking about GBC comparison, my PS4/PS5 comparison was also about positioning and messaging,
basically "here is new generation of Switch, but all we will continue to sell current Switch models and 90% all games will work on Switch devices for at least 2 years".
 
Last edited:
-Context of new console was about chip shortages, revision with existing chip dont change nothing about talk of launch new console that have chip shortages, and we know that chip shortages mainly refers to CPU/GPU and in some capacity RAM, not about screen or antenna
-I dont talk about sitting on produced and finished consoles, Nintendo would rather want to secure better production instead have console launch with small number in time when they dont need to have new console launch
-We don't have any new info at moment, last one was few months ago, plenty things could change in meantime
It’s just very unlikely they would change process node on their almost finished chip, due to all the costs and delays that would entail. And if they didn’t feel 8nm was good enough for a next gen switch, they wouldn’t have gone for it in the first place.
 
It’s just very unlikely they would change process node on their almost finished chip, due to all the costs and delays that would entail. And if they didn’t feel 8nm was good enough for a next gen switch, they wouldn’t have gone for it in the first place.

Everything is possible, remember, all this talk is just speculation, and we talk about product that based on first rumors should be on market already.
Maybe they thought 2 years ago that 8nm is good enough, and that doesnt think like that now, plenty of things could change last 2 years..
 
Last edited:
Everything is possible, remember, all this talk is just speculation, and we talk about product that based on first rumors should be on market already.
Maybe they thought 2 years ago that 8nm is good enough, and that doesnt think like that now, plenty of things changes last 2 years..
Possible, but unlikely.
 
Last edited:
Comparing something like that with hardware speculation of (still) non existing console based on rumors!?
Good job.
Because we’re discussing Nintendo and you still have to speculate in some base of reality with them. You’re trying to take all the shackles off and saying anything can happen. No, not anything can happen. They’re a business with boundaries they will stick within for their hardware.
 
Because we’re discussing Nintendo and you still have to speculate in some base of reality with them. You’re trying to take all the shackles off and saying anything can happen. No, not anything can happen. They’re a business with boundaries they will stick within for their hardware.

1st, I was talking about Nintendo all the time.
2nd, "Everything is possible" was taken from context because context was that last two years Nintendo plans about next Switch hardware could change (because Covid and Chip shortages), and its seems that once were change because based on rumors Switch "Pro" should be on market already.
And reality is that with Covid and chip shortages plans are constantly changing, even Nate who is here most reliable source constantly saying that (for instance he was saying that "Pro" and last time he said its not sure any more if its a Pro or Swtich 2), while some people act like everything is set in stone for next Switch hardware.
 
Last edited:
-Context of new console was about chip shortages, revision with existing chip dont change nothing about talk of launch new console that have chip shortages, and we know that chip shortages mainly refers to CPU/GPU and in some capacity RAM, not about screen or antenna
-I dont talk about sitting on produced and finished consoles, Nintendo would rather want to secure better production and to gureantee some solid number of production instead have console launch with small number in time when they dont need to have new console launch because Switch is still selling like crazy
-We don't have any new info at moment, last one was few months ago, plenty things could change in meantime

Talking about GBC comparison, my PS4/PS5 comparison was also about positioning and messaging,
basically "here is new generation of Switch, but all we will continue to sell current Switch models and 90% all games will work on Switch devices for at least 2 years".
It's a common misconception that the things like SoC and RAM are the biggest bottlenecks for the global shortage. Now, for the PS5 having a huge die will cause it to have production issues simply from the amount of silicon required, but silicon itself is not the primary item that's unable to meet demand.


Anyway the 8nm info is more or less all we have to work on. Nvidia has no other upcoming product which can feasibly be used in a Nintendo console until 2025 so it's really a matter of Occam's Razor at this point. Insider tells us Orin and Dane at 8nm, 8 months later Nvidia announces Orin is 8nm, therefore it's exceedingly likely Dane is too.
 
It's a common misconception that the things like SoC and RAM are the biggest bottlenecks for the global shortage. Now, for the PS5 having a huge die will cause it to have production issues simply from the amount of silicon required, but silicon itself is not the primary item that's unable to meet demand.

I dont understand, what my conclusion is incorrect?
 
Sorry, should've specified, this part-

I dont see how that link proves "that chip shortages mainly refers to CPU/GPU and in some capacity RAM, not about screen or antenna" is wrong.
Not to mentioned that for instance AMOLED screen prices and availability are much better now than they were few years ago, thats why even lower priced phones going with AMOLED screen.
 
Last edited:
I dont see how that link proves "that chip shortages mainly refers to CPU/GPU and in some capacity RAM, not about screen or antenna" is wrong.
Basically the idea that the primary cause of chip shortages is just the brand new cutting edge SoC/RAM chips is entirely wrong. Minor components that are used in all sorts of electronic are also a major driver of shortages, things that are absolutely needed for the new OLED model.
The current wave of chip shortages is expected to last for the foreseeable future, particularly for a growing list of critical devices produced in mature process nodes.

Chips manufactured at mature nodes typically fall under the radar, but they are used in nearly every electronic device, including appliances, cars, computers, displays, industrial equipment, smartphones, and TVs. Many of these chips are hot and in tight supply with long lead times, while others are easy to find. It depends on the chip type, specification, and vendor.

In addition, some leading-edge chips, such as select memories and processors, are also in the same boat. Those chips tend to grab most of the attention, but the ones based on mature nodes are also important. Among the mature-node semiconductors in tight supply are CMOS image sensors, display driver ICs, flash memory controllers, microcontrollers (MCUs), power MOSFETs, and power management ICs (PMICs).

Meanwhile, several types of chip products at matures nodes are hot and facing a shortfall, including microcontrollers. MCUs are used in a multitude of systems, such as appliances, cars, communication equipment, and industrial products. Infineon, Microchip, NXP and Renesas, and others supply MCUs. Some have their own fabs. Many outsource parts to the foundries.

Providing the processing functions in systems, MCUs integrate several components, such as the CPU and memory, on the same chip. MCUs come in 4-, 8-, 16-, and 32-bit configurations. The application depends on the configuration. For example, 32-bit MCUs are used in cars, while 4-bit products are found in appliances.

MCUs are processed in fabs at various nodes, such as 28nm, 40nm, 65nm, 90nm and 180nm. Some are developing devices at advanced nodes.

Display driver ICs (DDICs) and power management ICs (PMICs) are also in tight supply. DDICs are used to power up a flat-panel display, while PMICs are geared to manage the power in systems.

“We see applications related to power, such as PMICs and DDICs, in high demand,” UMC’s Ng said. “DDIC and PMIC production spans mature technologies from the 150nm through the 22nm/28nm nodes.”

Used in nearly every system, PMICs are chips that control the flow and direction of electrical power. Dialog, Maxim, Qualcomm, Samsung, STMicroelectronics, and TI sell PMICs. Some vendors build PMICs in their own fabs, while others use foundries.
Other chip types, such as analog and RF, are also in short supply. So are passive components like capacitors.
 
Basically the idea that the primary cause of chip shortages is just the brand new cutting edge SoC/RAM chips is entirely wrong. Minor components that are used in all sorts of electronic are also a major driver of shortages, things that are absolutely needed for the new OLED model.

I mean CPU/GPU and RAM are most noticeable things and problem to get production deals and numbers, so most people refers to them when say chip shortages, that doenst mean that some other parts are not in tight for finding.


"Chips manufactured at mature nodes typically fall under the radar, but they are used in nearly every electronic device, including appliances, cars, computers, displays, industrial equipment, smartphones, and TVs. Many of these chips are hot and in tight supply with long lead times, while others are easy to find. It depends on the chip type, specification, and vendor".

So those things can be problem or maybe not, while we know that there is problem for CPU/GPU in any case.


Also from same article:
“Overall, we think the chip shortages will last at least until mid-next year,” said Samuel Wang, an analyst at Gartner.
 
Everything is possible, remember, all this talk is just speculation, and we talk about product that based on first rumors should be on market already.
Maybe they thought 2 years ago that 8nm is good enough, and that doesnt think like that now, plenty of things could change last 2 years..
The problem is that semiconductor fabrication plants have been operating at over 90% utilisation since Q2 2020. So I imagine the absolute latest Nintendo and Nvidia could decide to use a process node more advanced than Samsung's 8N process node, and launch in holiday 2022 or early 2023, is probably 1H 2021, especially securing enough EUV lithography machines, which are used for process nodes more advanced than Samsung's 8N process node (Samsung's 7LPP process node or more advanced, or TSMC's N6 process node or more advanced (TSMC's N7+ process node's technicallly the first process node from TSMC to use EUV lithography, but TSMC's N7+ process node is not IP compatible with TSMC's N7 process node, unlike TSMC's N6 process node, which is why the Kirin 990 5G is the only chip publicly confirmed to be fabricated using TSMC's N7+ process node), taking more than a year. I imagine securing enough capacity for a process node is a process that requires companies to plan at least a year in advance, which is probably why kopite7kimi mentioned GH100 using TSMC's N5 process node as early as 11 March 2020.
 
Last edited:
I mean CPU/GPU and RAM are most noticeable things and problem to get production deals and numbers, so most people refers to them when say chip shortages, that doenst mean that some other parts are not in tight for finding.
I'm specifically pointing out that the bolded is likely not true. Do you have any source that suggests it is the most noticeable problem?
"Chips manufactured at mature nodes typically fall under the radar, but they are used in nearly every electronic device, including appliances, cars, computers, displays, industrial equipment, smartphones, and TVs. Many of these chips are hot and in tight supply with long lead times, while others are easy to find. It depends on the chip type, specification, and vendor".

So those things can be problem or maybe not, while we know that there is problem for CPU/GPU in any case.


Also from same article:
“Overall, we think the chip shortages will last at least until mid-next year,” said Samuel Wang, an analyst at Gartner.
That was from earlier in 2021, the outlook has become much worse since then.
 
I'm specifically pointing out that the bolded is likely not true. Do you have any source that suggests it is the most noticeable problem?

That was from earlier in 2021, the outlook has become much worse since then.

I disagree. Do you have source suggests otherwise?

Exactly my hole point, that things can change and that could effect Nintendo plans, from same reason I said to you couple of times that will not be exactly same situation until 2027. currently its not same situation even compared to last year.
 
I disagree. Do you have source suggests otherwise?
... The source I just shared above pointed to dozens of different commonly used components on mature processes being a major factor for the supply shortage.
Exactly my hole point, that things can change and that could effect Nintendo plans, from same reason I said to you couple of times that will not be exactly same situation until 2027. currently its not same situation even compared to last year.
Yes, it got worse...

Again I fail to understand why they would choose to wait while this uncertainty can get either better or worse all the while still affecting their current production capacity all the same.

Waiting does literally nothing for them.
 
... The source I just shared above pointed to dozens of different commonly used components on mature processes being a major factor for the supply shortage.

Yes, it got worse...

Again I fail to understand why they would choose to wait while this uncertainty can get either better or worse all the while still affecting their current production capacity all the same.

Waiting does literally nothing for them.

I disagree that thats points I am wrong.

You cant know if their producing capacites would be same, even Nate said that launch timing depends from "supply constraints".
Like I wrote, for instance waiting and later launch, could mean more consoles for launch 1st year, also Switch is still selling like crazy, they could easily say, there is no need to rush launch with this kind sales numbers and chip shortages.
 
GBC was a stop-gap until they could release the GBA.
Yeah, GBC was a weird situation. Their initial plans for a Game Boy successor fell through, but thanks to Pokémon they were dealing with a revival of an 8-year-old platform and had to do something to take advantage quick. Versus this Dane device which has been planned for ages and seems significantly different from original Switch, unless a large number of people have just been blowing smoke up our asses for years. Sure seems like a lot of time and effort for something they don't intend to last long.
 
Yeah, GBC was a weird situation. Their initial plans for a Game Boy successor fell through, but thanks to Pokémon they were dealing with a revival of an 8-year-old platform and had to do something to take advantage quick. Versus this Dane device which has been planned for ages and seems significantly different from original Switch, unless a large number of people have just been blowing smoke up our asses for years. Sure seems like a lot of time and effort for something they don't intend to last long.
Through process node shrinkages, I would assume Nintendo always as a GBC like device for all their consoles in the wings all the time as an emergency back-up plan. Basically just an overclocked version of the platform with more RAM.

It's interesting we see echoes of the GBC repeated in the Wii and even the rumored cancelled Mariko Overclocked SKU.
 
0
I disagree that thats points I am wrong.

You cant know if their producing capacites would be same, even Nate said that launch timing depends from "supply constraints".
Like I wrote, for instance waiting and later launch, could mean more consoles for launch 1st year, also Switch is still selling like crazy, they could easily say, there is no need to rush launch with this kind sales numbers and chip shortages.
Sure, it's full of uncertainty. That's the point. They can wait a year or two to launch it and take their chances on it maybe improving. But it may not improve, and it may in fact get worse. Nintendo is a risk averse company, they will not want to take the chance that it might get worse, especially when base Switch sales will have slowed down naturally by then.

Delaying their plans by a substantial amount (i.e. more than 3-5 months) could wind up helping them sell an extra 10-20% or it could wind up absolutely destroying their yearly revenue. It makes zero sense for a company like Nintendo to take that chance.
 
Like I wrote, for instance waiting and later launch, could mean more consoles for launch 1st year, also Switch is still selling like crazy, they could easily say, there is no need to rush launch with this kind sales numbers and chip shortages.
Inventory shortages are unfortunately inevitable no matter when Nintendo decides to have a launch, especially with demand for ABF substrates, which is used for all modern chips, isn't expected to be met until 2026 at the absolute earliest.
 
My thinking is that we are in an unique situation with COVID and global chain disruptions, so this is the time were we can expect never done before changes in production and timings.

I was on board on Switch 4k in 2022 train until it became that the chip shortage will not be over any time soon.

Yeah, I'm not discounting the possibility that circumstances might line up for Nintendo to want to migrate manufacturing processes. Maybe the 8nm process is consuming too much power, or maybe they just figured the continued high sales of Switch warranted pushing things back. But absent of any evidence for this, I don't think we should expect it.

Also, I wouldn't say the chip shortage is much of a reason to migrate from 8nm, as it's probably the most easily available process for Nintendo and Nvidia by a comfortable margin. Nvidia already have a large majority of the 8nm manufacturing allocation, and they're migrating their consumer GPUs over to TSMC 5nm later this year, which will free up plenty of capacity for Dane.

Specifically for the OLED Switch I haven't seen one available for MSRP in any store or online retailer in my region since October. Personally I think for that you're underestimating the gap between supply and demand, as anecdotally it seems to be a similar situation globally.

Now, I personally don't feel the need to get an OLED Switch so I haven't been searching that hard for the entire 4 months it's been out but I did have a couple friends ask me to look for them for a period of probably 4-5 weeks and I had zero luck finding one at MSRP.

Maybe I'm placing too much weight to my own anecdotal experience, but I've personally not had much difficulty finding the OLED model when I've looked (in UK, for reference). Although to be honest I haven't been looking too often, I'm trying to avoid the impulse purchase! I came close once, but managed to pull myself back from the brink.

Technically, there's Intel's Rocket Lake desktop lineup that launched last year. Rocket Lake's Cypress Cove cores are mainly a backport of Sunny Cove cores from 10nm back to 14nm. And Sunny Cove was used in the Ice Lake laptops that launched in late 2019. Hmm, and I just remembered that Rocket Lake doesn't use the Gen 11 integrated graphics that Ice Lake used, but instead Xe-LP (introduced with the Tiger Lake laptops on 10nm SuperFin). So that's a backport of something on a different version of 10nm back to 14nm. But circumstances were... very not normal for Intel desktop then. Your point still holds.

Yeah, I think Intel's misadventures in moving on from 14nm would count as drastic alright! I think that's basically it, though, in that you'd only really look at migrating to another node if the one you're working on simply isn't hitting expectations, either in yields, performance or power consumption. Yields shouldn't be an issue for 8nm, and although power consumption is the limiting factor for Nintendo, there shouldn't be too many surprises on that front for a process as mature as 8nm.

2. I have a new Nintendo Switch hardware theory. It’s sorta evil which makes me want to believe it’s real.

A 2022 Switch Pro based on Dane/Ampere but it will be just barely worth it. About 4x performance w/DLSS - enough to take games like BotW to 1440p docked and existing 1080p games to 4k. It will support more next gen/more demanding games but only at like 540p-720p docked.

I think they can sell 20m-40m units over 2- 3-years and then launch next gen in 2024-2025 on Nvidia 5nm.

The pros of this include:
-$400 price point; Gets consumers used to more expensive Nintendo hardware.
- Leaves plenty of room to market improved performance between Pro and Switch 2.
-Helps them resell more evergreen software and recoup any work they need to port/transfer games to the new hardware.

I don't think this is "evil", it's just a standard mid-gen upgrade/"pro" model. The technical jump required to make DLSS practical also means the minimum viable hardware to get DLSS at 4K (even just at 30fps) is still a pretty comfortable upgrade. You're probably looking at at least 2x performance jump on the GPU side pre-DLSS, just for a GPU config with enough tensor core grunt to handle the DLSS, plus the general benefit you get from moving from Maxwell to Ampere. On the CPU front, even with just 3 cores available for games, which is the minimum for compatibility, and conservative clock speeds (say 1.2GHz), the jump from A57 to A78 is a big one, with a 2.3x clock-for-clock performance boost even before any clock speed increases. Combined with the use of DLSS, it's a big perceived jump in performance for a mid-gen upgrade, probably matching Xbox One to One X in visual fidelity increase, and far beyond when it comes to the jump in CPU performance.

Also, if we've got a system which is capable of hitting 4K output in any form (even only with DLSS, and only at 30fps), then there shouldn't be any reason to see games running at 540p-720p output resolutions in portable mode. That's a 9x decrease in pixels, and the performance difference between portable and docked shouldn't be anywhere near that much, more likely close to 2x, maybe 3x if they really push GPU clocks when docked. In that case, there should absolutely be performance to run DLSS in portable mode to a 1080p screen, although with closer to 720p pre-DLSS rendering resolution likely in most cases.

If they really wanted to be evil they'd release a much more powerful revision based on Dane, and give it an IPS screen when the OLED model's on the market with less powerful hardware!
 
Sure, it's full of uncertainty. That's the point. They can wait a year or two to launch it and take their chances on it maybe improving. But it may not improve, and it may in fact get worse. Nintendo is a risk averse company, they will not want to take the chance that it might get worse, especially when base Switch sales will have slowed down naturally by then.

Delaying their plans by a substantial amount (i.e. more than 3-5 months) could wind up helping them sell an extra 10-20% or it could wind up absolutely destroying their yearly revenue. It makes zero sense for a company like Nintendo to take that chance.

But going by first rumors we got, new stronger device should be released already, that points that their plans changed at least once already. Offcorse they are risk aware company, but with current sales there is no risk releasing stronger Switch sometime in 2023.

I dont see how 3-5 months (even 6-9 months) delay would destroying their yearly revenue when Switch is selling like crazy and where they cant keep with demand despite they will do 24m this FY, point that they will not be any bigger slow down any time soon (for around year).

You acting like I talking about 2 years delay, while only thing is that I said that I expect that stronger Switch will be released most likely sometime next year and not this year.



Inventory shortages are unfortunately inevitable no matter when Nintendo decides to have a launch, especially with demand for ABF substrates, which is used for all modern chips, isn't expected to be met until 2026 at the absolute earliest.

Nate said that launch timing depends from "supply constraints", it could easily be later launch means more inventory and consoles ready for launch and that Nintendo could easily afford internal delay because Switch is selling great and they will end FY with another record breaking year, espacially if they doing full next gen launch so they need more consoles for launch and 1st year.
 
But going by first rumors we got, new stronger device should be released already, that points that their plans changed at least once already. Offcorse they are risk aware company, but with current sales there is no risk releasing stronger Switch sometime in 2023.
Which rumors? The Bloomberg one from early last year that was very clearly debunked when the OLED model was announced?

Virtually everyone talking about this device before Bloomberg started mentioning 2022 as the likely release timing early last year.

I dont see how 3-5 months (even 6-9 months) delay would destroying their yearly revenue when Switch is selling like crazy and where they cant keep with demand despite they will do 24m this FY, point that they will not be any bigger slow down any time soon (for around year).

You acting like I talking about 2 years delay, while only thing is that I said that I expect that stronger Switch will be released most likely sometime next year and not this year.
You know, I did think you were talking about a 1-2+ year delay, yes. Maybe I misread your initial posts, sorry about that. I agree we might probably see a 3-6 month delay, maybe to next March probably at the latest. I don't see them pushing it to next holiday season and risking their total hardware sales to drop below ~15M since it will definitely be supply constrained.
 
Please read this new, consolidated staff post before posting.

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


Back
Top Bottom