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

[PedroNavajas]

Going from the 22/20nm to 16/14nm Finfet is not the same as going from 8nm DUV to a 7-5nm EUV. If there was a delay and a redesign, you wouldn’t see that chip really for 2023/24. Probably 25/26 or later in the current times where it’s not really easy to do this.

Your scenario assumes that Nvidia don't have already a path for newer nodes and start working on such a chips today. I think that scenario is unlikely. I'm pretty sure Nvidia already has designs for 7-5nm in the works and my point is that I think is likely that Nvidia/Nintendo already had a plan for a SoC in newer nodes, regardless of whenever Dane was aiming to release in 8nm this year or the next. I find very hard to believe that Nvidia cannot produce a mobile SoC at 7nm until 2026...

Also, Nintendo has done more radical "last minute" SoC changes before (3DS from Tegra 2 to custom Pica200 SoC).

 

[ReddDreadtheLead]

Your scenario assumes that Nvidia don't already has a path for newer nodes and start working on the such chip today. I think that scenario is unlikely. I'm pretty sure Nvidia already has design for 7-5nm in the works. I find very hard to believe that Nvidia cannot produce a mobile SoC at 7nm until 2026...

Also, Nintendo has done more radical "last minute" SoC changes before (3DS from Tegra 2 to custom Pica200 SoC).

NV has the chip set for certain customers, and if they can deliver what the customers request which is a super computer essentially for cars and medical fields, they don’t need to make it again on a 7nm or new node. The scenario is about the current “deadend” node that is 8nm. Unless you know of some other node that Samsung has which is a successor and also better than the 8nm or of some process that makes it relatively easy to move from DUV to EUV, there isn’t really anything they can just shrink Dane to.


The Erista chip you mentioned was released in 2017 when the 16nm node was already on the market, they got a shrink 2 years after the switch, which was 3 years after 16nm was on the market.

And I doubt Nintendo did a “last minute”, they likely already had that set up as a contingency for when or if the Tegra 2 failed to meet their goals. And comparing the situation surrounding the 3DS isn’t really similar to the situation surrounding the Switch and it’s successor system.

For one, the switch isn’t as “easy” as the 3DS and the chips it had inside of it (it was not a SoC really). Things have come a long way but they’ve also grown more complex.

And it is not that nVidia or Nintendo cannot have Dane on a 7 nm or 5 nm node, it’s that the cost of doing so would be so high that I don’t see them doing such a thing. To move it from the 8nm to 7/5nm, which would not be a trivial task let me remind you.

 

[PedroNavajas]

NV has the chip set for certain customers, and if they can deliver what the customers request which is a super computer essentially for cars and medical fields, they don’t need to make it again on a 7nm or new node. The scenario is about the current “deadend” node that is 8nm. Unless you know of some other node that Samsung has which is a successor and also better than the 8nm or of some process that makes it relatively easy to move from DUV to EUV, there isn’t really anything they can just shrink Dane to.


The Erista chip you mentioned was released in 2017 when the 16nm node was already on the market, they got a shrink 2 years after the switch, which was 3 years after 16nm was on the market.

And I doubt Nintendo did a “last minute”, they likely already had that set up as a contingency for when or if the Tegra 2 failed to meet their goals. And comparing the situation surrounding the 3DS isn’t really similar to the situation surrounding the Switch and it’s successor system.

For one, the switch isn’t as “easy” as the 3DS and the chips it had inside of it (it was not a SoC really). Things have come a long way but they’ve also grown more complex.

And it is not that nVidia or Nintendo cannot have Dane on a 7 nm or 5 nm node, it’s that the cost of doing so would be so high that I don’t see them doing such a thing. To move it from the 8nm to 7/5nm, which would not be a trivial task let me remind you.

I'm not saying is trivial or easy to do, I'm saying that I think is likely that such thing is in the works regardless of whenever the Switch 4K was planned to release using Dane 8nm or not. I find hard to believe that Switch 4k is doomed to stay at 8nm until its end of life. That hasn't happened to any modern console.

And it is widely speculated that Nvidia designed their Ampere architecture for TSMC 7nm and the shift to Samsung 8nm was a "last minute" change. And I think they even released the profesional chips at 7nm? The architecture might not be as dependant on 8nm or their designs as inflexible as you are suggesting.

 

[fwd-bwd]

Is shrinking Dane down the road a given though? For what—Dane Lite? There's no guarantee such a product would ever be released. It's a budget model mainly for kids and casuals; might as well keep Mariko and lower the price instead. And if Nintendo (partnering with Nvidia?) is to wade further into cloud gaming, there'd be no incentive to replace Lite's SOC.

Even if a Dane Lite is in the cards, Nintendo could use the same 8nm Dane and accept a subpar battery life. Reasonable consumers would understand the trade-off for a budget model.

 

[PedroNavajas]

Is shrinking Dane down the road a given though? For what—Dane Lite? There's no guarantee such a product would ever be released. It's a budget model mainly for kids and casuals; might as well keep Mariko and lower the price instead. And if Nintendo (partnering with Nvidia?) is to wade further into cloud gaming, there'd be no incentive to replace Lite's SOC.

Even if a Dane Lite is in the cards, Nintendo could use the same 8nm Dane and accept a subpar battery life. Reasonable consumers would understand the trade-off for a budget model.

Improved battery life, like what they did with Mariko. Better economics for not using a depreciated node and reduced power envelope. Even the possibility of a Pro version. Chip shrinking is not exactly rare for the industry. I would say is rather common, including nintendo devices.

 

[Hermii]

Is shrinking Dane down the road a given though? For what—Dane Lite? There's no guarantee such a product would ever be released. It's a budget model mainly for kids and casuals; might as well keep Mariko and lower the price instead. And if Nintendo (partnering with Nvidia?) is to wade further into cloud gaming, there'd be no incentive to replace Lite's SOC.

Even if a Dane Lite is in the cards, Nintendo could use the same 8nm Dane and accept a subpar battery life. Reasonable consumers would understand the trade-off for a budget model.

If Dane is the Switch 2, it would be the first console generation in history not to have a die shrink.

Incentive would be to release a cheaper handheld with full software compitability.

The OG lite woudnt have been physically possible on that form factor with a 20 nm x1. It would melt.

 

[PedroNavajas]

If Nintendo is indeed locked to 8nm for the generation, that would severely limit their succesfull decades-old strategy of iterative versions for their handhelds. I think this is so important for their business model that they would have looked at alternative designs if Nvidia doesn't has a solution for them.

 

[Hermii]

If Nintendo is indeed locked to 8nm for the generation, that would severely limit their succesfull decades-old strategy of iterative versions for their handhelds. I think this is so important for their business model that they would have looked at alternative designs if Nvidia doesn't has a solution for them.

Or. die shrinking 8 nm aint as difficult as its made out to be. I dare say, none of us are experts.

 

[Dekuman]

From the analyst Q&A (note this is David Gibson's transcription not an official record, and could be inaccurate/incomplete):

Came here to post this. It's worth noting David Gibson is translating on the fly and sometimes do get the translation a bit off, so just take that with a grain of salt.

Edit: i see you've made a similar disclaimer. my bad.

 

[Brofield]

Is my faith misplaced to still hope for an announcement of a revision/successor/pro before March 2023?

I don't even know why I let it affect me this much my friend is being super smug about this quarter's earnings between hardware shipped, global chip shortage and middle of lifecycle that he guarantees there's not a chance there will be another Switch for a long while.

I personally disagree looking at the precedent of PS4 Pro and Xbox One X (where he personally owns the latter), but those don't count. Why, I'm not sure

inb4 I need to touch grass

 

[Skittzo]

Is my faith misplaced to still hope for an announcement of a revision/successor/pro before March 2023?

I don't even know why I let it affect me this much my friend is being super smug about this quarter's earnings between hardware shipped, global chip shortage and middle of lifecycle that he guarantees there's not a chance there will be another Switch for a long while.

I personally disagree looking at the precedent of PS4 Pro and Xbox One X (where he personally owns the latter), but those don't count. Why, I'm not sure

inb4 I need to touch grass

Everything we've heard from unofficial sources still says whatever this is is launching before early 2023. Nothing has changed about that.

 

[Brofield]

Everything we've heard from unofficial sources still says whatever this is is launching before early 2023. Nothing has changed about that.

Good enough for me; I'll take what solace I can in that haha

 

[Z0m3le]

I was told about a process node debate happening here? I think it's important to go with information we have until updated. Dane is 8nm until it isn't, we can always discuss its merits, but until we get new news to support it, Dane hasn't shrunk, we could make up whatever we want, but if we want a discussion around the actual product to reflect that product, we should tie it to whatever we have.

I've said in the past that Samsung is making 4 or 5 components for the device, so it makes sense to talk about their ram and storage solutions, I haven't been updated on the pro model for a long time, so I'll just use what information we have for now.

Also, I don't see it being more than $400, it might not be on the market for 5 years, I'd expect it to be replaced in 2027.

 

[Dakhil]

And it is widely speculated that Nvidia designed their Ampere architecture for TSMC 7nm and the shift to Samsung 8nm was a "last minute" change. And I think they even released the profesional chips at 7nm? The architecture might not be as dependant on 8nm or their designs as inflexible as you are suggesting.

Nvidia have always used TSMC for the fabrication of datacentre GPUs.

And datacentre Ampere GPUs (GA100) have a noticeably different design from consumer Ampere GPUs (GA102, etc.), as shown below.
GA100:

GA102:

As for consumer Ampere GPUs, there's a report from the Korea Herald that mentioned that Yoo Eung-joon, the chief of Nvidia Korea, mentioned that Nvidia initially planned on using Samsung's 7LPP process node to fabricate consumer Ampere GPUs. So assuming the Korea Herald's report is accurate, there's a possibility that Nvidia has always planned on using Samsung for the fabrication of consumer Ampere GPUs. And there's a possibility that shortly after the Korea Herald's report was published that Nvidia ultimately decided to use Samsung's 8N process node to fabricate consumer Ampere GPUs due to Samsung's 7LPP process node having yield issues.

 

[Skittzo]

I was told about a process node debate happening here? I think it's important to go with information we have until updated. Dane is 8nm until it isn't, we can always discuss its merits, but until we get new news to support it, Dane hasn't shrunk, we could make up whatever we want, but if we want a discussion around the actual product to reflect that product, we should tie it to whatever we have.

I've said in the past that Samsung is making 4 or 5 components for the device, so it makes sense to talk about their ram and storage solutions, I haven't been updated on the pro model for a long time, so I'll just hold to what information I have for now.

Also, I don't see it being more than $400, it might not be on the market for 5 years, I'd expect it to be replaced in 2027.

I think the discussion about the node interestingly revolves around the discussion about timing. I think it started because Thraktor mentioned an 8nm Dane in 2023-2024 would be quite old, older than the 20nm Erista in 2017, and from there thoughts were along the line of "well if it's coming in 2023-2024+ then could the node be changed?"

IMO the easier way to resolve this is that, like every rumor has stated, Dane is coming in late 2022 or early 2023. 8nm wouldn't be super old at that point.

 

[Dekuman]

I think the discussion about the node interestingly revolves around the discussion about timing. I think it started because Thraktor mentioned an 8nm Dane in 2023-2024 would be quite old, older than the 20nm Erista in 2017, and from there thoughts were along the line of "well if it's coming in 2023-2024+ then could the node be changed?"

IMO the easier way to resolve this is that, like every rumor has stated, Dane is coming in late 2022 or early 2023. 8nm wouldn't be super old at that point.

I think that comparison isn't great. I don't know much about semiconductors but i distinctly recall 20nm being a not great node and Apple et al moved off to 16nm ASAP, leaving a lot of excess capacity. Nvidia themselves skipped it on their consumer GPUs, going from 28nm on Maxwell to 16nm on Pascal cards Nintendo themselves moved off of it with the new Switch and Lite revisions. so it was an intermediate node.

If 8nm is what Dane is on , it is also what Ampere consumer cards are on. That's probably a better metric than guessing the age of the process.
@Thraktor is usually amazing with his logic and reasoning in his posts but that bit about the age of the node was uncharacteristic.

 

[TheDok]

Everything we've heard from unofficial sources still says whatever this is is launching before early 2023. Nothing has changed about that.

What sources?

 

[Skittzo]

I think that comparison isn't great. I don't know much about semiconductors but i distinctly recall 20nm being a not great node and Apple et al moved off to 16nm ASAP, leaving a lot of excess capacity. Nvidia themselves skipped it on their consumer GPUs, going from 28nm on Maxwell to 16nm on Pascal cards Nintendo themselves moved off of it with the new Switch and Lite revisions. so it was an intermediate node.

If 8nm is what Dane is on , it is also what Ampere consumer cards are on. That's probably a better metric than guessing the age of the process.
@Thraktor is usually amazing with his logic and reasoning in his posts but that bit about the age of the node was uncharacteristic.

Yeah that's fair. I'm not expert so I really don't know how old or dated it is, or how much that matters compared to other things.

What sources?

Nate, Bloomberg, Grubb, Nikkei, probably others I'm forgetting at the moment.

 

[shoptroll]

Is my faith misplaced to still hope for an announcement of a revision/successor/pro before March 2023?

I don’t think so? I haven’t been following the story of what comes next that closely but I noticed on the KS page for Yacht Club’s new game that they’re not detailing platform availability yet with an estimated release of Dec 2023. Which makes me think they’re expecting some new hardware to drop (probably from Nintendo) before too long.

 

[PedroNavajas]

I agree with Thraktor that an 8nm device in 2024 would be dated, and would still not be great in 2023. And I think that is on Nintendo's best interest to either release it sooner or move to a more modern node.

Desktop ampere comparison is not great either. The analogue of that is 28nm, which what desktop maxwell used. Under that metric, 8nm would still be lagging.

 

[Z0m3le]

I agree with Thraktor that an 8nm device in 2024 would be dated, and would still not be great in 2023. And I think that is on Nintendo's best interest to either release it sooner or move to a more modern node.

Orin is on 8nm and isn't out yet. I agree that if Dane slipped to 2024, it would be too old, I think if that were to happen, they would just cancel the current Switch device and move to a 5nm one for late 2025 or even 2026, which might even have a Zelda release in the first 12 months.

Dane started in 2019, so a late this year, to a late next year release makes sense, that's a 3 to 4 year project which is about as long as console SoCs take to design. I think Dane could be this calendar year, but Nintendo might delay it if they have a mainline Mario for the movie's launch, the Mario movie was set for a summer release this year and was delayed to the holidays, a mainline Mario could be coming out this October.

 

[ReddDreadtheLead]

I'm not saying is trivial or easy to do, I'm saying that I think is likely that such thing is in the works regardless of whenever the Switch 4K was planned to release using Dane 8nm or not. I find hard to believe that Switch 4k is doomed to stay at 8nm until its end of life. That hasn't happened to any modern console.

And it is widely speculated that Nvidia designed their Ampere architecture for TSMC 7nm and the shift to Samsung 8nm was a "last minute" change. And I think they even released the profesional chips at 7nm? The architecture might not be as dependant on 8nm or their designs as inflexible as you are suggesting.

If you expect them to spend a large sun just to redesign the chip for a newer node, sure.

But ask yourself, would Nintendo help foot the bill to redesign a chip for a different node that isn’t going to be a successor but a revision that only gets advertised as having a longer battery life when such a task is A) not trivial, B) very expensive to do in this times with shortages and COVID and C) they would be the only customer for this?


What sense does it make to spend that much and only sell that as some sort of “revised” model that improves battery life?

And it’s not the architecture that’s dependent on the node, it’s that the lithography of both nodes is very different to the point where moving from 8 to 7 requires re-designing the entire chip for this new lithography process. This is a massive R&D here and only one customer would be using it at the time, being Nintendo, as it’s a chip custom made solely for their use case that is a game console.

 

[Alovon11]

Orin is on 8nm and isn't out yet. I agree that if Dane slipped to 2024, it would be too old, I think if that were to happen, they would just cancel the current Switch device and move to a 5nm one for late 2025 or even 2026, which might even have a Zelda release in the first 12 months.

Dane started in 2019, so a late this year, to a late next year release makes sense, that's a 3 to 4 year project which is about as long as console SoCs take to design. I think Dane could be this calendar year, but Nintendo might delay it if they have a mainline Mario for the movie's launch, the Mario movie was set for a summer release this year and was delayed to the holidays, a mainline Mario could be coming out this October.

I personally feel they can launch it alongside the Mario movie paired game as it would give just about as much momentum for Dane and both Dane and OG Switch will sell out in the holidays anyway.

Heck, it would give them two spikes of sales for Dane at least, one in Holiday 2022, another with BOTW2 in March 2023 if we follow this scenario.

 

[Alovon11]

If you expect them to spend a large sun just to redesign the chip for a newer node, sure.

But ask yourself, would Nintendo help foot the bill to redesign a chip for a different node that isn’t going to be a successor but a revision that only gets advertised as having a longer battery life when such a task is A) not trivial, B) very expensive to do in this times with shortages and COVID and C) they would be the only customer for this?


What sense does it make to spend that much and only sell that as some sort of “revised” model that improves battery life?

And it’s not the architecture that’s dependent on the node, it’s that the lithography of both nodes is very different to the point where moving from 8 to 7 requires re-designing the entire chip for this new lithography process. This is a massive R&D here and only one customer would be using it at the time, being Nintendo, as it’s a chip custom made solely for their use case that is a game console.

Honestly, I am considering that they launch Dane as is, then just make a slightly smaller version of Dane to match the Portable mode profile for a Dane-Lite as the budget SoC to maintain software support in the future primarily CPU-side (Series S vs Series X)

 

[ShadowFox08]

Just because Nintendo says the Switch it in it's half life, doesn't mean we'll have to wait until 2026-2027 for Switch 2.

For one, the market dictates that. If it sells well enough, it can definitely last that long. If it doesn't, well Nintendo won't have much of a choice then.
Secondly, this is nothing new for Sony and MS consoles that have lasted +7-9 years easily. It's been 8 years and PS4 and xbone are still getting support, and Xbox 360 and PS3 were also getting support for a similar period of time.

Nintendo just wants to flaunt this because for the first time in a long time, a Nintendo console has been this successful at this point in it's life vs having less than 6 years and being on life support for previous home consoles at this stage, and quickly lose software support after. But a lot of is due to it's hybrid nature and no separate dedicated handheld console being sold (Wii u and 3ds). Otherwise, I don't know if a switch home console could last nearly this long and lose support immediately after. Who knows. But hybrid is here to stay.

So no, this does not mean a 2023 switch model will be a pro model. It can still be a successor. Hell it will be a successor, whenever it gets released. All the tech is brand new, and with our speculation of it's power being in the PS4 to PS4 pro range while docked (basically steam deck with DLSS support), by all means it's a next gen console.

I don't know how I feel about Nintendo marketing this as a "pro" model.l even for the crust few heads. I think it could confuse the mass consumers. They can still support the switch 1 (and have backwards compatibility)and share several 1st party games for a number of years. Really not that big of a deal. Sony and MS have been doing this. But what makes it different from Sony and MS consoles, is that a successor console will be releasing much less sooner than theirs.

 

[NineTailSage]

Orin is on 8nm and isn't out yet. I agree that if Dane slipped to 2024, it would be too old, I think if that were to happen, they would just cancel the current Switch device and move to a 5nm one for late 2025 or even 2026, which might even have a Zelda release in the first 12 months.

Dane started in 2019, so a late this year, to a late next year release makes sense, that's a 3 to 4 year project which is about as long as console SoCs take to design. I think Dane could be this calendar year, but Nintendo might delay it if they have a mainline Mario for the movie's launch, the Mario movie was set for a summer release this year and was delayed to the holidays, a mainline Mario could be coming out this October.

I do think the biggest difference between TX1 and Dane would be that 20nm was never really a great node for mobile hardware and 8nm is still decent enough for Nintendo to bring their hardware well above the 1Tflop range of the last generation consoles.
As long as they can solve for sufficient RAM, fast storage and CPU performance that's a major step up over last-gen, Dane would still be a major improvement over the current Switch and both PS4/ XboxOne...

 

[PedroNavajas]

If you expect them to spend a large sun just to redesign the chip for a newer node, sure.

But ask yourself, would Nintendo help foot the bill to redesign a chip for a different node that isn’t going to be a successor but a revision that only gets advertised as having a longer battery life when such a task is A) not trivial, B) very expensive to do in this times with shortages and COVID and C) they would be the only customer for this?


What sense does it make to spend that much and only sell that as some sort of “revised” model that improves battery life?

And it’s not the architecture that’s dependent on the node, it’s that the lithography of both nodes is very different to the point where moving from 8 to 7 requires re-designing the entire chip for this new lithography process. This is a massive R&D here and only one customer would be using it at the time, being Nintendo, as it’s a chip custom made solely for their use case that is a game console.

I guess this depends largely on how much customization Nvidia actually did for Nintendo. History doesnt not give us much information since the Switch used TX1 as it is with no customization whatsoever, so Nintendo probably didn't spend much if at all on design, and, we all(?) believe that orin is too large and aimed a too different market for Nintendo use it as it is.

So, the question is, how much customization was done for Dane and such work is prohibitively expensive to repeat? I do believe that Nvidia must be working on SoCs at newer nodes. How much customization would those need to be used in lieu of Dane? I also stand by my statement that node shrinking is a very important for Nintendo, as history proves.

 

[Zedark]

Yeah that's fair. I'm not expert so I really don't know how old or dated it is, or how much that matters compared to other things.

Nate, Bloomberg, Grubb, Nikkei, probably others I'm forgetting at the moment.

Did Grubb also say something about a 2022 device?

 

[Skittzo]

Did Grubb also say something about a 2022 device?

In the middle of 2021 a few weeks/months after the OLED model was announced he tweeted that he had only ever heard of a 4k Switch for 2022.

I'd search for it if Twitter searching wasn't such a pain.

 

[Dakhil]

Did Grubb also say something about a 2022 device?

In the middle of 2021 a few weeks/months after the OLED model was announced he tweeted that he had only ever heard of a 4k Switch for 2022.

I'd search for it if Twitter searching wasn't such a pain.

 

[ITSMILNER]

Having devkits out in the wild since 2021 would indicate a device launching before 2023 would it not? How early do devs get hold of a systems hardware before launch?

 

[bloopland33]

In the middle of 2021 a few weeks/months after the OLED model was announced he tweeted that he had only ever heard of a 4k Switch for 2022.

I'd search for it if Twitter searching wasn't such a pain.

And then a few weeks ago on his stream he speculated there'd be no new hardware till 2023+ or something, correct? I think they were commenting on that analyst report that made headlines a few weeks ago (about when he believed Nintendo would have new hardware)

 

[Skittzo]

Having devkits out in the wild since 2021 would indicate a device launching before 2023 would it not? How early do devs get hold of a systems hardware before launch?

Devkits have been out since late 2020, reportedly.

 

[karmitt]

And then a few weeks ago on his stream he speculated there'd be no new hardware till 2023+ or something, correct? I think they were commenting on that analyst report that made headlines a few weeks ago (about when he believed Nintendo would have new hardware)

He did, but unlike before neither he or Mike said they’d heard anything new. I took it to mean he was less optimistic due to complete silence on the subject

 

[Look over there]

My curiosity is more about that NoA job posting for a senior engineer who knows their shit about audio and video processing (importantly, including AV1 because the TX1 near certainly can't satisfactorily handle that); what's the rough range of time lag can we expect between the posting and then seeing the actual product utilizing their work?

 

[ILikeFeet]

My curiosity is more about that NoA job posting for a senior engineer who knows their shit about audio and video processing (importantly, including AV1 because the TX1 near certainly can't handle that); what's the rough range of time lag can we expect between the posting and then seeing the actual product utilizing their work?

that's probably impossible to know because of too many variables

 

[Look over there]

Mm, true.
Do we remember what job listings related to the original Switch showed up and how far back before launch?

 

[MP!]

Devkits have been out since late 2020, reportedly.

that seems very un nintendo like aren't their dev kits usually out to devs like 6 months- 1 year before launch?
I wonder what happened here.

 

[Dakhil]

that seems very un nintendo like aren't their dev kits usually out to devs like 6 months- 1 year before launch?
I wonder what happened here.

I imagine only a very small group of third party developers that Nintendo trusts (e.g. Capcom, Square Enix, etc.) received devkits in late 2020, considering NateDrake mentioned that that the distribution of devkits before E3 2021 were limited, and that devkits were planned to become accessible to more third party developers during the week of E3 2021.

 

[Skittzo]

that seems very un nintendo like aren't their dev kits usually out to devs like 6 months- 1 year before launch?
I wonder what happened here.

On top of what Dakhil said these were very preliminary devkits that emulated the target hardware. I think typically they get the final, real hardware devkits out ~6 months before launch or so.

 

[MP!]

I imagine only a very small group of third party developers that Nintendo trusts (e.g. Capcom, Square Enix, etc.) received devkits in late 2020, considering NateDrake mentioned that that the distribution of devkits before E3 2021 were limited, and that devkits were planned to become accessible to more third party developers during the week of E3 2021.

On top of what Dakhil said these were very preliminary devkits that emulated the target hardware. I think typically they get the final, real hardware devkits out ~6 months before launch or so.

Yeah I suppose I was thinking of finalized dev kits.
Still seems pretty early to me but makes sense.

 

[9-Volt]

Everything we've heard from unofficial sources still says whatever this is is launching before early 2023. Nothing has changed about that.

I can only believe in two outcomes:

• Dane is a full succesor, coming in late 2023/early 2024
• Dane is a pro version of current Switch, coming in late 2022/early 2023

I just can't wrap my head around a situation where Nintendo would let one of their fastest selling platform go without having a third revision, Pro or not. To me the least likeliest situation is Dane being a full successor releasing very soon, Nintendo killing Switch early for no good reason. Switch will die almost instantly when a full successor arrives as almost all third party developers are looking forward to a better hardware that'll make porting process a lot easier. This is not a GBA/DS situation, keeping Switch alive after the launch of the successor is pretty much meaningless.

There's still life left in Switch and I expect Nintendo would not waste it by launching a full successor soon.

 

[Skittzo]

I can only believe in two outcomes:

• Dane is a full succesor, coming in late 2023/early 2024
• Dane is a pro version of current Switch, coming in late 2022/early 2023

I just can't wrap my head around a situation where Nintendo would let one of their fastest selling platform go without having a third revision, Pro or not. To me the least likeliest situation is Dane being a full successor releasing very soon, Nintendo killing Switch early for no good reason. Switch will die almost instantly when a full successor arrives as almost all third party developers are looking forward to a better hardware that'll make porting process a lot easier. This is not a GBA/DS situation, keeping Switch alive after the launch of the successor is pretty much meaningless.

There's still life left in Switch and I expect Nintendo would not waste it by launching a full successor soon.

There's also the third possibility, that it's neither a full successor nor a pro version and instead something more like a GBC that's meant to be the new baseline for the next 5-6 years.

 

[9-Volt]

There's also the third possibility, that it's neither a full successor nor a pro version and instead something more like a GBC that's meant to be the new baseline for the next 5-6 years.

Yeah that's what I meant by "Pro" version. GBC is what I had in mind, not New 3DS or DSi.

Hell I even expect different colored cartridges for the "Pro" Switch.

 

[Hermii]

I can only believe in two outcomes:

• Dane is a full succesor, coming in late 2023/early 2024
• Dane is a pro version of current Switch, coming in late 2022/early 2023

I just can't wrap my head around a situation where Nintendo would let one of their fastest selling platform go without having a third revision, Pro or not. To me the least likeliest situation is Dane being a full successor releasing very soon, Nintendo killing Switch early for no good reason. Switch will die almost instantly when a full successor arrives as almost all third party developers are looking forward to a better hardware that'll make porting process a lot easier. This is not a GBA/DS situation, keeping Switch alive after the launch of the successor is pretty much meaningless.

There's still life left in Switch and I expect Nintendo would not waste it by launching a full successor soon.

One thing that's certain, is that game development for Switch will continue for a long time after new hardware arrives. They will not abandoned the current 100 million install base lightly, especially not during a chip shortage.

 

[9-Volt]

One thing that's certain, is that game development for Switch will continue for a long time after new hardware arrives. They will not abandoned the current 100 million install base lightly, especially not during a chip shortage.

Third parties might. There are tons of studios and some publishers who are willing to ignore the massive install base of Switch just to be not limited by its outdated hardware. Devs of Desperados 3 said that they could have ported the game if they pushed hard enough but instead they chose to wait until a new and more powerful Nintendo console arrives. There are a lot more studios think like this, all Switch could get after the launch of Dane could be Cloud games from third parties. The age of "miracle port" will conclude.

Switch is already missing some important but possible titles, 100m sales point isn't exactly changing minds.

 

[Dakhil]

One thing that's certain, is that game development for Switch will continue for a long time after new hardware arrives. They will not abandon the current 100 million install base lightly, especially not during a chip shortage.

Just to add to 9-Volt's comment, there are also third party developers who seem to refuse to bring games to Nintendo platforms due to petty politics (e.g. Activision, EA, etc.).

 

[9-Volt]

I may be misremembering this, but Josef Fares said that Switch can't handle It Takes Two but he'd consider bringing it to a next system. There are lots of studios waiting for a Switch successor to save them from Switch's limitations, so it's very possbile that out of gate this Dane will have gigantic third party support.

Just to add to 9-Volt's comment, there are also third party developers who seem to refuse to bring games to Nintendo platforms due to petty politics (e.g. Activision, EA, etc.).

Eh, publishers are where money is but devs care about ease of development too. There are no grudges between corporations, what Switch is missing is purely because of technical limitations and false marketing researches.

Activision brought everything to Switch but Call of Duty and that was because they probably couldn't get a well running downport of their liking. EA on the other hand, gives their studios some freedom. Those who don't want headaches caused by aged hardware are allowed to ignore Switch.

 

[Dakhil]

EA on the other hand, gives their studios some freedom. Those who don't want headaches caused by aged hardware are allowed to ignore Switch.

I can understand technically demanding games not coming to the Nintendo Switch. But the problem is that games that aren't inherently technically demanding (e.g The Sims) are also being skipped on the Nintendo Switch. In fact, an investor asked EA why EA hadn't bother bringing The Sims to the Nintendo Switch.

 

[Tentacle-tropes]

I may be misremembering this, but Josef Fares said that Switch can't handle It Takes Two but he'd consider bringing it to a next system. There are lots of studios waiting for a Switch successor to save them from Switch's limitations, so it's very possbile that out of gate this Dane will have gigantic third party support.

And then they will find another reason why they can’t bring it over. I’ve heard and read this song and dance before from forums and third parties. I’m not expecting the Dane to open the floodgates on this front and anyone who is; are going to probably be disappointed.

Eh, publishers are where money is but devs care about ease of development too. There are no grudges between corporations, what Switch is missing is purely because of technical limitations and false marketing researches.

Activision brought everything to Switch but Call of Duty and that was because they probably couldn't get a well running downport of their liking. EA on the other hand, gives their studios some freedom. Those who don't want headaches caused by aged hardware are allowed to ignore Switch.

There are quite a few games that could come to Switch now that would not be limited by any sort of technical limitations. Third parties also consistently miss the boat with Nintendo systems quite frequently. They also say the weirdest things only for Nintendo systems. Crash wasn’t on the Switch because they believed there was no audience for it. The Switch keeps getting legacy versions of FIFA. I can only buy technical excuses for why games are not coming for so long especially when you have games like The Witcher getting ported. As for CoD it probably because they were too busy to actually start a port and deemed not worth it roi wise to port one rather than “not to their liking”. I’m sure EA will cotinue throwing legacy FIFA titles and smaller games at the system. Maybe they’ll throw a bone every now and then by porting a bigger game years after the fact.