• Hey everyone, staff have documented a list of banned content and subject matter that we feel are not consistent with site values, and don't make sense to host discussion of on Famiboards. This list (and the relevant reasoning per item) is viewable here.

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

I want to answer, but I'm missing a verb here, not unlike "I accidentally a piano bench".
*potentially improve of the rest of the hardware :p
?

I wasn't comparing to the OG PS4?

I was comparing to the PS4 Pro, the range of Dane docked likely falls between 70-90% of it's GPU power.

Therefore, the 1050 being stronger than the PS4 non-pro, puts Dane on the PS4 Pro side of that gap
But Dane would not be at a 1050TI before DLSS :p
 
*potentially improve of the rest of the hardware :p

But Dane would not be at a 1050TI before DLSS :p
They could potentially see Wi-Fi 6 boards come down in price so it would support 802.11ax, which isn't very likely right now.
Maybe that new joystick tech we saw a few pages back?
Further internal storage speed improvements as newer eUFS standards bring prices down for the rest of it?
Marginally better/cheaper batteries?

A delay likely means more options for cheaper and/or better parts for everything else unrelated to the SoC.
 
But I will say that OLED model released in the midst of this rise in component pricing and Nintendo still only pushed the price up by $50, a price bump intended to preserve their target profit margin over a standard Mariko Switch model.
And Nintendo mentioned that the profit margin of the OLED model wouldn't increase compared to the Nintendo Switch.

And I may also say, Orin likely will have a notable per-FLOP increase over Ampere due to the doubling of L2 and the 50% more L1 Cache which is a direct adressment of Ampere's inability to feed all it's shader cores fast enough.
Orin's GPU actually has 50% less L2 cache compared to GA102 (4 MB vs 6 MB). And although Orin's GPU does have 50% more L1 cache in comparison to GA102 (192 KB vs 128 KB), Orin's GPU actually has the same amount of L1 cache as GA100.

Further internal storage speed improvements as newer eUFS standards bring prices down for the rest of it?
As of 9 August 2018, UFS 4.0 is planned to be announced at 2H 2022. So UFS 4.0 probably won't start being used in flagship smartphones until late 2023 at the earliest.
 
Last edited:
And Nintendo mentioned that the profit margin of the OLED model wouldn't increase compared to the Nintendo Switch.


Orin's GPU actually has 50% less L2 cache compared to the GA102 (4 MB vs 6 MB). And although Orin's GPU does have 50% more L1 cache in comparison to GA102 (192 KB vs 128 KB), Orin's GPU actually has the same amount of L1 cache as GA100.


As of 9 August 2018, UFS 4.0 is planned to be announced at 2H 2022. So UFS 4.0 probably won't start being used in flagship smartphones until late 2023 at the earliest.
Well, yeah, that’s what I mean in case I wasn’t clear. In spite of launching in the midst of a component price hike and using a more expensive screen + greater-capacity eMMC chips + OLED driver chip, etc etc etc, all it took to preserve the same margin as an ordinary Switch was $50. That’s cheaper than I would expect in an market going through price hikes.
 
0
It won’t be Switch 4K because people without 4K TVs will think they don’t need it - you want everyone to want your hardware. A lot of games probably won’t hit 4K either so they won’t want to advertise it in that way.

I don't agree with this sentiment. They just launched OLED Model and it's selling as well as anyone could hope in this environment. It even caused Nintendo to not be the best-selling hardware the month prior to launch in the US (breaking a historical streak) because people were clearly anticipating this new model.

You guys are taking 4K too literally. The general public understands 4K to mean higher quality. That's all Nintendo needs it to mean too. Just like when people got HD TVs, the average person didn't care if they just watched SD content on it. It was new and "better" so that's what they purchased. The same is happening with 4K TVs. I bet most parents and your average adults can't tell you the difference between a 1080p and 4K show on Netflix.

If Nintendo put 4K in the name, you can bet they'd market it all the features. OLED Model is essentially an upgrade for portable mode, Nintendo still pushed the new dock with LAN port in their marketing. They will make sure all the new features of the next Switch are pushed hard. And even if it's not called "4K" something, they are going to push the hell out of the fact that games will look better on your 4K TV. Why wouldn't they? It's going to have improved graphics at higher resolutions than 1080. That's going to be the main selling point. What's an easy shortcut to say this system can do better graphics? Maybe use a term that is well integrated in the culture and is clearly understood as a higher fidelity like 4K.

I'm not saying they will use 4K, but after they decided to go with Nintendo Switch OLED Model, I wouldn't be dismissive of a name using a specific technology.
 
Curious if the new device will get the full RE Engine support. Current switch has a more customized form of other engines to make use of the lower spec profile that the device offers as best as possible. It has CryEngine as well, and even Frostbite I believe with Plants vs Zombies. ID Tech 7…. RED Engine as well.

Does anyone know what engines the current switch does not support in anyway? I know Luminous from Square Enix and RAGE from Rockstar are absent, but what else is absent?


Décima is clearly a no go, as it is a proprietary engine that only PC and PS will utilize.

But outside of that, these engine supports can help with multiplatform games in any way possible for developers.


I also already expect a more customized form of UE5 for the next model to have any porting process go as smooth as possible.
 
So today, IBM (alongside Samsung Electronics) released a press release announcing a breakthrough in semiconductor design with vertical transport field-effect transistors (VTFETs), which offers 2x improved performance or 85% reduced energy consumption than fin field-effect transistors (FinFETs), which allows for semiconductor device scaling to potentially continue beyond nanosheet transistors, as well as allow cell phone batteries to last for over a week, instead of for days, without being charged, and etc. And IBM has confirmed to use Samsung's 5 nm** process node for fabricating the IBM Power11.

Assuming that Samsung can achieve performance or power efficiency metrics very similar to the theoretical performance or power efficiency metrics during mass manufacturing, and assuming Nintendo will continue to work with Nvidia at least a decade from now, I think Nintendo and Nvidia would definitely benefit from continuing to use Samsung's process nodes for chip fabrication in the future, considering that Samsung will probably introduce the usage of VTFETs for process nodes beyond 2 nm** (or 20 Å**).
 
So today, IBM (alongside Samsung Electronics) released a press release announcing a breakthrough in semiconductor design with vertical transport field-effect transistors (VTFETs), which offers 2x improved performance or 85% reduced energy consumption than fin field-effect transistors (FinFETs), which allows for semiconductor device scaling to potentially continue beyond nanosheet transistors, as well as allow cell phone batteries to last for over a week, instead of for days, without being charged, and etc. And IBM has confirmed to use Samsung's 5 nm** process node for fabricating the IBM Power11.

Assuming that Samsung can achieve performance or power efficiency metrics very similar to the theoretical performance or power efficiency metrics during mass manufacturing, and assuming Nintendo will continue to work with Nvidia at least a decade from now, I think Nintendo and Nvidia would definitely benefit from continuing to use Samsung's process nodes for chip fabrication in the future, considering that Samsung will probably introduce the usage of VTFETs for process nodes beyond 2 nm** (or 20 Å**).
If IBM was still designing Nintendo's CPUs, then maybe we'd see this sooner in a Switch hehe just kidding
 
Curious if the new device will get the full RE Engine support. Current switch has a more customized form of other engines to make use of the lower spec profile that the device offers as best as possible. It has CryEngine as well, and even Frostbite I believe with Plants vs Zombies. ID Tech 7…. RED Engine as well.

Does anyone know what engines the current switch does not support in anyway? I know Luminous from Square Enix and RAGE from Rockstar are absent, but what else is absent?


Décima is clearly a no go, as it is a proprietary engine that only PC and PS will utilize.

But outside of that, these engine supports can help with multiplatform games in any way possible for developers.


I also already expect a more customized form of UE5 for the next model to have any porting process go as smooth as possible.
calling UE5 "customized" feels wrong. "optimized", sure. but the days of separate engines is dead. now RE Engine... I have no idea.
 
So today, IBM (alongside Samsung Electronics) released a press release announcing a breakthrough in semiconductor design with vertical transport field-effect transistors (VTFETs), which offers 2x improved performance or 85% reduced energy consumption than fin field-effect transistors (FinFETs), which allows for semiconductor device scaling to potentially continue beyond nanosheet transistors, as well as allow cell phone batteries to last for over a week, instead of for days, without being charged, and etc. And IBM has confirmed to use Samsung's 5 nm** process node for fabricating the IBM Power11.

Assuming that Samsung can achieve performance or power efficiency metrics very similar to the theoretical performance or power efficiency metrics during mass manufacturing, and assuming Nintendo will continue to work with Nvidia at least a decade from now, I think Nintendo and Nvidia would definitely benefit from continuing to use Samsung's process nodes for chip fabrication in the future, considering that Samsung will probably introduce the usage of VTFETs for process nodes beyond 2 nm** (or 20 Å**).
That's a fascinating breakthrough if it becomes easily mass marketable.
 
0
I wonder if by february the SoC will be taped out already, maybe we'll have an idea of how strong the device will end up being at that point, but on the same token, maybe Nintendo will want to announce it at that point if it's release was no more than 6-8 months away to prevent leaks. In the end it'll be what it'll be though.
 
I wonder if by february the SoC will be taped out already, maybe we'll have an idea of how strong the device will end up being at that point, but on the same token, maybe Nintendo will want to announce it at that point if it's release was no more than 6-8 months away to prevent leaks. In the end it'll be what it'll be though.
Probably a rough idea based on the physical hardware configuration (the number of CPU cores, the number of CUDA cores, etc.) after tape out. The frequencies however are probably being continuously tweaked until the last possible moment. And the amount of RAM probably can also be tweaked until the last possible moment, such as when Capcom asked Nintendo to increase the amount of RAM on the Nintendo Switch.
 
They do things they don't like all the time. Leaving out all the times they've sold consoles at a loss as I detailed earlier, they don't like announcing consoles early after what happened with N64, but they did it with "NX" because they felt that they needed to. Nintendo doesn't like doing price cuts (because no one does if they don't have to), but they have (this is actually the first generation Nintendo has gone without a single price cut).

Are people going to spend at least $100 more than an Xbox Series S for a portable hybrid device that underperforms a Series S and does not have its attendant market appeal with Game Pass? We have to seriously consider that part of why Switch has performed as well as it has is because of its price point.
Well folk could've bought a PS4/XB1 that outperforms a Switch from Day 1 at the same price, if not less with bundles or sales. The Switch OLED is so far doing well, despite being $350. Meanwhile we've had more affordable Nintendo systems just sell like shit. I don't think price has helped the Switch anymore than other systems
 
Probably a rough idea based on the physical hardware configuration (the number of CPU cores, the number of CUDA cores, etc.) after tape out. The frequencies however are probably being continuously tweaked until the last possible moment. And the amount of RAM probably can also be tweaked until the last possible moment, such as when Capcom asked Nintendo to increase the amount of RAM on the Nintendo Switch.
Hopefully we know by February or March then, though whatever we get to know will depend on when they plan on releasing it.
 
Hopefully we know by February or March then, though whatever we get to know will depend on when they plan on releasing it.
Reports from news outlets, such as Bloomberg, Wall Street Journal, Nikkei, etc., on when Nintendo plans to start mass manufacturing the DLSS model* is probably a good indication of when Nintendo plans to launch the DLSS model*.
 
Last edited:
So today, IBM (alongside Samsung Electronics) released a press release announcing a breakthrough in semiconductor design with vertical transport field-effect transistors (VTFETs), which offers 2x improved performance or 85% reduced energy consumption than fin field-effect transistors (FinFETs), which allows for semiconductor device scaling to potentially continue beyond nanosheet transistors, as well as allow cell phone batteries to last for over a week, instead of for days, without being charged, and etc. And IBM has confirmed to use Samsung's 5 nm** process node for fabricating the IBM Power11.

Assuming that Samsung can achieve performance or power efficiency metrics very similar to the theoretical performance or power efficiency metrics during mass manufacturing, and assuming Nintendo will continue to work with Nvidia at least a decade from now, I think Nintendo and Nvidia would definitely benefit from continuing to use Samsung's process nodes for chip fabrication in the future, considering that Samsung will probably introduce the usage of VTFETs for process nodes beyond 2 nm** (or 20 Å**).
Is there any estimate or news on mass production of chips with this technology? And most importantly, will this be an exclusive of Samsung's foundries and chips ordered by IBM or will this be licensed to other foundries? (based on past behaviour).

This doesn't sound like a technlogy that could be used in a Switch 2, but is there any vague roadmap of when we could start to see high end devices that use it for chips or batteries? (Like 2025 iphones/galaxies, just to understand if this is something concrete or is another "graphene will create infinite batteries" scenario).
 
Is there any estimate or news on mass production of chips with this technology? And most importantly, will this be an exclusive of Samsung's foundries and chips ordered by IBM or will this be licensed to other foundries? (based on past behaviour).

This doesn't sound like a technlogy that could be used in a Switch 2, but is there any vague roadmap of when we could start to see high end devices that use it for chips or batteries? (Like 2025 iphones/galaxies, just to understand if this is something concrete or is another "graphene will create infinite batteries" scenario).
This was just an announcement of "hey we did it in a lab" by IBM research team that work in collab with plenty of public and private entities, including Samsung and Intel. If they are able to implement it at a mass production scale, then for sure the other players will follow suit as it basically breathes more life into Moore's Law beyond 1 nm and it gives more breathing room to people whose jobs are to seriously ponder what comes after Moore's Law. No specifics for licensing and stuff, but if it's successful, everyone and everything in the semiconductor industry will benefit from it.
 
Last edited:
They do things they don't like all the time. Leaving out all the times they've sold consoles at a loss as I detailed earlier, they don't like announcing consoles early after what happened with N64, but they did it with "NX" because they felt that they needed to. Nintendo doesn't like doing price cuts (because no one does if they don't have to), but they have (this is actually the first generation Nintendo has gone without a single price cut).

Are people going to spend at least $100 more than an Xbox Series S for a portable hybrid device that underperforms a Series S and does not have its attendant market appeal with Game Pass? We have to seriously consider that part of why Switch has performed as well as it has is because of its price point.
We'll see.. But Nintendo is pretty confident with the current Switch sales. With how OLED is priced, I think a $400 console is likely at this point. I can't imagine a $50 price difference between OLED at $300 in 2023 and $350 for switch 2, even if storage is 128GB or less.

And what Switch 2 has that x series s doesn't, is portability. Mobile will always cost more than home anyway.
*potentially improve of the rest of the hardware :p

But Dane would not be at a 1050TI before DLSS :p
I think if Dane is at least 1.4 TFLOPs, there's a good chance it will match OG PS4 at least. I'm not thinking about PS4 pro just yet or any architecture gains from Orion that it may have over ampere .. not pre DLSS at least.
I don't agree with this sentiment. They just launched OLED Model and it's selling as well as anyone could hope in this environment. It even caused Nintendo to not be the best-selling hardware the month prior to launch in the US (breaking a historical streak) because people were clearly anticipating this new model.

You guys are taking 4K too literally. The general public understands 4K to mean higher quality. That's all Nintendo needs it to mean too. Just like when people got HD TVs, the average person didn't care if they just watched SD content on it. It was new and "better" so that's what they purchased. The same is happening with 4K TVs. I bet most parents and your average adults can't tell you the difference between a 1080p and 4K show on Netflix.

If Nintendo put 4K in the name, you can bet they'd market it all the features. OLED Model is essentially an upgrade for portable mode, Nintendo still pushed the new dock with LAN port in their marketing. They will make sure all the new features of the next Switch are pushed hard. And even if it's not called "4K" something, they are going to push the hell out of the fact that games will look better on your 4K TV. Why wouldn't they? It's going to have improved graphics at higher resolutions than 1080. That's going to be the main selling point. What's an easy shortcut to say this system can do better graphics? Maybe use a term that is well integrated in the culture and is clearly understood as a higher fidelity like 4K.

I'm not saying they will use 4K, but after they decided to go with Nintendo Switch OLED Model, I wouldn't be dismissive of a name using a specific technology.
I don't think they will flaunt switch 2 as a 4k device. More so on an individual game level for 1st party games maybe while the regular switch is still around. They haven't flaunted power since GameCube days.
~545GFLOPs of Maxwell in the Switch is loosely equal to 820 GFLOPs PS4 custom GCN which is loosely equal to 690-720 or so Ampere GFLOPs for 720p on a portable screen.


I should really stress that I use the term loosely here since this is not as simple to translate from one to another to another.

There’s a laundry list of factors that needed to be factored in, bandwidth? Bit bus width? Clock speed? Core count? Etc.

So the comparison is really more difficult to pinpoint.


I should mention that the PS4 was 300 dollars for a long time like the Switch and we see how that turned out :)
I'm curious... where did you hear about Maxwell being 50% more performant efficient than PS4's GCN GPU?

I've heard switch can perform 33-50% more than on paper from a combination of it's more efficient architecture and if the right engine is used and optimized (UE4) that take advantage of it mixed precision mode as well for games like Snake pass. But from second hand. from z0mbie who heard form a Ubisoft dev, if I recall 🤔 But it's really up in the air.
 
Last edited:
Is there any estimate or news on mass production of chips with this technology? And most importantly, will this be an exclusive of Samsung's foundries and chips ordered by IBM or will this be licensed to other foundries? (based on past behaviour).

This doesn't sound like a technlogy that could be used in a Switch 2, but is there any vague roadmap of when we could start to see high end devices that use it for chips or batteries? (Like 2025 iphones/galaxies, just to understand if this is something concrete or is another "graphene will create infinite batteries" scenario).
There's no official estimate or news about when chips designed with VTFETs are going to be mass manufactured. But assuming that Samsung starts to offer process nodes using VTFETs a couple of years after Samsung's 2GAP process node is available to consumers, I presume early 2028 at the absolute earliest, assuming Samsung doesn't encounter any problems.

And I think there's a possibility Samsung will be the first foundry company to offer process nodes using VTFETs to consumers in the same way that Samsung's the first foundry company to offer process nodes with GAAFETs to consumers with Samsung's 3GAE and 3GAP process nodes. But considering IBM has announced a research collaboration with Intel on next-generation logic and packaging technologies, and assuming Intel continues to have an aggressive process node roadmap after Intel 18A, I think there's also a possibility Intel will offer process nodes using VTFETs to consumers at around the same time as Samsung, or before Samsung.
I don't think they will flaunt switch 2 as a 4k device. More so on an individual game level for 1st party games maybe while the regular switch is still around. They haven't flaunted power since GameCube days.
Personally, I expect Nintendo to only mention that the DLSS model* can output to as high as 4K when in TV mode.
 
0
And Nintendo mentioned that the profit margin of the OLED model wouldn't increase compared to the Nintendo Switch.


Orin's GPU actually has 50% less L2 cache compared to GA102 (4 MB vs 6 MB). And although Orin's GPU does have 50% more L1 cache in comparison to GA102 (192 KB vs 128 KB), Orin's GPU actually has the same amount of L1 cache as GA100.


As of 9 August 2018, UFS 4.0 is planned to be announced at 2H 2022. So UFS 4.0 probably won't start being used in flagship smartphones until late 2023 at the earliest.

Again I've seen this brought up before but we have to use scale in this train of thought, that ultimately can't be excluded in this comparison.
Orin a 16SM part has 4MB of L2$ vs GA102 a 82SM part sharing 6MB is not even close in spec overall...
Even if Nvidia did absolutely nothing to the Ampere architecture and managed to scale up the increase in L2$ in equivalent proportion to Orin, GA102 on 8nm would have night and day performance gains over the current card (mind you it would be a monstrosity of an expensive chip).
 
Well folk could've bought a PS4/XB1 that outperforms a Switch from Day 1 at the same price, if not less with bundles or sales. The Switch OLED is so far doing well, despite being $350. Meanwhile we've had more affordable Nintendo systems just sell like shit. I don't think price has helped the Switch anymore than other systems
Doug Bowser confirmed that OLED model is like PS4 Pro - the majority of buyers are people/households who already own a Switch, and therefore is selling to people who understand the value proposition they’re getting and consider it worth it. That’s relatively easy, you’re selling them a more luxurious Switch experience they know and understand.
The same cannot be said for entirely new hardware, the value proposition resets. They need to incentivize buying it over a Switch (which would likely see a price drop by the time it’s introduced) or over other gaming options.
We'll see.. But Nintendo is pretty confident with the current Switch sales. With how OLED is priced, I think a $400 console is likely at this point. I can't imagine a $50 price difference between OLED at $300 in 2023 and $350 for switch 2, even if storage is 128GB or less.

And what Switch 2 has that x series s doesn't, is portability. Mobile will always cost more than home anyway.
This response tells me you never expect a Switch family price drop, which is… ambitious.
But again, I said price is PART of the reason for its success. Form factor, games, battery life, etc. also have their part to play. But it’s not difficult to imagine Switch in 2017 being $400 would have been a MUCH harder sell, so let’s not pretend that introductory pricing doesn’t play its part in things, or that Switch’s success does not make that an important consideration.

Hybrid design/portability is great, but is it “$100 more“ great? One has to seriously and legitimately consider that.
Nintendo has been fortunate enough that Xbox and PlayStation console availability has been significantly reduced for the past year to a point where their hardware could legitimately be the only thing on a shelf to buy in some circumstances and was also a great piece of hardware to weather the COVID storm with. What about when that’s not the case? Like, for example, around the time that they launch this new hardware? That’s something worthy of consideration instead of assuming the good times will roll forever.
 
calling UE5 "customized" feels wrong. "optimized", sure. but the days of separate engines is dead. now RE Engine... I have no idea.
I mean, in a way it is “customized” but that is more so just optimized for the different uArch that the Dane would have vs the other systems, not customized in the sense of how RE Engine is. So in the context I was referring to the customized form is the optimized form so to speak.


Even then, UE5 isn’t optimized quite yet for either either hardware.
Early dev kits could be anything from this to a modified 2060. Have we ever seen what pre-final dev kits looked like for last gen systems? I'd assume they were generic amd boxes
I remember it being mentioned by Richard that the Devkit for the PS5 had a PS4 Pro. I don’t remember what it was for the Series X but probably a similar scenario.
We'll see.. But Nintendo is pretty confident with the current Switch sales. With how OLED is priced, I think a $400 console is likely at this point. I can't imagine a $50 price difference between OLED at $300 in 2023 and $350 for switch 2, even if storage is 128GB or less.

And what Switch 2 has that x series s doesn't, is portability. Mobile will always cost more than home anyway.

I think if Dane is at least 1.4 TFLOPs, there's a good chance it will match OG PS4 at least. I'm not thinking about PS4 pro just yet or any architecture gains from Orion that it may have over ampere .. not pre DLSS at least.

I don't think they will flaunt switch 2 as a 4k device. More so on an individual game level for 1st party games maybe while the regular switch is still around. They haven't flaunted power since GameCube days.

I'm curious... where did you hear about Maxwell being 50% more performant efficient than PS4's GCN GPU?

I've heard switch can perform 33-50% more than on paper from a combination of it's more efficient architecture and if the right engine is used and optimized (UE4) that take advantage of it mixed precision mode as well for games like Snake pass. But from second hand. from z0mbie who heard form a Ubisoft dev, if I recall 🤔 But it's really up in the air.
The GCN in the PS4 does one or the other but not both at the same time despite having support for FP16 at an equal rate to FP32. Switch can do both at the same time, or it can do one or it can do the other. It does depend on the game, yes, but generally it can do above its weight. The feature it has is not found in other Maxwell GPUs but in Pascal GPUs, as Switch borrows some features from Pascal.

When looking at Ampere, it is doing both FP32 and FP16 at the same time, but executes them differently than the predecessor. It does the same amount of FP32 as the FP16 calculations per second unlike the Maxwell where it does double the FP16 calculations as the FP32 calculations.

It should also be noted that, Ampere and Maxwell are a lot more efficient with memory bandwidth than GCN of any type.


Curiously, I think for multi platform games they didn’t really use the FP16 on the PS4 as the XB1 lacked support for it I think. So the games were on FP32 only, maybe.


Edit: actually XBox one does support it, but the devices couldn’t use it at the same time, so they likely opted just for FP32 for their games. Makes sense though.
Again I've seen this brought up before but we have to use scale in this train of thought, that ultimately can't be excluded in this comparison.
Orin a 16SM part has 4MB of L2$ vs GA102 a 82SM part sharing 6MB is not even close in spec overall...
Even if Nvidia did absolutely nothing to the Ampere architecture and managed to scale up the increase in L2$ in equivalent proportion to Orin, GA102 on 8nm would have night and day performance gains over the current card (mind you it would be a monstrosity of an expensive chip).
Yeah, when looking at it from the lens of a console like this, it should have a lot more relatively speaking for what the device ultimately has (4MB for 16SMs vs 6MB for 82SMs).
 
Last edited:
Doug Bowser confirmed that OLED model is like PS4 Pro - the majority of buyers are people/households who already own a Switch, and therefore is selling to people who understand the value proposition they’re getting and consider it worth it. That’s relatively easy, you’re selling them a more luxurious Switch experience they know and understand.
The same cannot be said for entirely new hardware, the value proposition resets. They need to incentivize buying it over a Switch (which would likely see a price drop by the time it’s introduced) or over other gaming options.

This response tells me you never expect a Switch family price drop, which is… ambitious.
But again, I said price is PART of the reason for its success. Form factor, games, battery life, etc. also have their part to play. But it’s not difficult to imagine Switch in 2017 being $400 would have been a MUCH harder sell, so let’s not pretend that introductory pricing doesn’t play its part in things, or that Switch’s success does not make that an important consideration.

Hybrid design/portability is great, but is it “$100 more“ great? One has to seriously and legitimately consider that.
Nintendo has been fortunate enough that Xbox and PlayStation console availability has been significantly reduced for the past year to a point where their hardware could legitimately be the only thing on a shelf to buy in some circumstances and was also a great piece of hardware to weather the COVID storm with. What about when that’s not the case? Like, for example, around the time that they launch this new hardware? That’s something worthy of consideration instead of assuming the good times will roll forever.
Sure and I imagine a much more powerful iterative successor would do just that but my main point is that, from the very beginning, you had more powerful systems freely available to purchase that were at the same price, if not cheaper with bundles and sales, and that didn't harm the Switch at all so bringing up the starting price when the price for the majority of the PS4/XB1s lifespan was $300 anyways is rather pointless. Any system would have a harder time selling at a higher price point. That's nothing new. The PS4 being $400 instead of $500 like the XB1 or $600 like the PS3 helped it out as well, same with the PS5 being $500 instead of $600
 
Here's some rumours from China Times about Intel's plans with TSMC.



Assuming there are grains of truth in the rumours from China Times, Nintendo and Nvidia probably won't be using TSMC's process nodes any time soon.
 
0
Nintendo doesn't need cutting edge, they just need a sizable jump over what they have now. all of Qualcomm's problems with Samsung's 4nm doesn't mean shit to nintendo since they aren't looking for crumbs of gains every year
 
Doug Bowser confirmed that OLED mode
I mean, in a way it is “customized” but that is more so just optimized for the different uArch that the Dane would have vs the other systems, not customized in the sense of how RE Engine is. So in the context I was referring to the customized form is the optimized form so to speak.


Even then, UE5 isn’t optimized quite yet for either either hardware.

I remember it being mentioned by Richard that the Devkit for the PS5 had a PS4 Pro. I don’t remember what it was for the Series X but probably a similar scenario.

The GCN in the PS4 does one or the other but not both at the same time despite having support for FP16 at an equal rate to FP32. Switch can do both at the same time, or it can do one or it can do the other. It does depend on the game, yes, but generally it can do above its weight. The feature it has is not found in other Maxwell GPUs but in Pascal GPUs, as Switch borrows some features from Pascal.

When looking at Ampere, it is doing both FP32 and FP16 at the same time, but executes them differently than the predecessor. It does the same amount of FP32 as the FP16 calculations per second unlike the Maxwell where it does double the FP16 calculations as the FP32 calculations.

It should also be noted that, Ampere and Maxwell are a lot more efficient with memory bandwidth than GCN of any type.


Curiously, I think for multi platform games they didn’t really use the FP16 on the PS4 as the XB1 lacked support for it I think. So the games were on FP32 only, maybe.


Edit: actually XBox one does support it, but the devices couldn’t use it at the same time, so they likely opted just for FP32 for their games. Makes sense though.

Yeah, when looking at it from the lens of a console like this, it should have a lot more relatively speaking for what the device ultimately has (4MB for 16SMs vs 6MB for 82SMs).

l is like PS4 Pro - the majority of buyers are people/households who already own a Switch, and therefore is selling to people who understand the value proposition they’re getting and consider it worth it. That’s relatively easy, you’re selling them a more luxurious Switch experience they know and understand.
The same cannot be said for entirely new hardware, the value proposition resets. They need to incentivize buying it over a Switch (which would likely see a price drop by the time it’s introduced) or over other gaming options.

This response tells me you never expect a Switch family price drop, which is… ambitious.
But again, I said price is PART of the reason for its success. Form factor, games, battery life, etc. also have their part to play. But it’s not difficult to imagine Switch in 2017 being $400 would have been a MUCH harder sell, so let’s not pretend that introductory pricing doesn’t play its part in things, or that Switch’s success does not make that an important consideration.

Hybrid design/portability is great, but is it “$100 more“ great? One has to seriously and legitimately consider that.
Nintendo has been fortunate enough that Xbox and PlayStation console availability has been significantly reduced for the past year to a point where their hardware could legitimately be the only thing on a shelf to buy in some circumstances and was also a great piece of hardware to weather the COVID storm with. What about when that’s not the case? Like, for example, around the time that they launch this new hardware? That’s something worthy of consideration instead of assuming the good times will roll forever.
interestingly enough the PS4 pro GPU (Polaris) has mixed precision mode. Think there have been a few titles where it seemed to punch above it's weight and be closer in performance to x bone x, but held back by bandwidth..


@Terell
I did. I gave an example of OLED dropping to $300 (and the other models dropping $50) whenever switch 2 comes out, but that there would still be something like at least $100 price difference between switch 1 and the release.

I'm basing this off of despite being almost 5 years old now, Nintendo hasn't had a permanent price drop for switch at all. Still $300, and selling like crazy. The audacity of Nitnendo to not have a price drop and having the OLED at $350. They just have been throwing in Mario kart bundles during the holidays for $300, and let the retail outlets do the rest the rest. Nintendo is under new leadership and they are cocky with the switch selling well, but at the same time they aren't obligated to drop by any means at this time. A $100 price drop is inevitable, but I don't see that happening between now and holiday 2023. $50 is my guess. Would love to be proven wrong though. Of course I'd prefer to pay $350 for a switch 2, but I can't see them selling OLED at $300 *and switch 2 selling only $50 more at $350 at the same time. I certainly think we will get a price drop for OLED when switch 2 arrives, but I see a $100 price difference at least...
 
Last edited:
Sure and I imagine a much more powerful iterative successor would do just that but my main point is that, from the very beginning, you had more powerful systems freely available to purchase that were at the same price, if not cheaper with bundles and sales, and that didn't harm the Switch at all so bringing up the starting price when the price for the majority of the PS4/XB1s lifespan was $300 anyways is rather pointless. Any system would have a harder time selling at a higher price point. That's nothing new. The PS4 being $400 instead of $500 like the XB1 or $600 like the PS3 helped it out as well, same with the PS5 being $500 instead of $600
PS4 and Xbox One also happened to be 3 and a half years old when Switch launched, which was right before sales of both of them were beginning to slow down (PS4 went from ~20mil sales in CY 2017 to ~13mil the next year and continued sliding until the PS5's release), which meant that consumers were interested in new hardware but Sony and Microsoft were in little hurry to offer anything. Meanwhile, when they finally did, they were rare as hen's teeth and high targets for scalping through their entire first year, a trend likely to continue through to 2023.
While a lot of kudos go to Nintendo for an outstanding product, the conditions which it entered the market and market conditions through its life were also VERY favourable to them and it's worth considering not taking those conditions for granted as always being there.
@Terell
I did. I gave an example of OLED dropping to $300 (and the other models dropping $50) whenever switch 2 comes out, but that there would still be something like at least $100 price difference between switch 1 and the release.

I'm basing this off of despite being almost 5 years old now, Nintendo hasn't had a permanent price drop for switch at all. Still $300, and selling like crazy. The audacity of Nitnendo to not have a price drop and having the OLED at $350. They just have been throwing in Mario kart bundles during the holidays for $300, and let the retail outlets do the rest the rest. Nintendo is under new leadership and they are cocky with the switch selling well, but at the same time they aren't obligated to drop by any means at this time. A $100 price drop is inevitable, but I don't see that happening between now and holiday 2023. Would love to be proven wrong though. Of course I'd prefer to pay $350 for a switch 2, but I can't see them swelling OLED at $300, let alone $250.
Market conditions have allowed them to take advantage of the situation and continue selling Switch at its current price. And y'know, business wise, I can't fault them for that, especially when they made it plain that Switch's market strategy was "let's use this hardware cycle to make up as much of the money we lost from our previous hardware cycle as possible", which wasn't exactly small potatoes, if anyone happens to recall. So it's doing precisely what it's meant to do in Nintendo's eyes.

FY2021 will mark the first decline in yearly hardware shipments, and while it's super-easy to blame ALL of that on production challenges, Nintendo will not be so short-sighted as to think that's the only possible reason. Nintendo's projections have their sales for the second half of their FY at 15.72 million units. Coupled with the fact that Nintendo makes conservative estimates so that they look like achievers to their investors, should they miss their forecast, it will mark a downturn in sales beyond what can be blamed on shortages. And despite a lot of bullishness from several people, that's not a guarantee, but I don't know if getting into the real nitty-gritty of sales data is what anyone wants from this thread any more than I've already provided.

And about pricing... well, to us in the enthusiast set, $100 is no big deal, but you don't sell 27 million units of hardware in a year to enthusiasts alone. Also, what is realistically or logically not a big difference in price and what is perceived to be a big difference in price are 2 different things. We live in a world where a penny off a price tag psychologically tricks us into thinking something's cheaper than it is, one should not discount the power of perception. I've been setting retail prices for the store I help operate for the better part of 3 years now and... yeah, that shit isn't as simple as some think it is.

The object with new hardware is to keep and expand your marketshare, not take it for granted. And with the next hardware lacking some of the attendant advantages Switch had during its cycle, being all "oh yeah, $400, the market will absolutely and unquestionably accept that" isn't being mindful of everything that goes into a price determination and consumer price perceptions.
 
Nintendo doesn't need cutting edge, they just need a sizable jump over what they have now. all of Qualcomm's problems with Samsung's 4nm doesn't mean shit to nintendo since they aren't looking for crumbs of gains every year
I'm not talking about cutting edge process nodes, but rather relatively mature, but still advanced process nodes. So I'm primarily talking primarily about 7 nm** and 6 nm** process nodes.
TSMC's 7 nm** and N6 process nodes are obviously out of the question since Nvidia's exclusively using TSMC's N7 process node to fabricate datacentre chips; and AMD, Intel, Mediatek, and Qualcomm are using TSMC's N6 process node for chip fabrication in the foreseeable future, which probably leaves very little available capacity for TSMC's N6 process node that Nintendo and Nvidia can use. (The reason I didn't mention TSMC's N7P process node is because TSMC mentioned that most consumers will transition to TSMC's N6 process node, which means that TSMC's N7P process node's probably no longer be available to Nintendo and Nvidia, unless Nvidia already secured capacity for TSMC's N7P process node in the past.)
And considering that Nvidia originally planned on using Samsung's 7LPP process node before ultimately using Samsung's 8N process node for fabricating consumer Ampere GPUs, Andrei Frumusanu mentioning that the binning of the Exynos 990 SoCs is really bad, the Snapdragon 765 being the only chip from Qualcomm being fabricated using Samsung's 7LPP process node (the Snapdragon 765G and the Snapdragon 768G are binned versions of the Snapdragon 765), and IBM and Tesla are confirmed and rumoured respectively to use Samsung's 7LPP process node to fabricate IBM Telum and HW 4.0, which are probably not high volume chips, yields for Samsung's 7LPP process node are probably still pretty bad. And considering that no company, including Samsung, has announced any chip(s) that are fabricated using Samsung's 6LPP process node, considering that Samsung's 6LPP process node seems to only be available for Samsung, Samsung's 6LPP process is out of the question for Nintendo and Nvidia.
Also, Qualcomm's rumoured to use Samsung's 4LPX process node, which is apparently a custom variant of Samsung's 5LPP process node, to fabricate the Snapdragon 8 Gen 1, which I'm inclined to believe, considering the performance improvements of the Cortex-X2 on the Snapdragon 8 Gen 1 has over the Cortex-X1 on the Snapdragon 888 are not very high. And I believe Samsung's 4LPE process node, which is rumoured to be used to fabricate the Exynos 2200(?), is the process node with the bad yields, not Samsung's 4LPX process node.

And that's why I'm still inclined to believe that Nintendo and Nvidia are using Samsung's 8N process node to fabricate Dane. However, I don't believe that Nintendo's going to be doomed for using Samsung's 8N process node to fabricate Dane, since process nodes alone aren't enough to achieve performance and power efficiency improvements, since how the CPU and GPU architectures are designed is also very important.

And I've mentioned how the advantages of using cutting edge process nodes are becoming less and less obvious, for reasons I've mentioned here.
 
Last edited:
0
PS4 and Xbox One also happened to be 3 and a half years old when Switch launched, which was right before sales of both of them were beginning to slow down (PS4 went from ~20mil sales in CY 2017 to ~13mil the next year and continued sliding until the PS5's release), which meant that consumers were interested in new hardware but Sony and Microsoft were in little hurry to offer anything. Meanwhile, when they finally did, they were rare as hen's teeth and high targets for scalping through their entire first year, a trend likely to continue through to 2023. While a lot of kudos go to Nintendo for an outstanding product, the conditions which it entered the market and market conditions through its life were also VERY favourable to them and it's worth considering not taking those conditions for granted as always being there.
People don't buy new hardware just because. They don't go "Well these systems are a few years old now so I guess I'll look for a new system just to get a system that DOESN'T get my CoDs, Fifas, GTAs, etc. because it's new, I guess". Never mind that the Switch came off of the failing Wii U while the PS4/XB1 were at their height, just as cheap, if not cheaper, and well established at that point, which are all points you, for some reason, want to ignore. Plenty of factors that could've worked against the Switch as well as help it. Plus, despite the shortages, the PS5 and XSX are or were selling in line with the PS4/XB1 were at the same point in their life, so it's not as if folk aren't getting them nor is the shortage limited to them, seeing as the Switch also has to deal with it. Never mind that other systems being old sure didn't help out the Wii U or systems like the 3DS initially and Vita

I'll be frank, this just sounds like "The Switch is doing well because of good luck", which you could apply to any system. I could also say the PS4 did well because the XB1 was $500 initially and had that used games debacle and the Wii U was just a shittily designed system all around and they shouldn't expect such a favorable condition. That the PS5/XSX are so in demand because people are staying home during the pandemic and really wanted new hardware after the PS4/XB1 were around for a while and they shouldn't expect such a favorable condition again. If the products aren't desirable themselves, they're not gonna sell well
 
Last edited:
People don't buy new hardware just because. They don't go "Well these systems are a few years old now so I guess I'll look for a new system just to get a system that DOESN'T get my CoDs, Fifas, GTAs, etc. because it's new, I guess". Never mind that the Switch came off of the failing Wii U while the PS4/XB1 were at their height, just as cheap, if not cheaper, and well established at that point, which are all points you, for some reason, want to ignore. Plenty of factors that could've worked against the Switch as well as help it. Plus, despite the shortages, the PS5 and XSX are or were selling in line with the PS4/XB1 were at the same point in their life, so it's not as if folk aren't getting them nor is the shortage limited to them, seeing as the Switch also has to deal with it. Never mind that other systems being old sure didn't help out the Wii U or systems like the 3DS initially and Vita

I'll be frank, this just sounds like "The Switch is doing well because of good luck", which you could apply to any system. I could also say the PS4 did well because the XB1 was $500 initially and had that used games debacle and the Wii U was just a shittily designed system all around and they shouldn't expect such a favorable condition. That the PS5/XSX are so in demand because people are staying home during the pandemic and really wanted new hardware after the PS4/XB1 were around for a while and they shouldn't expect such a favorable condition again. If the products aren't desirable themselves, they're not gonna sell well
You're right, I'm sorry, how dare I analyze perfection. They'll launch at $400 and it will have absolutely zero negative impact on its saleability, because nothing can stop the Switch hype train ever. Hell, by what I'm reading here, it'll be bulletproof at $500. Why not? It's not like terrific products fail to sell for what seem like inconsequential reasons to some or anything. The good times will never possibly end, how dare I suggest to the contrary.

Can this conversation that started because I dared to say that price was (emphasis mine since it was part of the post everyone seemingly ignored) ONE OF MANY factors of consideration for a company wanting to successfully sell a product be over now?
 
Imho the Switch as a brand has way more price elasticity that people give it. A 400$ Switch 2 wouldn't hurt its sales, especially if they keep a 300$ OLED and 150$ lite (the only model that really needs a price cut imho).

If the original Switch had launched at 350 it would have performed similarly to what it has performed at 300. This in general is true for the console business. In the age of the iphone people aren't averse to paying high prices for technology. The whole scaremongering of "if the next gen consoles launch at high prices they'll fail" was very silly. 500$ consoles are selling on par (if not better) than 400$ ones of the previous gen and are only limited by supply.

Of course this doesn't mean that a 900$ switch would sell like the iphone, but 300 or bust is not true. I'm sorry if this sounds out of place in a tech thread.
 
You're right, I'm sorry, how dare I analyze perfection. They'll launch at $400 and it will have absolutely zero negative impact on its saleability, because nothing can stop the Switch hype train ever. Hell, by what I'm reading here, it'll be bulletproof at $500. Why not? It's not like terrific products fail to sell for what seem like inconsequential reasons to some or anything. The good times will never possibly end, how dare I suggest to the contrary.

Can this conversation that started because I dared to say that price was (emphasis mine since it was part of the post everyone seemingly ignored) ONE OF MANY factors of consideration for a company wanting to successfully sell a product be over now?
Lol buddy, you're the one who tried to bring up price as a big reason, as if that doesn't apply to literally every product in existence

If you're gonna be a baby about it, sure. People are allowed to disagree with you on a forum, you know that, right?
 
Imho the Switch as a brand has way more price elasticity that people give it. A 400$ Switch 2 wouldn't hurt its sales, especially if they keep a 300$ OLED and 150$ lite (the only model that really needs a price cut imho).

If the original Switch had launched at 350 it would have performed similarly to what it has performed at 300. This in general is true for the console business. In the age of the iphone people aren't averse to paying high prices for technology. The whole scaremongering of "if the next gen consoles launch at high prices they'll fail" was very silly. 500$ consoles are selling on par (if not better) than 400$ ones of the previous gen and are only limited by supply.

Of course this doesn't mean that a 900$ switch would sell like the iphone, but 300 or bust is not true. I'm sorry if this sounds out of place in a tech thread.
We may never know if it would sell the same at 350. Customers care about the price, the reason why PS5 sells so well at 500 is that it is a steal at that price. You would have a hard time building a matching PC at 1,000 dollars. Switch OLED was a good proposition at 350, a hybrid system with a great library of games. The OG Switch did not have the library, it had "only" a system seller game, it needed to prove itself.
 
0
Lol buddy, you're the one who tried to bring up price as a big reason, as if that doesn't apply to literally every product in existence

If you're gonna be a baby about it, sure. People are allowed to disagree with you on a forum, you know that, right?
Show me where I said it didn't effect other products.
I'm not being a baby about it, I'm just frustrated how people make assumptions and put words in my mouth to do it and then talk down to me in the process using arguments I never made. That's not a conversation, that's not a disagreement, that's a farce, and I'm over it.
 
0
It'll ve whatever it is the latest version, if it's 3.1 by the time the new console comes out, then it'll use DLSS 3.1. This is not only Nintendo trying to show something to the world, the next Switch will be the PERFECT troyan horse for Nvidia to push the wonders of DLSS into everyone's homes and for us to see and then need it, they won't let that chance passes by.
With you on this comment. I wrote in a past post that I'm a believer in something called the "Common Endeavour", and what you wrote here is one great example of that. To add to this, I would put DLSS under the umbrella of "Disruptive Innovations In Technology" and "Ideas that could solve multiple problems" - I'm confident that Nintendo and Nvidia will embrace it.
 
Switch 2 will be $299 Nintendo want's mass mainstream market not tech nerds like us

Current Switch is launched at $299, but plenty things changed and will change from 2017. to 2023/2024,
I wouldn't be surprised at all if Nintendo launch Switch 2 and Switch 2 Lite at same time with $399/299 price points, and in same time keep selling keep OLED Switch for $299.
 
0
I think Nintendo could launch Switch 2 at $350-$400 late next year and sell out while still selling their OLED Switches at a high price. I don't think they'll drop prices that much in a year and half after all.
 
0
Big thing is i believe IIRC Nintendo got the X1 SoC at launch for $50 a pop. Conventional thinking here is they are unlikely to secure the same pricing for this new chip, just because of the sweet deal from nvidia theygot initially.

If we conservatively assume $100 a pop, there's your $50 difference right there. I think $399 is most reasonable expectation , the OLED model is pointing the way.

Not to jinx it but they did something similar with the DS line, where successive revisions were pricier with the DSi XL being the most expensive sku iirc, and it was leading up to the $250 3DS. They'll probably not do the equivalent of that with a $450 on Switch 2, but $400 is entirely possible.
 
0
PS4 SOC was estimated to be around 100 dollars I think (forgot where I saw this was ages ago), and was a ~350mm^2 chip.


Dane clearly won’t be that huge.

TX1 was likely between 40-55 bucks and that was a ~120mm^2 chip.

Dane probably won’t be that small.

So, price range would be between 50-75 dollars I think just for the SoC….




….in a normal world

And XBox One was estimated to be 110 or so dollars for the soc and it was a bit bigger than the PS4 APU, at 363mm^2.

Again, I should stress, I read this years ago.


Edit: these are on the 20nm and 28nm process respectively. It’s different for the 10/8nm process now with respect to Dane.
 
PS4 SOC was estimated to be around 100 dollars I think (forgot where I saw this was ages ago), and was a ~350mm^2 chip.


Dane clearly won’t be that huge.

TX1 was likely between 40-55 bucks and that was a ~120mm^2 chip.

Dane probably won’t be that small.

So, price range would be between 50-75 dollars I think just for the SoC….




….in a normal world

And XBox One was estimated to be 110 or so dollars for the soc and it was a bit bigger than the PS4 APU, at 363mm^2.

Again, I should stress, I read this years ago.


Edit: these are on the 20nm and 28nm process respectively. It’s different for the 10/8nm process now with respect to Dane.
Would the 8nm process and chip shortage possibly push prices up?
 
We have to bear in mind that Nvidia will probably not try to jack up the price to keep a good relationship with one of the best partners they ever had for the Tegra line, specially because they probably want to keep working together and because Nintendo is probably the ones who will popularize DLSS, which may open a bigger market for it and Nvidia... or at least make it open quicker.
The chip's price will also change depending on how much it's costumized, which hopefully isn't a lot (we all want those juicy RT cores). Hopefully it won't be too expensive to keep the price down and sales up.
 
Please read this staff post before posting.

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


Back
Top Bottom