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

[Dakhil]

They definitely seem to be advertising this GTC more than usually, so hopefully he shows some needed details about the Orin family of chips.

I hope so, too. But I can think of two more reasons why Nvidia might be marketing GTC 2021 (November 2021) much more aggressively than usual.

1. There's a possibility that Nvidia might be unveiling Hopper during the GTC 2021 (November 2021) keynote. Not only was there a rumour that Hopper was taping out soon, but AMD's also planning on announcing CDNA 2 (and Genoa) on the first day of GTC 2021 (November 2021).
2. Nvidia's definitely going to talk more about the metaverse, especially with Tim Sweeny from Epic Games giving a presentation during GTC 2021 (November 2021) about the metaverse, and with four more GTC 2021 (November 2021) presentations dedicated to the metaverse. Nvidia (and other people/companies) would be crazy not to talk about the metaverse, especially with Facebook changing the company name to Meta.

~

Although this is not related to the DLSS model*, AMD's planning on going on a deep dive of the Van Gogh APU powering the Steam Deck during Steamworks Virtual Conference: Steam Deck on 12 November 2021, the day after GTC 2021 (November 2021)'s final day. (Being a Steamworks developers is required to attend Steamworks Virtual Conference: Steam Deck.)

 

[Terrell]

I apologize for the light digression from the topic of this thread.

The fiscal year ending March 2012, Nintendo posted $533 million in losses. First time ever. The previous fiscal year they posted ~$1 billion in profit.

This was entirely because of the 3ds.

Sure, Wii was in a steep decline that year, but that didn’t cause those losses. It just didn’t negate the losses like a 2009 Wii would.

Fiscal year ending March 2013, Nintendo reported $366 million in losses. This is still entirely because of the 3ds hardware fiasco. It wasn’t until August 2012 that Nintendo stopped hemmoraging on every 3ds sold…finally breaking even. They didn’t make a profit on 3ds until the end of 2012.

The losses during Fiscal year ending 2014, can indeed be atttibuted to the Wii U…since they had to aggressively cut its hardware prices in most territories during 2013 and its software sales weren’t enough to offset those losses.

From 2014 on, Nintendo started posting profits again.

But absolutely the great majority of the losses Nintendo had from 2011-2013 were because of the 3ds. The poor sales of the Wii U in 2013 had very little to do with this.

FY 2012 was the year of the 3DS price drop, that cannot be denied as being a part of that year's negative forecast. Let's look at the following year, though.

FY 2013:

(Main reason for the decrease in net sales)
The net sales were down due to a large decrease in the unit sales of the "Wii" and "Nintendo DS" hardware and software despite the "Wii U" launch and the growth of the sales of "Nintendo 3DS."

(Gross profit ratio)
The gross profit ratio was 22.1%. There were both positive and negative factors contributing to this result; while there was an improvement in the profitability of the "Nintendo 3DS" hardware due to a decline in production costs, there was a decrease in overall product profitability due to the production costs of the "Wii U" hardware, released this fiscal year, being higher than its selling price.

Between the 2 pieces of hardware, Nintendo themselves saw 3DS profit growth (though far less than they wanted) that Wii U flushed down the crapper; that was a recurring trend. 3DS has 267% more 1st-party software sold than Wii U, combined with profitable hardware sales while Wii U hardware refused to see a cost reduction due to low production quantities. While I admit that 3DS wasn't a great contributor to profits like past handheld systems had been and had a part to play in their financial situation at the time, the only negative contribution 3DS made to their financial situation beyond the early price dip was being unable to pick up the Wii U's pathetic performance at market.

It’s the same concept of the ps4 pro or Xbox X. The uptake doesn’t have to be fast. The purpose is to elongate the lifecycle, not replace.

There is no such thing as “languishing with slow sales” for the new Switch or even the OLED. They are enhancement models. The uptake rate isn’t important. Nintendo doesn’t make their profit on hardware sales, so that doesn’t matter. Their profit is software. If someone in 2023 buys a $450 4K Switch or a $300 OLED Switch or a $250 grey Switch or a $170 Lite…Nintendo doesn’t give a damn. They all play the same Nintendo published games, that’s what’s important.

The R&D is already paid for. By 4 years of Switch software/NSO revenue success. The price point of the new Switch will cover its marginal production increase.

And I never said Nintendo won’t ever do a price cut, of course they will. I was explaining why the hybrid model hadnt gotten one up to this point.

I don't get how you can't see the contradiction here. If this new hardware sells poorly, it's not elongating anything, or at least doing so incredibly marginally.
If people are still primarily buying standard Switches instead of this alleged 4K "enhancement", then Nintendo is spending a massive chunk of their profit for an absolutely marginal return on investment.

Let's put this idea of yours in practice. The R&D for this new hardware with a brand-new customized SoC is likely to be in the $500 million to $1 billion range, if not more. If this "4K Switch enhancement" sells 4 million units across 2 years (which is frankly optimistic, considering your suggested price points for this "enhancement" with no exclusive software as you outlined are close to the same amount as a PS5) while the Mariko models sell 36 million units in that same 2-year span, that means this "enhancement" is only responsible for 10% of the growth in Switch's overall install base in those 2 years (leaving aside how many who would buy it to replace a Switch they already own, let's keep this simple). When looked at across the whole install base by that time, it'd make up less than 4% of the install base in totality.
Even Switch Lite, which only needed a die shrink, was around a third of all Switches sold in its first year, and the R&D expense of the Mariko SoC had a knock-on effect for the standard Switch in that the die shrink allowed for more chips to be produced in a similar timeframe, reducing production costs across the entire lineup. This 4K "enhancement" will not provide those same knock-on benefits, even in software sold by those who bought one in lieu of other options.
Spending $500 million or more on new hardware that sells an average of 2 million units in a year means that the boost to their overall business would need to be at least $125-250 of pure profit per owner of that hardware just to break even on the R&D in the first 2 years. Given that Switch owners currently spent an average $250 at retail on software through all available retail channels in 2019 (55,000 yen gross sales per hardware unit, less the hardware sale itself) and Nintendo collects 35% of that gross sales amount as profit at best, selling hardware that slowly would be a net loss to Nintendo's profitability overall.

The French have a saying: "the game isn't worth the candle". And I can't think of a better example for this idiom than using up to a billion dollars of profit to make new hardware to, in the absolutely best-case scenario of what that 10% of per-year hardware sales the "enhancement" nets them would represent, earn $400 million from that investment in the first calendar year and not see a return on investment until partway through year 2 or in year 3.
For comparison and using the same 35% margin on software sold, even using the same R&D cost for the original Switch, at 10 million sold in its first year at break-even price and 63.51 million units of mostly exclusive 1st-party software, Switch made just over $1 billion in first-party software alone. Pretty solid RoI there. So....

Nintendo is going to sell more Switches this fiscal year than last…how is that waning demand?

The first reported quarter of this fiscal year had them reporting a 21% global hardware sales decline YoY and a 10% global software sales decline YoY. I'm willing to take actual monetary bets that the following quarter will report another YoY decline.

Take a look at what these Nintendo published exclusive examples you are talking about are. They are titles that Nintendo expects to be relatively low sellers anyways. I hate the term “niche”, but it’s apt here.

Feel free to point to a major AAA Nintendo published title exclusive to the 4K Switch from 2022-2026. Doubt you’ll find it though.

To the "niche" exclusives point... Oracle of Ages/Seasons, Super Mario Bros. Deluxe, Donkey Kong Country, Pokemon Crystal, Wario Land 3.... yeah, those sound like "niche" games alright. /s

And finally, I'm not gonna add much to the discussion of how it's ludicrous to ask someone to point to something that doesn't exist yet to prove a point (that has already been driven home by others), except to say that, in addition to arguing that Nintendo doesn't care about 3rd-party relations and constructing strawmen with regard to this discussion elsewhere, it doesn't indicate that you're interested in a good faith discussion on this subject. And considering that, I'm not sure that I shouldn't even bother indulging you beyond this response.

 

[Terrell]

Lite SoCs won't need to run at docked clocks. That means that they could actually use power binned chips for the Lite model, especially if they are aiming for 15W in docked mode.

This kinda begs the question: do they have to aim for 15W in docked mode? It seems to me that, by using Tegra X1 for Switch and it mostly being a done deal before they ever really got there, that's the TDP they sorta stumbled into. I can't see anything that would keep them from going with a 20W TDP while docked, but that would likely be the maximum (that was around the TDP of the Wii U's GPU), unless there were concerns with cooling or something else I don't know of.

The 15W number, as least as far as I understand it, has just been used primarily as a guide rail to allow us to figure out what we can expect from performance, to measure Orin on a performance per watt basis when the TDP is considered completely equal with the TX1. Which is a totally fair way to manage expectations.

 

[ILikeFeet]

This kinda begs the question: do they have to aim for 15W in docked mode? It seems to me that, by using Tegra X1 for Switch and it mostly being a done deal before they ever really got there, that's the TDP they sorta stumbled into. I can't see anything that would keep them from going with a 20W TDP while docked, but that would likely be the maximum (that was around the TDP of the Wii U's GPU), unless there were concerns with cooling or something else I don't know of.

The 15W number, as least as far as I understand it, has just been used primarily as a guide rail to allow us to figure out what we can expect from performance, to measure Orin on a performance per watt basis when the TDP is considered completely equal with the TX1. Which is a totally fair way to manage expectations.

The wii u's heat sink took up a bit of volume despite being 33W. I guess you could use a fan, but it would have to spin up quite loudly unless there is some other way to remove heat. Like a dock fan or something

 

[Terrell]

The wii u's heat sink took up a bit of volume despite being 33W. I guess you could use a fan, but it would have to spin up quite loudly unless there is some other way to remove heat. Like a dock fan or something

As I understand it, a lot of the Wii U’s design was dictated by its main board size and a need to accommodate an optical drive. It left a LOT of height to work with that likely explains the size of the heat sink and its overall lack of heat pipe to direct heat right to the blower.

Heat currently vents out the top and away from the dock, so don’t know what you mean by a dock fan.
There’s also an aluminum shield on the Switch’s back that dissipates as much heat at it can out the back chassis.
It’s also worth noting that the fan and heat pipe in use in the original Switch was, according to iFixIt, a bit overkill when used with the Mariko SoC, which is why the fan and heat pipe were slightly shrunk and the fan vents are a little thinner on the OLED model. So there’s enough to suggest that Nintendo over-compensated with heat dispersal on Switch and may not need much more than it had in 2019 to regulate heat while docked on hardware with a slightly higher TDP.

 

[ReddDreadtheLead]

As I understand it, a lot of the Wii U’s design was dictated by its main board size and a need to accommodate an optical drive. It left a LOT of height to work with that likely explains the size of the heat sink and its overall lack of heat pipe to direct heat right to the blower.

Heat currently vents out the top and away from the dock, so don’t know what you mean by a dock fan.
There’s also an aluminum shield on the Switch’s back that dissipates as much heat at it can out the back chassis.
It’s also worth noting that the fan and heat pipe in use in the original Switch was, according to iFixIt, a bit overkill when used with the Mariko SoC, which is why the fan and heat pipe were slightly shrunk and the fan vents are a little thinner on the OLED model. So there’s enough to suggest that Nintendo over-compensated with heat dispersal on Switch and may not need much more than it had in 2019 to regulate heat while docked on hardware with a slightly higher TDP.

I think he means a fan in the dock that helps push air out of the device when docked to keep it cooler, rather than just relying on the small fan in the switch. It would be by the vents that pull air in(bottom ones), rather than a vacuum that pulls the air out (the top vents)

 

[Dakhil]

According to TechInsights, the Tensor SoC has a die size of 108.26 mm². And the Tensor SoC's fabricated using Samsung's 5LPE process node.

So I think Dane's at least going to be ~120 mm², considering Dane's probably going to fabricated using Samsung's 8N process node or Samsung's 7LPP process node.

 

[Terrell]

I think he means a fan in the dock that helps push air out of the device when docked to keep it cooler, rather than just relying on the small fan in the switch. It would be by the vents that pull air in(bottom ones), rather than a vacuum that pulls the air out (the top vents)

If I remember correctly, the vacuum is actually the most efficient method of drawing in air in this design. The design is as such that heat pipes draw heat off the SoC, it travels to the end of the heat pipe where it's drawn to a set of fins that sit right next to the blower's exhaust, pushing it directly out the vent, and an aluminum shield is like an insurance policy to ensure the primary contact point of the heat pipe to the chip cannot heat up to the point of failure.
By pushing air in from the other end, the vacuum action is diminished, meaning the blower sitting right next to the fins cannot as efficiently push heat out at the point the heat is drawn to. You want the vacuum action in such a design. If you are pushing air into the device, you want a standard heat sink (the fins being directly above the chip being cooled) that is situated in the direct path of air intake and output, allowing a wind tunnel effect across the heat sink fins. But in a portable device, that design would inevitably sacrifice thinness.

I guess that leads back to my original query: if the heat dispersal in Switch was actually more than it required like iFixIt believed it to be, does that present the option to push the design to something closer to 20W TDP without adding bulk?

 

[ShadowFox08]

According to TechInsights, the Tensor SoC has a die size of 108.26 mm². And the Tensor SoC's fabricated using Samsung's 5LPE process node.

So I think Dane's at least going to be ~120 mm², considering Dane's probably going to fabricated using Samsung's 8N process node or Samsung's 7LPP process node.

I really hope Switch 2 uses 6nm or lower. Although 5nm would be perfect. Sigh we'll see.

Anyway, with a bit of research..
Through early benchmarks, the Google pixel 6's GPU has surprisingly scored the highest on Wildlife (not scored on actual games) when compared to current Android phones, although CPU performance worse than S21.

Google Tensor Chip Packs More GPU Power than Snapdragon 888 and Exynos 2100

According to recent GPU benchmark scores of the Pixel 6, the Google Tensor chip beats Qualcomm Snapdragon 888 and Samsung's Exynos 2100.

beebom.com

Doing a further bit of research on your link, Wikipedia, as well as this source (assuming it's correct):

ARM Mali-G78 MP20 - GadgetVersus

ARM Mali-G78 MP20 graphics card benchmarks and specs, with the number of execution units, shading units, cache, memory, the power consumption, the lithography, the architecture, the generation, as well as the performances.

gadgetversus.com

Pixel 6's GPU seems to runs around 1.9 TFLOPs max (not sure how many cores..20?), with bandwidth at 51.2GB/s using LPDDR5 RAM. Pretty exciting nonetheless.

edit: Also checked A15 on iPhone 13. Already know CPU is close to 2x as powerful as Pixel 6 and it'd running at a tdp of 6 watts. But it's GPU FP32 performance is 1.5 TFLOPs, which is interesting.

Apple A15 Bionic vs Google Tensor - GadgetVersus

Comparison between Apple A15 Bionic and Google Tensor with the specifications of the processors, the number of cores, threads, cache memory, also the performance in benchmark platforms such as Geekbench, Passmark, Cinebench or AnTuTu.

gadgetversus.com

Anyway, it's going to be really exciting to see Switch 2's specs, especially considering it's going to be using a higher TDP and faster GPU vs handheld in docked mode.

 

[Intoxicate]

I hope this time, with the success of the Switch in their pockets, Nintendo will go for an advanced SOC and process instead of an on stock/safe option.
It would be great to have the Switch benefits (hybrid design, portability) enhanced with more competitive graphics. On the other hand, the improvements for 3D graphic have deminishing returns when I look at what the PS4 Pro and PS5 improved compared to the PS4. I'm not to keen on 4K, that's marketing stuff for me, and doesn't improve anything for me when I'm sitting 7m from my TV away.

 

[Hermii]

I hope this time, with the success of the Switch in their pockets, Nintendo will go for an advanced SOC and process instead of an on stock/safe option.
It would be great to have the Switch benefits (hybrid design, portability) enhanced with more competitive graphics. On the other hand, the improvements for 3D graphic have deminishing returns when I look at what the PS4 Pro and PS5 improved compared to the PS4. I'm not to keen on 4K, that's marketing stuff for me, and doesn't improve anything for me when I'm sitting 7m from my TV away.

As far as its possible to confirm anything via rumours, the safest bet is a customised version of Nvidia Orin based on Samsung 8nm.

 

[ILikeFeet]

Anyway, with a bit of research..
Through early benchmarks, the Google pixel 6's GPU has surprisingly scored the highest on Wildlife (not scored on actual games) when compared to current Android phones, although CPU performance worse than S21.

Anandtech found the GPU to be underwhelming. They also don't test games as they're not exactly fair. Some games have phone exclusive graphical features

 

[NineTailSage]

Anandtech found the GPU to be underwhelming. They also don't test games as they're not exactly fair. Some games have phone exclusive graphical features

I think this just proves that yes the manufacturing node does matter to some extent, but most of these SoC's aren't designed for sustained performance. Even if at 8nm Dane would be more than capable of providing a variety of ways Nintendo can achieve gaming profiles in both handheld and docked that scale enough to make the chip much better than the current Switch does at this.

The early rumors of both RTX 40 series cards and AMD's RX 7000 series being super power hungry to achieve their more than 2x performance gain over their respective counter parts, just continues the problem of finding a way to achieve these gains and reduce power consumption significantly at the same time...

 

[ArchedThunder]

I hope this time, with the success of the Switch in their pockets, Nintendo will go for an advanced SOC and process instead of an on stock/safe option.
It would be great to have the Switch benefits (hybrid design, portability) enhanced with more competitive graphics. On the other hand, the improvements for 3D graphic have deminishing returns when I look at what the PS4 Pro and PS5 improved compared to the PS4. I'm not to keen on 4K, that's marketing stuff for me, and doesn't improve anything for me when I'm sitting 7m from my TV away.

According to rumor the Switch 4K is using a brand new and likely custom GPU.

 

[ILikeFeet]

I think this just proves that yes the manufacturing node does matter to some extent, but most of these SoC's aren't designed for sustained performance. Even if at 8nm Dane would be more than capable of providing a variety of ways Nintendo can achieve gaming profiles in both handheld and docked that scale enough to make the chip much better than the current Switch does at this.

The early rumors of both RTX 40 series cards and AMD's RX 7000 series being super power hungry to achieve their more than 2x performance gain over their respective counter parts, just continues the problem of finding a way to achieve these gains and reduce power consumption significantly at the same time...

the Anandtech article is a good read. node matters but this design has a number of issues not related to the node: poor thermal dissipation, the strange design choice with using A76s, and just general inefficiencies despite paper specs showing a beast of an SoC

 

[NineTailSage]

the Anandtech article is a good read. node matters but this design has a number of issues not related to the node: poor thermal dissipation, the strange design choice with using A76s, and just general inefficiencies despite paper specs showing a beast of an SoC

It's definitely a good read and sounds like an outdated design from team Google that does nothing spectacular or original over Samsung's Exynos design. It must be said that even using an advanced 5nm node, it's just not enough on its own to make for a competitive chip in comparison to the other top end designs (but again looking at Apple in comparison the difference maker seems to be how all in on the amount of cache they use).

 

[lemonfresh]

Do you think we will get new info in nintendo earnings report which comes out in a couple hours?

 

[Skittzo]

Do you think we will get new info in nintendo earnings report which comes out in a couple hours?

No. They're going to try very hard to not even remotely hint at new hardware coming when they have a new OLED model to sell this holiday.

The only possible tidbit we may get that probably won't even mean anything is their R&D spending.

 

[ArchedThunder]

Do you think we will get new info in nintendo earnings report which comes out in a couple hours?

The only reason they talked about NX so early on was because the Wii U was doing so poorly.

 

[Alovon11]

the Anandtech article is a good read. node matters but this design has a number of issues not related to the node: poor thermal dissipation, the strange design choice with using A76s, and just general inefficiencies despite paper specs showing a beast of an SoC

Again, I feel that some of the issues may be in part due to unoptimized software.

AKA: The benchmarks/tests they ran may not have taken kindly to the parallelized nature of the Dual X1 Core- Tensor SoC

Especially as most SoCs nowadays are 1+3+4, or 4+4, not 2+2+4.

It just feels to be the only way to rationalize the weird data they got sometimes.

 

[Dakhil]

I really hope Switch 2 uses 6nm or lower. Although 5nm would be perfect. Sigh we'll see.

I think TSMC's N6 process node's probably the absolute best case scenario when TSMC's concerned. But I think the probability of TSMC's N6 process node being used to fabricate Dane is rather low, considering that the demand for TSMC's N6 process node is absurdly high.

• The Snapdragon 680 4G, the Snapdragon 695 5G, and the Snapdragon 778G+ 5G (and the Snapdragon 778G) are confirmed to be fabricated using TSMC's N6 process node.
• Intel has confirmed to be fabricating Alchemist GPUs using TSMC's N6 process node.
• AMD's rumoured to be using TSMC's N6 process node to fabricate the Rembrandt APUs, the Ryzen Embedded V3000, Navi 33, and the I/O for Navi 31 & Navi 32. (There are also dubious rumours about a PlayStation 5 redesign planned for release in Q2 2022 or Q3 2022 running on a new, semi-custom 6 nm** APU from AMD and a Xbox Series S refresh planned for release in late 2022 potentially running on a higher spec 6 nm** APU.)
• The Dimensity 810, Dimensity 920, Kompanio 900T, and Kompanio 1300T are confirmed to be fabricated using TSMC's N6 process node.

And considering that Samsung's 6LPP process node isn't available for general consumers, and no company, including Samsung, has announced any chips being fabricated using Samsung's 6LPP process node, Samsung's 6LPP process node probably isn't an available option for Nvidia.

I also think the probability that a 5 nm** process node from Samsung or TSMC being used to fabricate Dane is rather low as well.

Business Korea reported on 5 July 2021 that some of the 5 nm** process nodes on Samsung's V1 fab at Hwaseong have yields below 50%. And Business Korea's report seems to be vindicated by a Xiaomi executive's comment about the Snapdragon 780G, which is fabricated using Samsung's 5LPE process node, being cancelled due to insufficient capacity. Digital Chat Station mentioned that the Snapdragon 888 will continue being used in smartphones releasing in 2022. And Ice Universe mentioned Samsung will soon release the Exynos 1280, a 5 nm** entry-level Exynos SoC, considering the Exynos 1280's specs are apparently lower than the Exynos 1080's specs.

As for TSMC's N5 process node or TSMC's N5P process node, the probability is very close to 0%.

• Apple's probably using TSMC's N5P process node to fabricate the Apple A15 Bionic, M1 Pro, and M1 Max probably until late 2022.
• kopite7kimi mentioned least the high-end Lovelace GPUs (e.g. AD102) and Hopper GPUs are being fabricated using TSMC's N5 process node, which I imagine Nvidia will heavily prioritise in terms of the amount of available capacity Nvidia has secured for TSMC's N5 process node, considering the products mentioned have high-profit margins.
• greymon55 mentioned that the Navi 31 GPU & the Navi 32 GPU are fabricated using a 5 nm** process node from TSMC. And AMD has confirmed Zen 4 will be fabricated using a 5 nm** process node from TSMC.
• Digital Chat Station mentioned that Qualcomm has secured enough capacity for a 5 nm** process node from TSMC, as well as TSMC's N6 process node, which I think is possible, considering that Digital Chat Station was correct about Qualcomm securing enough capacity for TSMC's N6 process node, with the announcements of the Snapdragon SoCs that are fabricated using TSMC's N6 process node above. And there was a rumour from Ronald Quandt about the Snapdragon SC8280, which I assume is the model number for the 8cx Gen 2 successor, having 4 Gold+ cores running at a frequency of 2.7 GHz, and 4 Gold core running at a frequency of 2.43 GHz, as well as having a NPU running at 15 TOPS, which seems to be vindicated by a Geekbench 5 score, with a single-core score of 1007 and a multicore score of 4456, uploaded on 28 July 2021, with the model listed as "Qualcomm QRD", the name listed as "Snapdragon 8cx Gen 3", and the base frequency listed as "2.69 GHz". And comparing the Geekbench 5 score from 28 July 2021 with a Geekbench 5 score from 21 March 2021, with a single-core score of 981 and a multicore score of 4918, I think there's a possibility Qualcomm's using a 5 nm** process node from TSMC for the 8cx Gen 2 successor.

Saying that, I do believe there's a possibility Samsung's 8N process node or Samsung's 7LPP process node could be used to fabricate Dane, and I've mentioned the reasons here.

** → a marketing nomenclature used by all foundry companies
~

the Anandtech article is a good read. node matters but this design has a number of issues not related to the node: poor thermal dissipation, the strange design choice with using A76s, and just general inefficiencies despite paper specs showing a beast of an SoC

Again, I feel that some of the issues may be in part due to unoptimized software.

AKA: The benchmarks/tests they ran may not have taken kindly to the parallelized nature of the Dual X1 Core- Tensor SoC

Especially as most SoCs nowadays are 1+3+4, or 4+4, not 2+2+4.

It just feels to be the only way to rationalize the weird data they got sometimes.

Samsung's implementation of Arm's Cortex-A CPU designs also seems to be quite terrible, going by what Andrei Frumusanu has said about the Cortex-A55 cores in the Tensor SoC, which no amount of software optimisation is going to fix.

On the side of the Cortex-A55 cores, things also aren't looking fantastic for the Tensor SoC. The cores do end up performing the equally clocked A55's of the Snapdragon 888 by 11% - maybe due to the faster L3 access, or access to the chip's SLC, however efficiency here just isn't good, as it uses almost double the power, and is more characteristic of the higher power levels of the Exynos chips' A55 cores. It's here where I come back to say that what makes a SoC from one vendor different to the SoC from another is the very foundations and fabric design - for the low-power A55 cores of the Tensor, the architecture of the SoC encounters the same issues of being overshadowed in system power, same as we see on Exynos chips, ending up in power efficiency that's actually quite worse than the same chips own A76 cores, and much worse than the Snapdragon 888. MediaTek's Dimensity 1200 even goes further in operating their chip in seemingly the most efficient way for their A55 cores, not to mention Apple's SoCs.

Finally, the A55 behavioural characteristic showcases that this chip is very much related to Samsung's Exynos SoCs, falling behind in efficiency compared to how Qualcomm or MediaTek are able to operate their SoCs.

~

Do you think we will get new info in nintendo earnings report which comes out in a couple hours?

I think Nintendo will definitely talk about how the global chip shortage is affecting the OLED model in terms of sales, which I think should give a rough idea of how the DLSS model* will fare in terms of the global chip shortage.

 

[Vladimir]

If the rumored Dane ends up being the SOC for a new more powerful Switch, without counting DLSS, how much stronger would it be compared to the current Switch?

 

[ILikeFeet]

If the rumored Dane ends up being the SOC for a new more powerful Switch, without counting DLSS, how much stronger would it be compared to the current Switch?

we don't know the configuration of this chip, so the estimated range is very wide. anywhere from 40%+ to close to Xbox One

 

[My Tulpa]

Obviously nothing currently announced is going to be exclusive to unannounced hardware, but all that stuff is probably going to be out by 2023 at the latest. I don't really expect "AAA" support for the current Switch to last significantly beyond that, with probably only historically more handheld aligned stuff sticking around longer.

Why wouldn’t you expect most Nintendo published games to be playable on the OLED Switch for the next 4-5 years?

I have seen no good argument as to why that would/should be.

If the 4K Switch launches at the same price as the OLED Switch and effectively replaces it and Nintendo discontinues OLED production right after the 4K Switch launches…then I agree Nintendo would move over to 4K Switch exclusivity pretty quickly.

But nothing I’ve seen or read in here would suggest this to be the case. I don’t see how the 4K Switch wont be a relatively expensive/premium product in the Switch ecosystem. The older (cheaper) models will still have to be viable options for a pretty good while. And that’s just to buy. The active userbase of people on the OG Switch, 2019 Switch, Lite, OLED Switch isn’t going to all shift to 4K Switch as fast as people here think they are going to. And Nintendo won’t be motivated to try and push people to the 4k model. Why would they need to? They aren’t Sony/Microsoft.

 

[Alovon11]

we don't know the configuration of this chip, so the estimated range is very wide. anywhere from 40%+ to close to Xbox One

More accurately, up to 25% faster than OG PS4 Pre-DLSS Is actually the current range.

 

[Dakhil]

Why wouldn’t you expect most Nintendo published games to be playable on the OLED Switch for the next 4-5 years?

If by "most Nintendo published games", you mean Nintendo's AAA games, then it's because as Nintendo's future AAA games become more and more ambitious, the OLED model will have a much harder time running the game optimally. This is already the case with Super Mario 3D World + Bowser's Fury, specifically the Bowser's Fury portion, where the game runs at a range of 720p ~ 792p at mostly 60 fps in TV mode, with handheld mode running the game at 720p, but at 30 fps, which is definitely not optimal, since Nintendo usually aims for a consistent 60 fps experience for Super Mario Bros games at all costs, whether it's in TV mode, or whether it's in handheld mode. I think the performance of the Bowser's Fury portion of Super Mario 3D World + Bowser's Fury is an indirect admission from Nintendo that the Nintendo Switch, and by extension the OLED model, is not quite powerful enough to develop the types of ambitious games Nintendo wants to develop, as far as AAA games are concerned.

 

[Deleted member 887]

Why wouldn’t you expect most Nintendo published games to be playable on the OLED Switch for the next 4-5 years?

@Pokemaniac didn't say "most", they said the AAA stuff - Nintendo publishes a lot of games, ~20 a year. In the GBA->3DS years, the majority were for the mobile system.

5 years from today would be just shy of a 10 year "generation" for the Switch - longer than Nintendo has said they're planning. The hybrid nature of the console + lengthening of console generations means I don't think anyone has absolute clarity on the matter (I expect even Nintendo doesn't, the way they planned on keeping the GBA and the 3DS around as exit strategies of the og DS or the Switch tanked). But I don't think the plan is to support Tegra devices with mainline Mario/Zelda/Mario Kart in 5 years.

If the Pro revision is super powerful, I expect Nintendo to try to essentially "cross-fade" the two devices, ramping up Pro selling exclusives and cutting price on the Tegra machines, repositioning them as budget devices.

If for some reason it isn't and it's just a "premium way to play Switch games", then I expect Nintendo to launch the Trü Successor in 2025 (an 8 year generation for the Switch, allowing for pandemic to push all plans back a year), and to publish at most One Big Game for both generations, ala Twilight Princess/BotW, and keep Switch alive at most with a bunch of side scrollers and Pokemon spinoffs.

The only way I expect to be playing the latest 3D Mario on my current machine in 2026 is if Nintendo tries the "cross-fade" strategy and it doesn't work - requiring that Nintendo extend the Switch lifecycle as long as possible while they execute a rethink.

edit: punctuation changed for clarity

 

[Pokemaniac]

Why wouldn’t you expect most Nintendo published games to be playable on the OLED Switch for the next 4-5 years?

I have seen no good argument as to why that would/should be.

If the 4K Switch launches at the same price as the OLED Switch and effectively replaces it and Nintendo discontinues OLED production right after the 4K Switch launches…then I agree Nintendo would move over to 4K Switch exclusivity pretty quickly.

But nothing I’ve seen or read in here would suggest this to be the case. I don’t see how the 4K Switch wont be a relatively expensive/premium product in the Switch ecosystem. The older (cheaper) models will still have to be viable options for a pretty good while. And that’s just to buy. The active userbase of people on the OG Switch, 2019 Switch, Lite, OLED Switch isn’t going to all shift to 4K Switch as fast as people here think they are going to. And Nintendo won’t be motivated to try and push people to the 4k model. Why would they need to? They aren’t Sony/Microsoft.

Based on past history, Switch OLED is only going to be on the market for around 3 years at most. The Lite will probably be the last representative of the TX1-derived Switches, with maybe a TV-only microconsole to accompany it if Nintendo ends up with extra chips they need to dump.

As for games, sure, the current Switch will continue to receive games from Nintendo for probably another 4-5 years, but support is going to start shifting over to Dane after that's been out for a bit. By the end of that 4-5 year period, the only first party the current Switch is going to be getting is Pokémon, some lower budget games, and probably some ports.

 

[ILikeFeet]

Pokémon

4 years for Pokemon to finally look like a Switch game instead of an HD Wii game. well, usually we have to wait until new hardware to see it, now we can see it on the system itself

 

[My Tulpa]

I apologize for the light digression from the topic of this thread.

FY 2012 was the year of the 3DS price drop, that cannot be denied as being a part of that year's negative forecast. Let's look at the following year, though.

FY 2013:

Between the 2 pieces of hardware, Nintendo themselves saw 3DS profit growth (though far less than they wanted) that Wii U flushed down the crapper; that was a recurring trend. 3DS has 267% more 1st-party software sold than Wii U, combined with profitable hardware sales while Wii U hardware refused to see a cost reduction due to low production quantities. While I admit that 3DS wasn't a great contributor to profits like past handheld systems had been and had a part to play in their financial situation at the time, the only negative contribution 3DS made to their financial situation beyond the early price dip was being unable to pick up the Wii U's pathetic performance at market.

And all of that is to say…had the 3ds not existed, Nintendo wouldn’t have posted losses from 2011-2013. They wouldn’t be huge profits because of the Wii U performance, but it wouldn’t be losses. That’s all I’m saying.

That’s why I get prickly when people blame the Wii U for Nintendo losses during that period.

Look, the GameCube was a very similar “failure”. The GameCube and the Wii U had the same 1st party software sales figures. While the GameCube hardware sales were slightly better than the Wii U, the profit margins were less. Don’t forget, the GC was slashed down to $99 after only 2 years to help buoy poor sales.

Neither system were money loss systems for Nintendo, despite their low sales.

The Wii U just didn’t do anything to offset the actual money loss system at the time…the 3ds.

The GBA wasn’t a money loss system despite it selling FAR less software than the 3ds did.

Nintendo posted profits 2001-2004 despite the GC performance. It wasn’t because the GC made more profit for Nintendo than the Wii U, it didn’t.

The revenue Nintendo pulled in 2011-2014 was more than they pulled in 2001-2004.

The 3ds was the reason for the losses.

I don't get how you can't see the contradiction here. If this new hardware sells poorly, it's not elongating anything, or at least doing so incredibly marginally.

If by “sells poorly” you mean how the ps4 pro sold compared to the ps4/slim…then yea, it can sell “poorly” and still elongate the lifecycle.

No one expects Nintendo to sell ~80 million 4K Switches in the next 4 years like they have the non 4K hybrids…do they? This isn’t a ps4->ps5 scenario.

The 4K Switch model will probably end up being 25-30% of total Switch sales. Like the ps4 pro was.

If people are still primarily buying standard Switches instead of this alleged 4K "enhancement", then Nintendo is spending a massive chunk of their profit for an absolutely marginal return on investment.

Why would you say that, though?

The investment return is exactly what I said…to keep a large chunk of gamers in the switch ecosystem longer than they otherwise would have. All those gamers who feel Switch graphics/performance is a bit long in the tooth. The same exact reason the Xbox X and ps4 pro existed.

Nintendo will sell more Nintendo games in 2024-2025 than they would have if the 4K enhancement didn’t exist. That’s return in their investment right there.

Let's put this idea of yours in practice. The R&D for this new hardware with a brand-new customized SoC is likely to be in the $500 million to $1 billion range, if not more. If this "4K Switch enhancement" sells 4 million units across 2 years (which is frankly optimistic, considering your suggested price points for this "enhancement" with no exclusive software as you outlined are close to the same amount as a PS5) while the Mariko models sell 36 million units in that same 2-year span, that means this "enhancement" is only responsible for 10% of the growth in Switch's overall install base in those 2 years (leaving aside how many who would buy it to replace a Switch they already own, let's keep this simple). When looked at across the whole install base by that time, it'd make up less than 4% of the install base in totality.
Even Switch Lite, which only needed a die shrink, was around a third of all Switches sold in its first year, and the R&D expense of the Mariko SoC had a knock-on effect for the standard Switch in that the die shrink allowed for more chips to be produced in a similar timeframe, reducing production costs across the entire lineup. This 4K "enhancement" will not provide those same knock-on benefits, even in software sold by those who bought one in lieu of other options.
Spending $500 million or more on new hardware that sells an average of 2 million units in a year means that the boost to their overall business would need to be at least $125-250 of pure profit per owner of that hardware just to break even on the R&D in the first 2 years. Given that Switch owners currently spent an average $250 at retail on software through all available retail channels in 2019 (55,000 yen gross sales per hardware unit, less the hardware sale itself) and Nintendo collects 35% of that gross sales amount as profit at best, selling hardware that slowly would be a net loss to Nintendo's profitability overall.

4 million in two years?? Naw it will be more than that. The “enthusiast gamer tech must have new” demand is higher than that lol.

The number of people who move over to 4K Switch in the 4-5 years after its launch will be much higher than that. The ps4 pro was what, around ~30 million more or less when all is said and done? It will be something more like that.

Even if it’s half that, it’s worth the investment.

We saw with the Wii what the perception of an aging, past it’s prime Nintendo system can do to software sales decline. The 4K Switch will do what a Wii HD should have done.

If the majority userbase buys 1-2 games a year (as Nintendo has suggested) you are talking 6-8 Nintendo games during the 4K Switch lifespan sold to millions of gamers who would have otherwise passed had a 4K Switch upgrade not existed.

Even on the extreme low ball end of this, that’s over $1 billion in 1st party software sales revenue 2022-2025 that might otherwise be lost without such an upgrade. There is obviously value for Nintendo to do this to keep the enthusiast gamer market engaged.

I think we both agree that there will be return value on what the 4K hardware provides to the Switch ecosystem…you just feel it needs to have “true successor” type sales to have value. I’m saying it has value even if it’s ps4 pro type sales.

The first reported quarter of this fiscal year had them reporting a 21% global hardware sales decline YoY and a 10% global software sales decline YoY. I'm willing to take actual monetary bets that the following quarter will report another YoY decline.

Oh I’m sure there will be a YoY decline for q2 as well. Can’t have Animal Crossing and a global pandemic shutdown every year, huh?

But this will reflect “only” ~3 million sales difference? I expect that to be made up in the last 2 quarters with the OLED hardware and a better software release schedule.

Conservative Nintendo is still projecting 25.5 million this fiscal year. This within the Nikkei report a few months ago saying Nintendo plans on producing 30 million for this fiscal year. (Compare this to the exact same time in 2020 when Nikkei reported Nintendo planning on 25 million produced last fiscal year and Nintendo projecting 19 million shipped)

Even if it’s off from last fiscal year, it won’t be by much. It will be seen as flat. Hardly a major decline.

To the "niche" exclusives point... Oracle of Ages/Seasons, Super Mario Bros. Deluxe, Donkey Kong Country, Pokemon Crystal, Wario Land 3.... yeah, those sound like "niche" games alright. /s

?

Donkey Kong Country was an exclusive on the SNES…that’s a true successor system. Nintendo published only about 5 games on the NES after the SNES launched. This is nothing similar to a mid gen revision release.

Super Mario Bros Deluxe was an enhanced port of an old game…ok, I can see Nintendo putting an enhanced port of an old games exclusively on the 4K switch, sure. (Like Xenoblade on the n3ds)

I’ll give you the other GBC games being popular new titles…but my god, we are talking about going from monochrome gaming to color lol. That’s a much major shift than going 1080p to 4K with boosted graphic slider IQ.

I can buy gamefreak dying to have Pokémon in color. I don’t buy them in desperate needing to suffer lower userbase exclusivity just to get 4K Pokémon models out there.

We’ll see, I suppose.

And finally, I'm not gonna add much to the discussion of how it's ludicrous to ask someone to point to something that doesn't exist yet to prove a point (that has already been driven home by others)

I meant to point it out when it happens. It ludicrous anyone read it any other way. Like you earlier wanting to bet on what hasn’t happened yet. lol

except to say that, in addition to arguing that Nintendo doesn't care about 3rd-party relations and constructing strawmen with regard to this discussion elsewhere, it doesn't indicate that you're interested in a good faith discussion on this subject. And considering that, I'm not sure that I shouldn't even bother indulging you beyond this response.

I never said Nintendo doesn’t care about 3rd-party relations. And I don’t see where I used any strawmen, tbh. I also don’t see what’s “bad faith” in how I’m arguing. It might be abrasive how strongly I put forth my opinions, I’ll give you that, but I’m speculating just like everyone else here. I try to argue against something pretty directly.

 

[YolkFolk]

And all of that is to say…had the 3ds not existed, Nintendo wouldn’t have posted losses from 2011-2013. They wouldn’t be huge profits because of the Wii U performance, but it wouldn’t be losses. That’s all I’m saying.

That’s why I get prickly when people blame the Wii U for Nintendo losses during that period.

Look, the GameCube was a very similar “failure”. The GameCube and the Wii U had the same 1st party software sales figures. While the GameCube hardware sales were slightly better than the Wii U, the profit margins were less. Don’t forget, the GC was slashed down to $99 after only 2 years to help buoy poor sales.

Neither system were money loss systems for Nintendo, despite their low sales.

The Wii U just didn’t do anything to offset the actual money loss system at the time…the 3ds.

The GBA wasn’t a money loss system despite it selling FAR less software than the 3ds did.

Nintendo posted profits 2001-2004 despite the GC performance. It wasn’t because the GC made more profit for Nintendo than the Wii U, it didn’t.

The revenue Nintendo pulled in 2011-2014 was more than they pulled in 2001-2004.

The 3ds was the reason for the losses.



If by “sells poorly” you mean how the ps4 pro sold compared to the ps4/slim…then yea, it can sell “poorly” and still elongate the lifecycle.

No one expects Nintendo to sell ~80 million 4K Switches in the next 4 years like they have the non 4K hybrids…do they? This isn’t a ps4->ps5 scenario.

The 4K Switch model will probably end up being 25-30% of total Switch sales. Like the ps4 pro was.



Why would you say that, though?

The investment return is exactly what I said…to keep a large chunk of gamers in the switch ecosystem longer than they otherwise would have. All those gamers who feel Switch graphics/performance is a bit long in the tooth. The same exact reason the Xbox X and ps4 pro existed.

Nintendo will sell more Nintendo games in 2024-2025 than they would have if the 4K enhancement didn’t exist. That’s return in their investment right there.



4 million in two years?? Naw it will be more than that. The “enthusiast gamer tech must have new” demand is higher than that lol.

The number of people who move over to 4K Switch in the 4-5 years after its launch will be much higher than that. The ps4 pro was what, around ~30 million more or less when all is said and done? It will be something more like that.

Even if it’s half that, it’s worth the investment.

We saw with the Wii what the perception of an aging, past it’s prime Nintendo system can do to software sales decline. The 4K Switch will do what a Wii HD should have done.

If the majority userbase buys 1-2 games a year (as Nintendo has suggested) you are talking 6-8 Nintendo games during the 4K Switch lifespan sold to millions of gamers who would have otherwise passed had a 4K Switch upgrade not existed.

Even on the extreme low ball end of this, that’s over $1 billion in 1st party software sales revenue 2022-2025 that might otherwise be lost without such an upgrade. There is obviously value for Nintendo to do this to keep the enthusiast gamer market engaged.

I think we both agree that there will be return value on what the 4K hardware provides to the Switch ecosystem…you just feel it needs to have “true successor” type sales to have value. I’m saying it has value even if it’s ps4 pro type sales.



Oh I’m sure there will be a YoY decline for q2 as well. Can’t have Animal Crossing and a global pandemic shutdown every year, huh?

But this will reflect “only” ~3 million sales difference? I expect that to be made up in the last 2 quarters with the OLED hardware and a better software release schedule.

Conservative Nintendo is still projecting 25.5 million this fiscal year. This within the Nikkei report a few months ago saying Nintendo plans on producing 30 million for this fiscal year. (Compare this to the exact same time in 2020 when Nikkei reported Nintendo planning on 25 million produced last fiscal year and Nintendo projecting 19 million shipped)

Even if it’s off from last fiscal year, it won’t be by much. It will be seen as flat. Hardly a major decline.



?

Donkey Kong Country was an exclusive on the SNES…that’s a true successor system. Nintendo published only about 5 games on the NES after the SNES launched. This is nothing similar to a mid gen revision release.

Super Mario Bros Deluxe was an enhanced port of an old game…ok, I can see Nintendo putting an enhanced port of an old games exclusively on the switch, sure. (Like Xenoblade on the n3ds)

I’ll give you the other GBC games being popular new titles…but my god, we are talking about going from monochrome gaming to color lol. That’s a much major shift than going 1080p to 4K with boosted graphic slider IQ.

I can buy gamefreak dying to have Pokémon in color. I don’t buy them in desperately needing to suffer exclusivity just to get 4K Pokémon models out there.

We’ll see, I suppose.



I meant to point it out when it happens. It ludicrous anyone read it any other way.



I never said Nintendo doesn’t care about 3rd-party relations. And I don’t see where I used any strawmen, tbh. I also don’t see what’s “bad faith” in how I’m arguing. It might be abrasive how strongly I put forth my opinions, I’ll give you that, but I’m speculating just like everyone else here. I try to argue against something pretty directly.

I can assure you that overall the 3DS made more money for Nintendo than the Wii U.

 

[My Tulpa]

Eh…just saw the new Nikkei report about Nintendo being able to only produce 24 million units this fiscal year. Down from the 30 million they said in May.

Bummer, I still say they would have sold ~28 million this fiscal year if supply was there.

 

[My Tulpa]

I can assure you that overall the 3DS made more money for Nintendo than the Wii U.

Well that I agree with!

That wasn’t what the discussion was though.

That’s like me saying the ps3 overall made more money than the Vita…as that somehow negates the reality that Sony hemmoraged billions of $ of deficit from the ps3 while the Vita didn’t.

Vita was a “failure” but it wasn’t the cause of Sony losses in the same way Wii U wasn’t the cause for Nitnendo’s

 

[Pokemaniac]

4 years for Pokemon to finally look like a Switch game instead of an HD Wii game. well, usually we have to wait until new hardware to see it, now we can see it on the system itself

There should be some significant changes when Gen 9 releases in 2023 or so. How far they go remains to be seen, but the second generation on a platform is usually a pretty radical shake-up visually.

 

[ILikeFeet]

There should be some significant changes when Gen 9 releases in 2023 or so. How far they go remains to be seen, but the second generation on a platform is usually a pretty radical shake-up visually.

I'd call Legends that. The world design scope change is a change we thought would happen with gen 9. The "second game jump" is looking pretty unprecedented given the hardware. They've made under the hood improvements before, but they generally resulted in better perfomance and art. On switch, I think they have a larger tech problem than just bad performance (which isn't that bad)

 

[LoneRanger]

I don’t tend to post here but at this point I assume the next Swtich hardware will be:

• Capable of running any PS4/XOne game at (dynamic?) 720p portable and (via DLSS) 1440p docked resolutions with the same quality assets and settings.
• Via backward compatibility (and patches) it will run any Switch game at 720p portable/4k (via DLSS) docked.
• More RAM (8GB, ~7GB usable) and much better CPU than previous gen consoles.
• The console will be capable of very some PS5 current gen ports.
• No new gimmicks, console form factor is more or less the same as Switch OLED.
• Nintendo is working hard with big publishers to bring lots of ports of late PS4/XOne games in the first year.
• Every 2022/23 first party title will be cross gen; Splat 3, Bayo 3, Kirby aTFL, Zelda BOTW2, MK9, Metroid Prime 1R/4, Pikmin 4, M+R2, Xenoblade 3, FE4R, Donkey Kong, Pokémon Gen 9, etc. Most of them will not run very well at the OG model. Starting 2024 we will see some exclusive next hardware games, although smaller projects or remasters will likely appear in both hardwares.
• The price will be $400/$450 and will launch between holiday 2022/early 2023.

 

[ArchedThunder]

I don’t tend to post here but at this point I assume the next Swtich hardware will be:

• Capable of running any PS4/XOne game at (dynamic?) 720p portable and (via DLSS) 1440p docked resolutions with the same quality assets and settings.
• Via backward compatibility (and patches) it will run any Switch game at 720p portable/4k (via DLSS) docked.
• More RAM (8GB, ~7GB usable) and much better CPU than previous gen consoles.
• The console will be capable of very some PS5 current gen ports.
• No new gimmicks, console form factor is more or less the same as Switch OLED.
• Nintendo is working hard with big publishers to bring lots of ports of late PS4/XOne games in the first year.
• Every 2022/23 first party title will be cross gen; Splat 3, Bayo 3, Kirby aTFL, Zelda BOTW2, MK9, Metroid Prime 1R/4, Pikmin 4, M+R2, Xenoblade 3, FE4R, Donkey Kong, Pokémon Gen 9, etc. Most of them will not run very well at the OG model. Starting 2024 we will see some exclusive next hardware games, although smaller projects or remasters will likely appear in both hardwares.
• The price will be $400/$450 and will launch between holiday 2022/early 2023.

1800p and 4K for some PS4/XBO games potentially. Should also be easier to port PS5/Xbox Series to Switch 4k than it was to port PS4/XBO games to the base Switch as well.

 

[Pokemaniac]

Via backward compatibility (and patches) it will run any Switch game at 720p portable/4k (via DLSS) docked.

Some existing Switch games may get patched with DLSS support, but it will probably be the exception rather than the rule, at least for stuff more than a year old when it launches. Even for stuff that gets patched, basic resolution bumps are a lot more likely for older games in most cases.

 

[Deleted member 645]

I honestly can’t wait for some real legit spec info to be released.

 

[ILikeFeet]

I honestly can’t wait for some real legit spec info to be released.

probably won't be until after the new year. or maybe some will get bold if Orin specs come out of GTC

 

[lemonfresh]

And all of that is to say…had the 3ds not existed, Nintendo wouldn’t have posted losses from 2011-2013. They wouldn’t be huge profits because of the Wii U performance, but it wouldn’t be losses. That’s all I’m saying.

That’s why I get prickly when people blame the Wii U for Nintendo losses during that period.

Look, the GameCube was a very similar “failure”. The GameCube and the Wii U had the same 1st party software sales figures. While the GameCube hardware sales were slightly better than the Wii U, the profit margins were less. Don’t forget, the GC was slashed down to $99 after only 2 years to help buoy poor sales.

Neither system were money loss systems for Nintendo, despite their low sales.

The Wii U just didn’t do anything to offset the actual money loss system at the time…the 3ds.

The GBA wasn’t a money loss system despite it selling FAR less software than the 3ds did.

Nintendo posted profits 2001-2004 despite the GC performance. It wasn’t because the GC made more profit for Nintendo than the Wii U, it didn’t.

The revenue Nintendo pulled in 2011-2014 was more than they pulled in 2001-2004.

The 3ds was the reason for the losses.



If by “sells poorly” you mean how the ps4 pro sold compared to the ps4/slim…then yea, it can sell “poorly” and still elongate the lifecycle.

No one expects Nintendo to sell ~80 million 4K Switches in the next 4 years like they have the non 4K hybrids…do they? This isn’t a ps4->ps5 scenario.

The 4K Switch model will probably end up being 25-30% of total Switch sales. Like the ps4 pro was.



Why would you say that, though?

The investment return is exactly what I said…to keep a large chunk of gamers in the switch ecosystem longer than they otherwise would have. All those gamers who feel Switch graphics/performance is a bit long in the tooth. The same exact reason the Xbox X and ps4 pro existed.

Nintendo will sell more Nintendo games in 2024-2025 than they would have if the 4K enhancement didn’t exist. That’s return in their investment right there.



4 million in two years?? Naw it will be more than that. The “enthusiast gamer tech must have new” demand is higher than that lol.

The number of people who move over to 4K Switch in the 4-5 years after its launch will be much higher than that. The ps4 pro was what, around ~30 million more or less when all is said and done? It will be something more like that.

Even if it’s half that, it’s worth the investment.

We saw with the Wii what the perception of an aging, past it’s prime Nintendo system can do to software sales decline. The 4K Switch will do what a Wii HD should have done.

If the majority userbase buys 1-2 games a year (as Nintendo has suggested) you are talking 6-8 Nintendo games during the 4K Switch lifespan sold to millions of gamers who would have otherwise passed had a 4K Switch upgrade not existed.

Even on the extreme low ball end of this, that’s over $1 billion in 1st party software sales revenue 2022-2025 that might otherwise be lost without such an upgrade. There is obviously value for Nintendo to do this to keep the enthusiast gamer market engaged.

I think we both agree that there will be return value on what the 4K hardware provides to the Switch ecosystem…you just feel it needs to have “true successor” type sales to have value. I’m saying it has value even if it’s ps4 pro type sales.



Oh I’m sure there will be a YoY decline for q2 as well. Can’t have Animal Crossing and a global pandemic shutdown every year, huh?

But this will reflect “only” ~3 million sales difference? I expect that to be made up in the last 2 quarters with the OLED hardware and a better software release schedule.

Conservative Nintendo is still projecting 25.5 million this fiscal year. This within the Nikkei report a few months ago saying Nintendo plans on producing 30 million for this fiscal year. (Compare this to the exact same time in 2020 when Nikkei reported Nintendo planning on 25 million produced last fiscal year and Nintendo projecting 19 million shipped)

Even if it’s off from last fiscal year, it won’t be by much. It will be seen as flat. Hardly a major decline.



?

Donkey Kong Country was an exclusive on the SNES…that’s a true successor system. Nintendo published only about 5 games on the NES after the SNES launched. This is nothing similar to a mid gen revision release.

Super Mario Bros Deluxe was an enhanced port of an old game…ok, I can see Nintendo putting an enhanced port of an old games exclusively on the 4K switch, sure. (Like Xenoblade on the n3ds)

I’ll give you the other GBC games being popular new titles…but my god, we are talking about going from monochrome gaming to color lol. That’s a much major shift than going 1080p to 4K with boosted graphic slider IQ.

I can buy gamefreak dying to have Pokémon in color. I don’t buy them in desperate needing to suffer lower userbase exclusivity just to get 4K Pokémon models out there.

We’ll see, I suppose.



I meant to point it out when it happens. It ludicrous anyone read it any other way. Like you earlier wanting to bet on what hasn’t happened yet. lol



I never said Nintendo doesn’t care about 3rd-party relations. And I don’t see where I used any strawmen, tbh. I also don’t see what’s “bad faith” in how I’m arguing. It might be abrasive how strongly I put forth my opinions, I’ll give you that, but I’m speculating just like everyone else here. I try to argue against something pretty directly.

Why whenever I go to this thread there is always these long walls of text

 

[ShadowFox08]

1800p and 4K for some PS4/XBO games potentially. Should also be easier to port PS5/Xbox Series to Switch 4k than it was to port PS4/XBO games to the base Switch as well.

Not holding my break on anything beyond 2k for ps4/Xbox base games.. Switch 4k games is the only thing I feel most certain about, and it's not guaranteed most are possible without DLSS.

 

[ArchedThunder]

Not holding my break on anything beyond 2k for ps4/Xbox base games.. Switch 4k games is the only thing I feel most certain about, and it's not guaranteed most are possible without DLSS.

I’m talking about with DLSS…

 

[My Tulpa]

If by "most Nintendo published games", you mean Nintendo's AAA games, then it's because as Nintendo's future AAA games become more and more ambitious, the OLED model will have a much harder time running the game optimally. This is already the case with Super Mario 3D World + Bowser's Fury, specifically the Bowser's Fury portion, where the game runs at a range of 720p ~ 792p at mostly 60 fps in TV mode, with handheld mode running the game at 720p, but at 30 fps, which is definitely not optimal, since Nintendo usually aims for a consistent 60 fps experience for Super Mario Bros games at all costs, whether it's in TV mode, or whether it's in handheld mode. I think the performance of the Bowser's Fury portion of Super Mario 3D World + Bowser's Fury is an indirect admission from Nintendo that the Nintendo Switch, and by extension the OLED model, is not quite powerful enough to develop the types of ambitious games Nintendo wants to develop, as far as AAA games are concerned.

I hear you, and I agree with all you said. I’m the first one to say the new DLSS Switch exists because Nintendo would prefer their own games have certain standards (steady performance being a biggie). It’s all about them wanting their own games presented better than they are with the older Switch hardware. (They don’t really care how it helps 3rd party porting, that’s just a nice by-product)

I still think they don’t mind having a version of BotW2 and MP4 running on both older models…despite the OLED version having more performance issues.

You can say Nintendo will only hold their nose and do that cause they promised people those games would appear on these old machines…but at the end of the day they want to release their big software to the biggest userbase they have.

The only reason Sony, for example, will stop putting their big titles on the ps4 in 2023, is because most of the active PsN userbase will have moved to the ps5 by then. I think people will be surprised how much (little) the active Switch userbase will be on the 4K Switch during the next 4-5 years.

Think of all the households that have multiple Switches and share games.

I am willing to be shocked and surprised when most their big titles for 2024 skip the OLED Switch and Lite though.

 

[Pokemaniac]

I hear you, and I agree with all you said. I’m the first one to say the new DLSS Switch exists because Nintendo would prefer their own games have certain standards (steady performance being a biggie). It’s all about them wanting their own games presented better than they are with the older Switch hardware. (They don’t really care how it helps 3rd party porting, that’s just a nice by-product)

I still think they don’t mind having a version of BotW2 and MP4 running on both older models…despite the OLED version having more performance issues.

You can say Nintendo will only hold their nose and do that cause they promised people those games would appear on these old machines…but at the end of the day they want to release their big software to the biggest userbase they have.

The only reason Sony, for example, will stop putting their big titles on the ps4 in 2023, is because most of the active PsN userbase will have moved to the ps5 by then. I think people will be surprised how much (little) the active Switch userbase will be on the 4K Switch during the next 4-5 years.

Think of all the households that have multiple Switches and share games.

I am willing to be shocked and surprised when most their big titles for 2024 skip the OLED Switch and Lite though.

Like I said before, nothing that Nintendo has already announced is skipping Switch 1, because they've already announced the games for the platform and backwards compatibility will remove any potential motivation to bring the games forward to make sure they have an audience. No one is debating that.

Where we disagree is what happens after those games. Nintendo is unlikely to make their exit from Switch 1 quickly, but realistically I don't think they can really afford to wait more than a year or two after Dane is out to get that started, and the transition will absolutely start with the high budget "AAA" games that stand to gain the most from the new hardware.

 

[SiG]

I don’t tend to post here but at this point I assume the next Swtich hardware will be:

• No new gimmicks, console form factor is more or less the same as Switch

I highly suspect a true successor would have a gimmick or two to entice people to upgrade.

 

[ILikeFeet]

I highly suspect a true successor would have a gimmick or two to entice people to upgrade.

the hard part is determining what that might be. maybe a camera again, for AR?

 

[Dakhil]

Although this is currently not explicitly related to the DLSS model*, I'd like to share a pie chart graph of Nintendo's market segmentation by percentage, courtesy of JackAvery at Install Base Forum based on Nintendo's recent Q2 earning release.

And here's another pie chart graph of Nintendo's market segmentation by percentage and U.S. dollars, alongside with a similar pie chart graph for Sony as a comparison, courtesy of ZhugeEX.


I personally think the amount of revenue Nintendo earns from hardware and physical games is insane.

 

[ILikeFeet]

I personally think the amount of profit Nintendo earns from hardware and physical games is insane.

it's less how much but the ratio. nintendo makes more from hardware and physical, but only by .25B. whereas that's 73% of Nintendo's revenue, while it's 29% of the revenue.

quite frankly, Nintendo needs to grow the other segments, like digital and IP licensing

 

[Skittzo]

Although this is currently not explicitly related to the DLSS model*, I'd like to share a pie chart graph of Nintendo's market segmentation by percentage, courtesy of JackAvery at Install Base Forum based on Nintendo's recent Q2 earning release.

And here's another pie chart graph of Nintendo's market segmentation by percentage and U.S. dollars, alongside with a similar pie chart graph for Sony as a comparison, courtesy of ZhugeEX.


I personally think the amount of profit Nintendo earns from hardware and physical games is insane.

Both of those charts are revenue though, not profit. Unless I'm misunderstanding something.