• 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.
  • Do you have audio editing experience and want to help out with the Famiboards Discussion Club Podcast? If so, we're looking for help and would love to have you on the team! Just let us know in the Podcast Thread if you are interested!

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

Steam Deck: Am a joke to you?
In sales figures?

Yes. Yes you are.

It's not that the Deck isn't popular, but Valve just doesn't have the capacity to meet demand, and moreover, the Switch has many benefits (far superior ease of use, automatic clock changes, automatic docking and undocking.). These are things the PC platform is inherently incapable of, as they are symptoms of its primary advantages, flexibility and compatibility.
 
Of course that’s true to a tiny extent when you realise “PC gamers” are estimated to be in the high hundreds of millions. It’s potentially as big of a market as mobile to Nintendo.

I don’t really care one way or another but if I was Nintendo as a business I’d put a selection of games on PC. That Smash clone has just hit 10 million players so I’d hate to imagine the kind of revenue they’d generate out of Smash and Kart alone on PC even ignoring the DLC.

The porting costs would also be minimal considering Switch’s power and architecture. Their games would run at 720p/60fps on laptops from five years ago.

"PC Gamers" are people who play ANY game on their PC... and yes, that is HUNDREDS of millions. Hundreds of millions of stay at home parents and grandparents playing Peggle 2, toddlers playing edutainment and Windows Pinball or Purple Place.

However the number of PC Gamers™️is roughly equivalent to the number of gaming PCs capable of running current gen games (Series S' power level and above.), where the pool is closer to 100M. There are only about as many people who actively use Steam to play games as there are Switch games in their entirety, and many, though not most, Steam users don't have "gaming capable" hardware and are using it for social features or lighter titles.

Why target that demographic by pandering to them directly with more development efforts when you could convert many of them to simply BUY your console, and now you have a captive audience which is way more profitable per user?

One Nintendo Switch owner makes orders of magnitude more money than a single Mario Kart Tour user.
 
Tabletop mode is probably one of the more underrated switch features.

How can that be improved if at all?🤔
Bigger, better, brighter screen, more power, higher refresh rate, charging port on the top, better Bluetooth audio. Not MUCH, but basically just refinements to what we have; really the Switch - OLED Model is pretty close to a "perfect" tabletop console. While the original Switch could do it, the OLED Model really, really delivers on a home console experience (Wireless controllers, sit back, don't have to hold the console, full sized controls, large screen.) to the point I can get genuinely absorbed into a game even when on a train. For lack of a better word, while it could be improved in some ways, tabletop mode is already incredibly immersive.
 
0
I probably should have been clearer in my perfect storm scenario, but in this situation I could see Samsung get extremely aggressive to land manufacturing of the Drake SoC by giving a great package deal to Nintendo on OLED displays, RAM and UFS storage.
Samsung would greatly want to be in the Nintendo Switch business because it's a guaranteed high volume seller for the next 4-5yrs and not necessarily on their cutting-edge process.
Bit of wishful thinking unfortunately, Samsung is relatively unintegrated, with its fabs(chip and memory manufacturing), glass plants(screen manufacturing), mobile division(product integration and software support.), these are all separate organisations that compete with one another. Why do you think Samsung Mobile is turning off the tap of Samsung Fab and only using TSMC until 2024? Or why would Samsung Fab and Samsung Display supply most of the parts of the iPhone? Why would Mobile choose the same chip vendor as Apple instead of their own? They're competing businesses, not a monolithic organisation (Which in some ways is a good thing, because if they were they would be one of the world's largest and most dangerous monopolies.)
 
What's this talk of TSMC 5nm and does it relate to Drake/Nintendo in any way?
TSMC has the capability to manufacture on a 5nm process. Nintendo's hardware partner is Nvidia, whose Tegra X1(+) chips, used in the Nintendo Switch, are made by TSMC. The newest Tegra destined for the Nintendo Switch is codenamed Drake, some speculate it may be manufactured by TSMC, others suggest 8nm Samsung. Either way the efficiency gains over 16nm FinFET(2019 Switch, Lite, OLED) will be gigantic, more than double, and more than quadrouple 20nm MOSFET(2017 Switch).
 
Hey, guys, I’m a complete dumbass when it comes to tech talk.

What would be the hesitancy to going lower nodes???

Cost???
Cost 👍

Edit: To expand on this a bit, there's a sweetspot for cost to performance needed for something like a console, especially a portable one where efficiency is also important. 16nm FinFET is currently more or less there that sweetspot is, it's not modern enough to be in demand for things like new phones but it's new enough that they still have production capacity for it. Supply and demand then dictates that 16nm fabrication gets cheaper, at least to a point. Once it becomes so cheap it would be more profitable to replace the line with a newer process, especially as demand for the older process wanes, it will be deprecated, and the sweetspot will move, my guess would be to 12nm or 10nm, so if Nintendo intends on keeping Nintendo Switch (V1, V2, Lite, OLED) alive for years to come there may yet be a V3, Lite V2, etc. with even better battery life due to this smaller process.
 
Many companies related to the technology industry are lowering their profit forecasts for this fiscal year, which in the end is beneficial for the consumer, as it would indicate that prices will be regulated again and may even tend to fall.

What decisions we will see Nintendo take in the medium term, 6 months from now. The sales curve is not sustainable not because they do not have good products, but because their presence is centralized and many products are sold at prices too high for certain economies.

The video game industry needs a boost, it would be the E3 event who would achieve it, but this can not wait until June 2023, I think it is time for the ESA to move the event to January as is already happening with CES, it should even be at a nearby date, with this many participants could do business with technology companies. Also the GDC should be rescheduled at this time in order to consolidate a greater participation of the sector.

If this were to happen, the most predictable thing would be that Nintendo would wait until this time to make its big announcements.
 
Hey, guys, I’m a complete dumbass when it comes to tech talk.

What would be the hesitancy to going lower nodes???

Cost???

Newer nodes are generally reserved for the most expensive high end chips whereas the older nodes get used for more low to mid-range tech. My guess would be that Nintendo will go for 8nm which will be less efficient and therefore less powerful then if they opted for 5nm.
 
Many companies related to the technology industry are lowering their profit forecasts for this fiscal year, which in the end is beneficial for the consumer, as it would indicate that prices will be regulated again and may even tend to fall.

What decisions we will see Nintendo take in the medium term, 6 months from now. The sales curve is not sustainable not because they do not have good products, but because their presence is centralized and many products are sold at prices too high for certain economies.

The video game industry needs a boost, it would be the E3 event who would achieve it, but this can not wait until June 2023, I think it is time for the ESA to move the event to January as is already happening with CES, it should even be at a nearby date, with this many participants could do business with technology companies. Also the GDC should be rescheduled at this time in order to consolidate a greater participation of the sector.

If this were to happen, the most predictable thing would be that Nintendo would wait until this time to make its big announcements.
iNess is at it again.
 
Smaller node size would mean more performance power in Drake, right? By getting to cram more of the thingies that make the Drake do prettier rendering effects, yeah?

Many companies related to the technology industry are lowering their profit forecasts for this fiscal year, which in the end is beneficial for the consumer, as it would indicate that prices will be regulated again and may even tend to fall.

What decisions we will see Nintendo take in the medium term, 6 months from now. The sales curve is not sustainable not because they do not have good products, but because their presence is centralized and many products are sold at prices too high for certain economies.

The video game industry needs a boost, it would be the E3 event who would achieve it, but this can not wait until June 2023, I think it is time for the ESA to move the event to January as is already happening with CES, it should even be at a nearby date, with this many participants could do business with technology companies. Also the GDC should be rescheduled at this time in order to consolidate a greater participation of the sector.

If this were to happen, the most predictable thing would be that Nintendo would wait until this time to make its big announcements.

Don't get me wrong, I think this would be cool as hell (#TeamJanuaryStaysWinning), but wouldn't the Osborne effect also conflict directly with their holiday marketing plans? How much can you sell to an even modestly educated consumer who knows to wait just a few more weeks to see how long they have to wait for the next shinier toy?
 
Hey, guys, I’m a complete dumbass when it comes to tech talk.

What would be the hesitancy to going lower nodes???

Cost???
It boils down to cost, yeah but timing is also a factor. Especially if they started designing it on a different node, they'd need to scrap some of the work they've done and partially start over on that new node which takes a lot more time.
 
0
Newer nodes are generally reserved for the most expensive high end chips whereas the older nodes get used for more low to mid-range tech.
those are just reserved to anyone that’s willing to pay for the price. Doesn’t really have anything to do with power or performance.
 
those are just reserved to anyone that’s willing to pay for the price. Doesn’t really have anything to do with power or performance.

I imagine if 5nm is more efficient when it comes to energy consumption then that would mean Nintendo could increase the speed of the chips as it wouldn't use as much battery power as an 8nm chip. Correct me if I'm wrong. I also feel that the newer Nodes will likely be used by the top tier companies.
 
Smaller node size would mean more performance power in Drake, right? By getting to cram more of the thingies that make the Drake do prettier rendering effects, yeah?
Maybe, but it just means there’s more available.

Because the chips are smaller.

And you can make more of it.

So, in theory, it should be easier to produce.
Don't get me wrong, I think this would be cool as hell (#TeamJanuaryStaysWinning), but wouldn't the Osborne effect also conflict directly with their holiday marketing plans? How much can you sell to an even modestly educated consumer who knows to wait just a few more weeks to see how long they have to wait for the next shinier toy?
If the price is right, then the Osborne effect is diminished. Or rather I should say, if the price is wrong. Exhibit A: PlayStation2 to PlayStation3.


But, if the price is right and it’s worth the price then the Osborne effect may not really be a big deal. If it’s an acceptable price but the product is just not worth that price they’re gonna get burned.

Plus, consider this fact: the switch is a nearly 6 year old product that is still selling, the people that buy the switch now are a different demographic than the people who bought the switch in the first two to three years.

It’s not exactly the spring chicken it once was.
 
How much can you sell to an even modestly educated consumer who knows to wait just a few more weeks to see how long they have to wait for the next shinier toy?
But that's what EVERY company does with CES already, especially with Android phones, and moreover, Nintendo used to attend AND the Switch presentation was in January. Don't forget New 3DS' announcement.

Modestly educated consumers DON'T shop for tech during the holidays. 🤣
 
I imagine if 5nm is more efficient when it comes to energy consumption then that would mean Nintendo could increase the speed of the chips as it wouldn't use as much battery power as an 8nm chip. Correct me if I'm wrong. I also feel that the newer Nodes will likely be used by the top tier companies.
This is correct and in some scenarios the smaller node could be cheaper per SOC.

Though at some point Nintendo/Nvidia have to commit.

And yes newer nodes tend to be bought up by Apple and others for a while.
 
I imagine if 5nm is more efficient when it comes to energy consumption then that would mean Nintendo could increase the speed of the chips as it wouldn't use as much battery power as an 8nm chip. Correct me if I'm wrong. I also feel that the newer Nodes will likely be used by the top tier companies.
They already are. Apple, Qualcomm, Nvidia, AMD, etc. are all using 5 nm or its variants.

As for the first part, they’d go for an agreeable battery life first and foremost.
 
I still think 5nm is on the table. Whether or not Drake uses it would depend on when Nvidia saw the warning signs of glut early on.

That said, whatever they make for 8nm would have been the goal, not the concession. So no, "if Nintendo spent more we could have gotten the TX2" shit here
 
I mean I don’t see how this thing can be priced below $400, and I know some people don’t want to accept that but $400 seems like the absolute floor for this device regardless if they went on 8nm or 5nm.


Like it was not gonna be cheap, at all.


Though cheap may be relative to whom it may concern, but to most it is not the “oh this is a comfy purchase 😍” territory.


Thinking about it as well, it sort of makes sense why people will believe that the Nintendo switch lite will remain for several more years. It kind of helps to remain in the cheap range even if it’s not the most performant, it actually lets you have access into the platform….. at a price.
 
What's this talk of TSMC 5nm and does it relate to Drake/Nintendo in any way?
So there are rumors that Nvidia's paid billions to secure capacity of the N5 family for their next generation of products. Potentially great enough that you can theoretically fit Drake production in there (thus, it's an option on the table, but we don't know for sure).
More recently, there's rumor that Nvidia wants to cut back on N5 capacity due to falling demand from the decline of the crypto craze.

Of course, the decision for what node Drake's on was made a while ago, but the ideal scenario I'm hoping for is that Drake's on N5 all along, and TSMC's unwillingness to make concessions means that Nvidia re-allocates capacity from GPUs to more Drake.
 
Maybe, but it just means there’s more available.

Because the chips are smaller.

And you can make more of it.

So, in theory, it should be easier to produce.

If the price is right, then the Osborne effect is diminished. Or rather I should say, if the price is wrong. Exhibit A: PlayStation2 to PlayStation3.


But, if the price is right and it’s worth the price than the Osborne effect may not really be a big deal. If it’s an acceptable price but the product is just not worth that price they’re gonna get burned.

Plus, consider this fact: the switch is a nearly 6 year old product that is still selling, the people that buy the switch now are a different demographic than the people who bought the switch in the first two to three years.
I mean, making more yet I still worry over it's actual availability when fighting with scalpers 😋 I suppose it would be pretty unlikely to have an acceptable price and the product isn't worth it, but should still consider that 0.1% chance of probability.

That last point is actually a really good thought I hadn't considered; I guess with the demographic caring about more core experiences already saturated, that would be the time to sell fancier hardware while leaving the OG Switch more than capable hardware for those who are more casually into gaming or a younger audience who's parents are looking for a first age-appropriate gaming experience for their children.

But that's what EVERY company does with CES already, especially with Android phones, and moreover, Nintendo used to attend AND the Switch presentation was in January. Don't forget New 3DS' announcement.

Modestly educated consumers DON'T shop for tech during the holidays. 🤣
I dunno, for some reason I can accept that mobile phones have an extremely high turnover rate despite the high cost and minimal jump in tech, but the Switch being six years old, about to experience a great leap in performance at a price half that of most modern flagship phone lines I just can't reconcile simply because of the swaths of angry parents who would be livid at finding out the toy they had just bought for their kid is about to be rendered obsolete (not that it's true, but the way the kids will whine about it to mummy and daddy will make it seem like Nintendo broke into their house and killed their puppy)

The only reason the Switch presentation was in January of 2017 was because the Wii U was deader than dead. 2015 Q3 sales were 1.87m, and 2017 Q3 sales were 0.8m with production killed off that same month as the presentation. Much as I and many of us appreciated a Zelda and Switch launch so far removed from the holidays making it easier for diehards not have to fight the mainstream masses to get one in November, that's only gonna happen again if they have another catastrophic commercial failure.
 
So there are rumors that Nvidia's paid billions to secure capacity of the N5 family for their next generation of products. Potentially great enough that you can theoretically fit Drake production in there (thus, it's an option on the table, but we don't know for sure).
More recently, there's rumor that Nvidia wants to cut back on N5 capacity due to falling demand from the decline of the crypto craze.

Of course, the decision for what node Drake's on was made a while ago, but the ideal scenario I'm hoping for is that Drake's on N5 all along, and TSMC's unwillingness to make concessions means that Nvidia re-allocates capacity from GPUs to more Drake.
Timelinewise, how recent was the "Nvidia wanting to cut back 5nm" rumor versus when Drake could be finalized/taped out? Do you think a delay to next year was to both alleviate any potential shortage (since Nvidia would once again be like "we have all these chips lying around") and have Drake on a higher end node?
 
So there are rumors that Nvidia's paid billions to secure capacity of the N5 family for their next generation of products. Potentially great enough that you can theoretically fit Drake production in there (thus, it's an option on the table, but we don't know for sure).
More recently, there's rumor that Nvidia wants to cut back on N5 capacity due to falling demand from the decline of the crypto craze.

Of course, the decision for what node Drake's on was made a while ago, but the ideal scenario I'm hoping for is that Drake's on N5 all along, and TSMC's unwillingness to make concessions means that Nvidia re-allocates capacity from GPUs to more Drake.
That wasn't a rumor, that came from Nvidia financial reports I think. Totally of $9B
Timelinewise, how recent was the "Nvidia wanting to cut back 5nm" rumor versus when Drake could be finalized/taped out? Do you think a delay to next year was to both alleviate any potential shortage (since Nvidia would once again be like "we have all these chips lying around") and have Drake on a higher end node?
No effect. If Drake wad on 5nm, its production allotment was already considered before the decision to cut back on wafers. A delay won't help it get on a different node
 
That wasn't a rumor, that came from Nvidia financial reports I think. Totally of $9B

No effect. If Drake wad on 5nm, its production allotment was already considered before the decision to cut back on wafers. A delay won't help it get on a different node
I see. So you're saying node choice has to be made first before Drake could come into fruition.

So I guess that just strengthens the argument that "Dane" was the previous node (Samsung 8nm).
 
I see. So you're saying node choice has to be made first before Drake could come into fruition.

So I guess that just strengthens the argument that "Dane" was the previous node (Samsung 8nm).
No, that's extremely unlikely. The guy who leaked the name "Dane" in the first place seemingly admitted he simply got the name wrong.

There is no tangible evidence that a chip codenamed "Dane" ever actually existed.
 
Quoted by: SiG
1
No, that's extremely unlikely. The guy who leaked the name "Dane" in the first place seemingly admitted he simply got the name wrong.

There is no tangible evidence that a chip codenamed "Dane" ever actually existed.
So all those Dannish jokes were a lie, then.
 
I dunno, for some reason I can accept that mobile phones have an extremely high turnover rate despite the high cost and minimal jump in tech, but the Switch being six years old, about to experience a great leap in performance at a price half that of most modern flagship phone lines I just can't reconcile simply because of the swaths of angry parents who would be livid at finding out the toy they had just bought for their kid is about to be rendered obsolete (not that it's true, but the way the kids will whine about it to mummy and daddy will make it seem like Nintendo broke into their house and killed their puppy)

The only reason the Switch presentation was in January of 2017 was because the Wii U was deader than dead. 2015 Q3 sales were 1.87m, and 2017 Q3 sales were 0.8m with production killed off that same month as the presentation. Much as I and many of us appreciated a Zelda and Switch launch so far removed from the holidays making it easier for diehards not have to fight the mainstream masses to get one in November, that's only gonna happen again if they have another catastrophic commercial failure.
I don't buy for one bit that the Switch was yet another "lightning in a bottle" argument. There was precedent for the Switch to happen and I do believe the Wii U's failure helped shape it.
 
I mean I don’t see how this thing can be priced below $400, and I know some people don’t want to accept that but $400 seems like the absolute floor for this device regardless if they went on 8nm or 5nm.


Like it was not gonna be cheap, at all.


Though cheap may be relative to whom it may concern, but to most it is not the “oh this is a comfy purchase 😍” territory.


Thinking about it as well, it sort of makes sense why people will believe that the Nintendo switch lite will remain for several more years. It kind of helps to remain in the cheap range even if it’s not the most performant, it actually lets you have access into the platform….. at a price.
I mean yeah, OLED Model launched at over 400$ in Europe. Since when was anything less than 399.99$ ever reasonable to expect when OLED Model is 349.99$?

I'd say 449.99$ would be the absolute minimum launch price of the next Switch unless OLED falls to 299.99 to fit a 399.99 Drake model into the lineup cleanly. This could mean the discontinuation of the V2 (speculated for some time.) This would explain the OLED Model's relatively tame advertising before launch. OLED Model is what V2 was meant to be, internally it's been price optimised to hell and back, with smaller circuit boards, smaller chips, smaller cooling system, lots of empty space. V2 by comparison is incredibly inefficiently designed with an oversized cooling system and support circuitry the Mariko doesn't need because the V2 just got it dropped in hot off the press with no optimisations. The actual "S" model (comparable to Xbox One S, or Xbox 360 S) of this generation is the OLED Model. The only difference from usual is that this redesign became more expensive, but that's not entirely unusual, the PSP Go was a cost reduced product sold at a "premium" price. It would also hugely simplify production going into a new model. Four forms of the "same" machine on sale at the same time is just unlikely and inefficient. By replacing V2 with the OLED in the market, it allows more flexibility with Drake pricing, could mean if Drake uses the same adaptor, dock and cables, they could even simplify production with only one kind of HDMI cable, one kind of dock, and one kind of adaptor being shipped, along with internal optimisations (OLED and Drake sharing the same hinge design, Joy-Con rails, possibly battery, then OLED and Lite using the same cooling assembly.)
 
Wouldn’t the node and foundry have been decided and locked down years ago at this point? I can’t recall from the NVIDIA leak but 8nm was indicated somewhere in the files, right?
 
Wouldn’t the node and foundry have been decided and locked down years ago at this point? I can’t recall from the NVIDIA leak but 8nm was indicated somewhere in the files, right?
There is no node information in the leak. And yea, the node is something you have to design for, not the other way around
 
0
I just can't reconcile simply because of the swaths of angry parents who would be livid at finding out the toy they had just bought for their kid is about to be rendered obsolete (not that it's true, but the way the kids will whine about it to mummy and daddy will make it seem like Nintendo broke into their house and killed their puppy)

I don’t see why this is ever going to be a concern for companies launching better hardware this far in, even the family focused Nintendo.

People only just buying the Switch are incredibly late - 5+ years out now. They are not buying titles on day one for their children, and they have the entire lineup of software from the generation available. Popular service games like Fortnite and Fall Guys, evergreen titles like and Minecraft etc. are going to be fully supported for quite some time as well.

Any outcry around a new $400-$500 system dropping months after they buy a new Switch in 2022/23 won’t even register at Nintendo, reputationally or monetarily.
 
Bit of wishful thinking unfortunately, Samsung is relatively unintegrated, with its fabs(chip and memory manufacturing), glass plants(screen manufacturing), mobile division(product integration and software support.), these are all separate organisations that compete with one another. Why do you think Samsung Mobile is turning off the tap of Samsung Fab and only using TSMC until 2024? Or why would Samsung Fab and Samsung Display supply most of the parts of the iPhone? Why would Mobile choose the same chip vendor as Apple instead of their own? They're competing businesses, not a monolithic organisation (Which in some ways is a good thing, because if they were they would be one of the world's largest and most dangerous monopolies.)

Literally stated a perfect storm scenario...
Samsung mobile aren't that unhinged from the foundry business that they reserve initial cutting-edge nodes for its own mobile chips.
All of these big business tech companies learn how to play the illusion competing without being seen as a threat to draw in attention from antitrust laws.
 
Literally stated a perfect storm scenario...
Samsung mobile aren't that unhinged from the foundry business that they reserve initial cutting-edge nodes for its own mobile chips.
All of these big business tech companies learn how to play the illusion competing without being seen as a threat to draw in attention from antitrust laws.
Easier to say Samsung Foundries is shit.

Cause they are
 
since nvidia is making their own cpu, would nintendo ditch ARM and go fully nvidia in the future?
Only if they are suitable for a gaming system. I’m going to go with no. They’ll stick with ARM regular CPUs until I see otherwise.
Easier to say Samsung Foundries is shit.

Cause they are
Hey now, their memory and storage is good.
 
0
Samsung seems less and less relevant to Nintendo's current or future needs if TSMC is already in bed with Nvidia. I also remain cautiously optimistic about any SEA/China politics getting in the way of progress and billions in revenue. At least for the foreseeable.
 
Samsung seems less and less relevant to Nintendo's current or future needs if TSMC is already in bed with Nvidia. I also remain cautiously optimistic about any SEA/China politics getting in the way of progress and billions in revenue. At least for the foreseeable.
They supply RAM and Storage which would be very relevant I feel. And the screen.
 
Why do you think Samsung Mobile is turning off the tap of Samsung Fab and only using TSMC until 2024? Or why would Samsung Fab and Samsung Display supply most of the parts of the iPhone?
Just to clarify, Samsung has decided to use future Snapdragon SoCs for future Samsung Galaxy smartphones until 2024. And Qualcomm's making decisions on which process node is being used to fabricate future Snapdragon SoCs, not Samsung.

And I don't know if that's necessarily always the case. According to iPhone 13 Pro Max teardowns, the only components provided by Samsung and are fabricated by Samsung are the OLED display and Qualcomm's X65 5G modem (here and here). Most of the other components on the iPhone 13 Pro Max are fabricated using foundries outside of Samsung (e.g. TSMC, Kioxia, etc.).

since nvidia is making their own cpu, would nintendo ditch ARM and go fully nvidia in the future?
If the question's if Nintendo's going to stop using CPUs designed around Arm's ISA, then I don't believe so.

Grace's still an Arm CPU since Grace's based on Arm's next-gen Armv9 Neoverse CPU design.

If the question's if Nintendo's going to stop using Arm's future Cortex-A CPU designs in favour of Nvidia's custom Arm based CPU designs, depends on if Nvidia's custom Arm based CPU designs are more performant and power efficient compared to Arm's future Cortex-A CPU designs.

Saying that, I'm curious about Grace-Next since Grace-Next seems to be a custom Arm based CPU design from Nvidia.
 
Last edited:
"PC Gamers" are people who play ANY game on their PC... and yes, that is HUNDREDS of millions. Hundreds of millions of stay at home parents and grandparents playing Peggle 2, toddlers playing edutainment and Windows Pinball or Purple Place.

However the number of PC Gamers™️is roughly equivalent to the number of gaming PCs capable of running current gen games (Series S' power level and above.), where the pool is closer to 100M. There are only about as many people who actively use Steam to play games as there are Switch games in their entirety, and many, though not most, Steam users don't have "gaming capable" hardware and are using it for social features or lighter titles.

Why target that demographic by pandering to them directly with more development efforts when you could convert many of them to simply BUY your console, and now you have a captive audience which is way more profitable per user?

One Nintendo Switch owner makes orders of magnitude more money than a single Mario Kart Tour user.
What a strange way of making your point by limiting possible Switch port PC players to Series S levels of hardware which costs around $1000 going by the last couple of years of PC hardware prices.

The Switch when it runs games at 720p/60fps is around 20x weaker than Series S so no you wouldn’t need anything like that sort of hardware to run Switch ports… Yes there are Peggle / Facebook game PC users etc but the figure who play games on hardware on par with Switch is probably closer to 300-500 million users.

I’m not going to keep derailing the thread though.
 
0
Going a bit off-topic here: I know the answer is probably "we don't know" and any speculations are pretty much just lucky guesses, but realistically, what kind of GPU clock speed should we expect from Drake? Is the same as the current switch docked clocks (768 MHz) realistic? Would going lower/higher maybe cause some issues with backwards compatibility or is it completely unrelated?
 
If the OLED is an indicator of any future Nintendo hardware, not for long.
I’m not sure I follow, only Samsung makes the mobile OLED displays these days.

Unless you mean that Nintendo will remove the OLED? I’m not sure why they’d do that though.


Going a bit off-topic here: I know the answer is probably "we don't know" and any speculations are pretty much just lucky guesses, but realistically, what kind of GPU clock speed should we expect from Drake? Is the same as the current switch docked clocks (768 MHz) realistic? Would going lower/higher maybe cause some issues with backwards compatibility or is it completely unrelated?
Not really off topic, perfectly on topic I’d say with the focus of the thread and all.

To answer your question, or rather to give you a safe guess only assume that it will operate at the current Nintendo switch clock speeds in both modes.

Portable is 307MHz, 384MHz and 460MHz.

Docked it is only 768MHz unless I’m mistaken.

If it’s higher than that, it’s gravy. If not then there’s not much to do about it but it starts to get into a territory that I call “why bother even turning it on at that point? You’re just wasting power”.



For your second question about backwards compatibility, Alovon told me once that some games can be tied to the GPU clock logic, so it can in theory. However I’m not aware of any switch game that is tied to that or heard of any switch game that is tied to that. Almost all are about better performance when used an overclocked Erista/Mariko chip. The ones that didn’t get a major perf boost were a result of the memory bandwidth being limited and that’s pretty much it.

It’s possible that Nintendo is working on a software solution for the backwards compatibility this time as nvidia has the software stack that can allow for it. But nothing definitive here. I don’t think a TX1 will be virtualized into the switch a la Wii in Wii U scenario.


If I were to be very optimistic with hopes for clock speeds, here’s what I’d like: 500MHz portable, 1000MHz docked.

in a world where it’s on the best node


Note: me having very optimistic hopes is not me being realistic, these are separate. Me being realistic is expecting switch clocks🫠, which I do.
 
I’m not sure I follow, only Samsung makes the mobile OLED displays these days.
I think Nintendojitsu's saying there's no guarantee Nintendo's going exclusively Samsung for the RAM and internal flash storage since Nintendo's using Micron's LPDDR4X modules for the OLED model and Toshiba's (Kioxia's) eMMC 5.1 chip for the Nintendo Switch.

 
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