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

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

It's not just this one guy though. It's also Kopite, MLID, and even Rich from DF. Now maybe it's possible that they are all getting this info from the same few NVidia related sources who happen to be wrong about the node size, but that seems unlikely so far.

There are of course posters dismissing things because they don’t like what they hear - but at the same time you give too much credit to some of the dismissed sources, and not enough to the thread.

Enough posters in here have done heavy lifting explaining why sources like MLID should be taken with a ton of salt, and how claims like Samsung 8NM don’t make sense. In fact as far as I can tell every single source saying “it’s Samsung 8NM” seemingly does so on the basis of two things: kopite7kimi saying so a while back, and “because cheap” (which seems to be debatable).

This includes Digital Foundry. They have taken the approach to accept what kopite7kimi said as a constant, and have done much of their speculation / theorizing based off of it. Do you really expect this thread to just nod and say, “yeah I guess they’re right” when it feels like they aren’t viewing things with the same lens? Until they take the approach to remove it as a constant and consider some of the work done here I doubt they’ll be viewed all that favorably.

I agree that there’s a long history of people disagreeing with posts around dates. People get heated around those topics, but they haven’t been technical* disagreements, just emotional ones.

*for the most part - aside from talks of where T239 could not have been mass produced at a given time.

Edit: tone.
 
Last edited:
I think unless someone has very close sources to hardware manufacturers, its very hard to know the node process of SoC.

Software sources can know stuff about clock speeds/flop’s performance and such, with that someone can speculate about node process.

But everything is purely speculation.

Honestly, we should focus on just knowing clock speeds/performance. The rest is somewhat irrelevant for us. Battery life at worst will probably be like Switch 2017, at best as V2 model.
 
An insider can be reputable but have incorrect info on a specific project because of their sources.

I am aware that MLID has been accurate in the past, but they may be off the mark here.

It seems very easy for a game of telephone to happen, where a developer speaks to a tester speaks to a friend speaks to their cat and discussions like "I'm not sure what it is, but it's related to Orin, so maybe 8nm" become "there's a chance it's 8nm" becomes "I'm thinking it's 8nm".

It's different from other types of leaks like having access to design documents and reporting info directly from there.
 
Nope, that was an assumption based on site code, there’s not actually any last gen versions.
Oh, ok thanks for the correction.
The lpddr5x discovery and arguably the dock fan kind of matches up with what mlid said about 4 tflops docked.
Was I the only one who saw the actual video of MLID? He was guessing that it would have fan in the dock.
 
In term of "tech insiders" talking about the node, I only know of 2. One is kopite7kimi, who is certainly a legitimate inside source, but also is self-admittedly not 100% on T239's process node, and MLID, a person who seems like their sources are at best shakey. Every other comment on process node seems to originate with one of those two.

If I thought 8nm worked, then I would be assuming 8nm. And in fact I did for the longest time - Orin is on 8nm, RTX 30 is on 8nm, it shouldn't be surprising that their sibling is also on 8nm. Kopite said so, too. I would have been wildly surprised.

But the power numbers just don't make sense. We don't have enough data, but the Jetson power curves basically match up with power testing I've seen of Nvidia gaming cards, so I don't see any reason to not extrapolate from what we've got. Until someone has a better analysis, I am cautiously pessimistic about the ability to have Drake run on 8nm even irresponsibly low clock speeds.

"Well, maybe Nintendo will just clock under that!" By irresponsible clock speed, I mean that Nintendo is clock at a level so low that they could get the same performance by clocking fewer cores (and thus a cheaper chip) higher. And at the same - or better - power draw. Its irresponsible because it is a decision that isn't cheap and isn't powerful all for a chip that isn't backwards compatible. This is "because Nintendo" taken to the point where it's something Nintendo has never done.

Cautiously pessimistic - I'm not so arrogant that I believe back of the envelope calculations are outsmarting Nvidia's team of electrical engineers. And the moment a reliable source comes out and says 8nm, I'm ready to believe it. I don't think that it's a bad decision if it works. I just don't currently buy the argument that it works.

In term of "tech insiders" talking about the node, I only know of 2. One is kopite7kimi, who is certainly a legitimate inside source, but also is self-admittedly not 100% on T239's process node, and MLID, a person who seems like their sources are at best shakey. Every other comment on process node seems to originate with one of those two.
The only thing I like about MLID is that he sometimes brings genuinely interesting and knowledgeable people to his podcasts. I watch his content regularly, but as I said before, everything he says or shows I do not take seriously. I've seen him make false predictions and when proven wrong, quietly delete videos/posts to wipe off any evidence. At any rate, I don't mean to shit-talk about someone I don't really care for, but I do find his content somewhat entertaining and it helps me stay upto date with the word on the street.

If I thought 8nm worked, then I would be assuming 8nm. And in fact I did for the longest time - Orin is on 8nm, RTX 30 is on 8nm, it shouldn't be surprising that their sibling is also on 8nm. Kopite said so, too. I would have been wildly surprised.
When I used to think 8nm was a real possibility (because I thought Switch 2 was coming out in 2023) I too had these concerns, going over every variation of Orin over and over again and rationalizing how Nintendo could make it happen. At one point I thought they'd vastly raise the power envelope and go the route of the Steam Deck with the Orin NX, and at other times I thought they might go with the Orin Nano and double the memory instead. But I always knew AGX Orin wasn't feasible for one simple reason - die size, power and heat be damned.

Now that it's (probably) launching in 2025, and we have concrete evidence from the Nvidia leak, there isn't a single plane of existence where SEC8N makes sense. Anyone who still claims otherwise is either a) delusional or b) has a hate boner or, most likely, c) is farming rage bait
 
I think unless someone has very close sources to hardware manufacturers, its very hard to know the node process of SoC.

Software sources can know stuff about clock speeds/flop’s performance and such, with that someone can speculate about node process.

But everything is purely speculation.

Honestly, we should focus on just knowing clock speeds/performance. The rest is somewhat irrelevant for us. Battery life at worst will probably be like Switch 2017, at best as V2 model.
This.

It's not a bit of information listed on documentation for devkits or anything, as it doesn't matter to development. Getting such information on the SoC would be tough and limited to a very select group of potential sources. Will some have such access? Of course; but most will not & the vast majority of SoC process node discussion will be speculation based on varying levels of belief and self justification/thought.
 
Saying to someone to get their head checked and to get a psychiatric evaluation because they discussed an unpopular (potentially debunked) theory/opinion doesn't seem like the most constructive thing to do. Neither does the "cheerfulness" those comments received from others, but that's just my opinion. Ignoring or not answering every other comment about something that has already been debated is just as easy. Again, maybe that's just me.
 
Someone should compiled a list of supposed specs by leakers. Would be a good for a retrospective. And to know who's credible for Switch 3 Speculation


I'll start

MLID
8xA78C
8nm
LPDDR5
4TF Dock
"Very low" clock of 800mhz or less undocked

RedGamingTech
A78AE
10SM
Samsung 8LPE
8GB RAM
7"; 720p; OLED; 120Hz
64GB eMMC
DLSS 2 and FSR 3
Roughly on par with a PS4 when docked
Similar Switch OLED design, says who? Extremely unlikely.
Full backward compatible game & accessories,
And that is exactly why RGT is one of the least reliable soothsayers out there. Dude literally spits out incoherent gibberish in the middle of a gazillion "basically." Back before Ampere and RDNA2's launch I watched some of his stuff, and after seeing him miss more often than not I straight up blocked him. Hasn't shown up on my feed since.
 
Last edited:
There are of course posters dismissing things because they don’t like what they hear - but at the same time you give too much credit to some of the dismissed sources, and not enough to the thread.

Enough posters in here have done heavy lifting explaining why sources like MLID should be taken with a ton of salt, and how claims like Samsung 8NM don’t make sense. In fact as far as I can tell every single source saying “it’s Samsung 8NM” seemingly does so on the basis of two things: kopite7kimi saying so a while back, and “because cheap” (which seems to be debatable).

This includes Digital Foundry. They have taken the approach to accept what kopite7kimi said as a constant, and do most (all?) of their experiments and theorizing based off of it. Do you really expect this thread to just nod and say, “yeah I guess they’re right”, because of their production values and platform? Cause that’s all their current predictions offer us until they take the approach to remove it as a constant and repeat or verify the work done here.

I agree that there’s a long history of people disagreeing with posts around dates. People get heated around those topics, but they haven’t been technical* disagreements, just emotional ones.

*for the most part - aside from talks of where T239 could not have been mass produced at a given time.
I think the reason most people are upset when some mention MLID and Kopite7Kimi is probably the amount of outlets and popular thech youtubers has mention them as prove. Like the way digital foundry found it to be realistic is quite upsetting, but understandable in the point of view that Nintendo is the same as they were 10 years ago and hyping up Nintendo hardware can make people question your credibility of someone who analyse tech.

Like from all the evidence we gather is that 8nm seems more and more unlikely through the years with all the evidence we have gather (well the great people who investigate the information and well worded speculation that has come true).

The day that 12GB ram and the usage of LPDDR5X has pretty much giving me all the confidence that Nintendo aren't cheapening out on the hardware and specs.
Also the way Nintendo partnered with Nvidia up and asked for a custom chip with DLSS and all of Nvidia magic is something i would have never expected.

Overall i'm extremely excited since, this might be one of the last ,,true generational leap'' I'll see in my life time, which is quite upsetting.

the only worrying thing might be the name of the system, but if it has good marketing, like the Switch, then we're in pretty good.
 
And that is exactly why RGT is literally one of the least reliable soothsayers out there. Dude literally spits out incoherent gibberish in the middle of a gazillion "basically." Back before Ampere and RDNA2's launch I watched some of his stuff, after seeing him miss more often than not I straight up blocked him. Hasn't shown up on my feed since.
RGT as in RGT85??
That guy is a jokester
What’d he say??
 
Saying to someone to get their head checked and to get a psychiatric evaluation because they discussed an unpopular (potentially debunked) theory/opinion doesn't seem like the most constructive thing to do. Neither does the "cheerfulness" those comments received from others, but that's just my opinion. Ignoring or not answering every other comment about something that has already been debated is just as easy. Again, maybe that's just me.
I also don't think it's cool to judge the intention of something written by a non-native speaker of the language and translated by online tools, we have no idea what the guy's speech connotes in Chinese.
 
RGT as in RGT85??
That guy is a jokester
What’d he say??
Assuming you're talking about Switch 2 specs shared by RGT:

A78AE
10SM
Samsung 8LPE
8GB RAM
7"; 720p; OLED; 120Hz
64GB eMMC
DLSS 2 and FSR 3
Roughly on par with a PS4 when docked
Similar Switch OLED design, says who? Extremely unlikely.
Full backward compatible game & accessories,
 
I was wondering, since Devkits are already being distributed, and in fact we are in the preparation phase for mass production, would it be possible to track the T239s produced themselves? Maybe finding out at least whether it's being made by Samsung or TSMC would help a lot.
 
Assuming you're talking about Switch 2 specs shared by RGT:

A78AE
10SM
Samsung 8LPE
8GB RAM
7"; 720p; OLED; 120Hz
64GB eMMC
DLSS 2 and FSR 3
Roughly on par with a PS4 when docked
Similar Switch OLED design, says who? Extremely unlikely.
Full backward compatible game & accessories,
Would make sense for a Switch pro, but not the successor.
 
it does feel like every piece of external information that isn't 100% 'verifiable' is instantly debunked. go back to when the magnet thing first gets traction via unverifiable sources and straight away a whole army of posters have 'debunked' it saying it's literally impossible and couldn't work.

i guess it's nice to keep an open mind, something might be unlikely but not impossible. smart as many of the posters are here we don't know everything, it's certainly not worth getting emotionally invested in the need to disprove things before the full facts come to light.
 
It's not just this one guy though. It's also Kopite, MLID, and even Rich from DF. Now maybe it's possible that they are all getting this info from the same few NVidia related sources who happen to be wrong about the node size, but that seems unlikely so far.
it's super likely in this case. Not to jump on you, just saying.

Kopite has said that they've presumed the node based on Orin, and think (possibly correctly) that if it wasn't on 8nm they would have heard.

MLID has quoted (supposed) Nvidia sources on the 8nm question in this videos... and none of them have said 8nm. One said "I haven't worked on T239 since the beginning, we pitched several chips, Nintendo went with a cost reduced Orin" which says nothing about the node. A second person who says they were not on the T239 project says "8nm is a good choice for Nintendo because of the price, and you know Nintendo wants cheap."

MLID's own quotes from his sources show that his sources don't know what the node is, and don't say it's 8nm anyway.

And Rich has said that there are multiple node possibilities, recently talking about TSMC 7nm as a possibility. He is, to my knowledge, going off of kopite's statements, as he is a known Nvidia leaker with a good track record. If Rich has another source - and he might - clearly even Rich is reserving the possibility that his source is incorrect.

Looking at the exact three you've quoted there is only one primary source - kopite - who has admitted to not being 💯 on it. I'm completely open to it being 8nm. If a new, reliable source comes forward, or if kopite makes a definitive statement, or if there is strong evidence that Rich has an additional source, yeah, that is where my assumption goes. If I didn't have power draw information, it would still be there.

Basically I have strong data for 8nm being literally not possible and very weak support for 8nm being the selected node. That's why I say I'm betting on Not 8nm, but only with 60% confidence.
 
Across Xbox Series X|S and Xbox One, Xbox apparently has about a hundred million active users.

Across PS4 and PS5, PlayStation apparently has about a hundred and ten million active users.
Consider Xbox hardware has sold ~half as well as PlayStation these last generations, that seems hard to believe. Like, either every PS5 user gained is a PS4 user lost while Xbox One and Series have no (even negative?) overlap, or every Xbox machine is shared by twice as many people as every PlayStation machine.
 
I also don't think it's cool to judge the intention of something written by a non-native speaker of the language and translated by online tools, we have no idea what the guy's speech connotes in Chinese.
I considered that too but after looking at other messages and considering that it wasn't in just one and written in one way I removed the part about a potential language barrier from my comment. If it were just "look at your head" it would've been different.
 
Assuming you're talking about Switch 2 specs shared by RGT:

A78AE
10SM
Samsung 8LPE
8GB RAM
7"; 720p; OLED; 120Hz
64GB eMMC
DLSS 2 and FSR 3
Roughly on par with a PS4 when docked
Similar Switch OLED design, says who? Extremely unlikely.
Full backward compatible game & accessories,
Aah.
Can’t wait to get home and watch his latest YouTube hijinks!
RGT85 is not to be taken seriously.
He makes up imaginary scenarios of people disagreeing with him and attacks them in his videos. His YouTube channel is all skits to me..kinda like angry video game nerd lol.
 
Consider Xbox hardware has sold ~half as well as PlayStation these last generations, that seems hard to believe. Like, either every PS5 user gained is a PS4 user lost while Xbox One and Series have no (even negative?) overlap, or every Xbox machine is shared by twice as many people as every PlayStation machine.
Microsoft also includes PC users, meanwhile Sony doesn't, that's probably the reason they wanted to force PC users to have a Playstation account to boost it's numbers and impress investors.

If we're only talking about Xbox console itself, then i can maybe see 70 million if we include Xbox one and Series X.
 
Aah.
Can’t wait to get home and watch his latest YouTube hijinks!
RGT85 is not to be taken seriously.
He makes up imaginary scenarios of people disagreeing with him and attacks them in his videos. His YouTube channel is all skits to me..kinda like angry video game nerd lol.
I could be mistaken but I think it's just the same ones he shared before (pretty old). Someone here was just going over "previous leaker claims" and comparing those with what we know about Switch 2 specs now, including RGT's claim.

Then you asked what RGT said, so I shared what RGT shared previously (which I think was months ago).

If you haven't seen THAT video yet by RGT, then yeah, beware :)
 
0
it does feel like every piece of external information that isn't 100% 'verifiable' is instantly debunked. go back to when the magnet thing first gets traction via unverifiable sources and straight away a whole army of posters have 'debunked' it saying it's literally impossible and couldn't work.

i guess it's nice to keep an open mind, something might be unlikely but not impossible. smart as many of the posters are here we don't know everything, it's certainly not worth getting emotionally invested in the need to disprove things before the full facts come to light.
I don't recall the magnet info being 'debunked', just facing heavy scrutiny with its sources - which were, I believe, ninspider and Mobapad the accessory manufacturer. Which I think is fair to do. One is just a random person who has never posted here before and gave enigmatic hints, and the other has an incentive to kick off a little marketing campaign. I don't remember if there were other sources, please remind me if there were. There was definitely skepticism about how the mechanism could work safely but there were many posts saying "Nintendo has a massive engineering team and millions of dollars spent in R&D, they could probably figure it out", and there are devices out there that have magnetic detachable joy-con like controllers so it's definitely not impossible.

In the case of 8nm here the sources are also facing heavy scrutiny, and that combined with the amount of technical analyses of the known components and power curves and such is why people are hesitant to accept that information on face without some plausible explanation - like Nintendo and Nvidia doing some insane power saving tricks on 8nm, or the chip cutting the # of SMs. I think skepticism and scrutiny are always the right approach, and it makes sense with the types of dubious claims being made here (4 TFLOPS & 8nm with LPDDR5 RAM). We exercised the same approach with the shipping data and NVN2 leak as well, even though those were fairly clear cut, because it's worth it to be certain. That's why LiC has written out comprehensive essays on those two datamines.

8nm is not impossible, but I would like a strong reason to believe it if the given data about GPU size, estimated dimensions, rumored performance, and Nintendo's approach to handhelds suggest otherwise.
 
Since the specs leak the discussion seems to mainly revolve about this node thing. Can some help me understand why that's so important? I guess it's something to do with clock speeds? What would be ideal between 4nm and 8nm and which one looks most likely right now?
 
I just had a thought. Is there any possibility of analog triggers this time around? Those Joy Cons look a decent big larger than the current ones, surely it isn't all battery inside, right?

And if it is, could that possibly be used to bolster the main tablet's total runtime?
 
Since the specs leak the discussion seems to mainly revolve about this node thing. Can some help me understand why that's so important? I guess it's something to do with clock speeds? What would be ideal between 4nm and 8nm and which one looks most likely right now?
More than just clock speed.

8nm means shorter battery life. More bulky. More cooling required, gets hotter.

Because of those, that gives downward pressure to clock speed, which therefore also means downward pressure on performance of Switch 2.

Which all in itself does not really rule out 8N.

However, from nvidia leaks we know there's 12SMs. 12SMs and 8N simply doesn't seem to math/physic with each other at all. If you haven't seen this yet, highly recommended reading!
 
Consider Xbox hardware has sold ~half as well as PlayStation these last generations, that seems hard to believe. Like, either every PS5 user gained is a PS4 user lost while Xbox One and Series have no (even negative?) overlap, or every Xbox machine is shared by twice as many people as every PlayStation machine.
Ah, I've mixed up my numbers! 100 was the total for Xbox Network, which I forgot now includes PC! As of last available data I can find, it's topped 110M.

As of 2022, the number of Xbox Live users on console is 51.3M, with the number of Xbox Live Gold and/or Game Pass Users at 39.7, these figures per Statista. I don't think we have more up to date figures.

As of 2023, Nintendo Switch Online is at 38 million paying subscribers. Since you need to pay to play online on both platforms, this implies that the "online userbase" of both is extremely similar. Neither I would consider unhealthy.

For reference, Playstation Plus across all offerings is at 47.4M as of 2023.

While 50 million users is half the other platforms, it's apparent from the number of subscribers that the Xbox ecosystem has a very high level of engagement. On Xbox, you're more likely to subscribe to one or more Xbox subscriptions than you are to ignore them, while on PlayStation and Nintendo Switch, these figures are flipped.

All in all, my point is that Xbox is a healthy and competitive ecosystem. I believe the apparent lack of faith in it from Microsoft is less to do with it being an unprofitable, unsustainable business model, and much more with their desire, and failure, to monopolise the industry through sheer spending.

The Xbox development environment also has the benefit of being a Windows development environment. If you build for Xbox it's trivial (relatively speaking) to build for Windows. This gives Xbox an "ecosystem leg up" in a sense, you can develop once and build for 100M users across console and PC, and reach even more with Steam (albeit with huge overlap).

Between this and the comparatively similar number of people playing online (since you must pay to play online), I think my stance on the three ecosystems as not as different as we tend to think remains valid.
 
Aah.
Can’t wait to get home and watch his latest YouTube hijinks!
RGT85 is not to be taken seriously.
He makes up imaginary scenarios of people disagreeing with him and attacks them in his videos. His YouTube channel is all skits to me..kinda like angry video game nerd lol.
RGT is this case is Red Gaming Tech not RGT85 lol
 
More than just clock speed.

8nm means shorter battery life. More bulky. More cooling required, gets hotter.

Because of those, that gives downward pressure to clock speed, which therefore also means downward pressure on performance of Switch 2.

I see, thanks! So it's between 8 and 4 now? Which one looks likes it's going be?
 
I see, thanks! So it's between 8 and 4 now? Which one looks likes it's going be?

Just in case you haven't seen my edit to other comment yet, check it out (I added link to Thraktor's writeup).

Yes - so the reason why it's mostly debate between 8N and 4N is because nvidia products are predominantly fabbed on either Samsung 10nm (optimized) or TSMC 5nm (optimized).

This doesn't rule out that they could use a node process that's not SEC8N or TSMC 4N, but that would be kind of surprising considering it's probably cheaper for nvidia to just fab T239 on the same foundry as their other products.
 
What would be ideal between 4nm and 8nm and which one looks most likely right now?
TSMC 4N is the process that the Ada Lovelace GPUs are manufactured on, and it is actually 5nm. Yeah, it's confusing.

There is no hard evidence either way, but based on analyses, 5nm looks more likely as it allows the large GPU indicated by the Nvidia leak to take up less physical space while being clocked higher. The RAM being LPPDR5X may also indicate it - as 5X is higher bandwidth, more bandwidth can feed a system running at higher speeds better.
 
I just had a thought. Is there any possibility of analog triggers this time around? Those Joy Cons look a decent big larger than the current ones, surely it isn't all battery inside, right?
I wish. I think it's possible, even just to have more "feature parity" with other platform's controllers.
 
Since the specs leak the discussion seems to mainly revolve about this node thing. Can some help me understand why that's so important? I guess it's something to do with clock speeds? What would be ideal between 4nm and 8nm and which one looks most likely right now?
Smaller the node, the higher the clock speed while wasting less energy. Higher clock speed = better performance
 
Someone should compiled a list of supposed specs by leakers. Would be a good for a retrospective. And to know who's credible for Switch 3 Speculation


I'll start

MLID
8xA78C
8nm
LPDDR5
4TF Dock
"Very low" clock of 800mhz or less undocked

RedGamingTech
A78AE
10SM
Samsung 8LPE
8GB RAM
7"; 720p; OLED; 120Hz
64GB eMMC
DLSS 2 and FSR 3
Roughly on par with a PS4 when docked
Similar Switch OLED design
Full backward compatible game & accessories,
Impressive how this thread was able to at least partially debunk both of these claims with publicly available information. Given that the ram amount was more or less expected, it will be interesting to see what other expectations get met (or not).
 
Overall i'm extremely excited since, this might be one of the last ,,true generational leap'' I'll see in my life time, which is quite upsetting.

It’s only positive news for me.

I’ve been waiting for that threshold where my major issues with game visuals are resolved - lower resolutions being near the top, followed by poor framerates. For Nintendo’s purposes, I really think this is it. This is where I’m expecting real diminishing returns moving forward, and that’s a great thing. I don’t want the next Wind Temple to have me reeling as the image turns to garbage during the crescendo of a boss fight. I’m still a little bitter about Tears of the Kingdom having to be enjoyed in that state.

Maybe I’m being overly optimistic, but I expect to care far less about technical advancements on the next-next hardware from Nintendo, at least with respect to Nintendo’s own games and AA game quality. The AAA third party scene is reaching an impasse of sorts, and for all we know the majority are forced to stop raising the bar because it’s just costing too much. That could see Switch 2/3 being right where it needs to be, even if it’s not at the bleeding edge.
 
Since the specs leak the discussion seems to mainly revolve about this node thing. Can some help me understand why that's so important? I guess it's something to do with clock speeds? What would be ideal between 4nm and 8nm and which one looks most likely right now?
4nm is around double performance per watt. Which in this case, probably more or less equals actual performance. The gap is far to wide to recoup by tanking the battery life.
 
I don't recall the magnet info being 'debunked', just facing heavy scrutiny with its sources - which were, I believe, ninspider and Mobapad the accessory manufacturer. Which I think is fair to do. One is just a random person who has never posted here before and gave enigmatic hints, and the other has an incentive to kick off a little marketing campaign. I don't remember if there were other sources, please remind me if there were. There was definitely skepticism about how the mechanism could work safely but there were many posts saying "Nintendo has a massive engineering team and millions of dollars spent in R&D, they could probably figure it out", and there are devices out there that have magnetic detachable joy-con like controllers so it's definitely not impossible.

In the case of 8nm here the sources are also facing heavy scrutiny, and that combined with the amount of technical analyses of the known components and power curves and such is why people are hesitant to accept that information on face without some plausible explanation - like Nintendo and Nvidia doing some insane power saving tricks on 8nm, or the chip cutting the # of SMs. I think skepticism and scrutiny are always the right approach, and it makes sense with the types of dubious claims being made here (4 TFLOPS & 8nm with LPDDR5 RAM). We exercised the same approach with the shipping data and NVN2 leak as well, even though those were fairly clear cut, because it's worth it to be certain. That's why LiC has written out comprehensive essays on those two datamines.

8nm is not impossible, but I would like a strong reason to believe it if the given data about GPU size, estimated dimensions, rumored performance, and Nintendo's approach to handhelds suggest otherwise.
maybe debunked is the wrong word but there does seem to be a trend to knee-jerk dismiss things or deem them unlikely if a source is deemed not trustworthy enough. or maybe they got something wrong in the past so it's easier to ignore it if the information is 'out there' or doesn't align with the most positive outcome ie the node. that's fine and all but sometimes it's ok to believe and be wrong, likewise the skeptics can (and have) been wrong too. it's much easier to be a skeptic when we're dealing with things that can't be immediately proven. i guess i see things more as ok that's possible and work backward from there ie the magnet thing which sounded very Nintendo and believable from the get.
 
About the scalping issue, we have to think outside the box. Scalpers exist because people buy from them and they make profit. These scalpers are only there for the money. Now, let's think "who" buys from them. Typically the diehard fans and tech enthusiasts that couldn't buy the product, basically the hardcore side of the consumer. What do these guys most likely have? A Nintendo account. What if Nintendo makes the Switch 2 available for preorders online (from their store, making Nintendo accounts able to link with Amazon, etc) and prioritize orders from accounts that are ~6 months old, while limiting 1 console per account? This way, they can decrease the amount of people that would potentially buy from scalpers.
 
Just in case you haven't seen my edit to other comment yet, check it out (I added link to Thraktor's writeup).

Yes - so the reason why it's mostly debate between 8N and 4N is because nvidia products are predominantly fabbed on either Samsung 10nm (optimized) or TSMC 5nm (optimized).

This doesn't rule out that they could use a node process that's not SEC8N or TSMC 4N, but that would be kind of surprising considering it's probably cheaper for nvidia to just fab T239 on the same foundry as their other products.

Got it, I read the post! I think I have a satisfactory grasp at it now. It's very exciting that it seems that the best option is the most likely! Nintendo seems to be really investing on this one.
 
Please read this new, consolidated staff post before posting.

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


Back
Top Bottom