• 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!)

There's one big flaw in your reasoning here; the idea that Nintendo's original plans resembled my speculation at the time. If there's one thing Nintendo have committed to time and time again, it's that they'll never do anything I predict! Or maybe they saw my posts on the matter, and after realising I had predicted what hardware they were using they immediately changed their plans completely just to make sure I was wrong. That seems plausible.
Ahh the good old World Championship Wrestling (WCW) response to AOL Chatroom speculation figuring out future plans. Sneaky those Nintendo-Ninjas. Crazy, but sneaky.
 
Let’s not forget the staff post on attacking users. We don’t need further thread casualties :o
 
Seriously, though, I'm a bit late to respond to this, but I've been thinking about it, and I'm coming around to the possibility of the cancelled device being based on an older version of the T239. Mainly because an overclocked Mariko would be a bit of a crap mid-gen upgrade, and there aren't really many other alternatives. It would require certain assumptions, though.
Another way of making this narrative track isn't that T239 got rethought late, but that devkits shipped early. Nintendo pushes out devkits early - or at least ot more devs than usual - as a reaction to pandemic slowing development which simply exposed more devs to the whiplash of hardware development.

I'm not sure I 100% buy this story, but I can't square the size of the leaked device and SEC 8nm. Nor does an Ampere based 5nm device launching after Lovelace make clear sense to me either. Putting aside the tumult of the last month on "cancelled devices", the hard facts we have about Drake don't entirely add up despite being undeniable. There is no explanation that doesn't break my credulity somewhere.

Gaming out multiple narratives help make clear what sort of additional information would or would not fit with the picture we have now
 
A Mariko-based Switch Pro probably would have been a bit disappointing, but it's worth remembering we are talking about a cancelled system here. Would it really be that unreasonable that started work on such a system, to the point where some developers probably got briefed on it, but decided to cancel it fairly early specifically because the chip, while fine for the Switch Lite, didn't scale up how they wanted?
Like Ampere on 8nm, Maxwell on 16nm/12nm really wouldn't have been a surprise (Pascal already existed, which was basically that), so I can't imagine Nintendo would have expected anything more than we got with Mariko.

If we're talking a system that was cancelled early, then may guess is what I wrote in last paragraph; something that would have been more powerful than Mariko, but was cancelled before it got out of the planning stages. A case of Nintendo asking Nvidia whats kind of chip they can make using the Maxwell architecture to get higher-resolution Switch games in terms power, performance and cost, and asking developers how interested they'd be in developing for it. They shelve it early due to being too expensive, or too power hungry, or not being enough to get more third parties on board, and then move on to working on Drake (and the device that's using it) instead.
 
I don't see the benefit of a 1080p screen at 7in. In fact, I think there's a pretty decent chance they use the same display as the OLED model

That feels too low, no? If your rendering resolution is too low, the upscaling will look ugly; DLSS isn't a magic bullet. I think 1080p to 4K or 720p to 1440p is more likely

I actually tinkered around using a Retina calculator. The basic principle of "Retina" displays as apple described them is devices with such pixel density that individual pixels will not be noticed by people looking at them with average eyesight.

After writing the switch (OLED) native res and dimensions into the calculator:
tes-2.png

It seems like the Switch reaches Retina Display quality at a viewing distance of 16 inches (40 cm), which after a quick test is less than the distance I hold the switch at when playing handheld while sitting. Differences in people's behavior and positions aside (like the infamous bed mounts) it seems like the switch OLED holds pretty well as a Retina display for games that output the full 720p native resolution. For V1, V2 and Lite switches the distance is even lower, accounting for 14.5" for the first two and 12.88" for the latter. This is why games that output at the full native resolution seem so crisp and nice, and why games outputting at less than native res seem so subpar/jagged. It also doesn't help when the output resolution can't be upscaled nicely to a 720p display (for example, with resolutions that are not perfect divisors of 1280x720).

If we assume a similar sized display with a 1080p output res:
test1.png


We obtain better results than V1 switch, but not enough to radically change the perception of the image's crispness. Especially when 7in OLED displays are proven to look great at 720p. Considering that using a 1080p display requires more battery for a marginal increase in image quality, I bet they will stay using 720p displays, only with more games hitting that 720p goal instead of relying on sub-native resolutions. For me personally I don't have much problems with games in portable mode, except when games hit 480p or lower. Crash Nsane Trilogy was a pain to play due to this reason IMO.
 
Another way of making this narrative track isn't that T239 got rethought late, but that devkits shipped early. Nintendo pushes out devkits early - or at least ot more devs than usual - as a reaction to pandemic slowing development which simply exposed more devs to the whiplash of hardware development.

I'm not sure I 100% buy this story, but I can't square the size of the leaked device and SEC 8nm. Nor does an Ampere based 5nm device launching after Lovelace make clear sense to me either. Putting aside the tumult of the last month on "cancelled devices", the hard facts we have about Drake don't entirely add up despite being undeniable. There is no explanation that doesn't break my credulity somewhere.

Gaming out multiple narratives help make clear what sort of additional information would or would not fit with the picture we have now

Just thinking about Imran's comments from Era and Twitter on the hardware from March of 2021:

Expect this thing to be primarily for FPS boosts and resolutions, not a Switch 2.

Think more Pro than Super
It's funny, I was talking to some people today who were like "There's enough people with dev kits that leaks should happen any day now."

Guess it didn't even take that long.

After the OLED announcement, and even moreso after the Nvidia ransomware leak in 2022, these comments were disregarded by many here with members saying 'he knows nothing', citing that there's no way T239 would be just for FPS boosts. Just wondering with DF's comments and all the recent theorizing if his comments fit well into some of these narratives.

They kind of (?) indicate that physical hardware was in the hands of developers in early 2021.
 
Just thinking about Imran's comments from Era and Twitter on the hardware from March of 2021:

After the OLED announcement, and even moreso after the Nvidia ransomware leak in 2022, these comments were disregarded by many here with members saying 'he knows nothing', citing that there's no way T239 would be just for FPS boosts. Just wondering with DF's comments and all the recent theorizing if his comments fit well into some of these narratives.

They kind of (?) indicate that physical hardware was in the hands of developers in early 2021.
I still fully believe the bolded. Even if it had less SMs, an Ampere Tegra upgrade would have been a bonkers way to pursue mere "FPS boosts and resolutions." You'd be paying for a bunch of extra silicon (tensor and RT cores at minimum) that's going to do pretty much nothing for you, while having to architect a BC solution because Maxwell-based Switch games won't just run on it*, plus make the OS compatible with it, deal with boot security issues again, the list goes on. Those costs would have been better spent on upgrading the existing SoC with more SMs, higher clocks, or both, if you were trying to make a mid-gen refresh.

As for the devkit issue, I don't think that Orin itself was ready for use in 2020. And if Orin wasn't ready, then whatever T239 chip or other Orin derivative we're imaging might have existed definitely couldn't have been ready for use in devkits either.

And I have the rest of the same thoughts on this idea as I had before.

* Edit: And then what, are you going to rely on what is essentially BC to run future games too? Or are you going to make developers build and test two versions of their games for different CPU and GPU architectures just to be playable on two models of what's supposed to be the same system? And how does DLSS and RT fit into that picture?
 
Last edited:
This has to be the weirdest, most unique lead up to new/next-gen hardware in recent history, right? All the way back from "4K ready" to "new chip with DLSS" to the OLED criss-cross, to the 11 devs article & Nintendo's refutals, the Nvidia NVN2 hack, all the uncles, SPLOL3D false alarm, and the linux updates. And that's not even mentioning the overarching silence from developer sources since the 11 devs article on Sep 30, 2021. 1 year and 3 months (going on 4) and numerous gaming events that some expected might lead to chatter (GDC, Hot Chips, E3?, TGA, etc.) have passed by without so much as an "unofficial" official (by which I mean an article from a news outlet) whisper.

Like Switch is currently the hottest and best selling system of the past few years and is now approaching 6 years since the original launch, but it feels like no one outside of this thread is expecting or even theorizing on new hardware. It always seems to be coming in 202X (where X = current year + 2) and any discussions end with "well NSW is selling so well, so I don't think Nintendo's going to release anything anytime soon". Instead, the best we've gotten is various outlets repeating the same "the Switch is getting a little long in the tooth and showing it's age" and "X shows that we NEED a Switch Pro/2".

And even the little hints about dev kits or new HW from other reporters were relegated to passing mentions in tweets and podcast Q&As. "Yeah lots of dev kits out there" "It's neat" "Not a 2; mostly for res/fps boosts" "a mid-gen update was canceled internally" "Oh btw I heard RDR2 was in dev for NSW and 'Pro'" "Nintendo's gonna be quiet after Zelda" "maybe we hear something by 2024, I think" "2 early" Like Jeff Grub knew details about the SMB movie months before release, knew about Jedi Survivor among other 3P games, Nintendo's E3 2022 "plans", but AFAIK still isn't even aware of dev kits for Drake being out in the wild? Despite I believe Jason S. or Imran K. stating he heard from a dev that "there were so many dev kits out I wouldn't be surprised if something leaked"?

And the even crazier part is that we know work has been done on the T239 chip at least as of a year ago and has seen continued updates on the linux side of things. The hard data (FW datamines) have been shown to be indicative of future plans, so this NVN2 and linux stuff on top of the "late 2022 at the earliest" and "late 2022 - Q1 2023" comments from Nate should be pointing to something releasing sometime soon right? I mean, does Nintendo now have the tightest control on HW info and the strictest NDAs in the tech industry, when even Apple has stuff leaking constantly, year after year?

I'm just saying the past 2-3 years have been weird. Did we just happen to find out about this thing way too early? Could it also be that we just happen to know everything about this device outside of a concrete release date and specific software? Why does it feel like we're the only ones who care about the next HW from Nintendo? I'm just asking for the equivalent of the NX/Switch LKD daily rumors, Eurogamer NX concept scoop, DQXI NX whoopsie, all the Wii U ports coming year 1 (Smash, Splatoon, etc.), price guesstimations, and more like we got throughout 2016. Why does it feel like other journalists/insiders stopped looking into this thing over a year ago?

Does RDR2 coming to NSW or "Pro" (at the time) not warrant a full investigative article? Would Nate stating that Drake would receive exclusive games not spark others to do some inside sleuthing and research to get a scoop on new info? I understand Nate wanting to solidify his info before reporting on it, but what about everyone else? It's like the 11 devs-Zynga article made the whole industry quieter regarding NSW2/Drake, rather than doing what many would've assumed- pouring gasoline on a growing fire, resulting in new investigations/discussions over the next year.

I don't want this to sound like the leaks and info we have heard isn't substantial. I do appreciate everything we've heard from all sources. I'm just saying that the almost year and 4 months following MZ's "11 devs" article has been weird. Sure, maybe the lead up to NX/NSW would've felt exactly like this without Nintendo announcing NX or its release timing/window. But it all feels too quiet and it's mostly (again, I understand the NVN2+Linux info were and are massive) felt that way for too long.

Last thought: If Nvidia never got hacked, would we really have gone a full year without new info until we discovered the linux kernel updates in September 2022? I still think about how excruciating the past year+ would've been without that hack. Or maybe it would've been a butterfly effect and some insider was just about to leak info on Drake mid-late March 2022 and discarded it because of the hard, conclusive info of the early March Nvidia hack.
 
This has to be the weirdest, most unique lead up to new/next-gen hardware in recent history, right? All the way back from "4K ready" to "new chip with DLSS" to the OLED criss-cross, to the 11 devs article & Nintendo's refutals, the Nvidia NVN2 hack, all the uncles, SPLOL3D false alarm, and the linux updates. And that's not even mentioning the overarching silence from developer sources since the 11 devs article on Sep 30, 2021. 1 year and 3 months (going on 4) and numerous gaming events that some expected might lead to chatter (GDC, Hot Chips, E3?, TGA, etc.) have passed by without so much as an "unofficial" official (by which I mean an article from a news outlet) whisper.

Like Switch is currently the hottest and best selling system of the past few years and is now approaching 6 years since the original launch, but it feels like no one outside of this thread is expecting or even theorizing on new hardware. It always seems to be coming in 202X (where X = current year + 2) and any discussions end with "well NSW is selling so well, so I don't think Nintendo's going to release anything anytime soon". Instead, the best we've gotten is various outlets repeating the same "the Switch is getting a little long in the tooth and showing it's age" and "X shows that we NEED a Switch Pro/2".

And even the little hints about dev kits or new HW from other reporters were relegated to passing mentions in tweets and podcast Q&As. "Yeah lots of dev kits out there" "It's neat" "Not a 2; mostly for res/fps boosts" "a mid-gen update was canceled internally" "Oh btw I heard RDR2 was in dev for NSW and 'Pro'" "Nintendo's gonna be quiet after Zelda" "maybe we hear something by 2024, I think" "2 early" Like Jeff Grub knew details about the SMB movie months before release, knew about Jedi Survivor among other 3P games, Nintendo's E3 2022 "plans", but AFAIK still isn't even aware of dev kits for Drake being out in the wild? Despite I believe Jason S. or Imran K. stating he heard from a dev that "there were so many dev kits out I wouldn't be surprised if something leaked"?

And the even crazier part is that we know work has been done on the T239 chip at least as of a year ago and has seen continued updates on the linux side of things. The hard data (FW datamines) have been shown to be indicative of future plans, so this NVN2 and linux stuff on top of the "late 2022 at the earliest" and "late 2022 - Q1 2023" comments from Nate should be pointing to something releasing sometime soon right? I mean, does Nintendo now have the tightest control on HW info and the strictest NDAs in the tech industry, when even Apple has stuff leaking constantly, year after year?

I'm just saying the past 2-3 years have been weird. Did we just happen to find out about this thing way too early? Could it also be that we just happen to know everything about this device outside of a concrete release date and specific software? Why does it feel like we're the only ones who care about the next HW from Nintendo? I'm just asking for the equivalent of the NX/Switch LKD daily rumors, Eurogamer NX concept scoop, DQXI NX whoopsie, all the Wii U ports coming year 1 (Smash, Splatoon, etc.), price guesstimations, and more like we got throughout 2016. Why does it feel like other journalists/insiders stopped looking into this thing over a year ago?

Does RDR2 coming to NSW or "Pro" (at the time) not warrant a full investigative article? Would Nate stating that Drake would receive exclusive games not spark others to do some inside sleuthing and research to get a scoop on new info? I understand Nate wanting to solidify his info before reporting on it, but what about everyone else? It's like the 11 devs-Zynga article made the whole industry quieter regarding NSW2/Drake, rather than doing what many would've assumed- pouring gasoline on a growing fire, resulting in new investigations/discussions over the next year.

I don't want this to sound like the leaks and info we have heard isn't substantial. I do appreciate everything we've heard from all sources. I'm just saying that the almost year and 4 months following MZ's "11 devs" article has been weird. Sure, maybe the lead up to NX/NSW would've felt exactly like this without Nintendo announcing NX or its release timing/window. But it all feels too quiet and it's mostly (again, I understand the NVN2+Linux info were and are massive) felt that way for too long.

Last thought: If Nvidia never got hacked, would we really have gone a full year without new info until we discovered the linux kernel updates in September 2022? I still think about how excruciating the past year+ would've been without that hack. Or maybe it would've been a butterfly effect and some insider was just about to leak info on Drake mid-late March 2022 and discarded it because of the hard, conclusive info of the early March Nvidia hack.
I think it's "unique" only because we're seeing movement behind the curtain that we never saw before. imagine all this shit we're going through now, but for switch launch. imagine Digital Foundry, prior to switch reveal, talk about a cancelled successor (project indy), and all that shit
 
I still fully believe the bolded. Even if it had less SMs, an Ampere Tegra upgrade would have been a bonkers way to pursue mere "FPS boosts and resolutions." You'd be paying for a bunch of extra silicon (tensor and RT cores at minimum) that's going to do pretty much nothing for you, while having to architect a BC solution because Maxwell-based Switch games won't just run on it, plus make the OS compatible with it, deal with boot security issues again, the list goes on. Those costs would have been better spent on upgrading the existing SoC with more SMs, higher clocks, or both, if you were trying to make a mid-gen refresh.

As for the devkit issue, I don't think that Orin itself was ready for use in 2020. And if Orin wasn't ready, then whatever T239 chip or other Orin derivative we're imaging might have existed definitely couldn't have been ready for use in devkits either.

And I have the rest of the same thoughts on this idea as I had before.

I understand your stance on T239, and I'm not questioning it - I was only retreading the timeline a bit before asking my question.
Just wondering, with DF's comments and all the recent theorizing, if his comments fit well into some of these narratives.

They kind of (?) indicate that physical hardware was in the hands of developers in early 2021.

I don't personally think it's reasonable to assume that Imran was lying or even misinformed about the existence of some devkits in early 2021. He's a professional journalist with reasonable rapport in the industry, and the statement implies more than one contact is familiar with developers having kits in their hands at the time. Now what he didn't commit to or even comment on is anything about 4K, or T239, or even a release date.

If what you're saying is true, that nothing related to Orin was ready for consumption in late 2020, then my approach would not be to throw away the statement, but rather to assume that whatever device was being planned at the time was unrelated. Perhaps there really was an upclocked Mariko device that was cancelled?

If we try to fit Mochizuki's comments in, it's worth noting that he did mention 4K and DLSS in his March article - but that could have been conflation of refresh planned for that year, and early rumblings of T239. Seeing as Aula datamines also did mention the inclusion of some kind of 4K, and theorized using Mariko for improved clocks, we can see how things could get a bit murky in the reporting. Lastly, Nintendo's rejection of Mochizuki's claims saying they did not distribute a 4K kit for Nintendo Switch might have been explained away just as easily by the fact that T239 was always planned for a yet unnamed successor system.

Feel free to rip this one apart everybody!
 
If what you're saying is true, that nothing related to Orin was ready for consumption in late 2020, then my approach would not be to throw away the statement, but rather to assume that whatever device was being planned at the time was unrelated. Perhaps there really was an upclocked Mariko device that was cancelled?
by late 2020, I'd say that's a good time for Orin to be out in the wild. remember orin was announced for automotives, which require a long testing period. Orin was announced back in late 2019, when the first DLSS tests for NVN started. so there was probably preliminary hardware to send out to devs in 2020. binned Orin chips that didn't past muster for automotive testing, for instance
 
Quoted by: LiC
1
One thing I'm starting to think is that after the Nvidia hack, I should have taken a more critical look at my assumption that what were we seeing had to be the same as the hardware rumors that went on in 2020 and 2021. It's a natural assumption to make, rather than immediately going to kind of crazy theories we're debating now, but in retrospect the "Pro" talk and N3DS comparisons should have been bigger red flags. But we had to weigh that against the 4K and DLSS talk too. So the conclusion I came to was that everybody was talking about the same hardware, they were just doing a bad job of it, having heard about 4K and DLSS but being unsure of positioning and timeline, and inaccurately filling in the gaps. For instance, because of the OLED confusion, if they had heard about upgraded hardware early on and then conflated it with the 2021 release date of the OLED, that would bias perceptions toward it being more N3DS-like just because of when it was being released.

I don't think that's a totally unrealistic read on the situation now. But looking back on it, it does seem more and more that the "FPS boost" and "New 3DS" talk was totally incompatible with the reality of the hardware seen in the leak.

And I have to admit that a big reason for wanting to believe was release timing. The development timeline of NVN2 and T239 in the leak alone wasn't enough to know when it would release, but if we had reliable sources telling us new hardware was launching at the end of 2022 or beginning of 2023, then that's going to bias my own beliefs toward a unifying theory that means the hardware in the leak is the hardware with that release window.

So now, like I've been saying especially with respect to the idea of 2020 and even 2021 devkit hardware, I don't believe those past timelines make sense with the current timeline, so I have to conclude that these sources were not talking about the same hardware in the leak. As to theories for how those sources stretched the timeline for their hardware so far forward into what's now the timeline for T239 hardware, before learning that the former was cancelled, or theories for why that hardware was said to have features that the T239 hardware actually has and don't make sense for a refresh -- like I said before, those are at best topics to pass the time guessing at, because what I really care about is the hardware that is still happening.

Which brings me back to what I think was the first post I made after the whole "cancellation" mess -- the hardware that we've been discussing for the past year isn't cancelled. We don't even have any reason to think it's delayed. But its release timeline is somewhat more in question than it was, since a big part of what we believed about it was coming from the wrong hardware.

That said, to me, the evidence totally outside that insider info -- what we got from the leak, and the public progress we've been following, even the original leak of the name T239 -- it all still says "sometime in 2023."

I think it's "unique" only because we're seeing movement behind the curtain that we never saw before. imagine all this shit we're going through now, but for switch launch. imagine Digital Foundry, prior to switch reveal, talk about a cancelled successor (project indy), and all that shit
The difference there is that Project Indy never left Nintendo's doors (not counting a rumor about the screen which probably came from the screen supplier as usual). Third parties never knew about it. A lot of Nintendo departments probably didn't even know about it. That's not the same as something that supposedly had devkits out and games in development before all that was suddenly thrown into question.

by late 2020, I'd say that's a good time for Orin to be out in the wild. remember orin was announced for automotives, which require a long testing period. Orin was announced back in late 2019, when the first DLSS tests for NVN started. so there was probably preliminary hardware to send out to devs in 2020. binned Orin chips that didn't past muster for automotive testing, for instance
The only public statement on this is that Nvidia said Orin would start sampling sometime in 2021. So it's way too big of a stretch to say that a custom Ampere Tegra could have been in devkits at the end of 2020, and probably too big of a stretch to say Orin itself was. And that's just timing-wise. I've talked before about why Orin doesn't actually make any sense to use in devkit hardware even if it was ready.
 
One thing I'm starting to think is that after the Nvidia hack, I should have taken a more critical look at my assumption that what were we seeing had to be the same as the hardware rumors that went on in 2020 and 2021. It's a natural assumption to make, rather than immediately going to kind of crazy theories we're debating now, but in retrospect the "Pro" talk and N3DS comparisons should have been bigger red flags. But we had to weigh that against the 4K and DLSS talk too. So the conclusion I came to was that everybody was talking about the same hardware, they were just doing a bad job of it, having heard about 4K and DLSS but being unsure of positioning and timeline, and inaccurately filling in the gaps. For instance, because of the OLED confusion, if they had heard about upgraded hardware early on and then conflated it with the 2021 release date of the OLED, that would bias perceptions toward it being more N3DS-like just because of when it was being released.

I don't think that's a totally unrealistic read on the situation now. But looking back on it, it does seem more and more that the "FPS boost" and "New 3DS" talk was totally incompatible with the reality of the hardware seen in the leak.

And I have to admit that a big reason for wanting to believe was release timing. The development timeline of NVN2 and T239 in the leak alone wasn't enough to know when it would release, but if we had reliable sources telling us new hardware was launching at the end of 2022 or beginning of 2023, then that's going to bias my own beliefs toward a unifying theory that means the hardware in the leak is the hardware with that release window.

So now, like I've been saying especially with respect to the idea of 2020 and even 2021 devkit hardware, I don't believe those past timelines make sense with the current timeline, so I have to conclude that these sources were not talking about the same hardware in the leak. As to theories for how those sources stretched the timeline for their hardware so far forward into what's now the timeline for T239 hardware, before learning that the former was cancelled, or theories for why that hardware was said to have features that the T239 hardware actually has and don't make sense for a refresh -- like I said before, those are at best topics to pass the time guessing at, because what I really care about is the hardware that is still happening.

Which brings me back to what I think was the first post I made after the whole "cancellation" mess -- the hardware that we've been discussing for the past year isn't cancelled. We don't even have any reason to think it's delayed. But its release timeline is somewhat more in question than it was, since a big part of what we believed about it was coming from the wrong hardware.

That said, to me, the evidence totally outside that insider info -- what we got from the leak, and the public progress we've been following, even the original leak of the name T239 -- it all still says "sometime in 2023."


The difference there is that Project Indy never left Nintendo's doors (not counting a rumor about the screen which probably came from the screen supplier as usual). Third parties never knew about it. A lot of Nintendo departments probably didn't even know about it. That's not the same as something that supposedly had devkits out and games in development before all that was suddenly thrown into question.


The only public statement on this is that Nvidia said Orin would start sampling sometime in 2021. So it's way too big of a stretch to say that a custom Ampere Tegra could have been in devkits at the end of 2020, and probably too big of a stretch to say Orin itself was. And that's just timing-wise. I've talked before about why Orin doesn't actually make any sense to use in devkit hardware even if it was ready.

I feel like, especially from the bolded, that we're kind of proposing the same thing. I might just be going about it in a more ham-fisted way.
 
Just to make my ramble even longer: In hindsight there were other red flags too, like the idea that people could have learned in 2020 about a project that Nintendo had only started in 2020, which wasn't going to be released for another 2-3 years at least, and the idea that third parties would have been brought on board that early. That all made a lot more sense at first when it was Mochizuki saying something was going to happen in the next year or so, the same as his previous reports on new Switch models. But once the timeline slipped past that it and started being "2022, maybe 2023," it seems foolish now to make excuses and try to establish continuity between those early reports and the hardware from the leak. Kopite didn't even post about T239 until June 2021! How was Mochizuki or any third party going to get info about that project in 2020?

I understand your stance on T239, and I'm not questioning it - I was only retreading the timeline a bit before asking my question.

I don't personally think it's reasonable to assume that Imran was lying or even misinformed about the existence of some devkits in early 2021. He's a professional journalist with reasonable rapport in the industry, and the statement implies more than one contact is familiar with developers having kits in their hands at the time. Now what he didn't commit to or even comment on is anything about 4K, or T239, or even a release date.

If what you're saying is true, that nothing related to Orin was ready for consumption in late 2020, then my approach would not be to throw away the statement, but rather to assume that whatever device was being planned at the time was unrelated. Perhaps there really was an upclocked Mariko device that was cancelled?

If we try to fit Mochizuki's comments in, it's worth noting that he did mention 4K and DLSS in his March article - but that could have been conflation of refresh planned for that year, and early rumblings of T239. Seeing as Aula datamines also did mention the inclusion of some kind of 4K, and theorized using Mariko for improved clocks, we can see how things could get a bit murky in the reporting. Lastly, Nintendo's rejection of Mochizuki's claims saying they did not distribute a 4K kit for Nintendo Switch might have been explained away just as easily by the fact that T239 was always planned for a yet unnamed successor system.

Feel free to rip this one apart everybody!
I feel like, especially from the bolded, that we're kind of proposing the same thing. I might just be going about it in a more ham-fisted way.
Well, the post I responded to you on was mostly separate from my general rant. That arm of the discussion where I responded to you started with Thraktor's and oldpuck's theories for how there could have been some other Ampere Tegra, so that's why I brought up the point about Drake and Orin. If your connection to Imran's comments wasn't about that, then I apologize.

So, on your actual topic of there being some other upgraded hardware, not necessarily T239 or Ampere Tegra. I'm not convinced, to be honest! Yes, my rant is predicated on the idea that the reporting wasn't about T239, because I no longer believe my own previous rationalizations for how it could all be the same device. But what exactly they were talking about is another matter. To me, there's a push-pull here: The less distinct this supposed hardware was, e.g. overclocked Mariko, the more it fits with the idea of something that could be cancelled and get a throwaway mention by DF and not seriously damage Nintendo, but the less it fits with the reporting that insisted it had 4K via DLSS (which Nate said sometime around the beginning of 2021, and Mochizuki said in March 2021) and later RT and exclusives to boot.* The more distinct it was, the less believable its existence is at all, with no evidence in the Nvidia leak, not fitting with the idea of a mid-gen refresh, no parade of pissed-off third parties leaking that Nintendo cancelled all their games.

* Obviously this is where conflation with T239 would come into play, as you mentioned, but like I said in the rant, I don't really get how that confusion could have persisted even to the point that Mochizuki talked about his 11 devkits, or Nate in 2022 saying there had been no changes (prior to the "heavy info").

It's hard for me to rule out the idea that between the existence of two real projects, the OLED and T239, the mere idea of "Switch with a new Tegra" took on a life of its own and became a rumor that conflated aspects of both the real models. And/or, if there ever was any kind of mid-gen refresh planned, whether or not it lined up with the timeline as reported, that could have further impeded accurate reporting.

I just can't reconcile this kind of "everyone was right all along" narrative that seems to be sought after. And the more energy I expend on it, the more I feel like it distracts from the only hardware info that actually needs real reporting on right now, which is the status of the decidedly non-"refresh" project Nvidia began in 2020 and was working on throughout 2022, and when that's going to release.

But to be clear, I'm not saying we should stop. What else are we going to talk about? :p
 
One thing I'm starting to think is that after the Nvidia hack, I should have taken a more critical look at my assumption that what were we seeing had to be the same as the hardware rumors that went on in 2020 and 2021. It's a natural assumption to make, rather than immediately going to kind of crazy theories we're debating now, but in retrospect the "Pro" talk and N3DS comparisons should have been bigger red flags.

In your defense, you and I went back and forth pretty hard on this and you convinced me. We really did consider the other possibility, and it wasn’t just your smooth talking that pulled me around. It was the simplest, cleanest assumption.
But we had to weigh that against the 4K and DLSS talk too. So the conclusion I came to was that everybody was talking about the same hardware, they were just doing a bad job of it, having heard about 4K and DLSS but being unsure of positioning and timeline, and inaccurately filling in the gaps. For instance, because of the OLED confusion, if they had heard about upgraded hardware early on and then conflated it with the 2021 release date of the OLED, that would bias perceptions toward it being more N3DS-like just because of when it was being released.

I don't think that's a totally unrealistic read on the situation now. But looking back on it, it does seem more and more that the "FPS boost" and "New 3DS" talk was totally incompatible with the reality of the hardware seen in the leak.

And I have to admit that a big reason for wanting to believe was release timing. The development timeline of NVN2 and T239 in the leak alone wasn't enough to know when it would release, but if we had reliable sources telling us new hardware was launching at the end of 2022 or beginning of 2023, then that's going to bias my own beliefs toward a unifying theory that means the hardware in the leak is the hardware with that release window.
The release window also roughly corresponds with Orin release timing, and every indication is that Orin and Drake was/are deeply related. I don't think this is entirely confirmation bias talking.

Where it does maybe start breaking down (and worth considering in the context of Drake timing) was that late 2022 window. It's hard to imagine an Orin based/related device making 2022 ever, unless it was fabbed prior to Orin release. Which, not inconceivable.

Drake as a piece of hardware still doesn't make total sense to me, and the release schedule factors in there, part of why I would like to nail something down
 
0
I think the entire Hyrule ground level overworld will be structured more like the initial Great Plateau, in order to differentiate it from BOTW. Think about it. If you can go anywhere at anytime in TOTK's overworld the same way you could in BOTW, would that be exciting? I think the sky (and perhaps more) will provide that extreme level of openness, while the Hyrule we know and are familiar with will have road blocks, similar to past Zeldas (except more "natural" and less OoT "you can't pass through these gates until you get a letter from the princess!").

A giant tornado vortex is preventing access to some place near the bird village? How do I break through or dissipate it? Hyrule castle is inaccessible due to it 1. floating way above a giant crater and 2. intense malice flames? How do I either A. bring the castle down B. safely go through the malice C. get Gdorf to lower the intensity enough for me to get through? Similarly, Death Mountain will need a new way to access it. Perhaps we get the equivalent of Mole Mitts and can enter the volcano from underneath? Maybe Zora's Domain will have frozen over and you'll need to swim underneath the ice to access a cavern or dungeon?

These scenarios will also help to give a reason to play BOTW and not make it "obsolete" as many have pondered. BOTW will give you the completely open (sans the GrPl) experience for Hyrule with complete freedom, while TOTK will provide more of that GrPl tutorial level of guided restrictiveness that allows for a story that takes place in the present.

edit: ah shit, wrong thread. w/e I'll leave it up since it's been so slow recently
 
Last edited:
Is pointer controls that bad with Gyro? I assume LG OLED TVs magic remote have gyro too and not IR. Having used the LG OLED TV remote for pointing I find it quite fast.
I personally find it unusably bad.
 
0
I think the entire Hyrule ground level overworld will be structured more like the initial Great Plateau, in order to differentiate it from BOTW. Think about it. If you can go anywhere at anytime in TOTK's overworld the same way you could in BOTW, would that be exciting? I think the sky (and perhaps more) will provide that extreme level of openness, while the Hyrule we know and are familiar with will have road blocks, similar to past Zeldas (except more "natural" and less OoT "you can't pass through these gates until you get a letter from the princess!").

A giant tornado vortex is preventing access to some place near the bird village? How do I break through or dissipate it? Hyrule castle is inaccessible due to it 1. floating way above a giant crater and 2. intense malice flames? How do I either A. bring the castle down B. safely go through the malice C. get Gdorf to lower the intensity enough for me to get through? Similarly, Death Mountain will need a new way to access it. Perhaps we get the equivalent of Mole Mitts and can enter the volcano from underneath? Maybe Zora's Domain will have frozen over and you'll need to swim underneath the ice to access a cavern or dungeon?

These scenarios will also help to give a reason to play BOTW and not make it "obsolete" as many have pondered. BOTW will give you the completely open (sans the GrPl) experience for Hyrule with complete freedom, while TOTK will provide more of that GrPl tutorial level of guided restrictiveness that allows for a story that takes place in the present.
I have been saying similar things to this. Especially when people bring up the "Will Totk be better than Botw? " I keep saying I don't necessarily think, so I think it will be different and more of a traditional style Zelda but still very similar to Botw. Botw is the first Zelda for alot of people these days and for them to make it a little more traditional but still very open will get people accustomed to traditional Zelda's. Sure I think totk will be better in some regards like bug fixes and quality of life improvements, but I don't think it will necessarily live in Botw shadow because it will be very similar but different kind of like MM to oot.
 
Is pointer controls that bad with Gyro?

Well, I can deal with it most of the time, but the gyro drift drives me nuts in the end. It really stresses me out. But for me it's more complex than just the gyroscope being the only culprit here. When I use the pro controller on my PC through Steam it works much better than on the Switch. For me it feels like as if Nintendo is applying a drift correction algorithm on an OS level (because every game I've tried has it in a very severe way), which to me is more of a hindrance than a help. Of course, on steam it'll also eventually drift but there's no such correction for really small or slow movements like on the switch (I first noticed it playing splatoon on the wii u actually...)

Anyway, I wouldn't like the wii tech again (very low FoV). I would prefer an USB camera (or a camera on the switch itself) and a IR LED ring to connect on the joy-con so be tracked (like a VR controller). That would make me so happy to play any shooter... No drift anymore.
 
If your connection to Imran's comments wasn't about that, then I apologize.

My fault for replying to oldpuck’s post as my starting point. Probably could / should have been standalone; A question to anybody.


I just can't reconcile this kind of "everyone was right all along" narrative that seems to be sought after.

My approach is that some things are less likely to be totally mistaken than others, and that reputable journalists aren’t going to be pulling stuff out of thin air. They can and will make mistakes on some things. Broader strokes like saying they’ve confirmed many people have dev kits in early 2021 (or late 2020) does not feel like a mistake to me. It’s only after we add in the other inconsistent or ambiguous details from different sources that things get really murky; 4K support is ambiguous, and DLSS was incongruous with the timelines proposed (launching in 2021).

Since the beginning it’s always been safe and common to assume that not everything everybody has said was correct. Nothing’s really changed in that respect. We’re just seeing people pivot around what details might have been misunderstood or misrepresented by the reporter, or the source were it second-hand. For the last two years, it was taken almost as gospel in this thread that Bloomberg conflated the OLED Model with Drake, hence the 4K/DLSS claims**. It was also assumed Mochizuki was clutching at straws when he speculated that the device he heard about may have been shelved. When we’re now faced with evidence of T239 being near ready to ship (?), and at the same time Digital Foundry saying a mid-gen refresh was cancelled and no clear timeline as to when, it’s absolutely a worthwhile exercise to reevaluate previously dismissed claims and those understood to be true. And again - the goal isn't to prove everybody to be right.

As for how I'm choosing to reconcile things like this:
Obviously this is where conflation with T239 would come into play, as you mentioned, but like I said in the rant, I don't really get how that confusion could have persisted even to the point that Mochizuki talked about his 11 devkits

** I could see Mochizuki having gotten wind of T239/successor kits in the works at the time of his March 2021 reporting from some sources, and conflating it with details around a version of the OLED that was enhanced from other sources, scheduled for 2021 and yet never made it to market. Look it's all a bit tin-foil-hat when I read it out loud, but it's what I'm running with for now, as it jives better with my sensibilities than assuming either T239 is not making it to market, or that he was following countless bad leads.
 
Wait , did the new Nate the Hate talking about Switch succ? I had a look yesterday and seems like it was Xbox content

It was not the focus of the episode, but MVG teased his thoughts about it at the end. Namely, he is skeptical it'll be backwards compatible -- and questioned if it will even be a Switch 2.

At the very end, Nate said there would be another speculation episode next week for another company, but said we would have to wait to see for which company. Personally, I think it will be for Sony, because a few people have said a State of Play is coming in January. I would love to be wrong though, and it end up being Nintendo.
 
Why is MVG skeptical about BC? I'm not convinced that implementing some kind of solution is this insurmountable challenge that some are framing it as. Or that Nintendo would deliberately not have it, so they could (as I've heard too many times around the 'net) sell Deluxe ports of the entire library.
 
Last edited:
Why is MVG skeptical about BC? I'm not convinced that implementing some kind of solution is this insurmountable challenge that some are framing it as. Or that Nintendo would deliberately not have it to (as I've heard way too many times around the 'net) sell Deluxe ports of the entire library.

He gave no details. That's a podcast for another day.

I 100% agree with you, though; it is a solvable problem. But, devil's advocate here: what if it is not called Switch, and they change the game media significantly (from carts to something else or just really different carts)? And they take these steps to clearly delimitate the two systems (I e. Avoid the generational confusion that plagued the 3DS launch). In that scenario, I could see Nintendo feeling like they can skip BC -- even though technically they could make it work cause the architecture is similar.
 
My fault for replying to oldpuck’s post as my starting point. Probably could / should have been standalone; A question to anybody.

My approach is that some things are less likely to be totally mistaken than others, and that reputable journalists aren’t going to be pulling stuff out of thin air. They can and will make mistakes on some things. Broader strokes like saying they’ve confirmed many people have dev kits in early 2021 (or late 2020) does not feel like a mistake to me. It’s only after we add in the other inconsistent or ambiguous details from different sources that things get really murky; 4K support is ambiguous, and DLSS was incongruous with the timelines proposed (launching in 2021).

Since the beginning it’s always been safe and common to assume that not everything everybody has said was correct. Nothing’s really changed in that respect. We’re just seeing people pivot around what details might have been misunderstood or misrepresented by the reporter, or the source were it second-hand. For the last two years, it was taken almost as gospel in this thread that Bloomberg conflated the OLED Model with Drake, hence the 4K/DLSS claims**. It was also assumed Mochizuki was clutching at straws when he speculated that the device he heard about may have been shelved. When we’re now faced with evidence of T239 being near ready to ship (?), and at the same time Digital Foundry saying a mid-gen refresh was cancelled and no clear timeline as to when, it’s absolutely a worthwhile exercise to reevaluate previously dismissed claims and those understood to be true. And again - the goal isn't to prove everybody to be right.

As for how I'm choosing to reconcile things like this:

** I could see Mochizuki having gotten wind of T239/successor kits in the works at the time of his March 2021 reporting from some sources, and conflating it with details around a version of the OLED that was enhanced from other sources, scheduled for 2021 and yet never made it to market. Look it's all a bit tin-foil-hat when I read it out loud, but it's what I'm running with for now, as it jives better with my sensibilities than assuming either T239 is not making it to market, or that he was following countless bad leads.
When I say "everybody was right all along" I just mean "all prior information is reconciled." It's not about proving people wrong or right, we agree on that.

The problem with the bolded is that Mochizuki never speculated that anything may have been cancelled. He only said that "Nintendo could still decide not to release a 4K Switch" which just seems like a gesture at the notion that plans change, i.e. justification for getting things wrong before. I really don't think that can be read as speculation that it had been or looked like it would be cancelled, since that was the same article that said 11 companies had the tools from Nintendo and that they "expect[ed] to release their 4K Switch games during or after the second half of [2022]." He unhelpfully didn't explain how he could have thought a 2021 release was happening if information from 11 developers said 2022 or later -- did they plan games for 2021 and have to delay them? can we get a crumb of context here?? hello??? -- and this is where our current speculation over whether he was talking about two different things comes in, but either way, the information on the hardware as he understood it was very certain about its continued existence.

I've touched on this in response to the question of whether developers were given one thing and then had it taken away, if it was just meant for feedback or something, or if plans could have changed or evolved or whatever: Publishers don't develop and schedule games for platforms they aren't sure will be releasing! I have no idea how much or how frequently Nintendo communicates with third parties in the early stages of hardware development (and I'm sure it varies), but I'm certain that it's enough for them to know when to commit to actually developing titles.
 
Why is MVG skeptical about BC? I'm not convinced that implementing some kind of solution is this insurmountable challenge that some are framing it as. Or that Nintendo would deliberately not have it, so they could (as I've heard too many times around the 'net) sell Deluxe ports of the entire library.
His comments on BC for Drake never made much sense to me. It’s not like having a work around for the one incompatibility issue for the GPU would be some impossible feat, especially with Nvidia themselves working closely with Nintendo on both the hardware and software. A compatibility layer always seemed like a relatively simple solution to me.

He gave no details. That's a podcast for another day.

I 100% agree with you, though; it is a solvable problem. But, devil's advocate here: what if it is not called Switch, and they change the game media significantly (from carts to something else or just really different carts)? And they take these steps to clearly delimitate the two systems (I e. Avoid the generational confusion that plagued the 3DS launch). In that scenario, I could see Nintendo feeling like they can skip BC -- even though technically they could make it work cause the architecture is similar.
This flies in the face of their comments about wanting a smooth transition and to not have to start over with their next platform. Hell, Miyamoto even has a recent quote talking about how BC would be easier than ever now.
 
His comments on BC for Drake never made much sense to me. It’s not like having a work around for the one incompatibility issue for the GPU would be some impossible feat, especially with Nvidia themselves working closely with Nintendo on both the hardware and software. A compatibility layer always seemed like a relatively simple solution to me.


This flies in the face of their comments about wanting a smooth transition and to not have to start over with their next platform. Hell, Miyamoto even has a recent quote talking about how BC would be easier than ever now.
I wonder how many people believe nintendo is "perfection or bust". Wii U and 3DS offered perfect BC by the nature of their hardware, but Drake won't. but we already see with NSO and the previous VC that Nintendo really doesn't care about perfection
 
It was not the focus of the episode, but MVG teased his thoughts about it at the end. Namely, he is skeptical it'll be backwards compatible -- and questioned if it will even be a Switch 2.

1:31:25:
NateDrake: Then I had a $1 donation from Liam Warner writes in: "If a new Smash or a Ultimate port would require renegotiation of third party contracts, what's stopping Nintendo from just releasing updates for this game on the next-gen, assuming it's back compat? (MVG: Uh-huh.) I don't think it's likely, but is it the same game, so they could?"
NateDrake: Technically that would be a workaround. Yeah, they would just have to give whatever the contracts stipulate in terms of royalties or copies sold. So that would be a workaround to keep this game going forever.
MVG: A lot of assumptions about the next Nintendo hardware always going to be backwards compatible. I have some thoughts about that, but we'll talk about that when the time comes.
NateDrake: Yes, different topic for a different day.
Why is MVG skeptical about BC? I'm not convinced that implementing some kind of solution is this insurmountable challenge that some are framing it as. Or that Nintendo would deliberately not have it, so they could (as I've heard too many times around the 'net) sell Deluxe ports of the entire library.
Probably the same reasons he mentioned more than a year ago.
 
It was not the focus of the episode, but MVG teased his thoughts about it at the end. Namely, he is skeptical it'll be backwards compatible -- and questioned if it will even be a Switch 2.

At the very end, Nate said there would be another speculation episode next week for another company, but said we would have to wait to see for which company. Personally, I think it will be for Sony, because a few people have said a State of Play is coming in January. I would love to be wrong though, and it end up being Nintendo.

mvg.jpg
 
I wonder how many people believe nintendo is "perfection or bust". Wii U and 3DS offered perfect BC by the nature of their hardware, but Drake won't. but we already see with NSO and the previous VC that Nintendo really doesn't care about perfection
I imagine it’d generally be close enough to perfection for most people and they would release updates to fix issues as time goes on. I really think BC will be pretty comparable to PS5 and XSX.
 



Probably the same reasons he mentioned more than a year ago.

this is some nonsense waffling. we all know BC is possible if not 100% perfect, but Nintendo doesn't care for perfection, they care for solutions that encompass as many games as possible. which they can do.

I'm not gonna say MVG should justify his thoughts, but if he can't express why I should take his word into account, then he's not speaking anything worth listening to

I imagine it’d generally be close enough to perfection for most people and they would release updates to fix issues as time goes on. I really think BC will be pretty comparable to PS5 and XSX.
people are a lot more tolerant than folks on the internet think, so any issue, I suspect will be minor enough that most folks won't notice them
 
But, devil's advocate here: what if it is not called Switch, and they change the game media significantly (from carts to something else or just really different carts)? And they take these steps to clearly delimitate the two systems (I e. Avoid the generational confusion that plagued the 3DS launch).
Nintendo has historically been good with sticking with successful brand names like the Game Boy and DS, both lasting more than a decade with the GBA and 3DS lines. I'm doubtful they will scrap the Switch name as it's a more universal and appealing term that has become its own brand.

Was there a significant generational confusion during the 3DS launch? I have vague memories of it, but it seemed fairly clear that it was a more powerful DS followup capable of more advanced 3D. Regardless, Sony has managed to avoid generational confusion for more than two decades despite keeping the PlayStation brand. Even Nintendo was able to do it with the Game Boy Advance, by making it clear it was a new-gen Game Boy that could be a portable SNES. Calling Drake 'Switch 2' or even some 'powerful' suffix like Super, Ultra, or Advance and marketing it as clear as possible with next-gen exclusives will go a long way.

If it's a hybrid then I expect the same medium. I do expect the carts to look different, like the 3DS.

In that scenario, I could see Nintendo feeling like they can skip BC -- even though technically they could make it work cause the architecture is similar.

Nintendo went out of their way to include GBA BC in at least the DS and DS Lite despite different cartridge sizes and needing to include GBA hardware on board. It benefited them to have their new portable still be capable of playing an already popular and established games library.

Drake having BC means they can continue to sell the existing library at full price, with their new console immediately having access to a huge library. I believe - if they can make it work, then they will.

I recall this post from Thraktor from a long time ago about changes to the shader assembly for Drake, thought I'd bring it up again since it's relevant to this discussion.

So there's some additional instructions or features which Drake has over Orin, which aren't necessarily pulled from Ada. I've been thinking about this for a while, and I think the most likely reason for this is actually what I was discussing above. Namely, that Nvidia has added or changed some number of instructions on the Drake shader ISA to more easily facilitate translation-based BC with TX1 shaders. There's not really much else that I can think of which would warrant any change over Orin. The only SM-level changes that could have been back-ported from Ada would have been the updated tensor cores and RT cores, but I think we have good indications from the leak that we're looking at standard Ampere tensor and RT cores. Meanwhile one of the benefits of designing a custom SoC for your new console is that you can make these kinds of changes to achieve better backwards compatibility with existing games.

I assume translation based BC is similar to what the PS5 is doing?
 
Why is MVG skeptical about BC? I'm not convinced that implementing some kind of solution is this insurmountable challenge that some are framing it as. Or that Nintendo would deliberately not have it, so they could (as I've heard too many times around the 'net) sell Deluxe ports of the entire library.
I forgot where he said it, but he said that it was the Switch's internet browser for the eShop that was the issue and that it would be difficult to carry over without compromises. There's definitely a lot that I don't remember, though, so you might have to ask someone else.

Edit: Just saw the link that Dakhil put up and oof was I off base. Apparently, the GPU was the problem and that it might require some (imo) impractical solutions.
 
Last edited:
I forgot where he said it, but he said that it was the Switch's internet browser for the eShop that was the issue and that it would be difficult to carry over without compromises. There's definitely a lot that I don't remember, though, so you might have to ask someone else.
I mean, that's maybe one issue he talked about, but the biggie that he talked about at length last year is the fact that Drake's GPU uses a different architecture than the current Switch. And since Switch games are precompiled to run on the current Switch's specific architecture, every game would have to be recompiled for the new architecture (by the companies who made them, some of whom don't even exist anymore) or else popping an old Switch game onto new Switch just won't work.

Granted this thread has talked at length about possible solutions to that problem, and Nvidia and Nintendo are likely working on solutions so I'm not worried. Just saying that's the sort of obstacle MVG has talked about.
 
Nintendo went out of their way to include GBA BC in at least the DS and DS Lite despite different cartridge sizes and needing to include GBA hardware on board. It benefited them to have their new portable still be capable of playing an already popular and established games library.
They didn't go quite out of their way on that. The main reason is that the GBA hardware is used for DS games too, though not really programmable to the devs in that mode. The ARM7 gets a boost from 16Mhz to 32Mhz, and is in charge of audio processing, WiFi, and a few other things, leaving the 64Mhz ARM9 to process the games themselves. Even though the DSi took out the GBA slot, the GBA hardware was still intact. With the 3DS, both ARM9 and ARM7 were carried over to accompany the dual-core ARM11.
 
Nintendo going off the rails with a brand new concept with their next console is my biggest concern. I just want a more powerful Switch. They can do a third pillar thing if they want to try something else, but don't fuck up their transition again.
 
They didn't go quite out of their way on that. The main reason is that the GBA hardware is used for DS games too, though not really programmable to the devs in that mode. The ARM7 gets a boost from 16Mhz to 32Mhz, and is in charge of audio processing, WiFi, and a few other things, leaving the 64Mhz ARM9 to process the games themselves. Even though the DSi took out the GBA slot, the GBA hardware was still intact. With the 3DS, both ARM9 and ARM7 were carried over to accompany the dual-core ARM11.
I don't think this really goes against what Serif was saying. Merely having any old coprocessor for the DS's use would not have been enough to play GBA games without Nintendo "going out of their way" (such as it is) to use the ARM7TDMI, include the GBA BIOS, and allow the rest of the board to work in an ARM7-only GBA-like mode.
 
They didn't go quite out of their way on that. The main reason is that the GBA hardware is used for DS games too, though not really programmable to the devs in that mode. The ARM7 gets a boost from 16Mhz to 32Mhz, and is in charge of audio processing, WiFi, and a few other things, leaving the 64Mhz ARM9 to process the games themselves. Even though the DSi took out the GBA slot, the GBA hardware was still intact. With the 3DS, both ARM9 and ARM7 were carried over to accompany the dual-core ARM11.
Isn't it plausible to that a motivation to include the ARM7TDMI in the first place was to facilitate GBA BC, and they were able to multi-purpose it ? If they decided from the beginning to make it not BC, would the ARM7 still have been there? I also recall the GBA had a Sharp CPU for GB/C compat, I'm not sure if that ended up being used for any GBA games.

That and the inclusion of the fatter GBA card slot is why, despite the DS being a new brand and having a different storage medium, is why I said "out of their way" i.e. they could have certainly kept the DS non-compatible as their third-pillar distinct from the GBA line. It's why I'm skeptical that if BC with the Switch is feasible, that Nintendo somehow wouldn't do it for other reasons.
 
Correct me if I'm wrong but I feel like the Switch has had the most aggressive speculation about it's successor of any other previously released Nintendo console. Right from release people were already thinking about a Pro version using a new custom Nvidia chip rather than just a TX1.

I wonder if this, and the rampant Switch Pro rumors has actually resulted in less leaks, or less reported leaks. I guess to further elaborate, Nintendo is being extra careful because they know how widespread discussion is. Outlets are also likewise less likely to report on things given how long speculation has been going on, and at least in recent years probably very skeptical of their sources given the whole Switch Pro/OLED conflation.

It's impossible to know, but maybe in a world where things hadnt turned out quite like they did, we'd be going into 2023 with more outlets confidently reporting on a console launching this year. Regardless, this has been an incredibly weird generation for future hardware speculation.
 
I'm just saying that perhaps GBA BC was always planned from the beginning because they were also planning on using it for other things, similar to how the GBA had the chip from the GB/C that provided the additional sounds generators. It's just how Nintendo dealt with their portables by reusing what they already had, up until the Switch of course.
 
Why is MVG skeptical about BC? I'm not convinced that implementing some kind of solution is this insurmountable challenge that some are framing it as. Or that Nintendo would deliberately not have it, so they could (as I've heard too many times around the 'net) sell Deluxe ports of the entire library.
MVG comes from more of a hobbyist/emulator background, which for some reason seems to predispose people to the view that the shader issue is somehow insurmountable without sticking in the old hardware like consoles used to do. I'm not sure what exactly causes it, because it seems somewhat counterintuitive. There's something deeply ironic about one of the main Switch emulator devs (who, to be clear is not MVG) claiming that Nintendo and Nvidia couldn't possibly deal with it in software.
 
Correct me if I'm wrong but I feel like the Switch has had the most aggressive speculation about it's successor of any other previously released Nintendo console. Right from release people were already thinking about a Pro version using a new custom Nvidia chip rather than just a TX1.

I wonder if this, and the rampant Switch Pro rumors has actually resulted in less leaks, or less reported leaks. I guess to further elaborate, Nintendo is being extra careful because they know how widespread discussion is. Outlets are also likewise less likely to report on things given how long speculation has been going on, and at least in recent years probably very skeptical of their sources given the whole Switch Pro/OLED conflation.

It's impossible to know, but maybe in a world where things hadnt turned out quite like they did, we'd be going into 2023 with more outlets confidently reporting on a console launching this year. Regardless, this has been an incredibly weird generation for future hardware speculation.
I dunno, NX speculation was wild.
 
Switch 2 not being fully BC with OG Switch games is a big deal breaker for me I think. I’ve invested in a lot of software for the Switch this generation (think I have over 400 games on the device) it’s my main indie game machine so I hope these all carry over to the next console. Hell, it would be nice as well if some of them also got a bit of a boost from some kind of system level booster like PS4 Pro did but that would just be a nice extra.
 
Switch 2 not being fully BC with OG Switch games is a big deal breaker for me I think. I’ve invested in a lot of software for the Switch this generation (think I have over 400 games on the device) it’s my main indie game machine so I hope these all carry over to the next console. Hell, it would be nice as well if some of them also got a bit of a boost from some kind of system level booster like PS4 Pro did but that would just be a nice extra.


I think it will be for at least one gen. Switch 3 will probably cut off Switch 1 games/cartridges. Nintendo is usually good at keeping software from at least one previous gen when they start a new concept (except the Wii U to switch and n64 to GCN, both previous consoles didn't sell very well).
 
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