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

[Dakhil]

Like for Switch 2?

If you believe Nintendo's next model is a successor, not a mid-gen refresh, then yes.

I personally believe the DLSS model*'s an iterative successor in the similar vein to the iPhones (e.g. the iPhone 13 is an iterative successor to the iPhone 12).

 

[ILikeFeet]

If there is no Switch Pro then it should be for Switch 2, correct?

many of us here considered the Pro and the 2 to be the same thing

 

[Meelow]

many of us here considered the Pro and the 2 to be the same thing

It's possible that the Switch Pro turned into the Switch 2 after all this time.

 

[ILikeFeet]

It's possible that the Switch Pro turned into the Switch 2 after all this time.

naming is something that can be nailed down damn near whenever. I don't put too much stock into names because they don't really tell you anything

 

[Pokemaniac]

Something that I wanted to point out for a while but wasn't sure if it was right or not, but there is a possibility that the Dane name is not for the final silicon, but for the devkit silicon that comes before the final silicon, which is not farfetch'd.

PS5 had codenames for two chips at one point in time if I remember right, the Ariel soc which was in the early devkits with Navi-based GPUs, while the final devkits had the Oberon which was with the unique Navi-based uArch in the PS5 that has the actual final hardware features. Ariel was also taped out along with Oberon for its use case, and Oberon is what we see in the case of the PS5 APU. Before Ariel the devkits were based off of the PS4 Pro.

I do not think banking on "Dane" getting taped out actually means that the final hardware will come out soon, it could just as equally be for something like the pre-final silicon codename while they are still hard at work with the actual final silicon.

Just to put that out there.

Series X has Dante I think but its final name was Scarlett for the soc.

It's just odd for there not to be 2 codenames for silicon that pertains to Nintendo and only one name exists right now.

this is only based on what I remember from online discussion.

I don't think that really gels with the way kopite7kimi was originally describing the chip. Plus it seems unnecessary when they can just strap together a heavily throttled desktop Ampere card (which coincidentally launched right around the same time the devkits supposedly went out initially) with one of Nvidia's existing Jetson boards to get a good enough approximation for a pre-silicon devkit.

It's possible that the Switch Pro turned into the Switch 2 after all this time.

I don't think it's super likely Dane ever shifted in purpose from what we've heard. I think there's a strong possibility that a more traditional "Switch Pro" was a potential path that the Mariko hybrid could have taken, but that definitely would have been a separate project from what we've been discussing more recently.

 

[ShadowFox08]

as someone on InstallBase put it, people can take a wait, but a sustained delay in getting a system would put them off on it

---

in other news, an outlet had an interview with someone with a Steam Deck dev kit. not much here, but I found this relevant

Steam Deck Developer Kit Impressions - Boiling Steam

We had a chance to have a quick Q&A with a developer, who asked to remain anonymous, with hands-on access to the developer kit of the Steam Deck. Because of a non-disclosure agreement (NDA), they couldn’t comment on every question… <a...

boilingsteam.com

How is it that the micro SD cards aren't far behind of the flash storage? The NVME on steam deck has a bandwidth of 3 GB/s, though I'm not sure that correlates directly to read speeds.

Around a PS4 in portable mode (in raw theoretical performance), and around the ballpark of an Xbox Series S in docked. Of course the performance of a Series S in a Switch form factor probably gives you an idea of the cost!

In reality bandwidth would probably be the major bottleneck. Even with 128-bit LPDDR5X you'd be pretty severely bottlenecked, I'd imagine. Of course, if money is no object, then a 24GB stack of HBM3 should probably do the job (might want to clock it down a bit, though, 819GB/s is a tad overkill).



It's fair in the sense that GA104 is a 17.4 billion transistor, 392.5mm2 chip, and Orin's a 21 billion transistor, 460mm2 chip. GA104 has a lot more GPU logic, but Orin has CPU cores, deep learning accelerators, vision accelerators, etc., all of which consume power. The RAM situation is definitely different, though, that GDDR6 is hungry.

My point was more that chips of that size are usually GPUs or big server CPUs, designed to consume as much as 200W. The clocks we have for Jetson AGX Orin are for a 50W envelope, and they could certainly clock a lot higher if it wasn't limited to such a low power budget. I'd say CPU clocks around 2.8GHz and GPU around 1.5GHz should probably be doable. In fact, Nvidia advertise 254 TOPS for the Drive AGX Orin (vs 200 for the Jetson Orin), which would suggest some kind of increased clock and power draw in that use-case, although they don't specify whether it's an increase on the GPU or DLA side (or both).


Yeah, I could see them getting a 25%-33% improvement in capacity, which would be a nice bump and might allow closer to Mariko battery life with closer to TX1 power draw, but I don't expect it to make a major change to the capabilities of the console. Basically I'm just trying to caution against unrealistic expectations that things like solid state batteries will come along and triple battery life any time soon.

There is the HBME2 or the HBM2 that offers 256GB/s to 460.8 GB/s. Absolutely crazy they only use 1.2 V

Speaking of, Sony has confirmed to Bloomberg that Sony will continue manufacturing PlayStation 4 consoles throughout 2022 due to the shortage of PlayStation 5 consoles. So I think the cross-gen period for the current-gen consoles (PlayStation 5, Xbox Series X|S), as well as the DLSS model*, will last much longer than people would have anticipated.

---

Anyway, Jordan Klein, an analyst from Mizuho Securities USA Inc., has apparently heard from an Inspur executive that AMD's increasing the price for datacentre customers by 10-30% in 2022.


Although I still believe Nintendo and Nvidia are probably going to use Samsung's 8N process node to fabricate Dane, in a hypothetical scenario where Nintendo and Nvidia decide to use an advanced process nodes from TSMC (TSMC's N7 process node or more advanced) to fabricate Dane, I can see the DLSS model*'s MSRP be higher than people would have anticipated (≥$449.99).

I hope it fails ufbut sells for they much, unless it's like a 512GB model and they sell a 128 or 256GB for $400 or cheaper.

If this thing is coming out after Steam Deck and its using TSMC N7, it better have a competitive price and performance to it. Steam Deck with 256GB SSD is $529 with a carrying case. But its a february release with 4 cores at 3.6Ghz, 1.6 TFLOP GPU (exactly 2.5x slower than X box series S), and 16GB LPDDR5. Switch 2 likely won't match the CPU speed or RAM amount (16GB), but it likely has a deeper discount/deal with Nvidia then Steam does with AMD... Aand its coming out like a year later at least.

It's possible that the Switch Pro turned into the Switch 2 after all this time.

it's going to have multiple generation jumps in CPU and GPU, as well as bandwidth. I don't know how anyone can say it's a pro model. It's a successor if the rumors of ampere tech with DLSS are true. pro models will at least have the same CPU. But anyway, the time for a pro model has long sailed.

I don't understand the concept of the DLSS model being an iterative successor personally. Certainly not the concept of a significant upgrade every 2 years or so like a mobile phone, which is definitely not gonna be happening, which I don't think anyone here is implyinh. But I do see them being supported with mostly the same games for the first 2 or 3 years.. I feel like we have already been getting there with PS4 and PS5 and xbone and x series x when the latter consoles came out.

IMO, next gen switch will be similar to current. I'm sticking with Switch 2 and refusing it to call it Switch DLSS as if it's a pro model. I think Switch 2 will be released in 2023 (2022 Q4 at the earliest, but it seems less likely to me unless they release it with botw 2 this yr), while Switch 1 will be supported for a few more years with mostly the same games (1st party for sure). Switch 2 will have multiple models like lite l, XL models etc, and likely multiple skus with more storage space.

 

[Z0m3le]

Dane isn't a Nintendo codename, it's Nvidia's codename for the custom Tegra chip T239 that Nvidia is making for Nintendo, which has the same architecture and chip process as Orin. It's not the name for a devkit from Nintendo or something developers might even have heard, the information comes from Kopite7 who gave it alongside Orin information that turned out 100% accurate.

Nate was able to confirm developers were working on games for a Nintendo platform that had DLSS functionality, which requires a new Tegra chip, he said they were targeting holiday 2022/early next year. He was on record saying this while Bloomberg was saying that the "pro" model with new cpu/gpu was coming last year, because they knew about the new OLED screens being shipped and got it all confused.

As soon as Zelda botw sequel was announced, rumors starting popping up that it was going to come out along side the pro model, this is something I've heard through many sources over the years now. Now like I said in my last post, I sort of dropped off, so I haven't heard from anyone for about half a year now, but I did reach out this week and did get interesting news, someone who is working on software for this new model had their game "moved" from end of this year to early next year, they don't know if it is suppose to launch with new hardware or not but it is targeting new hardware specs for Nintendo. The pandemic does cause constant delays, even Elden Ring slid a month recently, but this info which is now over a month old, could indicate that Dane has been moved from this calendar year but still on track for end of next FY (March 2023).

Honestly, I think we are sort of in a tough waiting period, we know a lot of the spec side of things, but we haven't seen a lot of Orin benchmarks and when those kits get into people's hands, they can literally test the performance we are looking at, at say 5 watts, 10 watts and 15 watts, of course software will be an issue just like it is with Xavier devkits, but it can give us an indication of clock speeds, which can give raw performance numbers. Though it being "first quarter" 2022, it will likely be released after some other solid info. That ultimately still leaves us with holiday 2022 - Sping 2023, but because there is no big indication of millions of new chips/parts for this new device yet seen by a news source like bloomberg, we don't know if the device will release at all, we can be sure that it exists, but ultimately it can be canceled up to the point where Nintendo is buying large quantity of new parts.

 

[SiG]

Dane isn't a Nintendo codename, it's Nvidia's codename for the custom Tegra chip T239 that Nvidia is making for Nintendo, which has the same architecture and chip process as Orin. It's not the name for a devkit from Nintendo or something developers might even have heard, the information comes from Kopite7 who gave it alongside Orin information that turned out 100% accurate.

Nate was able to confirm developers were working on games for a Nintendo platform that had DLSS functionality, which requires a new Tegra chip, he said they were targeting holiday 2022/early next year. He was on record saying this while Bloomberg was saying that the "pro" model with new cpu/gpu was coming last year, because they knew about the new OLED screens being shipped and got it all confused.

As soon as Zelda botw sequel was announced, rumors starting popping up that it was going to come out along side the pro model, this is something I've heard through many sources over the years now. Now like I said in my last post, I sort of dropped off, so I haven't heard from anyone for about half a year now, but I did reach out this week and did get interesting news, someone who is working on software for this new model had their game "moved" from end of this year to early next year, they don't know if it is suppose to launch with new hardware or not but it is targeting new hardware specs for Nintendo. The pandemic does cause constant delays, even Elden Ring slid a month recently, but this info which is now over a month old, could indicate that Dane has been moved from this calendar year but still on track for end of next FY (March 2023).

Honestly, I think we are sort of in a tough waiting period, we know a lot of the spec side of things, but we haven't seen a lot of Orin benchmarks and when those kits get into people's hands, they can literally test the performance we are looking at, at say 5 watts, 10 watts and 15 watts, of course software will be an issue just like it is with Xavier devkits, but it can give us an indication of clock speeds, which can give raw performance numbers. Though it being "first quarter" 2022, it will likely be released after some other solid info. That ultimately still leaves us with holiday 2022 - Sping 2023, but because there is no big indication of millions of new chips/parts for this new device yet seen by a news source like bloomberg, we don't know if the device will release at all, we can be sure that it exists, but ultimately it can be canceled up to the point where Nintendo is buying large quantity of new parts.

So essentially, we have a "Project Atlantis" in our hands.

The successor could very well shift to using a newer SoC/chip (the only logical conclusion if Nvidia/Nintendo decides to cancel Dane production in favor of a more recent and perhaps more producable chip) and we wouldn't even know it since we aren't developers...

Or maybe we're putting too much faith into Kopite7's assertion that Dane is an Nvidia chip specifically built for Nintendo.

I can still see the chip shortage as drastically infuencing both Nvidia and Nintendo's production plan.

It's possible that the Switch Pro turned into the Switch 2 after all this time.

This wouldn't be the first time such things happened at Nintendo.

 

[Hermii]

So essentially, we have a "Project Atlantis" in our hands.

The successor could very well shift to using a newer SoC/chip (the only logical conclusion if Nvidia/Nintendo decides to cancel Dane production in favor of a more recent and perhaps more producable chip) and we wouldn't even know it since we aren't developers...

Or maybe we're putting too much faith into Kopite7's assertion that Dane is an Nvidia chip specifically built for Nintendo.

I can still see the chip shortage as drastically infuencing both Nvidia and Nintendo's production plan.

This wouldn't be the first time such things happened at Nintendo.

The fact that Nvidia hasn’t publicly unveiled t239 does suggest kopite was on the money.

 

[Z0m3le]

Yeah, I actually tend to think of Dane as GBC, but it could end up being like Atlantis and resurrect itself as GBA down the line with even better hardware. That hasn't come to pass and as far as a month ago, people were still working on Dane projects, remember I haven't really been in the loop for ~4 months, so while I'm saying Dane might never come to market, nothing has changed that I'm aware of outside of a 3rd party studio with a single game being delayed a few months during covid. That isn't a direct indication that Dane is delayed or canceled, it is just worth remembering that nothing has been promised to us yet.

 

[ILikeFeet]

How is it that the micro SD cards aren't far behind of the flash storage? The NVME on steam deck has a bandwidth of 3 GB/s, though I'm not sure that correlates directly to read speeds.

The most likely story is that their game isn't hitting storage all that much. I don't see this device as being popular with anyone but indies, after all

 

[LoneRanger]

Do you think this next hardware switch console was originally a 2021 launch (before Covid), was replaced by OLED and delayed to 2023 as positioned as a full successor?

Remember, basically all nintendo titles got massive delays of between 3 months to 1 year. Its very very likely games such as Splatoon 3, Zelda botw 2, xenoblade 3, pokemon legends arceus were originally 2021 games

I find all these situation really strange. I mean, if really Nintendo is planning a new shift in hardware for next year, why not delayed Splatoon 3 to being a new next gen title? Why the power boost of next Switch HW I’m sure Nintendo team can do a massive game.

Maybe its happening one or both of these two things:
1) They think people will upgrade it even if first party software is cross-gen, just like Xbox or Playstation users. IMO, they need at least one first party exclusive day one lol.
2) They pushing hard to get a lot of exclusive (not playable on actual HW) third party titles for the first 6 months/1 year. Some are PS4/XOne ports (example RDR2), others are current-gen day 1 games.

I also think this next hardware will launched at september 2023. They will gave an preliminary announcement the day (or maybe a bit before) their FY earnings (late april). Then, a few days before E3 they revealed the console itself, and in E3 they showed the games. In September they do their traditional direct to hype gamers for 2024, revealing more exclusive Switch successor titles.

 

[Z0m3le]

So, here has been my thoughts this morning:
If you are Nintendo and you have a new more powerful Switch, you have this big Mario movie with other movie commitments already in the works, and you have both Mario and Zelda ready for this year, what do you release this holiday?

My heart says all of it, but my brain says you release a brand new Mario title along side the new movie and move Dane and Zelda to Feb/March next year when you need them. Mario sells better during the holidays, it will sell even more next to its movie, meanwhile Zelda and Dane would sell in any month, moving it to next year gives you an animal crossing effect... It boosts the entire platform's sales and gives current owners a reason to double dip that didn't buy the OLED model. It virtually Guarantees another 20M+ sales for next year, putting it above 140M sales by the end of March 2024. I don't think Nintendo plans to release another platform for years, and that a "pro" model always seems to come on the back half of Nintendo's handheld hardware.... GB's 11 year life span, had GBC in the last 3 years. GBA's ~4 year life span, had GBA SP launch in its 3rd year. DS's 8 year life span, had DSi in it's 5th year, New 3DS is the only exception here and I'd argue it is because of how badly the 3DS launched, the Wii U did and the price of the Switch.

I think because of Switch's capabilities as a cloud gaming device, that a more powerful Switch could replace it without being a new platform, developers can release Dane Switch games and offer cloud versions to users who haven't upgraded.

Do you think this next hardware switch console was originally a 2021 launch (before Covid), was replaced by OLED and delayed to 2023 as positioned as a full successor?

Remember, basically all nintendo titles got massive delays of between 3 months to 1 year. Its very very likely games such as Splatoon 3, Zelda botw 2, xenoblade 3, pokemon legends arceus were originally 2021 games

I find all these situation really strange. I mean, if really Nintendo is planning a new shift in hardware for next year, why not delayed Splatoon 3 to being a new next gen title? Why the power boost of next Switch HW I’m sure Nintendo team can do a massive game.

Maybe its happening one or both of these two things:
1) They think people will upgrade it even if first party software is cross-gen, just like Xbox or Playstation users. IMO, they need at least one first party exclusive day one lol.
2) They pushing hard to get a lot of exclusive (not playable on actual HW) third party titles for the first 6 months/1 year. Some are PS4/XOne ports (example RDR2), others are current-gen day 1 games.

I also think this next hardware will launched at september 2023. They will gave an preliminary announcement the day (or maybe a bit before) their FY earnings (late april). Then, a few days before E3 they revealed the console itself, and in E3 they showed the games. In September they do their traditional direct to hype gamers for 2024, revealing more exclusive Switch successor titles.

December 2019, Nvidia announced Orin for 2022. Orin and Dane are the same architecture and process node, they are sister chips and use the same technology (according to the source that gave us Dane's existence anyways). It is unlikely to me that Nintendo planned to release a custom Tegra chip based on the same technology as Orin as much as a year before Orin's release, so no, I think OLED model was always coming as a premium model, much like the XL models of previous hardware, that predated the more powerful versions of those consoles. Dane was always meant to come out FY 2022, and I think Zelda was always meant to launch with it.

 

[Dakhil]

But its a february release with 4 cores at 3.6Ghz, 1.6 TFLOP GPU (exactly 2.5x slower than X box series S), and 16GB LPDDR5.

Regardless of which foundry company Nintendo and Nvidia decide to work with, I very highly doubt the CPU and/or GPU is going to run at a frequency as high as 3.6 GHz. 3 GHz is currently the highest frequency the Cortex-A78 runs at, and that's with a smartphone SoC fabricated using TSMC's N6 process node. And keep in mind that smartphones run at high CPU frequencies in short bursts rather than for long, sustained periods of time. I very highly doubt the CPU on Dane is going to run at a sustained frequency of 3 GHz for thermal and yield reasons.

I don't understand the concept of the DLSS model being an iterative successor personally. Certainly not the concept of a significant upgrade every 2 years or so like a mobile phone, which is definitely not gonna be happening, which I don't think anyone here is implyinh. But I do see them being supported with mostly the same games for the first 2 or 3 years.. I feel like we have already been getting there with PS4 and PS5 and xbone and x series x when the latter consoles came out.

I was talking more about the cross-gen period for an iterative successor being much longer than the cross-gen period for a traditional successor, rather than the rate of hardware improvements, when talking about an iterative successor.

The reason I mentioned that the iPhone 13 is an iterative successor to the iPhone 12 is because all of the software the iPhone 12 can run can also run on the iPhone 13; and the iPhone 12 and iPhone 13 run on the same OS. But obviously, there will be some software that will run exclusively on the DLSS model* and not on the Nintendo Switch from day one, which is why I said "similar" and not "same".

Speaking of, Microsoft confirmed to the Verge that all Xbox One models were discontinued since late 2020.

Do you think this next hardware switch console was originally a 2021 launch (before Covid), was replaced by OLED and delayed to 2023 as positioned as a full successor?

I very highly doubt that's the case since firmware datamines from around late 2020 mentioned that the OLED model (codenamed "Aula") was using the Tegra X1+ (codenamed "Mariko"), and that Nintendo have been working on it for around 1.5 - 2 years.

 

[Thraktor]

There is the HBME2 or the HBM2 that offers 256GB/s to 460.8 GB/s. Absolutely crazy they only use 1.2 V

Well, if we're treating money as no object we may as well go with the latest and greatest. Slightly more seriously, though, I think some form of HBM could be a good solution for a (high-end) portable gaming device, but it would probably want to be a tweaked low-power variant. HBM2E is already pretty much the most power efficient memory out there, at around 4 pJ/bit (picoJoules per bit), but scaled up to 460.8GB/s, that's still just under 15W for the package. Much less than something like GDDR6 for the same bandwidth, and absolutely worthwhile for the high-end HPC GPUs it's used on, but far too much for a portable console.

HBM itself is inherently efficient for two main reasons. The first is physical, in that it's connected on package via an interposer rather than over PCB traces, and the second is, because of that interposer connection, it's able to use a very wide bus (1024 bit per package, vs 16-64 bit for other memory standards) at a lower clock, and wide and slow is more efficient than narrow and fast. However, the latest standards like HBM2E and HBM3 are basically wide and fast. HBM3 chips are going to hit 6.4Gb/s per pin, which is the same as LPDDR5, and faster than almost all currently available DDR5. Again, this makes sense for the kind of server/HPC applications that need this much bandwidth, and it's still no doubt the most power-efficient way to get it, but it's going in quite a different direction than a portable gaming device would require.

That said, if let's say Nvidia and Samsung were to work together to make the most efficient HBM possible, I'd say it could be very efficient indeed. Nvidia actually published a proposal for something called FGDRAM back in 2017 (PDF link), which is a variant of HBM which can significantly reduce power consumption, to around 2 pJ/bit. This is still targeting very high performance of around 1TB/s (basically it's an alternative to what would become HBM3), so I have no doubt you could improve efficiency further by bringing clocks down considerably and going back to a wide and slow approach.

Let's say they could hit 1.5 pJ/bit with FGDRAM and much-reduced clocks. As far as I'm aware the LPDDR4 in the original Switch consumed around 2W to provide 25.6GB/s, and with this memory you could hit around 165GB/s in the same 2W envelope. If you pushed the power envelope up to 5W (let's say for docked mode) you could get to 415GB/s, which is not bad at all for a device the same form-factor as a Switch. A device which would be vastly more expensive than the Switch, but the same form-factor nonetheless.

 

[Skittzo]

Like for Switch 2?

Edit: Ohh right, I forgot about Dane. I remember hearing about it.

If there is no Switch Pro then it should be for Switch 2, correct?

Do you think this next hardware switch console was originally a 2021 launch (before Covid), was replaced by OLED and delayed to 2023 as positioned as a full successor?

I don't quite understand why anyone should think that the positioning of this model has changed in any way.

 

[Z0m3le]

I think part of why it makes sense for Nintendo to delay Dane and Zelda to early next year is because of how packed this year is looking to be:

Jan: Pokemon Legends
Feb: (Probably windwaker/twilight princess HD), maybe a new NSO game.
Mar: Kirby
Apr: Advance Wars
May, June, July (Speculation: Not sure which game goes in which month) Splatoon 3, Fire Emblem, A nintendo family game, hopefully Mario Strikers 3
August: Metroid Prime Remake?
September: Mario+Rabbids 2?
October: Bayonetta 3
November: Big new Mario game.
December: Monolith's new game?
20M+ sales

Just feeling out what 2023 would possibly be

Jan 2023: Star Fox Zero port?
Feb 2023: new NSO game? (Nintendo Remix, like NES Remix, but it goes through multiple generations of hardware)
March 2023: Dane Switch and Zelda
April 2023: Mario Kart 10
May 2023: Yoshi game
June 2023: rhythm heaven?
July 2023: Paper Mario game?
August 2023: Platinum game of some sort? Astral Chain 2?
September 2023: Donkey Kong game / Pokemon game
October 2023: Luigi's Mansion 4
November 2023: Metroid Prime 4
December 2023: Pikmin 4
25M+ sales (rides Mario right into new hardware and Zelda)

 

[Vantell]

Do you think this next hardware switch console was originally a 2021 launch (before Covid), was replaced by OLED and delayed to 2023 as positioned as a full successor?

Remember, basically all nintendo titles got massive delays of between 3 months to 1 year. Its very very likely games such as Splatoon 3, Zelda botw 2, xenoblade 3, pokemon legends arceus were originally 2021 games

I find all these situation really strange. I mean, if really Nintendo is planning a new shift in hardware for next year, why not delayed Splatoon 3 to being a new next gen title? Why the power boost of next Switch HW I’m sure Nintendo team can do a massive game.

Maybe its happening one or both of these two things:
1) They think people will upgrade it even if first party software is cross-gen, just like Xbox or Playstation users. IMO, they need at least one first party exclusive day one lol.
2) They pushing hard to get a lot of exclusive (not playable on actual HW) third party titles for the first 6 months/1 year. Some are PS4/XOne ports (example RDR2), others are current-gen day 1 games.

I also think this next hardware will launched at september 2023. They will gave an preliminary announcement the day (or maybe a bit before) their FY earnings (late april). Then, a few days before E3 they revealed the console itself, and in E3 they showed the games. In September they do their traditional direct to hype gamers for 2024, revealing more exclusive Switch successor titles.

I think the OLED was meant to be more than it is. What I suspect that happened is that Nintendo was preparing for the Switch Pro. Ordering all new components etc. Then COVID hit and we got issues in the logistics of these things. Nintendo already got hold of the new screens and thus decided to strip their original idea to what the OLED is now.

 

[Intoxicate]

I think the OLED was meant to be more than it is. What I suspect that happened is that Nintendo was preparing for the Switch Pro. Ordering all new components etc. Then COVID hit and we got issues in the logistics of these things. Nintendo already got hold of the new screens and thus decided to strip their original idea to what the OLED is now.

The OLED Switch has a complete revised system board, the dock has changed completely etc. I don't think it was a general supply issue.

 

[Mr Swine]

The OLED Switch has a complete revised system board, the dock has changed completely etc. I don't think it was a general supply issue.

Nintendo probably was thinking of overclocking the Oled model but went against it since it would cost more to have the same Heat pipes and fan as the 2017 model thus getting less profit from it?

 

[Z0m3le]

Nintendo probably was thinking of overclocking the Oled model but went against it since it would cost more to have the same Heat pipes and fan as the 2017 model thus getting less profit from it?

No, originally there was talk of overclocking the 2019's red box model with the same Mariko chip, the cooling might have been the reason they didn't do it, but from my understanding, it seems they couldn't get the clocks high enough, so they decided to do no overclocking with the red box model instead. OLED model was long after they decided not to overclock Mariko, so I think it was just a chance to sell people an "XL" version of the Switch with a bigger, more vibrant screen and a better kickstand. The OLED model was always going to be what we got in 2021.

 

[Skittzo]

I think the OLED was meant to be more than it is. What I suspect that happened is that Nintendo was preparing for the Switch Pro. Ordering all new components etc. Then COVID hit and we got issues in the logistics of these things. Nintendo already got hold of the new screens and thus decided to strip their original idea to what the OLED is now.

Nah we know for a fact this is not the case. Aula was datamined right at the start of the pandemic and therefore must have been officially in the works long before it hit.

 

[Dakhil]

No, originally there was talk of overclocking the 2019's red box model with the same Mariko chip, the cooling might have been the reason they didn't do it, but from my understanding, it seems they couldn't get the clocks high enough, so they decided to do no overclocking with the red box model instead. OLED model was long after they decided not to overclock Mariko, so I think it was just a chance to sell people an "XL" version of the Switch with a bigger, more vibrant screen and a better kickstand. The OLED model was always going to be what we got in 2021.

Just to add to this, there was a rumour about a Tegra X1 sample that was fabricated using TSMC's N7 process node. And although Nvidia and TSMC warned Nintendo about the difficulty of exceeding the Cortex-A57's max frequency of 1.9 GHz, the Cortex-A57's frequency was increased from 1.9 GHz to 2.52 GHz, which resulted in the power consumption being significantly higher than expected.

 

[ShadowFox08]

Regardless of which foundry company Nintendo and Nvidia decide to work with, I very highly doubt the CPU and/or GPU is going to run at a frequency as high as 3.6 GHz. 3 GHz is currently the highest frequency the Cortex-A78 runs at, and that's with a smartphone SoC fabricated using TSMC's N6 process node. And keep in mind that smartphones run at high CPU frequencies in short bursts rather than for long, sustained periods of time. I very highly doubt the CPU on Dane is going to run at a sustained frequency of 3 GHz for thermal and yield reasons.

I was talking more about the cross-gen period for an iterative successor being much longer than the cross-gen period for a traditional successor, rather than the rate of hardware improvements, when talking about an iterative successor.

The reason I mentioned that the iPhone 13 is an iterative successor to the iPhone 12 is because all of the software the iPhone 12 can run can also run on the iPhone 13; and the iPhone 12 and iPhone 13 run on the same OS. But obviously, there will be some software that will run exclusively on the DLSS model* and not on the Nintendo Switch from day one, which is why I said "similar" and not "same".

Speaking of, Microsoft confirmed to the Verge that all Xbox One models were discontinued since late 2020.


I very highly doubt that's the case since firmware datamines from around late 2020 mentioned that the OLED model (codenamed "Aula") was using the Tegra X1+ (codenamed "Mariko"), and that Nintendo have been working on it for around 1.5 - 2 years.

I was just mentioning the steam Deck's CPU speed/specs. I don't expect the Switch 2 to run at those speeds at all.

 

[Dakhil]

I was just mentioning the steam Deck's CPU speed/specs. I don't expect the Switch 2 to run at those speeds at all.

Fair enough. Valve mentioned that the 4 cores/8 threads Zen 2 CPU on the Steam Deck runs at a range of 2.4 - 3.5 GHz, which seems to suggest that the Steam Deck probably won't be running at 3.5 GHz for sustained periods of time.

 

[ReddDreadtheLead]

Let's say they could hit 1.5 pJ/bit with FGDRAM and much-reduced clocks. As far as I'm aware the LPDDR4 in the original Switch consumed around 2W to provide 25.6GB/s, and with this memory you could hit around 165GB/s in the same 2W envelope. If you pushed the power envelope up to 5W (let's say for docked mode) you could get to 415GB/s, which is not bad at all for a device the same form-factor as a Switch. A device which would be vastly more expensive than the Switch, but the same form-factor nonetheless.

iirc, it clocks down the RAM and targets 21GB/s in handheld mode.

I don't think that really gels with the way kopite7kimi was originally describing the chip. Plus it seems unnecessary when they can just strap together a heavily throttled desktop Ampere card (which coincidentally launched right around the same time the devkits supposedly went out initially) with one of Nvidia's existing Jetson boards to get a good enough approximation for a pre-silicon devkit.

thing is that the recent rumor, as per Nate, describes it as an actual tablet with a battery where the battery/thermals were impacted more negatively than anticipated with RT enabled, not like the AGX devkits that have an Xavier chip and an RTX GPU in a case. Ariel was taped out well over a year before the release of the PS5, which had its APU taped out months before release and what was in the final devkits.

 

[jimnastics]

Feb: (Probably windwaker/twilight princess HD)

Is there a solid source for this, or just your instinct? I'd be so bloody happy if that happened, but feels like we'd know by now surely?

 

[Mr Swine]

No, originally there was talk of overclocking the 2019's red box model with the same Mariko chip, the cooling might have been the reason they didn't do it, but from my understanding, it seems they couldn't get the clocks high enough, so they decided to do no overclocking with the red box model instead. OLED model was long after they decided not to overclock Mariko, so I think it was just a chance to sell people an "XL" version of the Switch with a bigger, more vibrant screen and a better kickstand. The OLED model was always going to be what we got in 2021.

Couldn’t they just do a 25% overlock on the CPU and GPU and call it a day or did they honestly hoped to run at Max clocks?

 

[JoshuaJSlone]

Couldn’t they just do a 25% overlock on the CPU and GPU and call it a day or did they honestly hoped to run at Max clocks?

Maybe, but that small a change wouldn't make for much of a selling point. "Buy the New Nintendo Switch, and games will drop from 30fps less often!"

 

[ILikeFeet]

Couldn’t they just do a 25% overlock on the CPU and GPU and call it a day or did they honestly hoped to run at Max clocks?

25% cpu overclock is pointless when games can use a 1.78GHz boost for loading. Having that at all times would be a worthwhile boost

 

[Dakhil]

thing is that the recent rumor, as per Nate, describes it as an actual tablet with a battery where the battery/thermals were impacted more negatively than anticipated with RT enabled, not like the AGX devkits that have an Xavier chip and an RTX GPU in a case. Ariel was taped out well over a year before the release of the PS5, which had its APU taped out months before release and what was in the final devkits.

I think there's potentially a small possibility GA107 could be used for the handheld devkits.

---

Anyway, NateDrake mentioned during the most recent episode of Nate the Hate that Nintendo and Monolith Soft wanted to announce Xenoblade Chronicles 3(?) in early 2021, with the release date being the end of 2022. But that ultimately didn't happen due to Xenoblade Chronicles 3(?) having optimisation issues and other development problems related to COVID-19. NateDrake expects that Nintendo's and Monolith Soft's plans were delayed a full year, therefore NateDrake thinks Xenoblade Chronicles 3(?) could be announced in 1H 2022 and be released in December 2022, which would mirror Nintendo's and Monolith Soft's plans for 2021 with respect to Xenoblade Chronicles 3(?). However, NateDrake mentioned that he can't dismiss the possibility that the release date of Xenoblade Chronicles 3(?) could easily be delayed to early 2023 depending on how Monolith Soft overcomes the optimisation issues and other development problems related to COVID-19.


(The Xenoblade Chronicles 3(?) segment starts at 28:24.)

Assuming the DLSS model* is launching in holiday 2022, and if Nintendo and Monolith Soft have no choice, but to delay releasing Xenoblade Chronicles 3(?) from December 2022 to early 2023, I would prefer that Nintendo and Monolith Soft do so, since I would like to see Xenoblade Chronicles 3(?) have at least DLSS support (and maybe ray tracing support).

 

[ReddDreadtheLead]

Is there a solid source for this, or just your instinct? I'd be so bloody happy if that happened, but feels like we'd know by now surely?

Pretty sure that’s just an example considering the other part of what was included in that February line up for the year of what can fit in that slot.

Couldn’t they just do a 25% overlock on the CPU and GPU and call it a day or did they honestly hoped to run at Max clocks?

You’d be wasting your time with that as the issues with the switch aren’t GPU related but memory and CPU related.

And the Mariko memory isn’t that big a jump for that.

Nah we know for a fact this is not the case. Aula was datamined right at the start of the pandemic and therefore must have been officially in the works long before it hit.

We knew about Aula before that.

 

[manfromthedam]

My heart says all of it, but my brain says you release a brand new Mario title along side the new movie and move Dane and Zelda to Feb/March next year when you need them.

1. That sounds reasonable.
2. It hurts.
3. New 2D Mario with great art and music would be a great balm.
4. But it'd still hurt. xD

A point Emily (?) made last year on Twitter was that Zelda 2 and Dane weren't coupled per se. That could give hope for Zelda remaining in 2022 (besides Nintendo undoubtedly knowing Zelda's infamy for delays, which might bring more weight to them saying 2022).

Another point a user here raised recently was that a tie-in Mario game wouldn't be a necessity per se, with so many Mario titles already being on the market. That would leave the tentpole position open for Zelder. Still, I personally think a 2D Mario makes sense from a planning and business standpoint, so yeah... Pain .

 

[Skittzo]

We knew about Aula before that.

We did? I thought it was first datamined in April 2020.

 

[Look over there]

How is it that the micro SD cards aren't far behind of the flash storage? The NVME on steam deck has a bandwidth of 3 GB/s, though I'm not sure that correlates directly to read speeds.

Hitting that 3 GB/s in sequential read probably also requires the drive going at full blast.
Since NVMe allows for a controller to support a whole bunch of power states (pre-set power levels), I'm assuming that what's in the Steam Deck by default has some states set to very low wattages.

Although, the question that comes to my mind now is, what about user replacement drives? IIRC, the memory controller's on the drive, so I assume that whatever a user buys off the shelf and sticks into the deck will have power states assumed for desktop usage.
Or would the Steam Deck have its own list of power states?

 

[karmitt]

.

So, here has been my thoughts this morning:
If you are Nintendo and you have a new more powerful Switch, you have this big Mario movie with other movie commitments already in the works, and you have both Mario and Zelda ready for this year, what do you release this holiday?

If Nintendo has a great new 3D Mario under wraps, ready to go end of this year, I can definitely wait another 4-6 months for Zelda.

Do we have any reason to believe Mario is being made other than banking on a movie tie-in? That would be a very fun surprise.

A point Emily (?) made last year on Twitter was that Zelda 2 and Dane weren't coupled per se. That could give hope for Zelda remaining in 2022 (besides Nintendo undoubtedly knowing Zelda's infamy for delays, which might bring more weight to them saying 2022).

Unless you're talking about something else, I believe the point being made was not that Zelda could arrive before a revision, but that a revision does not need any killer app at launch. People were arguing that a 2021 hardware release didn't make sense as Zelda wasn't going to hit holiday 2021. This was after her post and the Bloomberg article had said that the 4K Switch was launching in 2021, but before the OLED Model had been announced.

The sentiment remains true - a revision does not need a Zelda tier release at launch. I'm sure it'll have some great software around it's launch, but it's going to sell out regardless of what the titles are.

 

[ReddDreadtheLead]

We did? I thought it was first datamined in April 2020.

I’m pretty sure that it was in 2019 that the news came out as it was two years prior to Aula, ie the OLED model releasing.

Unless you're talking about something else, I believe the point being made was not that Zelda could arrive before a revision, but that a revision does not need any killer app at launch. People were arguing that a 2021 hardware release didn't make sense as Zelda wasn't going to hit holiday 2021. This was after her post and the Bloomberg article had said that the 4K Switch was launching in 2021, but before the OLED Model had been announced.

The sentiment remains true - a revision does not need a Zelda tier release at launch. I'm sure it'll have some great software around it's launch, but it's going to sell out regardless of what the titles are.

Thing is, they clearly moved on from “pro” and will have it as the “2”, best to drop the Pro nomenclature at this point because of how late it is unless Nintendo is bullish enough to have a GBC situation again but more ambitious.

Anyway, the NSW2 would need some launch title, this isn’t a third party driven platform, it’s a first party driven but what title would actually be ready for it? Unless they just decide to go the XBox route, a platform driven by the third party content first and services then the first party stuff, I’m not sure how great that would pan out.

If it’s not a Zelda title, the likely candidate is actually a 3D Mario title.

Metroid would never be a good fit because it isn’t a system seller, it depends on the backs of Zelda and Mario to get it’s jumpstart.

Lots of EPD projects should be finishing or are finished, but there’s like 1-2 teams that aren’t finished and are working on something.


I’m only speculating here.

Then again, there’s that core enthusiast, like PlayStation and Xbox have, so does Nintendo. They would buy it regardless if it had a Nintendo launch title or not as they are the fanatics. See: XBox Series X|S and PS5 (to an extent) as an example


There is a benefit of early adopters.

I’m curious if they will have the better battery life model again or if they’ll go for a Mariko battery life this time from the start.

 

[karmitt]

I don't see there being any good reason to think that this device won't still be a revision. That's the last we'd heard from Nate and others, and I don't think a year is going to change that.

For now I do think we're looking at a more ambitious GBC situation as you've phrased it.

 

[Waluijob]

I don't see there being any good reason to think that this device won't still be a revision. That's the last we'd heard from Nate and others, and I don't think a year is going to change that.

For now I do think we're looking at a more ambitious GBC situation as you've phrased it.

More of a DSi/New Nintendo 3DS, if you will, that actually gets support for its expanded hardware.

 

[Procheno]

With the possibility of Xenoblade 3 being potentially delayed into 2023 now on the table...

Q1 2023 Dane please

 

[karmitt]

With the possibility of Xenoblade 3 being potentially delayed into 2023 now on the table...

Q1 2023 Dane please

Xenoblade 2 and Astral Chain suffered from resolution more than any other Switch titles to me.

Xenoblade 3 deserves better.

 

[bloopland33]

Yeah, I actually tend to think of Dane as GBC, but it could end up being like Atlantis and resurrect itself as GBA down the line with even better hardware. That hasn't come to pass and as far as a month ago, people were still working on Dane projects, remember I haven't really been in the loop for ~4 months, so while I'm saying Dane might never come to market, nothing has changed that I'm aware of outside of a 3rd party studio with a single game being delayed a few months during covid. That isn't a direct indication that Dane is delayed or canceled, it is just worth remembering that nothing has been promised to us yet.

Am I reading this right that you yourself know of a third party working on a Dane game? Or are you referencing Nate's comments some weeks back.

Not asking you to divulge any more info or anything, just sniffing out any crumb of hopium. That Grubb comment the other day got me spooked haha

 

[Instro]

Am I reading this right that you yourself know of a third party working on a Dane game? Or are you referencing Nate's comments some weeks back.

Not asking you to divulge any more info or anything, just sniffing out any crumb of hopium. That Grubb comment the other day got me spooked haha

What did Grubb say?

 

[karmitt]

What did Grubb say?

Quick summary: They're responding to recent articles on the analyst expecting there will be no Switch Pro, and that we'll see a 2024 Switch successor. Both Mike and Jeff both think that the ship has sailed for a Switch Pro, so whatever we see next is a Switch 2 / 'successor'. Jeff expects that we start to see very early rumblings later this year, and something more concrete in 2023.

As I said, it sounds speculative. Jeff does have a lot of connections, but this doesn't come off like he's reporting anything from a trusted source.

 

[Procheno]

Xenoblade 2 and Astral Chain suffered from resolution more than any other Switch titles to me.

Xenoblade 3 deserves better.

I think Nintendo and Monolith Soft supposedly delaying Xenoblade 3 for almost a year (not including covid) all because of optimization issues means they're starting to take the technical issues more seriously now and are working hard to enhance their engine to be more top tier. It was clear Xenoblade 2 suffered because the engine wasn't as optimized for the Switch as it could have been. They sacrificed that aspect for a shorter deadline. But now with 4-5 years of development, they can finally hit a big one. And I'm praying for a Dane launch to coexist

 

[Deleted member 645]

I personally think it makes since for another revision “4K” switch. Get a couple more years our of the switch name. I’m not sure if their “next-gen” console will be in the “switch” family. I personally am expecting a brand new console with some sort of VR capabilities. That makes a lot of sense to me. We shall see though. They have big decisions to make.

 

[Terrell]

So essentially, we have a "Project Atlantis" in our hands.

The successor could very well shift to using a newer SoC/chip (the only logical conclusion if Nvidia/Nintendo decides to cancel Dane production in favor of a more recent and perhaps more producable chip) and we wouldn't even know it since we aren't developers...

Or maybe we're putting too much faith into Kopite7's assertion that Dane is an Nvidia chip specifically built for Nintendo.

I can still see the chip shortage as drastically infuencing both Nvidia and Nintendo's production plan.

This wouldn't be the first time such things happened at Nintendo.

Well, at this point, the only thing coming after Orin is a possible process node change for those chips (which wouldn't provide enough of a performance boost to matter that much) and the next generation Atlan chips, but those aren't going to be available until 2025 at the earliest according to Nvidia (and who knows what's going to happen now thanks to chip shortages).

So it's gonna be a Dane chip as we've already discovered or an Atlan variation for 2025/2026 release, take your pick.

 

[Skittzo]

We knew about Aula before that.

Define "we". As I understand it it only became pubic knowledge in April 2020 but I guess there may have been insider discussions about it beforehand?

 

[fwd-bwd]

Sony filed a patent for a collapsable control stick:

This obviously is intended for a pocketable device. But is it for a future Xperia gaming phone, or a revival of PSP/Vita?

Edit: See @Dakhil's reply below. I forgot about that one.

 

[Dakhil]

Sony filed a patent for a collapsable control stick:

This obviously is intended for a pocketable device. But is it for a future Xperia gaming phone, or a revival of PSP/Vita?

I think the collapsible analog stick patent's probably related to Sony's fairly recent patent on a mobile controller for smartphones.