• Hey everyone, staff have documented a list of banned content and subject matter that we feel are not consistent with site values, and don't make sense to host discussion of on Famiboards. This list (and the relevant reasoning per item) is viewable here.
  • Do you have audio editing experience and want to help out with the Famiboards Discussion Club Podcast? If so, we're looking for help and would love to have you on the team! Just let us know in the Podcast Thread if you are interested!

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

The DLSS implementation can be customized with Switch 2 since Nvidia knows exactly what the hardware and can inject it directly into the API. Any game that is developed using the NVN2 API will likely have DLSS applied automatically. DLSS requires Tensor cores to work but it doesn't seem to require a hefty amount of them seeing as how the RTX3050M has 64 tensor cores and implements DLSS just fine. A customized version of DLSS should be even more efficient.
 
Quoted by: LiC
1
The DLSS implementation can be customized with Switch 2 since Nvidia knows exactly what the hardware and can inject it directly into the API. Any game that is developed using the NVN2 API will likely have DLSS applied automatically. DLSS requires Tensor cores to work but it doesn't seem to require a hefty amount of them seeing as how the RTX3050M has 64 tensor cores and implements DLSS just fine. A customized version of DLSS should be even more efficient.
It won't be automatic. The data required for DLSS is not something NVN could ever determine; the developer has to specify it, as well as choosing when where and how in their graphics pipeline they want it to take effect.

I also doubt how "customized" the DLSS implementation for Nintendo will be. It will be optimized (by the compiler) for ARM CPUs and Ampere GPUs, but that's no different from what the Windows drivers would already do today, and I don't imagine there's all that much that could be done on the algorithm side to improve it.
 
It won't be automatic. The data required for DLSS is not something NVN could ever determine; the developer has to specify it, as well as choosing when where and how in their graphics pipeline they want it to take effect.

I also doubt how "customized" the DLSS implementation for Nintendo will be. It will be optimized (by the compiler) for ARM CPUs and Ampere GPUs, but that's no different from what the Windows drivers would already do today, and I don't imagine there's all that much that could be done on the algorithm side to improve it.
I wonder how close the in-memory representation can be, between NVN2 and DLSS. That's where I'd expect some small perf/memory increase. And possibly some latency drops from knowing how many tensor cores you've got at compile time, rather than run time.
 
0
It won't be automatic. The data required for DLSS is not something NVN could ever determine; the developer has to specify it, as well as choosing when where and how in their graphics pipeline they want it to take effect.

I also doubt how "customized" the DLSS implementation for Nintendo will be. It will be optimized (by the compiler) for ARM CPUs and Ampere GPUs, but that's no different from what the Windows drivers would already do today, and I don't imagine there's all that much that could be done on the algorithm side to improve it.

I could be wrong, but I'm suggesting that it will be implemented at a low level. When a game on PC utilizes DLSS it has to be done within the actually game software because there is no way of knowing what graphics card will be running the game. So perhaps not automatically, but developers will be able to very easily implement it into the rendering pipeline in Switch 2 because its injected directly into the NVN2 API. So any game that is ported to Switch will naturally come along with the ability to utilize DLSS without much effort.
 
Would it be possible for Nintendo to implement pressure sensitive trigger buttons for the next joycons? I guess analog triggers would be too small, but how expensive (or not) would it be to add them? Would open up new controls and allow for better GC compatibility.
 
0
I could be wrong, but I'm suggesting that it will be implemented at a low level. When a game on PC utilizes DLSS it has to be done within the actually game software because there is no way of knowing what graphics card will be running the game. So perhaps not automatically, but developers will be able to very easily implement it into the rendering pipeline in Switch 2 because its injected directly into the NVN2 API. So any game that is ported to Switch will naturally come along with the ability to utilize DLSS without much effort.
Yeah the API can make it easier for developers but there's no way it can be done at a system level automatically, since DLSS requires specific inputs from the game's specific engine regarding motion vectors. It's not something the API can have access to.
 
Maybe DlSS runs slightly faster on Drake due to being more tightly integrated into the api, but I doubt it’s a game
changer.
 
0
It's a rather fascinating discussion that I have seen it come up, but I think in short that while it would be appreciated, 1) Nintendo does currently sit on the board of the SD Association, and it wouldn't make sense to use a competing storage format, and 2) space constrictions might prove difficult to fit inside the current form factor of the Switch if no changes to the thickness of the console are being made

The part I've bolded seems to come up a lot for some reason, but it's not true. Here's the list of board members of the SD Association, and you'll note that Nintendo aren't on there. They are listed as General Members, which is the basic level of membership required for them to support SD cards, use the logo, etc. As per this page, being a general member they don't even have voting rights within the SD Association.
 
Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
I should address the previous post, I can perhaps see something for 2025/26 as enough for a workable PRO model for PS5 and XSX if they were to do one, and a new generation of systems for 2028/29.


I wonder how fast SSD speeds can get in the next 8-10 years though, that’ll matter for those consoles. Perhaps not a big enough MEM upgrade, but SSD speeds to help with that would be the thing I foresee.


Maybe… 24-32GB of RAM, but SSDs with >10-15GB/s speeds
 
The part I've bolded seems to come up a lot for some reason, but it's not true. Here's the list of board members of the SD Association, and you'll note that Nintendo aren't on there. They are listed as General Members, which is the basic level of membership required for them to support SD cards, use the logo, etc. As per this page, being a general member they don't even have voting rights within the SD Association.
I'm not sure where this meme came from. Nvidia is a the UFS adopter member, roughly the same level, and it has never been brought up in support of UFS as a storage solution
 
UFS card!
1080p OLED Screen!
Performance Mode DLSS!
1Gbit/S or more read speed from internal storage, UFS Card and Game Card slot!
Wide kickstand!
Games "only on Drake" go gold in Q1!
I want to believe!
 
I have a couple of questions:

1. Is it still possible that Drake might use a magnesium alloy shell? I was actually looking into the LG Gram that Dakhil posted about and noticed that it is built with magnesium alloy housing.

2. Does anyone have any idea when the first firmware was flashed onto Switches in their initial shipment?

I'm trying to line up potential firmware datamine dates (5.0.0 had lite, 10.0.0, had OLED/aula/abcd I belive), so I'm trying to see if 15.0.0, the one with the weird internal updates is the same timeframe away from release. Because 5.0.0 and 10.0.0 were both roughly 530 days on average from release. Taking the same general 530 days after Oct 11. 2022 (when 15.0.0 released) would place us at...March 23, 2024. I'm hoping that because lite/Mariko and OLED were just revisions that Nintendo had them in the firmware much earlier.

Was 1.0.0 or whatever the equivalent initial firmware flashed onto Switch systems say...around 153 days or ~5 months before launch- around Oct.1st of 2016?
 
1. Is it still possible that Drake might use a magnesium alloy shell? I was actually looking into the LG Gram that Dakhil posted about and noticed that it is built with magnesium alloy housing.
Definitely a possibility, especially since there's a good amount of metal used on the OLED model.
 
Honestly my expectations for the next system would be to go all-aluminium, rather than magnesium, due to price. We know from OLED Model that using a whole bunch of aluminium doesn't make it unwieldy, and it has a very good strength to weight ratio, something Nintendo engineers claim they value.

Though when I say "all aluminium", I don't mean ALL aluminium, just mostly. OLED Model's only surface without it (discounting Joy-Con) is the upper backplate (and under the kickstand). Doesn't seem like a huge jump to make that aluminium, but: Damage to the plastic Dock, and RF penetration by wireless modules are concerns. So I think it'll maybe use more metal but leave the bottom and top edges plastic.
 
I have a couple of questions:

1. Is it still possible that Drake might use a magnesium alloy shell? I was actually looking into the LG Gram that Dakhil posted about and noticed that it is built with magnesium alloy housing.

2. Does anyone have any idea when the first firmware was flashed onto Switches in their initial shipment?

I'm trying to line up potential firmware datamine dates (5.0.0 had lite, 10.0.0, had OLED/aula/abcd I belive), so I'm trying to see if 15.0.0, the one with the weird internal updates is the same timeframe away from release. Because 5.0.0 and 10.0.0 were both roughly 530 days on average from release. Taking the same general 530 days after Oct 11. 2022 (when 15.0.0 released) would place us at...March 23, 2024. I'm hoping that because lite/Mariko and OLED were just revisions that Nintendo had them in the firmware much earlier.

Was 1.0.0 or whatever the equivalent initial firmware flashed onto Switch systems say...around 153 days or ~5 months before launch- around Oct.1st of 2016?
What makes you think 15.0.0 is the relevant version? Unexplained changes that may or may not be related to unannounced hardware started appearing in 12.0.0 with the addition of stubbed syscalls and the "4kdp_preferred_over_usb30" setting.

And that's a pretty tenuous pattern to begin with, especially considering that Lite and OLED were mostly just new parameters/profiles within the existing OS, whereas the new hardware is likely to be a different branch/build of the OS entirely.

* Hidden text: cannot be quoted. *
Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
Honestly my expectations for the next system would be to go all-aluminium, rather than magnesium, due to price. We know from OLED Model that using a whole bunch of aluminium doesn't make it unwieldy, and it has a very good strength to weight ratio, something Nintendo engineers claim they value.

Though when I say "all aluminium", I don't mean ALL aluminium, just mostly. OLED Model's only surface without it (discounting Joy-Con) is the upper backplate (and under the kickstand). Doesn't seem like a huge jump to make that aluminium, but: Damage to the plastic Dock, and RF penetration by wireless modules are concerns. So I think it'll maybe use more metal but leave the bottom and top edges plastic.
I'm going to quote my post from last year on the problems with going "all aluminium".
Unfortunately, one caveat with having a complete aluminium or magnesium housing is that metals in general reflect and absorb radio waves, which can interfere with wireless connections (Wi-Fi, Bluetooth), resulting with poor wireless connections. And that can be very problematic for devices with poor antennae placement, such as the Nintendo Switch, HTC 10, etc.

Of course, Nintendo does seem to take the Nintendo Switch's issue with poor Wi-Fi reception seriously, considering Nintendo replaced the PCB antenna used on the Nintendo Switch with a LCD antenna on the OLED model. But in order to maintain reasonably consistent wireless connections on the OLED model, Nintendo opted to have most of the back of the OLED model use plastic instead of metal since the wireless antennae are located in the back of the OLED model.

Based on my personal experience, the Nintendo Switch can be very warm to the touch when taking the Nintendo Switch out of the dock after long sessions on TV mode. And trying to remove a device with a fully metal housing from a dock after long sessions on TV mode can be very difficult since it can be too hot to touch, especially with metals generally being very good heat conductors, which is very problematic, unless Nintendo plans on installing fans on the DLSS model*'s dock.
 
What makes you think 15.0.0 is the relevant version? Unexplained changes that may or may not be related to unannounced hardware started appearing in 12.0.0 with the addition of stubbed syscalls and the "4kdp_preferred_over_usb30" setting.

And that's a pretty tenuous pattern to begin with, especially considering that Lite and OLED were mostly just new parameters/profiles within the existing OS, whereas the new hardware is likely to be a different branch/build of the OS entirely.
Oh, was it 12.0.0 where the changes started? Don't mind me then; I'm just losing myself in all these dates I've been comparing. Guess I saw 5.0.0 (lite) > 10.0.0 (OLED) > 15.0.0 Drake? as some kind of pattern or thought of it as more than a coincidence. I'm def just looking into the firmware changes too much.

If anyone wants a peek at the date...uh data I've been collecting and comparing in an effort to line up a common trend for Nintendo HW releases, here's a look don't laugh

DRAKE launch lead up

MZ = Mochizuki

-2020-

991-779 days: Aug 24-Mar 23: first Bloomberg article referring to 4K, Drake concepting?, 4K ready > chip

-2021-

800 days- Mar 3: imran khan “dev kits are out there”
780 days- Mar 23: MZ “New nvidia chip with DLSS”
638 days- Aug 12: Stephen Totilo, RDR2 for Switch + "Pro"
589 days- Sep 30: MZ “11 devs and spices” + timing (377 DAYS TO LATE 2022)

***504 days:- Dec 24: dev kits en masse?

-2022-

534 days?
543 days: Nov 15
439 days - Mar 1: NVN2 LEAK
?410 days- Mar 28:
?400 days- Apr 7:
?286 days- July 30: ???
220 days- Oct 4: concept? leak (tape out, hw finalized prior- 9/20 linux kernel update T239)
214 days- firmware 15.0.0 ??? datamine
[557-543-492 days] = Feb 14, 2024 * Apr 5, 2024 * Apr 19, 2024 | Avg. 530 days = Mar 23, 2024
? 179 days- Nov 14: MZ timing
? 159 days- Dec 7 - lite backplate leak (possible hw prototype leak?)

? 134 days- Dec 29: general specs leak, MZ timing, etc. (around manufacture start)

2023-

? 114 days- Jan 18: MZ timing? (X1+ chip for v2)
74 days- Feb 27: clock speeds leak?
xx days- Feb 7 - Mar 30: ANNOUNCEMENT/REVEAL/DETAILS (Mar 22?)
0 days- May 12: LAUNCH?

SWITCH launch lead up

-2014-

Jun 16 + ~6 mo: NX starts dev

-2015-

638 days- Jun 4: rdr2 equiv.? ⬇️
584 days- July 28: DQ XI leaked
504 days- Oct 16: dev kits go out en masse via MZ WSJ

-2016-

400 days- Jan 28: MZ “likely to release this year” screen part orders
220 days- July 26: EG NX concept leak (tape out?, hw finalized prior)

134 days- Oct 20: TEASER, general specs leak, 1 week before financial results?

74 days- Dec 19: clock speeds leak (Switch lite announce)

-2017-

50 days- Jan 12: FULL REVEAL PRESENTATION
0 days- Mar 3: LAUNCH

-2018- Switch lite (Sep 20, 2019) + v2 (July 17, 2019)

557-492 days (525 avg)- Mar 12: firmware 5.0.0 datamine
351-286 days - Oct 4: MZ “new revision planned”

-2019- Switch lite (Sep 20, 2019) + v2 (July 17, 2019)

156 days- Apr 17: lite backplate leak
179-114 days- Mar 25: MZ “2 new models this summer” Bloomberg

-2020- OLED

543 days- Apr 13: firmware 10.0.0 datamine
410 days- Aug 24>Sep 8: MZ “upgraded console 2021” & 4K ready? Bloomberg

-2021- OLED

275 days- Jan 6: aula?
218 days- Mar 3: “bigger OLED screen model” MZ Bloomber
 
Last edited:
Oh, was it 12.0.0 where the changes started? Don't mind me then; I'm just losing myself in all these dates I've been comparing. Guess I saw 5.0.0 (lite) > 10.0.0 (OLED) > 15.0.0 Drake? as some kind of pattern or thought of it as more than a coincidence. I'm def just looking into the firmware changes too much.

If anyone wants a peek at the date...uh data I've been collecting and comparing in an effort to line up a common trend for Nintendo HW releases, here's a look don't laugh

DRAKE launch lead up


-2020-

991-779 days: Aug 24-Mar 23: Drake concepting, 4K ready > chip

-2021-

800 days- Mar 3: imran khan “dev kits are out there”
780 days- Mar 23: MZ “New nvidia chip with DLSS”
638 days- Aug 12: RDR2
589 days- Sep 30: MN1 “1 devs and spices” + timing (377 DAYS TO LATE 2022)
***504 days:- Dec 24: dev kits en masse?

-2022-

534 days?
543 days: Nov 15
439 days - Mar 1: NVN2 LEAK
?410 days- Mar 28:
?400 days- Apr 7: __”likely to release this year” x part orders
?286 days- July 30: ???
220 days- Oct 4: concept? leak (tape out, hw finalized prior- 9/20 linux kernel update T239)
214 days- firmware 15.0.0 ??? datamine
[557-543-492 days] = Feb 14, 2024 * Apr 5, 2024 * Apr 19, 2024 | Avg. 530 days = Mar 23, 2024
? 179 days- Nov 14: MZ timing
? 159 days- Dec 7 - lite backplate leak (possible hw prototype leak?)
? 134 days- Dec 29: general specs leak, MZ timing, etc. (around manufacture start)
MZ latest? article: dec 17-25:> 26

2023-

? 114 days- Jan 18: MZ timing? (X1+ chip for v2)
74 days- Feb 27: clock speeds leak
xx days- Feb 7 - Mar 30: ANNOUNCEMENT/REVEAL/DETAILS (Mar 22?)
0 days- May 12: LAUNCH*

SWITCH launch lead up

-2014-

Jun 16 + ~6 mo: NX starts dev

-2015-

638 days- Jun 4: rdr2 equiv.? ⬇️
584 days- July 28: DQ XI leaked
504 days- Oct 16: dev kits go out en masse via MZ WSJ

-2016-

400 days- Jan 28: MZ “likely to release this year” screen part orders
220 days- July 26: EG NX concept leak (tape out?, hw finalized prior)

134 days- Oct 20: TEASER, general specs leak, 1 week before financial results?

74 days- Dec 19: clock speeds leak (Switch lite announce)

-2017-

50 days- Jan 12: FULL REVEAL PRESENTATION
0 days- Mar 3: LAUNCH

-2018- Switch lite (Sep 20, 2019) + v2 (July 17, 2019)

557-492 days (525 avg)- Mar 12: firmware 5.0.0 datamine
351-286 days - Oct 4: MZ “new revision planned”

-2019- Switch lite (Sep 20, 2019) + v2 (July 17, 2019)

156 days- Apr 17: lite backplate leak
179-114 days- Mar 25: MZ “2 new models this summer” Bloomberg

-2020- OLED

543 days- Apr 13: firmware 10.0.0 datamine
410 days- Aug 24>Sep 8: MZ “upgraded console 2021” & 4K ready? Bloomberg

-2021- OLED

275 days- Jan 6: aula?
218 days- Mar 3: “bigger OLED screen model” MZ Bloomber

Your idea actually isn't that wild. 15.0.0 is a very weird release, and there was some speculation that it might be a version designed to get flashed to units in manufacturing. But like @LiC says, what goes onto the hardware will be a different build, just due to the nature of the hardware.
 
I have a couple of questions:

1. Is it still possible that Drake might use a magnesium alloy shell? I was actually looking into the LG Gram that Dakhil posted about and noticed that it is built with magnesium alloy housing.

2. Does anyone have any idea when the first firmware was flashed onto Switches in their initial shipment?

I'm trying to line up potential firmware datamine dates (5.0.0 had lite, 10.0.0, had OLED/aula/abcd I belive), so I'm trying to see if 15.0.0, the one with the weird internal updates is the same timeframe away from release. Because 5.0.0 and 10.0.0 were both roughly 530 days on average from release. Taking the same general 530 days after Oct 11. 2022 (when 15.0.0 released) would place us at...March 23, 2024. I'm hoping that because lite/Mariko and OLED were just revisions that Nintendo had them in the firmware much earlier.

Was 1.0.0 or whatever the equivalent initial firmware flashed onto Switch systems say...around 153 days or ~5 months before launch- around Oct.1st of 2016?
The "hardware config gets datamined" event historically happens well before the firmware is finalized, but it's also just not really a necessary step of the process, and hardware can launch without it ever happening.

If Drake launches in spring, as expected, it will probably have some 15.x version (not necessarily one existing Switches got) on it. In practice, the lead time seems to be around 6 months (give or take).
* Hidden text: cannot be quoted. *
Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
Your idea actually isn't that wild. 15.0.0 is a very weird release, and there was some speculation that it might be a version designed to get flashed to units in manufacturing. But like @LiC says, what goes onto the hardware will be a different build, just due to the nature of the hardware.
Thanks for the bit of validation. All the dates I've been comparing assume, of course, that Drake launches May 12, 2023. Do you happen to know when 1.0.0 was flashed onto launch Switch systems or when that first would have happened relative to initial manufacturing?
The "hardware config gets datamined" event historically happens well before the firmware is finalized, but it's also just not really a necessary step of the process, and hardware can launch without it ever happening.

If Drake launches in spring, as expected, it will probably have some 15.x version (not necessarily one existing Switches got) on it. In practice, the lead time seems to be around 6 months (give or take).
Do you think that lite/v2 (Mariko) and OLED (aula) being more simple revisions would explain why they were added to the firmware roughly 1.5 years (530 days) before launch? And that the original Switch and Drake would have closer gaps? Because 530 days before May 12 2023 would put us at around...Nov 28, 2021. And I don't think there were weird firmware changes around then or before?

Buuuut LiC is saying there were some changes going back to 12.0.0 with 4K stuff, which launched in April 5, 2021. That would be a 237 day difference, which if you assume was the timed "delayed" from late 2022 with Zelda to May 12, 2023, and if you go backwards from the current TotK release would place you around Sep 16, 2022. Now could Drake and Zelda have launched in late September/October of this year? Potentially. It would explain the weird Splatoon 3 /XBC3 date swap. Maybe without Zelda in October, they wanted a big seller around that time? I feel like Pokemon was always going to launch Nov 2022, so where else would Zelda have fit if XBC3 was initially intended for September?

I don't know. I'm just trying to make potential sense of firmware update shenanigans and it could be that I'm making mountains out of molehills.

tldr: I might be crazy and I might just be connecting dots (with regards to firmware updates) that don't apply the same way to brand, spanking, new hardware, but I'm stuck at home for 2 weeks recovering from a surgery and I've got a lot of free time on my hands...

Expecting Drake on May 12, 2023, but hoping it hasn't been pushed a year. In terms of Mochizuki & manufacturing timing leaks, going off of his past articles relative to launches...we've got a general window of late december to mid/late January for some kind of "imminent" article or rumblings to start before I would personally say to hang up the towel on May 12.
 
In short:
  • The CPU is always 1020 MHz.*
  • The GPU can be 307 MHz, 384 MHz, or 460 MHz in handheld/docked, or 768 MHz in docked only.
  • The RAM is always 1331 MHz in handheld, or 1600 MHz in docked.
* Unless it's overclocked in the "loading boost" mode, in which case it's 1785 MHz and the GPU is always 76 MHz.

Other combinations in that table, like the ones with the CPU at 1224 MHz are devkit only or no longer exist.
 
Last edited:
Thanks for the bit of validation. All the dates I've been comparing assume, of course, that Drake launches May 12, 2023. Do you happen to know when 1.0.0 was flashed onto launch Switch systems or when that first would have happened relative to initial manufacturing?

Do you think that lite/v2 (Mariko) and OLED (aula) being more simple revisions would explain why they were added to the firmware roughly 1.5 years (530 days) before launch? And that the original Switch and Drake would have closer gaps? Because 530 days before May 12 2023 would put us at around...Nov 28, 2021. And I don't think there were weird firmware changes around then or before?

Buuuut LiC is saying there were some changes going back to 12.0.0 with 4K stuff, which launched in April 5, 2021. That would be a 237 day difference, which if you assume was the timed "delayed" from late 2022 with Zelda to May 12, 2023, and if you go backwards from the current TotK release would place you around Sep 16, 2022. Now could Drake and Zelda have launched in late September/October of this year? Potentially. It would explain the weird Splatoon 3 /XBC3 date swap. Maybe without Zelda in October, they wanted a big seller around that time? I feel like Pokemon was always going to launch Nov 2022, so where else would Zelda have fit if XBC3 was initially intended for September?

I don't know. I'm just trying to make potential sense of firmware update shenanigans and it could be that I'm making mountains out of molehills.

tldr: I might be crazy and I might just be connecting dots (with regards to firmware updates) that don't apply the same way to brand, spanking, new hardware, but I'm stuck at home for 2 weeks recovering from a surgery and I've got a lot of free time on my hands...

Expecting Drake on May 12, 2023, but hoping it hasn't been pushed a year. In terms of Mochizuki & manufacturing timing leaks, going off of his past articles relative to launches...we've got a general window of late december to mid/late January for some kind of "imminent" article or rumblings to start before I would personally say to hang up the towel on May 12.
I don't think there's any particular significance to when the first firmware evidence of something shows up. The only really significant date is when the firmware starts getting flashed for manufacturing, but that's generally something we can't tell for sure until after the fact.
 
Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
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