• 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!)

So now we have to wait to see if a Slim gets announced this year , to validate this guys claims, but overall when it comes to the next Switch if we dont hear nothing by the end of September, thats when we go into panic mode
 
So now we have to wait to see if a Slim gets announced this year , to validate this guys claims, but overall when it comes to the next Switch if we dont hear nothing by the end of September, thats when we go into panic mode

It is interesting that quite a few of us are now allowing up to September before giving up for 2022.

Could they really pull of September announcement and October or November release? It feels like the announcement itself would be entirely out of their control - no way it's not largely leaked that close to launch.
 
I think if the announcement is September, it's a 2023 product. July-August is the window and seems like they used it up for the Splatoon OLED model.
I could be wrong of course, just my gut feeling.

As i've said before, I think/feel/believe this is a H2 2023 product, but willing to believe in earlier or later dates as information filters in.
 
Didn't the Mariko switch also get slightly faster Ram? It seemed like a decent upgrade all things considered.

Yeah it went from LPDDR4 to LPDDR4X IIRC.

36% more memory bandwidth.

Again, a die shrink incurs a higher clock frequency.


There is a reason for this, going below that is out of one’s own volition.

This is very complicated but in essence when you shrink a chip from one node to a better node what happens is that you increase the density of transistors that are available (not necessarily the amount), and as a result you increase the frequency of the 1 to0 and 0 to1 switching that occurs on the chip itself.
I'm 99% sure this was only for power savings. The memory clock is part of the performance profile system the Switch uses, and there's nothing beyond 1600 MHz in the firmware.
 
The reporting was about a new, cheaper model of Switch releasing in 2019 to act as a successor to the 3DS's market. I don't see how Mochizuki could interpret a supplier telling him 'They're buying STM32H7 chips that are no more powerful than the OG 3DS CPU and have no hardware accelerated graphics*, and obviously are incompatible with Switch software,'
I'm at: 'is this much and this Specific information what a supplier would tell him'. We know what it is now, but that does make it easy to inject knowledge and Information that may not have been there at the time, as well as project subject matter enthusiasm the source may not share. The supplier may just be there to be Sam the reliable supply man, who is kind enough to share, but doesn't care much if the products Nintendo or furby.

I feel like if he received such an exhaustive specifications list like this we would get much more specific and detailed leaks.

I feel it's more along the lines of 'oh there's also this product for a less powerful casual nostalgia system more like the old handhelds.

as 'Budget Switch coming later this year.' It's far more likely Mochizuki heard not just about the die shrink, but also about the smaller battery, or smaller screen, or the non-detachable controller and smaller shell, et cetera.

He would still know it's a die shrink that can clock higher while using less power.

Nintendo probably chose that particular chip for the G&W SMB because it was the cheapest**, most easily sourced microcontroller that could handle emulating the NES without even a hint of slowdown. I'm sure they could have managed with an M0+ based chip, based on posts online about people emulating NES on RP Pico, but going with an M7 probably saved more time (and thus money) not having to optimize the software as much as a weaker processor might require, than they would have saved on a slightly cheaper microcontroller. Remember, it was a limited edition release, so saving pennies each unit isn't as important as saving work-weeks of expensive engineering efforts.

I have no doubt at all that's why nintendo chose that chip. I agree with this 100%. I honestly don't think it's possible to get a cost effective chip that would not be overkill for what the new game and watch was, even if it's the cheapest in existence. Which I feel could easily be a source for potential misinterpretation of intent.

*No hardware accelerated graphics in the sense we tend to think of when discussing video game platforms. It does have a display controller of course, and a "graphic oriented memory management unit", as well as a JPEG codec, the latter of which could technically count as it is hardware accelerated and does relate to graphics, but unless they designed the emulator to render the framebuffer as a JPEG file for some god-forsaken reason I doubt the codec was used for anything. I'd guess it's mostly the same NES emulator as they used for the 3DS virtual console (since they're both 32-bit ARM systems, granted one is ARMv6 and the other ARMv7), but that's baseless speculation on my part.

**I couldn't find exact prices, but the STM32H7B0 used in the Game & Watch Super Mario Bros is listed as the most power efficient option in the 'value' section of the line-up, as well as the only 'value' option with the graphics optimized MMU.

Sorry for having such a long response to your rather innocuous theory, I just love geeking over Cortex M and this thread has never given me the chance to do so until now. So. . . Thanks!

Ha ha ha, no problem, glad I could help you geek out lol.
 
I think if the announcement is September, it's a 2023 product. July-August is the window and seems like they used it up for the Splatoon OLED model.
I could be wrong of course, just my gut feeling.

As i've said before, I think/feel/believe this is a H2 2023 product, but willing to believe in earlier or later dates as information filters in.

Yeah, as short as marketing windows have become, it's hard to see a September announcement turn into an October/November release, especially if it really is basically a Switch successor. I could see Q1 2023 (calendar year) from a September announcement, though.
 
Noice, is the source a Funcle or a Muncle, and is one more reliable than the other?
Also, is this the first leak of a PS5 Slim? I am sure everyone assumed it would happen eventually, but this seems a little early in the PS5 lifecycle for one.
Yea, feels a bit early to me, just in terms of potential power savings.
For products that can launch this year, the best that can happen for a PS5 slim, node-wise, is going from N7 to N5. TSMC advertises about -30% power draw. I do/did not think that is enough for a 'Slim'.
My expectation for a Slim was N3 (going from N7 to N3 would've cut power draw by like half, or close enough), and that starts volume production by the end of this year. So I thought a Slim wasn't really feasible to launch until next year at the earliest.
 
I think if the announcement is September, it's a 2023 product. July-August is the window and seems like they used it up for the Splatoon OLED model.
I could be wrong of course, just my gut feeling.

As i've said before, I think/feel/believe this is a H2 2023 product, but willing to believe in earlier or later dates as information filters in.
It's possible for it to be that late, but earlier is more likely if devkits were sent at the end of 2020. Nearing 3 full years with third parties having kits in hand is extreme.
 
I'm 99% sure this was only for power savings. The memory clock is part of the performance profile system the Switch uses, and there's nothing beyond 1600 MHz in the firmware.
Also, since plain LPDDR4 has been phased out of catalogues by sometime this year (when that happened, I don't know, my wayback machine-fu isn't up to it), there's probably also a long term sourcing aspect to it.
 
Yea, feels a bit early to me, just in terms of potential power savings.
For products that can launch this year, the best that can happen for a PS5 slim, node-wise, is going from N7 to N5. TSMC advertises about -30% power draw. I do/did not think that is enough for a 'Slim'.
My expectation for a Slim was N3 (going from N7 to N3 would've cut power draw by like half, or close enough), and that starts volume production by the end of this year. So I thought a Slim wasn't really feasible to launch until next year at the earliest.
Interesting point on the respective power savings.
Maybe the "Slim" is meant more for Sony's production managers, as in slim down that stress from sourcing, manufacturing and scalping :p
 
0
Also, since plain LPDDR4 has been phased out of catalogues by sometime this year (when that happened, I don't know, my wayback machine-fu isn't up to it), there's probably also a long term sourcing aspect to it.
Yeah. It was probably a no-brainer in terms of cost even back then.

But I don't think there's a single (non-homebrewed) case where Mariko-based Switches outperform the original Erista model. The clocks are set identically.
 
Yeah. It was probably a no-brainer in terms of cost even back then.

But I don't think there's a single (non-homebrewed) case where Mariko-based Switches outperform the original Erista model. The clocks are set identically.

Definitely, Nintendo was very specific about the parity.
 
0
It is interesting that quite a few of us are now allowing up to September before giving up for 2022.

Could they really pull of September announcement and October or November release? It feels like the announcement itself would be entirely out of their control - no way it's not largely leaked that close to launch.
It would definitely leak, but they seemingly didn't care with the Lite or OLED either so who knows. They could certainly swing it if they wanted to drop this thing heading into Thanksgiving/Black Friday week.
 
0
That’s not the only game, I remember always seeing a strategy game on the eShop and it was “TBA”, did that ever get released?

It was King’s something
King's Bounty II? It got released last August. I was interested until I saw the reviews.
* Hidden text: cannot be quoted. *
Thanks for sharing this. I was not able to view that screenshot. The info regarding Switch is as vague as it gets. As for the new PS5—this is only my opinion—it's probably less of a "Slim" but more of a cheaper-and-easier-to-make revision.
I think you're both agreeing on the point here, but just to be sure we're all on the same page, "eINK model" was just a joke about how the new hardware could be referring to anything, not that eINK is something that's actually happening for Switch.
🧐
 
Thanks for sharing this. I was not able to view that screenshot. The info regarding Switch is as vague as it gets. As for the new PS5—this is only my opinion—it's probably less of a "Slim" but more of a cheaper-and-easier-to-make revision.

This makes a lot more sense right now. For a product in very high demand, redesigning the entire package (and I assume lowering the price) feels like overkill right now.

Shame
 
0
Could they really pull of September announcement and October or November release? It feels like the announcement itself would be entirely out of their control - no way it's not largely leaked that close to launch.
A September announcement indeed makes no sense from multiple perspectives.
It is pure copium. Either we see it by august or it's not happening this year.
 
Thanks for sharing this. I was not able to view that screenshot. The info regarding Switch is as vague as it gets. As for the new PS5—this is only my opinion—it's probably less of a "Slim" but more of a cheaper-and-easier-to-make revision.
A PS5 Series S equivalent makes a lot of sense, considering the environment.
Could they really pull of September announcement and October or November release? It feels like the announcement itself would be entirely out of their control - no way it's not largely leaked that close to launch.
Why would they care about the leak? In this context, the “leak” would hit the enthusiast media who already own switches. A leak would be unlikely to damage sales. Nintendo would want to control the rollout, but a leak of the Lite level would likely have no negative impact.

I think September is too late because I think they’d want more chances to beat the drum and make sure demand is solid and the pitch on the new system is clear.
 
Thanks for sharing this. I was not able to view that screenshot. The info regarding Switch is as vague as it gets. As for the new PS5—this is only my opinion—it's probably less of a "Slim" but more of a cheaper-and-easier-to-make revision.
Yeah, I checked his historical posts and found that he claimed that "no new molds have been received so it may be an internal upgrade" back to March 2022. Things might change though.
 
A PS5 Series S equivalent makes a lot of sense, considering the environment.

I didn't think that's what was being implied. I thought it would be an internal revision for Sony's benefits, but nothing passed down the consumers.
 
Didn't the Mariko switch also get slightly faster Ram? It seemed like a decent upgrade all things considered.
But they downclocked it so it matched the og, right? So there was not much of a change. Drake will probably use LPDDR5 because LPDDR4X is in it's way out.
 
0
It's possible for it to be that late, but earlier is more likely if devkits were sent at the end of 2020. Nearing 3 full years with third parties having kits in hand is extreme.
It still remains possible that drake was delayed due to the pandemic effecting both demand for the switch (keeping its sales consistent for longer) as well as making it much harder to supply.
 
Regardless of your thoughts on any particular word, responding to someone politely explaining why they do not like it with "just a word" is needlessly dismissive. - hologram, Aurc, paranoodle, Donnie, BozPaggs
It still remains possible that drake was delayed due to the pandemic effecting both demand for the switch (keeping its sales consistent for longer) as well as making it much harder to supply.

I don’t think developers that have games ready for Drake is going to like Nintendo delaying the hardware just because Switch is still selling.
 
I think if the announcement is September, it's a 2023 product. July-August is the window and seems like they used it up for the Splatoon OLED model.
I could be wrong of course, just my gut feeling.

As i've said before, I think/feel/believe this is a H2 2023 product, but willing to believe in earlier or later dates as information filters in.
This is just a reasonable guess, but my guess is mid-September announcement so that Nintendo can relax some NDAs just prior to Tokyo Game Show, give exhibitors a chance to potentially talk about or show off what's coming. It's basically the last trade event until March 2023, I think(?), so it seems like a safe bet that this is what will happen if it's going to happen before 2023.
 
This is just a reasonable guess, but my guess is mid-September announcement so that Nintendo can relax some NDAs just prior to Tokyo Game Show, give exhibitors a chance to potentially talk about or show off what's coming. It's basically the last trade event until March 2023, I think(?), so it seems like a safe bet that this is what will happen if it's going to happen before 2023.
I honestly still think next week is fairly likely. It's a very conspicuous gap in the Nintendo calendar and the only real chance for Nintendo news that isn't Splatoon or Xenoblade until TGS. I would hardly believe they'd go radio silent for a week between Splatoon 2 tournaments and August will surely have a Splatoon 3 Direct, Testfire, then there's Gamescom.

All this plus the fact they skipped out on having a summer event for first party games and the amount of people saying Nintendo is having something in July, such as Jeff Grubb, who also believes this console will come out this calendar, or at least fiscal, year.

If they go from here to Splatoon 3's launch with basically nothing revealed in time for the holidays other than Pokémon, that would be very, very strange.

And if it's not a direct... Then I have to imagine it's a showcase. A showcase of what? I'd have to say, probably Drake.
 
Last edited:
Skipping a weekend Splatoon 2 tournament because they are announcing some far off hardware earlier in the week is not an idea I can take seriously, sorry. They even just announced hardware specifically for that particular audience last week.
If there is a reason, it will be a Splatoon related reason, whether that's a public testfire or something behind the scenes like server overhauling or whatever.
I guess, at a stretch, it could just be because everyone is playing Mario Kart that weekend!
 
Last edited:
Mariko is exactly more performant in game performance. By 25-30%.

It can and does solve a LOT of gpu/cpu performance issues making for play with stable frame rates remaining on the upper bounds of dynamic resolution. If you have a hard modded Mariko switch (PSA the team xecutor dude who cracked this one, got gat, by the ninteninjas)

He accurately stated exactly what it was, based on the specifications for what its normal operational range of use was. That Nintendo decided to instead underpower it to use it for more battery life, rather than a smoother higher res switch experience, is not something that can be predicted looking at a part.
That percentage is in docked mode. Its significantly more in handheld mode isn't it? If you clock it to the battery life of the original.
 
It's like saying the PS4 Slim and Xbox One S were designed to be more powerfull.

1. as another user pointed out, One S is faster than the original.

2, In addition to the die shrink, Mariko also upgraded its memory. Switches bandwidth bottleneck, could pretty much disappear in an instant if that's what Nintendo wanted.

So imo it does seem designed to handle higher clocks, and address the flaws of the original.
 
Last edited:
@Hermii I would be surprised if they start letting you run games at 60fps handheld only and 30fps on TV, but they did it the other way around (Bowser's Fury) so, maybe.
If they reworked X1 to be able to deal with LPDDR5 then it would be an all round good upgrade, but i guess they are skipping that idea.
 
Last edited:
That percentage is in docked mode. Its significantly more in handheld mode isn't it? If you clock it to the battery life of the original.

Yeah i stated docked, it's what I have memorized off the top of my head. You get more efficiency per watt at lower clocks, but I don't remember the curve.

It's definitely true the vastly most performant homebrew profile is setting docked clocks, or normal x1 operating range clocks while running in the handheld mode.
 
0
It really depends on what goal Nintendo has with this new hardware.

The goal seems to be offering a model that keeps software engagement high during the latter half of the Switch lifespan. Also, a stepping stone for Nintendo to start figuring out DLSS and ray tracing development.

I feel like it is a bit different this time. While the Lite or the OLED had mainly new buyers in mind, the Drake Switch would likely also be attractive to people who already own the switch and want more power. This is similar to the PS4 Pro, but it is possible that the interest in Drake is much higher.

Hm, I bet most OLED buyers as of now were owners of previous models.

Drake might sell more than the ps4 pro did, but that’s irrelevant. The point was to suggest that Nintendo only expects Drake to be a fraction of the ecosystem sales at the end of the day. Maybe not ~25% like the ps4 pro, but certainly less than 50%

Also with the investment in this new potential processor and specs Nintendo for sure wants to sell it to many people as possible.

I’ve argued this before, but the investment isn’t about hardware sales, it’s about software. The investment costs will pay dividends by:

1) selling wayyy more Switch software the next 4 years to enthusiasts who might have lost interest in aging hardware

And

2) allowing Nintendo to explore DLSS/tensor core/RT gaming development early and at a leisurely pace.

They often lament about entering HD development too late. Waiting till Wii engagement/appeal was dying, learning along a launch of a next gen console that they couldn’t support well.

The investment in Drake will be paid back 100 times over not just in software sales, but in being able to dabble in future, new development techniques while still also being able to fully support the current large userbase of the majority of platform holders (hybrid/Lite/OLED users)

Maybe they will really position it as iterative model with a few exclusives.

This would be much easier with an exciting new software title. It dosen‘t have to be BOTW 2 of course, but that title together on a new console would give it the right initial boost.

A 4K/60fps gaming Switch would sell out for at least a year without any software being tied to it. Even at $500.
 
1. as another user pointed out, Series S is faster than the original.

2, In addition to the die shrink, Mark also upgraded its memory. Switches bandwidth bottleneck, could pretty much disappear in an instant if that's what Nintendo wanted.

So imo it does seem designed to handle higher clocks, and address the flaws of the original.
One S not Series S. ;P
And Mariko.
 
Kinda curious to see if they can make the PS5 looks even uglier.
 
0
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