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

StarTopic Future Nintendo Hardware & Technology Speculation & Discussion |ST| (New Staff Post, Please read)

I think that the big takeaway for this thread is is their observation of the performance of the Asus ROG, considering the very high power consumption despite advanced node manufacturing. Even in 15w APU mode, the system gets into the high-20s watts of consumptions .

Then he cautions people on assuming anything close to a series S for the next Switch given that it will consume much less power

In before Arm vs x86
In before DLSS
And NVMe. DF has been consistent talking about Series S's power in terms of its overall architecture (see Ollie's PS4 Pro/Series S comparison) while the fan community tends to boil it down to teraflops.

I don't think 4TF is on the table for REDACTED's GPU, but I think even the most optimistic here about REDACTED's Docked performance would recognize that Series S's memory bandwidth, CPU clock speeds, and storage speed are off the table. A lot of the technical going around in circles is arguing which of these is the one that it most likely to be the prime bottleneck, and which Nintendo should blow whatever spare dollars/watts it has on.

"Current gen architecture, last gen performance" is unlikely to steer you wrong.
 
0
Switch U will be announced in January 2024 and will launch on May 10, 2024. The biggest technical achievement we'll see will be Monolith Soft's action game which will be a launch title for the Switch U.
I don't think they're going to announce right after the Holidays, when people will have gotten their new Switches.
 
What kind of gigantic fuck up would require Nintendo to pull the SDK from all parties?

I have no idea how in this day and age developers just sat there, as Nintendo claimed back the devkits on which people had spent X months developing/porting games. This is beyond me.
Doesn't need to be a fuck up.

Better tech or things along those lines present themselves. It's happened before with other hardware and their devkits -- better technology became a viable option, the other kits were recalled, and the original plans were cancelled/delayed in favor of the better, more attractive tech & options.
 
Doesn't need to be a fuck up.

Better tech or things along those lines present themselves. It's happened before with other hardware and their devkits -- better technology became a viable option, the early kits were recalled, and plans the original plans were cancelled/delayed in favor of the better, more attractive tech & options.

Obviously time for UFS 4.0

(UFS 4.0 consuming so little power at 1 GB/s does make it an intriguing option if Nintendo is concerned about supporting Unreal 5)
 
I don't think they're going to announce right after the Holidays, when people will have gotten their new Switches.
For the Holidays there will only be DLCs and third party games. Now with the launch of TotK is when more consoles will be sold. And if not in January then it will be in February, but the best month for a console announcement is January.
 
Quoted by: D36
1
For the Holidays there will only be DLCs and third party games. Now with the launch of TotK is when more consoles will be sold. And if not in January then it will be in February, but the best month for a console announcement is January.
Nintendo is not going to move 15 million Switches with just DLC, third-parties, and announcing their successor immediately after their most crucial sales period. I don't understand how you came to that conclusion.
 
Doesn't need to be a fuck up.

Better tech or things along those lines present themselves. It's happened before with other hardware and their devkits -- better technology became a viable option, the other kits were recalled, and the original plans were cancelled/delayed in favor of the better, more attractive tech & options.
Do you know what the motivation was in this particular case?
 
Nintendo is not going to move 15 million Switches with just DLC, third-parties, and announcing their successor immediately after their most crucial sales period. I don't understand how you came to that conclusion.
Watch them do just that.

Coastendo the fence sitter.

Do you know what the motivation was in this particular case?
The gimmick.
 
Nintendo is not going to move 15 million Switches with just DLC, third-parties, and announcing their successor immediately after their most crucial sales period. I don't understand how you came to that conclusion.
“Our goal of selling 15 million unit this fiscal year is a bit of stretch. But we will do our best to bolster demand going into the holiday season so that we can achieve the goal.”
 
“Our goal of selling 15 million unit this fiscal year is a bit of stretch. But we will do our best to bolster demand going into the holiday season so that we can achieve the goal.”
Even if it's a stretch, do you honestly believe that Nintendo will just coast the rest of the year on mostly DLC? That's not how you bolster demand for your current system, especially for the Holidays and Zelda can only carry them so far, when most people who want a Switch already have one.
Watch them do just that.

Coastendo the fence sitter.


The gimmick.
Aight, bruh, you're just trolling at this point 🤣
 
Nintendo is not going to move 15 million Switches with just DLC, third-parties, and announcing their successor immediately after their most crucial sales period. I don't understand how you came to that conclusion.
I know nintendo said they aren't planning to drop the price, but i have to imagine it's their best way to achieve 15 million switches sold this fiscal year.
 
I know nintendo said they aren't planning to drop the price, but i have to imagine it's their best way to achieve 15 million switches sold this fiscal year.
A new Mario will probably be their second best option. By the time that comes out, it'll have been a year since Sparks of Hope; when was the last time we had to wait a whole year for a Mario game?
 
Even if it's a stretch, do you honestly believe that Nintendo will just coast the rest of the year on mostly DLC? That's not how you bolster demand for your current system, especially for the Holidays and Zelda can only carry them so far, when most people who want a Switch already have one.

Aight, bruh, you're just trolling at this point 🤣
Mario Kart, Splatoon and Pokémon DLCs. TotK is the last push.
 
Quoted by: D36
1
Doesn't need to be a fuck up.

Better tech or things along those lines present themselves. It's happened before with other hardware and their devkits -- better technology became a viable option, the other kits were recalled, and the original plans were cancelled/delayed in favor of the better, more attractive tech & options.
Somehow I don't think that's a terribly common thing to happen only a year (or less) out from when the hardware is supposed to release (according to what you were saying), a time when, before the Switch, Nintendo would have already publicly announced it.

There's a fair amount of evidence Nintendo significantly retooled something recently, but there's much stronger evidence that it happened to Switch OLED, with its vestigial 4k support, than anything related to Drake, which we can reasonably guess was at least close to finalized last year.
 
Mario Kart, Splatoon and Pokémon DLCs. TotK is the last push.
That's not gonna yield any good results, even when Nintendo inevitably revises their forecast. I'm trying to understand how you got to this conclusion and why this makes any sense.
 
Aight, bruh, you're just trolling at this point 🤣
I’m really not, having read some parts of the Q&A, Nintendo‘s focus seems more about keeping consumer engagement high especially during the transition period, and if they can get that just by releasing DLC‘s of big game, such as tears of the Kingdom, or the next wave for Mario Kart, or the DLC for Pokémon scarlet and violet, and the DLC for Splatoon 3, or the new additions to the NSO, they can actually just coast all year, even into next year if they need to. Maybe release like a couple extra new games, but they aren’t starved for some things to actually keep product engagement high, they have things.

And they’ve already made a stance about keeping engagement high.

And people still buy software, even old ones. Some games even got a boost like Mario games got a boost from the movie.

And a final farewell with a Metroid Prime 4 or maybe a Mario Odyssey 2 sort of ordeal, for holiday 2024.

Or 2025.


Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
Quoted by: D36
1
I’m really not, having read some parts of the Q&A, Nintendo‘s focus seems more about keeping consumer engagement high especially during the transition period, and if they can get that just by releasing DLC‘s of big game, such as tears of the Kingdom, or the next wave for Mario Kart, or the DLC for Pokémon scarlet and violet, and the DLC for Splatoon 3, or the new additions to the NSO, they can actually just coast all year, even into next year if they need to. Maybe release like a couple extra new games, but they aren’t starved for some things to actually keep product engagement high, they have things.

And they’ve already made a stance about keeping engagement high.

And people still buy software, even old ones. Some games even got a boost like Mario games got a boost from the movie.

And a final farewell with a Metroid Prime 4 or maybe a Mario Odyssey 2 sort of ordeal, for holiday 2024.

Or 2025.


* Hidden text: cannot be quoted. *
I understand that, but nothing drives engagement more than consistent new software. I'm no business man, but having downloadable content be a supplement to full scale releases sounds like a bad business move, even if said DLC is from their evergreen titles. Hell, I think it's gonna take more than a big Mario game to get somewhere even close to where Nintendo wants. If they're making a stance on keeping engagement high, then you need a strong lineup, like this first half of the year, otherwise you worsen an already steady decline in sales.
 
I understand that, but nothing drives engagement more than consistent new software. I'm no business man, but having downloadable content be a supplement to full scale releases sounds like a bad business move, even if said DLC is from their evergreen titles. Hell, I think it's gonna take more than a big Mario game to get somewhere even close to where Nintendo wants. If they're making a stance on keeping engagement high, then you need a strong lineup, like this first half of the year, otherwise you worsen an already steady decline in sales.
Nintendo already admitted that 15M is going to be difficult to achieve, they’ll aim to try to hit it but if anything they’ll simply revise down.

They project 180M software this year which is lowballing it from them. Zelda is clearly not going to do all that or even half of that. It’ll do a solid chunk though.

Evergreens will continue to sell

They’ll continue to do DLC and make them some showpiece because it increases engagement. They’ll throw in first party because that increases engagement. They’ll throw in third party because that increases engagement.
 
Nintendo already admitted that 15M is going to be difficult to achieve, they’ll aim to try to hit it but if anything they’ll simply revise down.

They project 180M software this year which is lowballing it from them. Zelda is clearly not going to do all that or even half of that. It’ll do a solid chunk though.

Evergreens will continue to sell

They’ll continue to do DLC and make them some showpiece because it increases engagement. They’ll throw in first party because that increases engagement. They’ll throw in third party because that increases engagement.
So you agree that Nintendo isn't going to coast the rest of the year, then? Because I agree with the last paragraph; they need more than what was already shown to reach somewhere close to their sales target.
 
That's not gonna yield any good results, even when Nintendo inevitably revises their forecast. I'm trying to understand how you got to this conclusion and why this makes any sense.
Because they will release the Switch U in 1 year. It makes less sense to release new software when the new console is near. I even think they will release MP4 as a launch title for the Switch U.
 
I think that the big takeaway for this thread is is their observation of the performance of the Asus ROG, considering THE very high power consumption despite advanced node manufacturing. Even in 15w AU mode, the system gets into the high-20s.

Then he cautions people on assuming anything close to a series S for the next Switch given that it will consume much less power

In before Arm vs x86
In before DLSS
What is the performance/speed in 15w mode on CPU and GPU clocks?.
4.0 should come out Fall 2024 though.
Got dammit you're with RedDredd Redemption guy 😐
what if the world was made of pudding
Well ya see...
I believe the power consumption of the RAM is determined primarily by how high the RAM frequency is, not the amount of RAM. And the RAM frequency can be tweaked after the SoC's taped out.
You're saying 16GB of RAM running at the same frequency as 12GB of RAM, would have the same power consumption?..

In the case of increasing RAM, would increasing/doubling the bus width (let's say from 64 to 128), require another tape out? Not that it would apply to Switch 2.

Switch U will be announced in January 2024 and will launch on May 10, 2024. The biggest technical achievement we'll see will be Monolith Soft's action game which will be a launch title forthe Switch U.

Xenoblade X Remastered

Nintendo already admitted that 15M is going to be difficult to achieve, they’ll aim to try to hit it but if anything they’ll simply revise down.

They project 180M software this year which is lowballing it from them. Zelda is clearly not going to do all that or even half of that. It’ll do a solid chunk though.

Evergreens will continue to sell

They’ll continue to do DLC and make them some showpiece because it increases engagement. They’ll throw in first party because that increases engagement. They’ll throw in third party because that increases engagement.
It's weird that people seriously think Nintendo won't announce a game this summer to be released this holiday season. A big game, mind you. Yeah, they can technically coast, but I don't see them getting to 15 million switches sold with just DLC. Hell my backlog needs a break. But I am willing to bet my account a big game is coming out this holiday. I'm not holding my breath on MP4, but think a Mario game is a good candidate.
 
Because they will release the Switch U in 1 year. It makes less sense to release new software when the new console is near. I even think they will release MP4 as a launch title for the Switch U.
But what about the momentum? They have to keep releasing software to keep the momentum going, even with new hardware, as consumers may be hesitant to upgrade.
 
You're saying 16GB of RAM running at the same frequency as 12GB of RAM, would have the same power consumption?..

In the case of increasing RAM, would increasing/doubling the bus width (let's say from 64 to 128), require another tape out? Not that it would apply to Switch 2.
a neglectable amount of power increase

increasing bus width would require a new tape out, yes
 
Nintendo not releasing new software this year is the equivalent of Apple telling iPhone 14 lineup owners that their devices will no longer access social media apps, so entertain yourself with…?

Sure, Evergreen titles will keep growing, what with TOTK, Pokemon DLC, Mario Kart DLC, Splatoon DLC and Mario Movie promotion.

Buuutt… How can those move 15 million hardware units? Let’s say Nintendo revises the forecast down to 12 million. How can that lineup move 12 million units?

There will obviously be new software. Not just because it’s logical, but because Nintendo wants a smooth transition
 
Current Mood:

wallcover.jpg
 
So you agree that Nintendo isn't going to coast the rest of the year, then? Because I agree with the last paragraph; they need more than what was already shown to reach somewhere close to their sales target.
evergreens will take a portion, Pikmin will take its small portion. Zelda will take its sizable portion. 3rd party evergreens will also take their portion.

DLC will be provided. 2 new titles or 3 to settle it out, like a Mario party. But otherwise, I see this as coasting.
 
Quoted by: D36
1
evergreens will take a portion, Pikmin will take its small portion. Zelda will take its sizable portion. 3rd party evergreens will also take their portion.

DLC will be provided. 2 new titles or 3 to settle it out, like a Mario party. But otherwise, I see this as coasting.
If one of those three titles is something big like a 2D/3D Mario, I don't see that as coasting.
 
You're saying 16GB of RAM running at the same frequency as 12GB of RAM, would have the same power consumption?..

In the case of increasing RAM, would increasing/doubling the bus width (let's say from 64 to 128), require another tape out? Not that it would apply to Switch 2.
I probably misspoke. But I assume that if the supply voltage and the operating temperatures are the same between the 6 GB (48 Gb) LPDDR5 modules and the 8 GB (64 Gb) LPDDR5 modules, and assuming two of the same 64-bit LPDDR5 modules are used to obtain 128-bit LPDDR5, then I imagine the power consumption between the two different LPDDR5 modules should at least be very similar. But of course, not all LPDDR5 modules are created equal. So I suppose that depends on which LPDDR5 modules Nintendo chooses. The point I was trying to make is that the capacity of the RAM modules alone isn't the only contributor to the power consumption.

I imagine another tape out is required to increase the RAM bus width since the number of RAM controllers in the SoC probably has to increase. As shown in Locuza's annotated die shot of the Exynos 2200, the four 16-bit LPDDR5 controllers are integrated inside the SoC.
https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2Ff9f32098-0c8e-44a4-9b86-dbc6264cc431_1810x1778.jpeg
 
That is not going to get them where they want to be. It simply isn't.
With the Mario Switch bundle and the OLED TotK edition they can. They basically said "We might sell 15 million Switches, but it will be hard."
 
0
Nintendo not releasing new software this year is the equivalent of Apple telling iPhone 14 lineup owners that their devices will no longer access social media apps, so entertain yourself with…?

Sure, Evergreen titles will keep growing, what with TOTK, Pokemon DLC, Mario Kart DLC, Splatoon DLC and Mario Movie promotion.

Buuutt… How can those move 15 million hardware units? Let’s say Nintendo revises the forecast down to 12 million. How can that lineup move 12 million units?

There will obviously be new software. Not just because it’s logical, but because Nintendo wants a smooth transition
What? No that is not at all comparable to that.


If one of those three titles is something big like a 2D/3D Mario, I don't see that as coasting.
Why would you waste that on the Nintendo switch when you have a New Console on the Horizon? This is why I said many pages ago that this is a balancing act and not easy for Nintendo. You can’t just throw everything at the switch and then have the new system behind dry Especially for the first two years which is super crucial for it.

Nintendo isn’t developing for a console that is a switch, it’s for something that is a lot higher than the switch. Just like how dev times increased from Wii to Wii U, they’ll increase again from Switch to Switch 2
 
What? No that is not at all comparable to that.



Why would you waste that on the Nintendo switch when you have a New Console on the Horizon? This is why I said many pages ago that this is a balancing act and not easy for Nintendo. You can’t just throw everything at the switch and then have the new system behind dry Especially for the first two years which is super crucial for it.

Nintendo isn’t developing for a console that is a switch, it’s for something that is a lot higher than the switch. Just like how dev times increased from Wii to Wii U, they’ll increase again from Switch to Switch 2
The new system isn't gonna be left behind that easily, especially if it has cross-gen support. Besides, if Drake doesn't release this year, 2D Mario is probably the way to go to get the most engagement.
 
I probably misspoke. But I assume that if the supply voltage and the operating temperatures are the same between the 6 GB (48 Gb) LPDDR5 modules and the 8 GB (64 Gb) LPDDR5 modules, and assuming two of the same 64-bit LPDDR5 modules are used to obtain 128-bit LPDDR5, then I imagine the power consumption between the two different LPDDR5 modules should at least be very similar. But of course, not all LPDDR5 modules are created equal. So I suppose that depends on which LPDDR5 modules Nintendo chooses. The point I was trying to make is that the capacity of the RAM modules alone isn't the only contributor to the power consumption.

I imagine another tape out is required to increase the RAM bus width since the number of RAM controllers in the SoC probably has to increase. As shown in Locuza's annotated die shot of the Exynos 2200, the four 16-bit LPDDR5 controllers are integrated inside the SoC.
https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2Ff9f32098-0c8e-44a4-9b86-dbc6264cc431_1810x1778.jpeg
Just to make sure I understand correctly, if Nintendo for whatever reason only had 1 8GB Module and decided post-tape-out to change it to either 2 6GB Modules or 2 8GB Modules, they would likely have to make a new SoC?
 
The new system isn't gonna be left behind that easily, especially if it has cross-gen support. Besides, if Drake doesn't release this year, 2D Mario is probably the way to go to get the most engagement.
Nintendo doesn’t seem keen on delivering a cross gen though. Especially one like the other platforms that do 2-3 years of cross gen before straight dropping the older platform.
Just to make sure I understand correctly, if Nintendo for whatever reason only had 1 8GB Module and decided post-tape-out to change it to either 2 6GB Modules or 2 8GB Modules, they would likely have to make a new SoC?

If Nintendo is using a 128-bit interface in their SoC they will use RAM that equates up to that of the appropriate RAM interface.

If it is 1 8GB module, the RAM module will need to have 128-bit.

If they upgrade from 8GB to 12GB and the only way is via 6GB, they’ll need 2 64-bit 6GB RAM modules to do so.

However, 2 64-bit 6GB RAM modules are cheaper than 1 128-bit 8GB RAM module.

2 64-bit 8GB modules (for 16GB) might be just about or cheaper slightly than one 8GB 128-bit though.

There is no need to change the SoC for this. Since this is outside the SoC. So long as it is rhe right RAM kind, and it is the appropriate one, it’ll work.

If they move from LPDDR5 to LPDDR5X, that requires a new SOC. If they have a 128-bit interface and wish to have a 192-bit interface, that requires a new SOC.

If they take 8GB LPDDR5 memory and move to 2 6GB LPDDR5 memory, there’s no need for a new tape out.
 
What kind of gigantic fuck up would require Nintendo to pull the SDK from all parties?

I have no idea how in this day and age developers just sat there, as Nintendo claimed back the devkits on which people had spent X months developing/porting games. This is beyond me.
Nvidia did fuck up on the OG Switch allowing for a huge exploit so there is that.
 
The underperformance of the ROG Ally so far is making me give a second thought to this wild hypothetical I saw elsewhere: what if the Ally's the salvage bins of the Phoenix chips? I'm not necessarily believing it yet, but I dunno, I expected better.

Aside from that, I think that my takeaway is that it's actually pretty impressive that Zen is able to be pulled down into laptop power ranges and still perform so well as it is, but at the end of the day, that's not the primary segment they're for. IMO, as long as Intel's Core series continues the back and forth with Zen and keeps the pressure on AMD into needing to expand the core to keep up performance in the server segment, I'm not expecting a whole lot of improvement in specifically the 10 watt SoC space from Zen based chips. Not that AMD should be all that bothered by that; that's below typical laptop range. And handheld PC is still too small a market on its own. I don't yet see a need for AMD to come up with a separate CPU line just for low power.
And I'm restating that I think that progression for x86-based PCs in specifically 10 watt or under SoC range will have to rely on microarchitecture series that specifically target lower power in the first place. whispers Aaaaaattttoooommmm
 
I haven't followed much these days because I'm taken by the best video game of all time... I see however that the topic has progressed... Has there been any news?
 
0
Please read this new, consolidated staff post before posting.

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


Back
Top Bottom