• 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)

If Switch 2 comes out at webs of 2024 or early 2025, it will probably be using Samsungs 8nm node.

Nintendo is just going to coast 1.5-2 years with software. I do wonder if they will delay big Switch games for Switch 2 release year
 
46% of the total gross profit from video games, not counting mobile or playing cards since they are minuscule. It's in Nintendo's Q3 earnings explanatory material, slide #5. Let me know if I misread it.
I wouldn’t interpret that as percent of profit. The numbers don’t add up to 100% and the footnotes say it’s proportion of total sales though I’m not sure what it is actually supposed to indicate. As the Switch being less than 100% of hardware sales is odd, unless it’s proportion of the whole year.
 
Last edited:
Looking at that Geekbench 6 launch article again, I'm thinking about this:
"One big change for Geekbench 6 compared to Geekbench 5 and other versions is in the way that multi-core scores are calculated. Previously multiple individual tasks were created and measured to see how quickly they would complete. The more cores you had the quicker they would complete. However, in Geekbench 6, one workload is used and all the cores work together on that one shared objective. It is still true that the more cores you have, the quicker it will complete. However, there is now interaction between the cores."

I get the shift to the latter approach as general computing goes. But I am curious as to which approach would be expected to be more reflective of game programming.
 
If Switch 2 comes out at webs of 2024 or early 2025, it will probably be using Samsungs 8nm node.

Nintendo is just going to coast 1.5-2 years with software. I do wonder if they will delay big Switch games for Switch 2 release year
It will be TSMC 4N, Samsung 8nm its not good for device like Switch
 
It will be TSMC 4N, Samsung 8nm its not good for device like Switch
I tend to agree, but I would phrase it differently. I would say 8nm probably isn't good enough to run Drake satisfactory in handheld mode. The specific soc we know about with 12 ampere sm and 8 a78 cores.

A Switch device running a smaller SOC sure. But not Drake.
 
I tend to agree, but I would phrase it differently. I would say 8nm probably isn't good enough to run Drake satisfactory in handheld mode. The specific soc we know about with 12 ampere sm and 8 a78 cores.

A Switch device running a smaller SOC sure. But not Drake.
Indeed, idk why people still think it will be still on Samsung 8nm, TSMC 4N( basically 5nm but better) its also most likely beacuse TSMC will be shut down 7nm soon and will move all production to 6nm+
 
I like how easy it is for the "future hardware
Looking at that Geekbench 6 launch article again, I'm thinking about this:
"One big change for Geekbench 6 compared to Geekbench 5 and other versions is in the way that multi-core scores are calculated. Previously multiple individual tasks were created and measured to see how quickly they would complete. The more cores you had the quicker they would complete. However, in Geekbench 6, one workload is used and all the cores work together on that one shared objective. It is still true that the more cores you have, the quicker it will complete. However, there is now interaction between the cores."

I get the shift to the latter approach as general computing goes. But I am curious as to which approach would be expected to be more reflective of game programming.
I would like to see what the workload actually looks like. Is it work-stealing? Scatter gather? Is there shared memory? Are threads working on interdependent tasks?

Games, in general, are terrible at multi-threading. UE4 still has a thread model that looks like a Windows XP era application, which is a worst of all possible worlds situation. Threads blocking on each other is common, leaving all those Zen cores/hyperthreads under utilized, but also they don't have a lot of shared state that would be advantaged by things like ARMs L3 coherence in a cluster.

My understanding is that UE5 is working to improve the issue, and I have no sense of how Nintendo's internal engines work, but given how early Nintendo started embracing multicore over fast clocks - and given how lightweight thread instantiation is on HorizonOS, I would expect that they're at least as good at it as the rest of the industry, if not better.
 
0
Indeed, idk why people still think it will be still on Samsung 8nm
Because Orin is on 8nm, as is most of Ampere. I think there are enough questions on 8nm's power draw to think it might be viable. If I had to bet money, I'd say no, but it isn't a confident no.
 
If Switch 2 comes out at webs of 2024 or early 2025, it will probably be using Samsungs 8nm node.

Nintendo is just going to coast 1.5-2 years with software. I do wonder if they will delay big Switch games for Switch 2 release year
Switch 2 could have come out in 2022 on TSMC 4N.

This is an extraordinarily pessimistic view.
 
Wouldn’t that node be on its way out by then?
Samsung has said they want 8nm to live as long as 14nm, which is a decade old. It’ll likely still be around, and very cheap.

If hardware launches in 2025, then it will be extremely delayed. The SOC is ready now. It might be on 8nm in 2025, but only if it’s on 8nm now, and some colossal fuck up results in a multiyear delay. But it’s not clear to me that it is small enough to be on 8nm no matter what year it launches in.
 
0
we've heard a lot of stories at this point, but the one I'd really love to hear is what exactly is wrong with the fucking thing

iirc (and it's a BIG iirc) in 2021 brainchild alluded to power consumption being a serious concern at that time. does anyone else remember that? I have a tendency to imagine shit by mistake
 
we've heard a lot of stories at this point, but the one I'd really love to hear is what exactly is wrong with the fucking thing

iirc (and it's a BIG iirc) in 2021 brainchild alluded to power consumption being a serious concern at that time. does anyone else remember that? I have a tendency to imagine shit by mistake
About Drake? I think NateDrake said that about ray tracing. But I don't think those fears are credible
 
I wouldn’t interpret that as percent of profit. The numbers don’t add up to 100% and the footnotes say it’s proportion of total sales though I’m not sure what it is actually supposed to indicate. As the Switch being less than 100% of hardware sales is odd, unless it’s proportion of the whole year.
Thanks for checking. It reads to me that in fiscal Q1-Q3 (FY03/2023), 46.2% of gross profit from “total dedicated video game platform sales” was from hardware sales, down from 50.4% for the same period in FY03/2022. The software proportion (although undisclosed) presumably is 53.8%. I don’t see any components that should be added up to 100% on this slide. Am I missing something?
 
Hopefully Nintendo provides us with some factual news come May, because the waves of optimism and pessimism in this thread are astounding lol
I don't think that's Nintendo's fault. People just have unreasonable expectations or assume they understand how Nintendo works...

I do want to know though if their secrecy as a company surpasses that of several government intelligence agencies.
 
Hopefully Nintendo provides us with some factual news come May, because the waves of optimism and pessimism in this thread are astounding lol
Why would Nintendo mention new hardware before or around ToTK's release?
They're totally going to wait til June at the very least, and even then I don't really expect the first tease until Q4 2023.
 
0
we've heard a lot of stories at this point, but the one I'd really love to hear is what exactly is wrong with the fucking thing

iirc (and it's a BIG iirc) in 2021 brainchild alluded to power consumption being a serious concern at that time. does anyone else remember that? I have a tendency to imagine shit by mistake
There was a vague allusion to a problem, but I don't recall it being presented as especially serious.

Something important to keep in mind is that uncertainty in reporting and actual uncertainty behind the scenes are two very different things. There are many sources for the former that don't necessarily implicate the latter.
 
Another Alps Alpine analog stick patent. This one probably won't be commercialized, but is fun to see anyway.

On the back of a Joy-Con looking controller, a dial wheel is added:
img
img


The dial wheel controls a rotary cam:
img


When engaged, the cam mechanism locks the analog stick in place. E.g, you want Mario to always go right, so you tilt the stick to the right and lock it:
img


Another embodiment of the patent eschews the rotary cam and dial wheel, but houses a push-in cam within the stick assembly:
img


Similar to the other embodiment above, when engaged (pushed in), the cam mechanism locks the analog stick in place:
img


You: no-drift stick
Me, an intellectual: on-demand drift
 
Another Alps Alpine analog stick patent. This one probably won't be commercialized, but is fun to see anyway.

On the back of a Joy-Con looking controller, a dial wheel is added:
img
img


The dial wheel controls a rotary cam:
img


When engaged, the cam mechanism locks the analog stick in place. E.g, you want Mario to always go right, so you tilt the stick to the right and lock it:
img


Another embodiment of the patent eschews the rotary cam and dial wheel, but houses a push-in cam within the stick assembly:
img


Similar to the other embodiment above, when engaged (pushed in), the cam mechanism locks the analog stick in place:
img


You: no-drift stick
Me, an intellectual: on-demand drift
oh for a moment i thought its for changing the resistance of the stick to have it be be harder or easier to push (depeending if the action needs fine tuning or full on fast flicking)

i feel like if there is a situation in a game where i would need that functionality, then the game should just have a toggle in game to keep the direction locked?
 
we've heard a lot of stories at this point, but the one I'd really love to hear is what exactly is wrong with the fucking thing

iirc (and it's a BIG iirc) in 2021 brainchild alluded to power consumption being a serious concern at that time. does anyone else remember that? I have a tendency to imagine shit by mistake
In all likelihood nothing is wrong with it. Just spiralling rumours.
 
Pardon my ignorance but after my dreams being crashed for a Q1 release of the pro/2 are there any reasons for optimism?I decided not to enter in this region of the forum…
Well, Nintendo is probably going to release a new console eventually…
 
0
the issue with Returnal on PC and UE4 at large is an interesting one. RT and level streaming seemingly being on the same thread limits RT potential of the engine and that might be trickling down to consoles. devs would have to put in extra work splitting threads or be very mindful of their streaming to avoid bottlenecking. don't know if UE5 solves this (kinda doubt it to be honest), but it's gonna be paramount for RT usage to grow in the current generation
 
Because Orin is on 8nm, as is most of Ampere. I think there are enough questions on 8nm's power draw to think it might be viable. If I had to bet money, I'd say no, but it isn't a confident no.

Maxwell was primarily on the 28nm process but the Tegra X1 was designed on the 20nm process, so it wouldn't be out of the norm for Nvidia to target a smaller process for a mobile based SOC. As we have moved on from a first half 2023 release to second half at the earliest and 2024 being just as likely, the cost of better processes is coming down rapidly. Will it be using the cutting edge process that Apple will use with their latest Iphone? No, but something like 4N is very viable for a product releasing late 2023 or at any point in 2024.

Laptops with the RTX4050 are releasing here in a few days starting at $999. So the lineup of lower cost RTX40 cards are starting to roll out and they are all on the 4N process.

RTX4050 Laptop
 
Thanks for checking. It reads to me that in fiscal Q1-Q3 (FY03/2023), 46.2% of gross profit from “total dedicated video game platform sales” was from hardware sales, down from 50.4% for the same period in FY03/2022. The software proportion (although undisclosed) presumably is 53.8%. I don’t see any components that should be added up to 100% on this slide. Am I missing something?
As these are under the heading "Main Variable Factors", I believe they are metrics which contribute to changes in gross profit, rather than metrics of gross profit. The proportion of hardware sales impacts gross profit, as hardware has a lower margin than software. Similarly, the proportion of first party software sales will impact gross profit, as will the percentage of digital, as they're all measures of "how much of our business is coming from high-margin areas vs low-margin". FX rates will impact gross profit too, as they've got costs and revenues across multiple currencies.
 
0
Another Alps Alpine analog stick patent. This one probably won't be commercialized, but is fun to see anyway.

On the back of a Joy-Con looking controller, a dial wheel is added:
img
img


The dial wheel controls a rotary cam:
img


When engaged, the cam mechanism locks the analog stick in place. E.g, you want Mario to always go right, so you tilt the stick to the right and lock it:
img


Another embodiment of the patent eschews the rotary cam and dial wheel, but houses a push-in cam within the stick assembly:
img


Similar to the other embodiment above, when engaged (pushed in), the cam mechanism locks the analog stick in place:
img


You: no-drift stick
Me, an intellectual: on-demand drift
It's odd that this patent uses the Joy-con as its example body. If this patent doesn't have anything to do with Nintendo, why would that be the case? It's not even a case of using a stock model since Alps doesn't manufacture entire controllers, just sticks and smaller components.

I agree this is unlikely to be commercialized, and I highly doubt Nintendo has any use for it, but it's still interesting.
 
Maxwell was primarily on the 28nm process but the Tegra X1 was designed on the 20nm process, so it wouldn't be out of the norm for Nvidia to target a smaller process for a mobile based SOC. As we have moved on from a first half 2023 release to second half at the earliest and 2024 being just as likely, the cost of better processes is coming down rapidly. Will it be using the cutting edge process that Apple will use with their latest Iphone? No, but something like 4N is very viable for a product releasing late 2023 or at any point in 2024.
I was saying in my post that I'm betting it's on 5nm. But I don't think 8nm is ruled out.

Yes, Tegra has used smaller nodes in the past - but Orin is a Tegra product, and it is on 8nm. Yes, 5nm is viable for commercial products - in fact, Nvidia has a 5nm architecture, already released, and Drake doesn't use it. Yes, the power tool numbers for Orin don't look great, but Drake has fewer memory controllers and additional power saving tech that Orin doesn't have. And, we don't know what sort of clock scheme Nvidia and Nintendo are going for - we've been assuming static clocks, but a DVFS setup that balances CPU and GPU power depending on workload would radically change the math.

8nm is a long lived node, with good yields, that is only going to get cheaper. If Drake can run on 8nm, I suspect it will, just to keep costs down in the short term, and open up a 5nm revision when that is also nice and mature. I suspect it can't. If I throw every bit of optimism I can muster at Orin's numbers - cutting memory channels and assuming crazy savings from FLCG - I still can't get in spitting distance of the original Switch. But that's all handwavy math, I'm not an electrical engineer
 
0
It's odd that this patent uses the Joy-con as its example body. If this patent doesn't have anything to do with Nintendo, why would that be the case? It's not even a case of using a stock model since Alps doesn't manufacture entire controllers, just sticks and smaller components.

I agree this is unlikely to be commercialized, and I highly doubt Nintendo has any use for it, but it's still interesting.
eing how there are 2 designs, the first one seems to only make sense in a rather flat/shallow controller, making it unsutable for something like the Dual Sense 5. Thats probably the reason, joycons are just the closest to a "standard" form factor for those smaller controllers.
 
0
the issue with Returnal on PC and UE4 at large is an interesting one. RT and level streaming seemingly being on the same thread limits RT potential of the engine and that might be trickling down to consoles. devs would have to put in extra work splitting threads or be very mindful of their streaming to avoid bottlenecking. don't know if UE5 solves this (kinda doubt it to be honest), but it's gonna be paramount for RT usage to grow in the current generation

No one is doing anything serious with regards to raytracing before 2028 other than NVIDIA.

Consoles are garbage at it and the devs with the technical skill to do raytracing mostly don't care at all about optimizing for or focusing on PC.
 
I just don't know if I can think of a dev for 2D Zelda who fits the criteria

1. Wants to make 2D Zelda
2. Nintendo would trust them to make an original 2D Zelda
3. Could be hired by Nintendo to work on an original 2D Zelda
Outside Grezzo? 1-Up Studios. Nintendo owns them and they have worked on top-down Zelda in the past.
Grezzo and 1-Up Studios haven't designed top down Zeldas. Tri-Force Heroes was directed by Nintendo Employee Hiromasa Shikata, who also directed Link Between Worlds. Daiki Iwamoto directed the two DS games, and also supervised the Link's Awakening remake at Grezzo.

If Nintendo wants an original 2D Zelda, and doesn't want EPD to make it (and frankly, I think giving that team a chance to turn over a game in less than 5 years would be good for their soul), they can assign Kentaro Tominaga or some other hot Junior Director/Level Designer to take the reigns and simply use Grezzo as development staff. They clearly trust them with Zelda, but that doesn't mean letting Grezzo do whatever they want.
 


Price drop incoming?


holy shit I bet that's a Mario Switch for the movie

A price drop for a Special Edition? That seems a little much. But a price drop on the OLED and maybe sales for all the Mario games for Mario Day? Ok, that also seems a little much lol but I can see it happening. Was kind of expecting a price drop (and V2 being phased out like mentioned above).
 
A price drop for a Special Edition? That seems a little much. But a price drop on the OLED and maybe sales for all the Mario games for Mario Day? Ok, that also seems a little much lol but I can see it happening. Was kind of expecting a price drop (and V2 being phased out like mentioned above).
what it's not an OLED?
 
I think it's about the right time for OLED to drop to 300 and let the V2 be phased out. That's what I think this is, personally. The new SKU may be due to SCO, like a redesigned box, or just one choice of Joy-Con colour, etc.

Agree. I think initially they can drop the price of the V2 to $249, and then begin the phase out of that model. I'm sure with the current stock levels the phase out of V2 could be done over the course of the rest of the year, but certainly once Switch 2 is in the world, there is no reason to have three different OG Switch SKU's. I am firm in the belief that Nintendo will not want to go beyond $399 for Switch 2 and that it makes sense to have a significant price difference between the new and old consoles.
 
Grezzo and 1-Up Studios haven't designed top down Zeldas. Tri-Force Heroes was directed by Nintendo Employee Hiromasa Shikata, who also directed Link Between Worlds. Daiki Iwamoto directed the two DS games, and also supervised the Link's Awakening remake at Grezzo.

If Nintendo wants an original 2D Zelda, and doesn't want EPD to make it (and frankly, I think giving that team a chance to turn over a game in less than 5 years would be good for their soul), they can assign Kentaro Tominaga or some other hot Junior Director/Level Designer to take the reigns and simply use Grezzo as development staff. They clearly trust them with Zelda, but that doesn't mean letting Grezzo do whatever they want.
There is no way there is going to be a mainline 2D Zelda without any EPD involvement. EPD will be involved in some capacity. Even if its EPD doing the design and 1-UP doing the grunt work, so 1-UP Studios would be fine.
 
There is no way there is going to be a mainline 2D Zelda without any EPD involvement. EPD will be involved in some capacity. Even if its EPD doing the design and 1-UP doing the grunt work, so 1-UP Studios would be fine.
1-UP in it's current incarnation is probably not outfitted for that. They are primarily artists, and work under production management of other EPD teams, and has never developed a title as a solo team. It's either a EPD 3 project with 1-UP providing support under a whole bunch of EPD 3 staff, or it's an EPD3 director and producer managing Grezzo (with possible 1-UP support), who can provide the whole package, including staff who have designed Zelda dungeons and bosses before
 
0
Usually the special edition OLEDs get a $20 mark up or so, right? Something doesn't add up; the OLED getting a price drop to $300 is suspicious, but if this is the ZOLED then is standard OLED dropping to $280? I thought there was the math on price economics posted a dozen or so pages ago that described that a drastic price drop like that would hurt the bottom line more than anything.

I think this might not be what we think it is Nevermind I'm slow to people already suggesting a game bundle with V2 lol
 
Agree. I think initially they can drop the price of the V2 to $249, and then begin the phase out of that model. I'm sure with the current stock levels the phase out of V2 could be done over the course of the rest of the year, but certainly once Switch 2 is in the world, there is no reason to have three different OG Switch SKU's. I am firm in the belief that Nintendo will not want to go beyond $399 for Switch 2 and that it makes sense to have a significant price difference between the new and old consoles.
I agree. I think the most likely hardware strategy over the next 12 months is:

Price drop for OLED Model to 299.99,
Price drop for V2 to 249.99
Phase out V2 as they sell through without making more, clear production lines for the new device.

By time the Drake model launches for 399.99, there's no more V2, plenty of production capacity, and a nice, simple pricing structure:

199.99 for the entry level model with no TV mode.
299.99 for the middle of the road, wants TV Mode but not 4K
399.99 for the full blown next gen model, 4K, exclusives and all. Maybe with some cloud ports for the other models.
 
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