• 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| (Read the staff posts before commenting!)

Thanks for the clarification, even 75% of PS5's CPU would be closer than what was originally expected in these early speculation talks.
Nintendo hasn't enjoyed a jump in performance this large both in CPU and GPU since the GameCube days.
Right, and linking back to this Thraktor post from the old place, the gulf in performance for A57 vs. Jaguar is a lot larger (numbers are single-core performance):

Incidentally, as someone posted a few pages back, there are now benchmarks for the AMD 4700S, which is a repurposed PS5 chip (running at slightly higher clocks, though). We've also now got A78 benchmarks, so I though it might be interesting to do a comparison of how the A78 would compare to console CPUs past and present.

Specifically, I'm going to use the following Geekbench 5 single-core benchmarks:
Nvidia TX1 - Switch SoC - A57 @2GHz - 280
AMD A9-9820 - Xbox One APU - Jaguar @2.35GHz - 294
AMD 4700S - PS5 APU - Zen 2 @4GHz - 998
Mediatek Dimensity 1200 - A78 @3GHz - 975
The math works out to a per-core ratio of 294/140 = 2.1 vs. PS4 Jaguar cores at 2.35 GHz (Switch A57 at 1.0 GHz, not 2.0 GHz). And then there is a 2.33x difference in core count (7 vs. 3), giving a theoretical gulf of 4.89x, or alternatively putting Switch 1 at 20.4% of PS4. The math is fuzzy, but ballpark suggests that we are much closer next gen than current gen
 
Last edited:
Can't watch that right now, but is that podcast talking about Nates podcast, or is Nate actually a guest in that specific podcast episode?

Also, for me as a non-native english speaker, reveals means more than announcement. It means something's being shown to me.
Nate: “All I can say is what I said on my show is that the Switch 2 appears to be poised and ready for a reveal or an announcement in March. Nintendo will announce Switch 2 in March.”
 
Thanks for the clarification, even 75% of PS5's CPU would be closer than what was originally expected in these early speculation talks.
Nintendo hasn't enjoyed a jump in performance this large both in CPU and GPU since the GameCube days.

I think on a better process node than the 8xA78AE (Samsung N8), the relative power/efficiency will be a lot better and thus the gap between the PS5 and 8xA78C in single core could be a lot closer (~10%).

Also, you can observe in Switch OC videos with the switch, how the CPU load is spread across all cores. BOTW and TOTK are (with not much surprise), quite well spread in CPU load across all cores (of course not 100%, because that's not possible, nor desirable). I wish I had the time to line them all up as I think most first-party titles have very high utilisation (with not much surprise).
 
Really? It's horrible. Mine's always dropping out. I just stopped using them.
Are they disconnecting, or is it like some inputs don't get recognised? I had issues in the past but figured out it was my own fault. In my previous house I had the Switch docked partially behind the TV, and I noticed if I was ever sitting with my feet on top of the chair and held my left hand beside my left leg the left Joy-Con wouldn't properly send inputs, basically because I was blocking the signal from actually getting to the Switch. I'd look into that, just in case it's something similar happening to you. It happens occasionally with my pro controller too depending on how I'm sitting, but it's not as bad as with the Joy-Cons.
 
Honestly, DF should stay out of Switch 2 speculation.
I disagree. Why should anyone stay out of Switch 2 speculation? Speculation is easy. Everyone is free to offer their opinions of what might be. I appreciate those with some expertise in the areas of hardware and software weighing in. It is ALL good IMO. I certainly believe DF to be pretty well qualified to guess just like anyone else... maybe more?
 
I disagree. Why should anyone stay out of Switch 2 speculation? Speculation is easy. Everyone is free to offer their opinions of what might be. I appreciate those with some expertise in the areas of hardware and software weighing in. It is ALL good IMO. I certainly believe DF to be pretty well qualified to guess just like anyone else... maybe more?
We, the Famigods, are the only ones qualified and powerful enough to speculate on Switch 2. If John‘d join, maybe he could also become qualified.
 
I disagree. Why should anyone stay out of Switch 2 speculation? Speculation is easy. Everyone is free to offer their opinions of what might be. I appreciate those with some expertise in the areas of hardware and software weighing in. It is ALL good IMO. I certainly believe DF to be pretty well qualified to guess just like anyone else... maybe more?
some speculation is alarmist. just pure fantasy or there is no logic sometimes in what is being said. Now, everyone has a right to say their opinion but DF ahould understand what they has some sort of influence. I have no hate against DF they are awesome at what they do but unfounded speculation does harm to a community.

As an example look Xbox meltdown this week, this was all done with I would consider alarmist speculation. Not every speculation is healthy. Feel free to disagree, these are just my thoughts.
 
Yeah, I can’t see that ever happening. The controller itself is $70.
That's the MSRP yes, but that's not the cost of manufacturing. There's probably a set markup that they could ignore if it was bundled with the Switch 2, but that said... yeah I don't think it's likely. Even if the markup was at 40% or something at retail (so $50 x 1.4 = $70), that's still an additional 50 dollars of value added to the base system. Not the wisest decision if you want people to buy a cheap console.
 
Does that mean there's a higher chance of any leaks within the next few weeks or at the beginning of March?
Not really.

It really needs to be stressed, but even with the Switch 1 there was a lot of mystery about the actual solid information we had for it. Hell, the same day it was going to be revealed, the only reason why we knew about it's upcoming reveal was thanks to Emily Rogers predicting it would be "that week". Nintendo's secrecy is an artform in of itself. It's actually weird that leakers like Pyoro have cropped up and got stellar accuracy in the recent Nintendo presentations.
 
Insane to me that we're trying to gatekeep who should and shouldn't be involved in Switch 2's discussion. Digital Foundry is one of the most knowledgeable sources about gaming hardware in the industry, let them speculate if they want
 
Insane to me that we're trying to gatekeep who should and shouldn't be involved in Switch 2's discussion. Digital Foundry is one of the most knowledgeable sources about gaming hardware in the industry, let them speculate if they want

I think it's not people having issues with them speculating, but rather due to it being Digital Foundry, expect a better "quality" of speculation.
 
2028 will be the year that Nintendo feels like Apple:
• Switch 2 - 8” LCD screen
• Switch 2 Lite - same as Lite, but better
• Switch 2 OLED - 6” OLED screen, remisnicient of the first Switch
 
0
No joy con point brought up by Nate again but in a livestream? Some people here about to be real disappointed!
Im just trying to wrap my head around whether Nate has heard any rumours of a Switch successor with no Joy-Cons, or that he is moreso going off of anecdotes and speculation since he doesn't have any statistical data on how people play with their Switch.

Edit: Nate is also aware of his audience pushing back on his speculation so I'm surprised he's still pretty confident in Nintendo ditching the Joy-Cons, and packaging a pro-controller instead. Also, even if Nintendo is selling compatible Joy-Cons still, would people have to charge them with their Switch, alongside their Pro-Controller, and the system itself? Or is there a separate charging dock that they plan on offering. It's just sounds convoluted.
 
Last edited:
I think it's not people having issues with them speculating, but rather due to it being Digital Foundry, expect a better "quality" of speculation.

They specify that they're speculating based on anecdotal evidence, who cares if it's outlandish? I could see people having an issue if they claimed it was happening without a doubt
 
They specify that they're speculating based on anecdotal evidence, who cares if it's outlandish? I could see people having an issue if they claimed it was happening without a doubt

Thing is, again because it's Digital Foundry, a lot of people who're not following this whole situation as much as we in this thread do, actually do take what they speculate as gospel.

And i do say, if you're earning your living with such things, i don't think it's unfair or something "outrageous" to expect a higher standard from them.
 
The original Nintendo Switch development kits had Joy-Con Controllers but didn't allow them to be attached. Honestly I wouldn't be too surprised if some of this is coming from people seeing (but not working with) development kits of the new system that have the usual weird restrictions. Like controllers screwed onto the console so that they can't be separated but handheld mode can be tested, alongside a Switch Pro Controller, especially as controllers likely wouldn't have been totally finalised by time kits went out.

I wonder if Nintendo still manufactures a small volume of Wii Classic Controller Pro for use with SDEV kits...
 
I have no problem with speculation or even insider information coming from DF, but I miss the quality and responsibility that was attributed to them.
The DLA's situation was quite annoying, several "specialized media" sites reporting something that they clearly didn't say badly, but speculated without any basis.
 
It's very likely those were used because Nintendo still has some stock of them leftover from the Wii/U days, and not that they're newly manufactured.
On the one hand, maybe, both Wii Mini and Nintendo Switch were on the market simultaneously, so a certain amount of Wii accessories would have been being manufactured for years and years.

On the other hand, however, the Wii remote external extension connector was built into SDEV- and they never got rid of it, as far as I know. They'd need to keep an amount of WCCP around not just for any new SDEVs sent out (which, I admit, may be zero, being superceded by EDEV and ADEV.), but support for people with existing SDEV kits. Leftovers, maybe, but ongoing support for the kits, and designing the kit around the connector, that would imply the need for the occasional round of new ones.

Now I wonder even more, how fast, if ever, was SDEV discontinued, how long will replacement parts (like WCCP) be available, if at all(instead telling developers to return the SDEV for an EDEV). How does the Wii Remote external extension connector interface with Nintendo Switch on the systems its installed in, and was the use of the connector a technical necessity, stock convenience, or otherwise?

I have to imagine the next gen kit just has USB ports for Pro Controllers, even early on, as be it relying on new old stock or low volume manufacturing, WCCP is a lot less viable nowadays, and of course has less features you can test.
 
Thing is, again because it's Digital Foundry, a lot of people who're not following this whole situation as much as we in this thread do, actually do take what they speculate as gospel.

And i do say, if you're earning your living with such things, i don't think it's unfair or something "outrageous" to expect a higher standard from them.

If we're talking about the Joy Cons, then John didn't officially report or speculate about that through Digital Foundry, they speculated as a guest in another podcast.

If we're referring to 8nm concerns, then I truly believe that John and Rich must have a source as they've doubled down heavily on this
 
GBA/DS early dev boards may have used newly manufactured SFC controllers
I think that's somewhat unlikely; Super Famicom was discontinued in 2003, and SNES (UK and Ireland) in 2005.

So GBA and early DS kits could use the same lines as the actual controllers and eat up the last scraps of production, especially since controller production may have continued after the console's discontinuation for people buying replacements, especially since technical support for SNES (UK and Ireland) and SFC continued for several years after discontinuation, and that meant having the parts on hand.
 
If we're talking about the Joy Cons, then John didn't officially report or speculate about that through Digital Foundry, they speculated as a guest in another podcast.

If we're referring to 8nm concerns, then I truly believe that John and Rich must have a source as they've doubled down heavily on this
Wouldn't John and Rich have mentioned that they have heard rumours from a source by now, as I'm pretty sure they've said in the past that their 8nm belief was based on speculation(only person who has actual sources is kopite and he has a worse track record with Tegra SOCs). Also the likelihood of them knowing is pretty low and we probably won't even know the node until we actually get our hands on the device and tear it down, or unless there's a huge leak of the device specs, in which that case DF wouldn't be the only one with that data.
 
Nate is a guest in the podcast and he says that the March reveal is not just based on what he heard back in August, but a information that has resurged in the past weeks.
Guess there’s smoke coming for March, meaning that the last week of February is when mass production will go full blast
 
Wouldn't John and Rich have mentioned that they have heard rumours from a source by now, as I'm pretty sure they've said in the past that their 8nm belief was based on speculation(only person who has actual sources is kopite and he has a worse track record with Tegra SOCs). Also the likelihood of them knowing is pretty low and we probably won't even know the node until we actually get our hands on the device and tear it down, or unless there's a huge leak of the device specs, in which that case DF wouldn't be the only one with that data.

The reason why I'm assuming that they must have a source is that when Rich was presented with evidence as to why 8nm wouldn't make sense given the information that came out of the NVIDIA hack, he agreed and said it didn't. However, this hasn't stopped him and John from continuously mentioning it.
 
If production is starting soon (lets say in ~1 month) we are 90% likely to get a new leak from an Asian outlet like Nikkei or Bloomberg.
 
Guess there’s smoke coming for March, meaning that the last week of February is when mass production will go full blast
I don't think this is an unreasonable assumption.

Hidden content is only available for registered users. Sharing it outside of Famiboards is subject to moderation.
 
Nintendo NDA’s are serious business. The podcast was a bit….weird but still fun to listen to. Speculating on this sort of thing on a predications based episode seems super silly but…. IF there is some red herring stuff going on…your theory is on point.
Yeah, what tripped was the joycon. He knows it is crucial for backward compatibility, so something was off with that one.

That's an insane question lol
I mean we have a lot of technically knowledge people here. It is nice to exchange theories and ideas.
after resetera, I think he had enough of message boards
Wow, was it that bad?
 
Technical query: Could force feedback work well with joycons? I feel like their small form and battery limitations might be an issue if they're pursuing something like that.
 
The reason why I'm assuming that they must have a source is that when Rich was presented with evidence as to why 8nm wouldn't make sense given the information that came out of the NVIDIA hack, he agreed and said it didn't. However, this hasn't stopped him and John from continuously mentioning it.
I mean, 8nm obviously isn't completely ruled out, it's just that it's simply an inefficient choice. John and Rich also seem to be operating off of the fact that the Switch will have a bigger screen so it must have 8nm, which has been argued against by several people on Fami.

To your point though, on Nate's podcast, Nate offered no push back to John talking about 8nm, which is surprising considering Nate is aware of the arguments made for 5nm/4N on Fami
 
I think that's somewhat unlikely; Super Famicom was discontinued in 2003, and SNES (UK and Ireland) in 2005.

So GBA and early DS kits could use the same lines as the actual controllers and eat up the last scraps of production, especially since controller production may have continued after the console's discontinuation for people buying replacements, especially since technical support for SNES (UK and Ireland) and SFC continued for several years after discontinuation, and that meant having the parts on hand.
This is true; I also recently found out that the AV Famicom was still manufactured until 2003, and as such you can find genuine, first-party Famicom SoCs and PPUs with 2003 datecodes.
 
If, again if, mass production does start with the reveal in early March (or shortly before at the end of Feb), then i don't see that launching any later than September.

Heck, as a proud Team H1 member, this would make me more confident in June actually.

Personally, in this case i would expect the reveal to be meaty, and not just a simple video like Switch October 2016.

we gonna be getting some purposeful entertainment soon?

You can watch Masters of the Universe - Revolution RIGHT NOW on Netflix!
 
Or we could just directly compare them on a benchmark. Looks like that gap doesn't scale all the way to those two core configurations
CoreGeekbench 6 Single CoreGeekbench 6 Multicore
8x A78AE @ 2.19Ghz10015449
8x Zen 2 @ 3.6Ghz13958240
Percent difference71.8%66.1%
Note: The AE cores definitely underperform the C core (the one in T239) when it comes to multicore. I expect the CPU to clock beneath this (personal estimate is closer to 2Ghz, maybe as low as 1.75?), but their isn't anything to sneeze at when it comes to 75% of a PS5's CPU crammed into a tablet.
Yeah, this amazing to hear. Usually people would think a console CPU is 2 to 3 time more powerful than a handheld.

And I think when it comes to gameplay, we pretty much peaked right? Maybe some creative stuff I haven't seen myself. It just seems that the bulk of the work is for NPC AI or helping the GPU and other resources
 
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