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

[Link_enfant]

Thanks a lot for the clarification, very exciting indeed!
And there's still a possibility they go for a 'better' option than 8nm Samsung for instance, which doesn't seem crazy if they want the best battery life (among other things) right?

 

[ILikeFeet]

Thanks a lot for the clarification, very exciting indeed!
And there's still a possibility they go for a 'better' option than 8nm Samsung for instance, which doesn't seem crazy if they want the best battery life (among other things) right?

Given the config, they have to unless they want to do what all these other handheld do and say "fuck battery life". But Nintendo wouldn't do that

 

[Skittzo]

Thanks a lot for the clarification, very exciting indeed!
And there's still a possibility they go for a 'better' option than 8nm Samsung for instance, which doesn't seem crazy if they want the best battery life (among other things) right?

This kinda tips the scales for me from "it's probably still 8nm" to "there's a good chance it's better now"

Nothing about the process node has been really talked about by anyone outside of Kopite who is an Nvidia leaker that mentioned it was a derivative of Orin last year, Orin being on Samsung 8nm. But so far he has not claimed to know the node this was on I believe.

 

[Serif]

Given 8 cores and possibly 5/6 nm, how might Drake compare to other consoles like PS4 or Series S?

 

[arsen]

Given 8 cores and possibly 5/6 nm, how might Drake compare to other consoles like PS4 or Series S?

We have no information about the clocks yet, which is an important factor for such comparison.

 

[Cybergatuno]

8 CPU cores in a single cluster points to the best possible configuration for Switch "2", basically we are somehow in a new era for Nintendo, where they actually want to compete with technology, 8 cores of one of the fastest ARM chips available is amazing IMO. It's something I've been saying they needed to do because it has become the standard in this industry for the past 10+ years to have 8 cores for game developers, and here it is.

Furukawa did warn us in 2020.

Nintendo used to look at conventional technology that enabled a lower price and appeal to users. However, it is now exploring cutting-edge technology.

He also said the Switch was "in the middle of its life-cycle" back then

 

[Bonejack]

I hope it's not to shoon for this:

 

[Z0m3le]

I hope it's not to shoon for this:

This info indicates that the chip exists in a production state. You don't really need Linux for chips that haven't been completed, there wouldn't be a point, so yeah, soon enough millions of these SoC will likely exist and be in production lines, this lines up with a spring release, as the original Switch begun full production around this time in 2016 for a March release.

 

[ReddDreadtheLead]

Given 8 cores and possibly 5/6 nm, how might Drake compare to other consoles like PS4 or Series S?

Depends on the clocks!

Let’s assume it is same speed as the PS4 CPU of 1.6GHz


Over 3x as performant for around 2W.


Series S is out of the question since that’s a desktop CPU. But I think it would be almost twice as performant though (at best) to Drake if 8 A78C @ 1.75GHz I think



Which is smaller than the current generation really.


Well, last gen.

And, 1.75Ghz would make it a little over 3X the CPU grunt of the XBox One S/VCR

 

[sadsquid]

 

[Gay Bowser]

This kinda tips the scales for me from "it's probably still 8nm" to "there's a good chance it's better now"

Same.

I’m still not sure I can see N5. Apple’s still using that for a lot of their products (they haven’t even moved all their products from 7nm to 5nm yet) and with AMD just starting to use it for their products this fall I think it’s still going to be more supply constrained than Nintendo would like. But N7 or N6 do seem very possible. AMD moving to 5nm is going to free up some capacity there, as is Apple switching products like the base iPad from 7nm to 5nm (which might not actually happen this year; I could see them keeping the A13 for one more go-around if they’re improving the screen and design of the iPad this time, which they reportedly are).

 

[Bonejack]

I love how, after so many "leakers" and info and hacks ... all it took for a major new "happening" moment here was a new user (welcome btw) being bored and using Google.

Let this sentence work in your mind.

 

[KMStwo]

I’m gonna Ignore the Linux stuff, and focus on one aspect: if real then this has 8 cores as a single cluster.

This would mean 8 homogeneous CPU cores.

Perhaps… the A78C as that’s the bare minimum for a single 8 core cluster.


How did you even find this?

So 8 core CPU is basically confirmed??

 

[Deleted member 1324]

Xenoblade doesn't use FSR. It uses a custom scaler presumably built by Monolith.

It can be difficult at times to tell if something is actually using FSR, because there are reasons the code may be present, but not in use, but disproving its use is often quite simple because using the code without disclosing it would be a license violation.

Yeah my bad. I’m guessing Tears of the Kingdom will use similar IQ tech to Xenoblade 3.

 

[ILikeFeet]

So 8 core CPU is basically confirmed??

Pretty much, yes

 

[Dark Cloud]

damn it’s happening.

 

[KMStwo]

Furukawa did warn us in 2020.

He also said the Switch was "in the middle of its life-cycle" back then

He also said that in 2021, but for some reason this year he said nothing

 

[Thraktor]

Hi guys,

First post here

It seems NVidia is adding support for Tegra239 into the Linux Kernel.


Well, this seems like a confirmation.

Well done on spotting this!

For anyone unsure about what this means, the important part is this:

Adding support for Tegra239 SoC which has eight cores in a single cluster

We know from a variety of sources that the T239 chip is Drake, ie the one being used in the new Switch model. It having "eight cores in a single cluster" means it has 8 CPU cores (obviously), but doesn't strictly speaking limit what those cores can be, as ARM's DynamicIQ supports clusters with mixes of different cores (eg 4x A78 and 4x A55). However, as the code would almost certainly have to be changed to support these hetergeneous clusters (ie different core types would run at different frequencies), and the commit here doesn't make such changes, we can reasonably assume that the cores are all of the same type. The only big ARM cores which can be configured with 8 cores in a single cluster are the A78 (in the A78C variant) and later, and given the T234/Orin SoC uses A78 cores, it seems very likely that T239 is using an 8 core A78C CPU.

Furthermore, the fact that T239 support is being added to the Linux kernel means that Nvidia plan to use it in products other than the new Switch. This isn't particularly surprising, as the Mariko chip (T210B01) was also created for Nintendo, but used in other products (Jetson Nano), so Nintendo don't seem to have any problem with their semi-custom parts being used elsewhere. It likely allows them to get a better deal from Nvidia.

Lastly, code being updated to support T239 likely means that there is actually hardware to run this code on. That is, at the very least, engineering samples of T239 already exist, meaning full scale production of the chip is likely to start soon if it hasn't already.

Very nice find!

This seems like Nvidia's planning on releasing a Nvidia Shield TV model that's equipped with Drake?

It could be a Shield product, or Jetson (although that would have some overlap with Jetson Orin NX in performance), or it could be a third party product, like an Android tablet or Chromebook or whatever. Hard to tell, but they definitely expect to use this in non-Nintendo products.

 

[Eclissi]

Can anyone report these news on Kopite's Twitter and ask him what he thinks?

 

[Raccoon]

holy shit news

May 12 is on the menu?

 

[Dakhil]

Can anyone report these news on Kopite's Twitter and ask him what he thinks?

kopite7kimi already knows.

 

[ReddDreadtheLead]

Well done on spotting this!

For anyone unsure about what this means, the important part is this:

We know from a variety of sources that the T239 chip is Drake, ie the one being used in the new Switch model. It having "eight cores in a single cluster" means it has 8 CPU cores (obviously), but doesn't strictly speaking limit what those cores can be, as ARM's DynamicIQ supports clusters with mixes of different cores (eg 4x A78 and 4x A55). However, as the code would almost certainly have to be changed to support these hetergeneous clusters (ie different core types would run at different frequencies), and the commit here doesn't make such changes, we can reasonably assume that the cores are all of the same type. The only big ARM cores which can be configured with 8 cores in a single cluster are the A78 (in the A78C variant) and later, and given the T234/Orin SoC uses A78 cores, it seems very likely that T239 is using an 8 core A78C CPU.

Furthermore, the fact that T239 support is being added to the Linux kernel means that Nvidia plan to use it in products other than the new Switch. This isn't particularly surprising, as the Mariko chip (T210B01) was also created for Nintendo, but used in other products (Jetson Nano), so Nintendo don't seem to have any problem with their semi-custom parts being used elsewhere. It likely allows them to get a better deal from Nvidia.

Lastly, code being updated to support T239 likely means that there is actually hardware to run this code on. That is, at the very least, engineering samples of T239 already exist, meaning full scale production of the chip is likely to start soon if it hasn't already.



It could be a Shield product, or Jetson (although that would have some overlap with Jetson Orin NX in performance), or it could be a third party product, like an Android tablet or Chromebook or whatever. Hard to tell, but they definitely expect to use this in non-Nintendo products.

Considering it has T234 as “.maxcpus_per_clister = 4”

And “.num_clusters = 3”


Unlike what it says for T239:

“.maxcpus_per_clister = 8”

“.num_clusters = 1”

I think that would be important and it seems to delineate them as more than one cluster or not.

 

[Mr Swine]

Doesn’t Nvidia have a contract with Samsung that they have bought production lines on their 8nm fabrication?

I guess it would only be used for their Car automation SoC and not for gaming as they have moved back to TMSC for the 40xx series?

 

[Bonejack]

I hereby officially request Cybergatuno getting a custom tag: "1st Post Hall of Fame"

holy shit news

May 12 is on the menu?

Maybe even sooner? (Speculation and legit question, i don't know shit about this whole scene anymore)

E: (Clarification: Sooner, but not as in CY 2022, that is out of the question.)

 

[Dark Cloud]

We’re getting oh so much closer to the reveal. I just want to know when it’s going to be revealed

 

[Thraktor]

Considering it has T234 as “.maxcpus_per_clister = 4”

And “.num_clusters = 3”


Unlike what it says for T239:

“.maxcpus_per_clister = 8”

“.num_clusters = 1”

I think that would be important and it seems to delineate them as more than one cluster or not.

It's a single cluster, but the ARM DynamicIQ cluster can contain more than one core type. That is, when you have a phone with 1 X1 core, 3 A78 core and 4 A55 cores, they're all contained within a single DynamicIQ cluster, they're not treated as 3 different clusters.

However, as I mentioned in my post, if the cluster was actually heterogeneous, say 4x A78 and 4x A55, then they would almost certainly have to make much bigger code changes to accommodate multiple core types within the cluster, so I'd still say it's a safe bet that they're homogenous, and therefore A78C.

 

[Simba1]

Given 8 cores and possibly 5/6 nm, how might Drake compare to other consoles like PS4 or Series S?

Like people said, clock is big factory.
Drake CPU clock will probably be somewhere around 1.5GHz, that should mean its noticeable stronger than PS4 Jaguar 1.6GHz CPU,
but noticeable weaker than Xbox Series S Zen 2 3.6GHz CPU (thats basically full desktop CPU).

 

[LoneRanger]

Octa Core CPU is the best possible outcome. Hopefully there is 1 for OS and 7 for developers.

But I would still as always said temper my expectation regarding clocks. My expect for CPU/GPU are the same Switch clocks. 1GHZ for CPU and 768/307 MHz for GPU in docked/portable modes. I know that would be a 8x jump in CPU + 6x jump in GPU + DLSS + new architecture + 3x RAM jump

 

[ReddDreadtheLead]

It's a single cluster, but the ARM DynamicIQ cluster can contain more than one core type. That is, when you have a phone with 1 X1 core, 3 A78 core and 4 A55 cores, they're all contained within a single DynamicIQ cluster, they're not treated as 3 different clusters.

However, as I mentioned in my post, if the cluster was actually heterogeneous, say 4x A78 and 4x A55, then they would almost certainly have to make much bigger code changes to accommodate multiple core types within the cluster, so I'd still say it's a safe bet that they're homogenous, and therefore A78C.

Just wanted to address the first part: is it? I mean there’s some geekbench that would outright delineate as three clusters but sees it as 1 processor 8 cores.

Like this for example:

Wouldn’t this also apply here?

 

[Z0m3le]

Like people said, clock is big factory.
Drake CPU clock will probably be somewhere around 1.5GHz, that should mean its noticeable stronger than PS4 Jaguar 1.6GHz CPU,
but noticeable weaker than Xbox Series S Zen 2 3.6GHz CPU (thats basically full desktop CPU).

depends on the node used, if it's 4N, I think we could easily see over 2GHz, but yes it will be weaker than Series S' CPU, however games pushing these CPUs to 100% is very unlikely, ports should be pretty easy with 7 A78 cores running at 1.6GHz+.

Octa Core CPU is the best possible outcome. Hopefully there is 1 for OS and 7 for developers.

But I would still as always said temper my expectation regarding clocks. My expect for CPU/GPU are the same Switch clocks. 1GHZ for CPU and 768/307 MHz for GPU in docked/portable modes.

Portable mode is actually 460MHz, 307MHz is just the minimum game clock, there are big portable games like Zelda and Mario that use 460MHz which means 1.41TFLOPs is likely the minimum handheld spec.

 

[Procheno]

This all sounds super exciting. Hoping for the best with this hardware.

I do want to double check though for precaution: Are we 100% certain about the leaked NVIDIA specs and 8 core cluster speculation being up to date and accurate? Could it be that the specs leaked from NVIDIA were just dev kits specs with extra power (like how most dev kits have more ram than what the actual hardware comes with)?

And with everything seemingly lining up in terms of "expectations", is it possible we're assuming a logical outcome of what hardware Nintendo would "have to use" in unison with these other things but Nintendo may adjust the specs to be cheaper despite less efficient battery power or obvious bottleneck?

And the T239, are we 100% sure it's for the Switch Drake?

I'm just covering the bases to keep my expectations in check because this all sounds so good.... too good in fact lol. I want to belieeeve.

 

[Stinky Horse]

This all sounds super exciting. Hoping for the best with this hardware.

I do want to double check though for precaution: Are we 100% certain about the leaked NVIDIA specs and 8 core cluster speculation being up to date and accurate? Could it be that the specs leaked from NVIDIA were just dev kits specs with extra power (like how most dev kits have more ram than what the actual hardware comes with)?

And with everything seemingly lining up in terms of "expectations", is it possible we're assuming a logical outcome of what hardware Nintendo would "have to use" in unison with these other things but Nintendo may adjust the specs to be cheaper despite less efficient battery power or obvious bottleneck?

And the T239, are we 100% sure it's for the Switch Drake?

I'm just covering the bases to keep my expectations in check because this all sounds so good.... too good in fact lol. I want to belieeeve.

Drake is just the codename for the T239 SoC.
If Nintendo are still updating their platform next year it will use that.

 

[Bonejack]

How about a new OT subtitle? "Octocore Traveler"?

 

[kvetcha]

This all sounds super exciting. Hoping for the best with this hardware.

I do want to double check though for precaution: Are we 100% certain about the leaked NVIDIA specs and 8 core cluster speculation being up to date and accurate? Could it be that the specs leaked from NVIDIA were just dev kits specs with extra power (like how most dev kits have more ram than what the actual hardware comes with)?

And with everything seemingly lining up in terms of "expectations", is it possible we're assuming a logical outcome of what hardware Nintendo would "have to use" in unison with these other things but Nintendo may adjust the specs to be cheaper despite less efficient battery power or obvious bottleneck?

And the T239, are we 100% sure it's for the Switch Drake?

I'm just covering the bases to keep my expectations in check because this all sounds so good.... too good in fact lol. I want to belieeeve.

Drake is just the internal codename for T239.

The NVidia leak showed that NVN2 was expecting a full 12SM GPU. No wiggle room. So that's more or less in stone.

The 8-core cluster commit was just made, like, yesterday.

How about a new OT subtitle? "Octocore Traveler"?

sickos.jpg

 

[Simba1]

depends on the node used, if it's 4N, I think we could easily see over 2GHz But yes it will be weaker than Series S' CPU, however games pushing these CPUs to 100% is very unlikely, ports should be pretty easy with 7 A78 cores running at 1.6GHz+.

Regardless node, I dont think Nintendo will use 2GHz or more in any case, IMO around 1.5GHz is most realistic expectation.
Current cross gen games still definitely not pushing PS5/XsX/ CPUs, but by time more and more games will start pushing CPU,
saying that, having on mind rest of Drake configuration, Drake hardware should be well rounded and capable without big bottlenecks.

 

[Z0m3le]

This all sounds super exciting. Hoping for the best with this hardware.

I do want to double check though for precaution: Are we 100% certain about the leaked NVIDIA specs and 8 core cluster speculation being up to date and accurate? Could it be that the specs leaked from NVIDIA were just dev kits specs with extra power (like how most dev kits have more ram than what the actual hardware comes with)?

And with everything seemingly lining up in terms of "expectations", is it possible we're assuming a logical outcome of what hardware Nintendo would "have to use" in unison with these other things but Nintendo may adjust the specs to be cheaper despite less efficient battery power or obvious bottleneck?

And the T239, are we 100% sure it's for the Switch Drake?

I'm just covering the bases to keep my expectations in check because this all sounds so good.... too good in fact lol. I want to belieeeve.

I'll just make it simple. Nvidia was hacked, from that hack, we saw that Switch 2 will use T239 codenamed "Drake" (this is a Nvidia codename). This 8 core leak is from September 5th, just 2 weeks ago, and is real, this is the linux kernal, which also means that this is a chip in production right now, as os support for a CPU points to real hardware.

From the above, we can know that the specs are
1536 Cuda cores (Steam Deck has 512 cores for comparison) also 48 Tensor cores and 12 RT cores.
8 core A78C CPU (better IPC than Ryzen 2, but will have a lower clock than current gen consoles)
128bit memory bus, 88GB/s - 102GB/s for LPDDR5, which is the memory type based on Orin's relation to T239.

 

[kvetcha]

Considering the miracles a lot of devs have worked with the Erista/Mariko CPU I can't imagine that Drake's CPU is going to be a major bottleneck in 99% of games.

 

[LoneRanger]

Portable mode is actually 460MHz, 307MHz is just the minimum game clock, there are big portable games like Zelda and Mario that use 460MHz which means 1.41TFLOPs is likely the minimum handheld spec.

DF first reported 307MHz clock for portable GPU. I always understand that nintendo’s originally idea was to use that frequency number but wasn’t enough for some game and decided to let developers bump it to 460MHz if they want.

Now next Switch HW has a more comfortable HW so I think they will let developers clock GPU from 307 to 460MHz as OG Switch. That gives a range from 0.94 to 1.41 TFlops as you mentioned. We don’t know next HW screen resolution. If its still 720p, that’s enough power to render all actual Switch titles at max resolution with lowest clock, and therefore, saved more battery life. I think with DLSS it would also be possible to port any PS4 game that render between 900-1080p native to next Switch HW at 720p portable resolution. In docked, it would be 4k with DLSS.

 

[mjayer]

Hey guys. Still team 2022 reveal over here, with a small chance of holiday 2022 release! The hype never dies. (Well, it’ll die after October )

Regardless node, I dont think Nintendo will use 2GHz or more in any case, IMO around 1.5GHz is most realistic expectation.
Current cross gen games still definitely not pushing PS5/XsX/ CPUs, but by time more and more games will start pushing CPU,
saying that, having on mind rest of Drake configuration, Drake hardware should be well rounded and capable without big bottlenecks.

It can’t be regardless of node! No one just leaves untappped potential lying on the ground. If there is overhead to go higher for the best efficiency of the chip, it will.

 

[Bonejack]

Considering the miracles a lot of devs have worked with the Erista/Mariko CPU I can't imagine that Drake's CPU is going to be a major bottleneck in 99% of games.

Mihoyo devs reading this info and immediately calling Nintendo for devkits. (I think Genshin Impact is very CPU-demanding)

 

[Ojoloco]

 

[Z0m3le]

Regardless node, I dont think Nintendo will use 2GHz or more in any case, IMO around 1.5GHz is most realistic expectation.
Current cross gen games still definitely not pushing PS5/XsX/ CPUs, but by time more and more games will start pushing CPU,
saying that, having on mind rest of Drake configuration, Drake hardware should be well rounded and capable without big bottlenecks.

Yeah, I disagree with the 2GHz number, we have heard in the past that Nintendo wanted to push A57 above 2GHz, but more importantly, we've seen them push A57 on 20nm to 1.78GHz, that is pretty close to 2GHz, so Nintendo does want the CPU performance, it's also worth noting that because of DLSS, the GPU's performance just isn't as important as the CPU's IMO.

DF first reported 307MHz clock for portable GPU. I always understand that nintendo’s originally idea was to use that frequency number but wasn’t enough for some game and decided to let developers bump it to 460MHz if they want.

Now next Switch HW has a more comfortable HW so I think they will let developers clock GPU from 307 to 460MHz as OG Switch. That gives a range from 0.94 to 1.41 TFlops as you mentioned. We don’t know next HW screen resolution. If its still 720p, that’s enough power to render all actual Switch titles at max resolution with lowest clock, and therefore, saved more battery life. I think with DLSS it would also be possible to port any PS4 game that render between 900-1080p native to next Switch HW at 720 portable resolution.

Sure it could possibly go lower, but when discussing the specs of a device like this, the higher number that is possible, is the one that should be used, so if they allow 460MHz, that is the spec we would use for handheld mode, if it's not accessible for developers to use, then that is another story, however because Switch games should run on this device, letting it run at the proper clock speed the game was designed for, should help with compatibility of poorly coded games, given Switch's giant catalog, I have to assume some badly coded games exist.

 

[Dakhil]

Thank you for sharing my tweet. (I'm not being sarcastic here.)

 

[olobolger]

I can't wait to get my hands on this device, pretty exciting stuff!!

 

[Skittzo]

This all sounds super exciting. Hoping for the best with this hardware.

I do want to double check though for precaution: Are we 100% certain about the leaked NVIDIA specs and 8 core cluster speculation being up to date and accurate? Could it be that the specs leaked from NVIDIA were just dev kits specs with extra power (like how most dev kits have more ram than what the actual hardware comes with)?

And with everything seemingly lining up in terms of "expectations", is it possible we're assuming a logical outcome of what hardware Nintendo would "have to use" in unison with these other things but Nintendo may adjust the specs to be cheaper despite less efficient battery power or obvious bottleneck?

And the T239, are we 100% sure it's for the Switch Drake?

I'm just covering the bases to keep my expectations in check because this all sounds so good.... too good in fact lol. I want to belieeeve.

The Nvidia hack showed the API, which is the interface devs use when making games on the device. Well, one of them.

This API sees that the end target has 12SMs, so there is no reason to think that configuration could be just for devkits. It's specifically showing that the development software itself looks for that configuration for any games developed on the device.

 

[merp]

No, you didn't. That shit is indeed fire and deserves to be in this thread

I actually did mean to post it elsewhere, in the Nintendo Gaming Discussion, because that's where you can post "off-topic" stuff.

 

[LinkURL]

is there currently any game on the market that would need the power of a PS4 + DLSS for 720p?

 

[Procheno]

Drake is just the codename for the T239 SoC.
If Nintendo are still updating their platform next year it will use that.

Drake is just the internal codename for T239.

The NVidia leak showed that NVN2 was expecting a full 12SM GPU. No wiggle room. So that's more or less in stone.

The 8-core cluster commit was just made, like, yesterday.


sickos.jpg

I'll just make it simple. Nvidia was hacked, from that hack, we saw that Switch 2 will use T239 codenamed "Drake" (this is a Nvidia codename). This 8 core leak is from September 5th, just 2 weeks ago, and is real, this is the linux kernal, which also means that this is a chip in production right now, as os support for a CPU points to real hardware.

From the above, we can know that the specs are
1536 Cuda cores (Steam Deck has 512 cores for comparison) also 48 Tensor cores and 12 RT cores.
8 core A78C CPU (better IPC than Ryzen 2, but will have a lower clock than current gen consoles)
128bit memory bus, 88GB/s - 102GB/s for LPDDR5, which is the memory type based on Orin's relation to T239.

Okay, and stupid question, but there's also no chance the T239 representing Drake has changed since the leaks occurred? Probably because I assume it's way too big of a shift to do this late and/or no odd signs have been shown to show that indeed be the case?

And also the 8 core thing is not for some other device that happens to use the same tech?

Really pulling for that H1 2023

 

[Skittzo]

Okay, and stupid question, but there's also no chance the T239 representing Drake has changed since the leaks occurred? Probably because I assume it's way too big of a shift to do this late and/or no odd signs have been shown to show that indeed be the case?

And also the 8 core thing is not for some other device that happens to use the same tech?

Really pulling for that H1 2023

Do you mean the codename itself has changed or is now referring to something else? That seems extremely unlikely.

Could the specs of T239 have changed since February? Maybe, but that's also fairly unlikely considering how far along it seemingly was. But possible.

 

[Z0m3le]

Okay, and stupid question, but there's also no chance the T239 representing Drake has changed since the leaks occurred? Probably because I assume it's way too big of a shift to do this late and/or no odd signs have been shown to show that indeed be the case?

And also the 8 core thing is not for some other device that happens to use the same tech?

Really pulling for that H1 2023

The Nvidia leak was up to date as of mid February this year.
The 8 core CPU leak is from 2 weeks ago.
They specify the same chip (Tegra239) which is the known SoC for Switch 2 as we explained in the other posts, so no, it's not a different chip, they both identify as T239, which is one specific Tegra chip that we have confirmed from that Nvidia leak, is the SoC Nintendo is using in the next device... and it's in production right now, this lines up with rumors from Nate about the new Switch and software for it coming in 1H 2023. Switch actual begun production at the end of September 2016 for a March 2017 release, so if production of Switch "2" started right now, it would be ready for Spring next year, my guess is alongside Zelda.

Do you mean the codename itself has changed or is now referring to something else? That seems extremely unlikely.

Could the specs of T239 have changed since February? Maybe, but that's also fairly unlikely considering how far along it seemingly was. But possible.

Because of testing and tape out, Feb info would be the current spec for finished hardware which is what this leak is telling us. The chip is already in production, so no, it's not just unlikely, there is no turn around time fast enough to go through testing and release a month+ ago in time for this email to make sense... specs couldn't have changed IMO.