• 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!)

Was Dane even in existence? Does that mean Drake is TSMC?

Well, its always possible Nintendo was working on 2 projects in parallel under the t239 moniker, before they decided to go all in on the more powerful version. But more likely, kopite was just wrong about the code name Dane, and 8nm.
Dane shows up in the leaks, so yea, it definitely existed. I assume Kopite7Kimi's info at the time was that dane was based on Orin. given when Orin was first created, a branch of it as a successor to Mariko was tested. but then Nvidia ditched Samsung due to foundry issues and explored better performing designs with TSMC, leading to Drake
 
My understanding is that games have to be built around NPLN - it replaces the underlying network protocol - and that NPLN is being introduced slowly into Nintendo games and some key partners until it can become the "default."

Even with NPLN upgrading the core technology, there is also the question of how much money Nintendo has sunk into buying capacity from Google.
If I’m not mistaken, NPLN runs on AWS, no?
 
Dane shows up in the leaks, so yea, it definitely existed. I assume Kopite7Kimi's info at the time was that dane was based on Orin. given when Orin was first created, a branch of it as a successor to Mariko was tested. but then Nvidia ditched Samsung due to foundry issues and explored better performing designs with TSMC, leading to Drake
While it was mentioned a single time in a single file scrubbing certain strings, imo that’s very thin evidence to make all these assumptions.

They changed process node and basically created an entirely new chip, while still keeping t239?

Imo the Occam’s razor says it was always 12sm.
 
It seems safe to just assume that Dane existed and was 8nm but was turned into Drake on a newer process.
I don't think there's any reason to believe Kopite was wrong.
 
0
They changed process node and basically created an entirely new chip, while still keeping t239?
It's not basically an entirely new chip. A lot of Dane was likely the same, due to it being based on Orin and having the same form factor and functionality of the switch. It's more accurate to say Drake is a modified version of Dane, assuming Dane ever existed. The reason why we believe Dane did exist is because Kopite has been right about everything else.
 
It's not basically an entirely new chip. A lot of Dane was likely the same, due to it being based on Orin and having the same form factor and functionality of the switch. It's more accurate to say Drake is a modified version of Dane, assuming Dane ever existed. The reason why we believe Dane did exist is because Kopite has been right about everything else.
I’m not dying on this hill. Signing out.
 
Hey if you doubt Kopite that's fine by me man, no need for an argument there. I think he's legit though he really has gotten everything right so far.

It is a new chip even if it’s based on the same Uarch. Ee don’t know anything about Dane other than being based on Orin and on Samsungs 8nm node. It’s like saying that the RTX3060 is the same as the 3080
 
While it was mentioned a single time in a single file scrubbing certain strings, imo that’s very thin evidence to make all these assumptions.

They changed process node and basically created an entirely new chip, while still keeping t239?

Imo the Occam’s razor says it was always 12sm.
Yeah, there's no evidence T239 was ever anything but Drake. Changing something like the SM count while keeping the chip as T239 seems highly unlikely. And as for the process node, I find it silly to bother speculating when there's no evidence at all about the current node (besides Kopite seemingly guessing based on T234) or the notion that it changed at some point.

We should also remain highly skeptical of using the open-scare-quotes "performance of Drake" close-scare-quotes as justification for any theories about its development, when we don't actually know what its performance will be like.
 
since it's largely up to Nvidia, methinks the choice was already made for Nintendo. that could explain why Drake is so much more performant than we expected. Dane may have been an 8nm design (unless the leak has details about that) prior to the decision to move
They also have an article on the Ada leaks here, mostly around estimating die sizes for each of the chips. They do think that the increased density of TSMC N4 will result in lower costs for Nvidia, despite higher wafer costs:
The rest of the Ada lineup comes away a lot tamer in terms of die sizes and overall BOM. Performance should generally be above Ampere at the same power with a decently lower cost to fabricate despite much higher wafer costs.
I think Samsung's 5LPE process node is still a possibility since Samsung's 5LPE process node seems decent enough where Qualcomm seems to have used Samsung's 5LPE process node for the entire duration of the Snapdragon 888's lifecycle, which can't really be said about the Snapdragon 8 Gen 1 since Qualcomm's rumoured to switch to TSMC's N4 process node for the fabrication of the Snapdragon 8 Gen 1 for 2H 2022. And kopite7kimi did say that T239 is a custom version of Orin, which the illegal Nvidia leaks does at least seem to imply. Considering Orin's probably fabricated using Samsung's 8N process node, I imagine there's definitely a non-trivial amount of money saved by re-using Samsung's IPs. And I think Samsung's 5LPP process node and more advanced process nodes seem to be more problematic than Samsung's 5LPE process node when yields are concerned, especially with Samsung's 4LPX process node that Qualcomm used for the fabrication of the Snapdragon 8 Gen 1 not really having any 4 nm** features in comparison to Samsung's 4LPE process node used for the fabrication of the Exynos 2200, which does imply Samsung's 4LPX process node's probably a custom version of Samsung's 5LPP process node. And rumours say that the yield rate for the Snapdragon 8 Gen 1 is ~35%, with the Exynos 2200 having a yield rate lower than 35%, which is definitely very problematic for high volume chips, which I think Drake definitely is. But saying that, I think the probability of Samsung's 5LPE process node being used for the fabrication of Drake is not as high as I've originally thought.

And although this seems to not be directly related to the DLSS model*, I wonder if Nvidia's going to take a little bit of a page from Nvidia's plans with the Pascal GPUs and indirectly from AMD's plans with the RDNA 3 GPUs, where the high-end and/or mid-range Ada GPUs are fabricated using TSMC's 4N process node, which kopite7kimi seems to imply is closer to TSMC's N5P process node than TSMC's N4 process node, which isn't necessarily a bad thing, since TSMC's N5P process node is slightly more power efficient than TSMC's N4 process node; and the mid-range and/or entry-level Ada GPUs are fabricated using TSMC's N5 process node, especially with kopite7kimi mentioning that all Ada GPUs are fabricated using TSMC's N5 process node, and TSMC's 4N process node is technically still part of TSMC's 5 nm** family. The reason why I'm asking is because assuming Nintendo has plans to die shrink Drake in the future, the process node Nvidia uses for the mid-range and/or entry-level Ada GPUs could be the same process node used to die shrink Drake in the future, especially if TSMC's N6 process node is used for the fabrication of Drake initially. TSMC's N6 process node used for the fabrication of Drake.

~* → a tentative name that I use~
~** → a marketing nomenclature used by all foundry companies~

They changed process node and basically created an entirely new chip, while still keeping t239?
I don't think the illegal Nvidia leaks have mentioned anything about which process node is being used. All the talk about which process node is being used has been purely educated speculations.
 
@NateDrake

Hey Nate, you mentioned you want to discuss FSR on Switch before the Leak, will you pack this into the Leak episode or another own dedicated one?

Plus: Can you tell us, when we can expect the release of this highly anticipated episode? I think the most of us are dying to hear the new info 😂😂👍😊
 
@NateDrake

Hey Nate, you mentioned you want to discuss FSR on Switch before the Leak, will you pack this into the Leak episode or another own dedicated one?

Plus: Can you tell us, when we can expect the release of this highly anticipated episode? I think the most of us are dying to hear the new info 😂😂👍😊
FSR will be its own dedicated episode, but I'm thinking of adding a secondary talking point to it. We need to schedule a recording time for it still and set a time that'll work for the intended guest.
 
FSR will be its own dedicated episode, but I'm thinking of adding a secondary talking point to it. We need to schedule a recording time for it still and set a time that'll work for the intended guest.


Cool - thank you!

Regarding the Episode release time:

I rather meant the NVN2 episode ( correctly mentioned)? 😅

I personally expect them - if it will be a Late 2022 release - to show it / announce it just like the OLED version last year in June.
 
Cool - thank you!

Regarding the Episode release time:

I rather meant the NVN2 episode ( correctly mentioned)? 😅

I personally expect them - if it will be a Late 2022 release - to show it / announce it just like the OLED version last year in June.
NVN2 will not be addressed until I get confirmation/deconfirmation on the info I'm checking in on. While I have a solid lead on it, I'd like more to assure no mishap.
 
Dane shows up in the leaks, so yea, it definitely existed. I assume Kopite7Kimi's info at the time was that dane was based on Orin. given when Orin was first created, a branch of it as a successor to Mariko was tested. but then Nvidia ditched Samsung due to foundry issues and explored better performing designs with TSMC, leading to Drake
I thought they can’t use Samsung foundry designs on TSMC and vice versa due to legal issues? (Proprietary foundry processes used with the design, etc)

What about costs? Would TSMC make it that much more of sn expensive chip? Not to mention all of the arguments prior as to why Nvidia/Nintendo wouldn’t likely go TSMC…
 
FSR will be its own dedicated episode, but I'm thinking of adding a secondary talking point to it. We need to schedule a recording time for it still and set a time that'll work for the intended guest.
Zack Snyder’s Nate the Hate. 4 hours of pure Drake discussion. Presented in 4:3 aspect ratio to preserve speculative intent.
 
I thought they can’t use Samsung foundry designs on TSMC and vice versa due to legal issues?

What about costs? Would TSMC make it that much more of sn expensive chip? Not to mention all of the arguments prior as to why Nvidia/Nintendo wouldn’t likely go TSMC…
they would have to redesign the chip. but if Nintendo never intended on a successor for a couple years from when they first started the design phase, this probably wouldn't matter too much.

as for costs, they could be amoratized if using the same node as Lovelace. might be more expensive than 8nm, but they save time by not having to make a jump in the future, as well as simply getting a significantly better product for the price
 
0
Considering NVidia has engineers at both Samsung and TSMC foundries, therefore offering flexibility, a chip can be built at either foundry. Whether NVidia/Nintendo intended it to be at TSMC or at Samsung at the start is a different question.


As an aside, I believe that while these chips likely will not clock very high or close to high, as in +2.5GHz for the CPU or +1.5GHz for the GPU in the docked mode, the efficiency and yield concern of Samsung foundries should be less apparent.

Even though SEC has yielded less than stellar results, I believe it it due to theirs not being meant for clocking so high but can have an acceptable and comparable curve to the TSMC equivalent.

So, this is just a theory so do not quote me as an exact with this, but SEC 5nm with CPU @ 2GHz is probably equal to TSMC 5nm at 2GHz in power draw and performance… while being pretty close in terms of heat…. Maybe?

But SEC 5nm with CPU at 2.5GHz is probably closer to TSMC 7nm CPU at 2.5GHz than a 5nm equivalent from TSMC in terms of power draw and performance. While also being hotter, maybe?


Again, I’m not very knowledgeable with this aspect and it is only in theory, in essence about the curve and the drop off of the SEC equivalent compared to a TSMC counterpart.

It could be that I’m pretty off on this deduction. But this is of course still just speculation.
 
Zack Snyder’s Nate the Hate. 4 hours of pure Drake discussion. Presented in 4:3 aspect ratio to preserve speculative intent.
Sounds like the perfect episode for people to hate.

I got enough hate last night for saying the BotW2 topic of it being too much for Switch was an unneeded conversation since DF presented it as opinion and speculation.

Ironic thing being: so many within the fan community say Nintendo doesn't need new hardware; yet, they are the most vocal saying the footage is not on Switch. Claiming the latter is admittance that new hardware must be soon.
 
I thought they can’t use Samsung foundry designs on TSMC and vice versa due to legal issues? (Proprietary foundry processes used with the design, etc)

What about costs? Would TSMC make it that much more of sn expensive chip? Not to mention all of the arguments prior as to why Nvidia/Nintendo wouldn’t likely go TSMC…
You're right about the legal issues. The work around is to employ separate groups of people to work with the different foundries' IPs. IE, throw money at the problem.
Of course, remember, back before the Nvidia hack revealed the API's looking at 12 SMs, one of the guiding assumptions we operated under was 'try to minimize cost'. So designing in parallel wouldn't have crossed our minds as an option.

Whether opting for a more recent TSMC node would make it more expensive than our default Samsung 8 nm assumption kind of goes back and forth... Thraktor did roughly calculate that in a post-12 SM reveal light, the density improvements actually tilt things in favor of the best node(s). That said, the balance may change again if Samsung foundries really has lost so much leverage to resort to charging by the usable die instead of by the wafer.
Of course, the initial design cost will be higher; they always are with each successive node. But it still may pay for itself in the long run.
 
0
Sounds like the perfect episode for people to hate.

I got enough hate last night for saying the BotW2 topic of it being too much for Switch was an unneeded conversation since DF presented it as opinion and speculation.

Ironic thing being: so many within the fan community say Nintendo doesn't need new hardware; yet, they are the most vocal saying the footage is not on Switch. Claiming the latter is admittance that new hardware must be soon.
Unless you are referring to the same person or persons holding both those views, my only comment is the fan community is quite large and there is definately a 'no new switch is needed' clique , i found this out posting on reddit where any reference to a Pro or a more powerful switch gets immediate downvotes. I think they are a small minority however, maybe very active online. In anycase most people usually hold internally consistent views. Like if I want to see a new Switch , I would be fairly excited at DF speculation BOTW2 footage could be from the new hardware.
 
So, this is just a theory so do not quote me as an exact with this, but SEC 5nm with CPU @ 2GHz is probably equal to TSMC 5nm at 2GHz in power draw and performance… while being pretty close in terms of heat…. Maybe?

But SEC 5nm with CPU at 2.5GHz is probably closer to TSMC 7nm CPU at 2.5GHz than a 5nm equivalent from TSMC in terms of power draw and performance. While also being hotter, maybe?
The first paragraph seems about right, considering when comparing the Cortex-A55 cores on the Snapdragon 865 vs on the Snapdragon 888, with the same 1.8 GHz frequency, the same amount of L2 cache (128 KB per CPU core), and with the only difference being the process node being used (TSMC's N7P process node for the Snapdragon 865 vs Samsung's 5LPE process node for the Snapdragon 888), Samsung's 5LPE process node does seem to practically be on par with TSMC's N7P process node in terms of performance and power efficiency, but probably when the frequency is ≤2 GHz.

An interesting comparison to make – and probably one of the rare ones we're actually able to achieve today, is the comparison between the Cortex-A55 cores inside of both the Snapdragon 865 and the new Snapdragon 888. Both SoCs feature the same IP cores, clock them at the same 1.8GHz frequency, and both feature the same amount of L2 cache, with their only real difference being their process nodes.

Using SPEC's 456.hmmer – because it's a workload that primarily resides in the lower cache hierarchies and thus, we avoid any impact of the possibly different memory subsystem, we can see that both SoCs' power consumption indeed is almost identical, with performance also being identical with a score of 6.84 versus 6.81 in favour of the new Snapdragon 888.
But for the second paragraph, that's hard to say with complete certainty when power efficiency is concerned, especially when dealing with different amounts of L2 cache. Although Qualcomm has doubled the amount of L2 cache on the Cortex-A78 cores on the Snapdragon 888 vs on the Cortex-A77 cores on the Snapdragon 865 (512 KB per CPU core vs 256 KB per CPU core), with the same frequency of 2.42 GHz, the Cortex-A78 cores on the Snapdragon 888 are 4.9-8.9% more performant than the Cortex-A77 cores on the Snapdragon 865, but at the cost of consuming 24% more power. Although I don't know how big the power consumption decrease would be if Qualcomm decided to have the Cortex-A78 cores on the Snapdragon 888 have the same amount of L2 cache as the Cortex-A77 cores on the Snapdragon 865 (256 KB per CPU core), I imagine the power consumption increase of the Cortex-A78 cores on the Snapdragon 888 vs the Cortex-A77 cores on the Snapdragon 865 would still be considerably high. So my guess is that Samsung's 5LPE process node would consume ≥10% more power than TSMC's N7P process node when the frequency is ~2.5 GHz.

The Cortex-A78 cores of the Snapdragon 888 are 4.9% and 8.9% faster than the Cortex-A77 middle cores of the Snapdragon 865. The power consumption comparison here isn't apples-to-apples due to the new cores doubling up on the L2 cache. Arm states the A78 has an +7% IPC improvement and a -4% power reduction versus the A77. The Snapdragon 888's middle cores however use +24% more power. Excluding the theory that that doubled L2 cache significantly raises power, we're probably still seeing a notable process node power efficiency difference between Samsung's 5LPE node and TSMC's N7P node, with the Samsung node still falling behind.
 
Last edited:
Unless you are referring to the same person or persons holding both those views, my only comment is the fan community is quite large and there is definately a 'no new switch is needed' clique , i found this out posting on reddit where any reference to a Pro or a more powerful switch gets immediate downvotes. I think they are a small minority however, maybe very active online. In anycase most people usually hold internally consistent views. Like if I want to see a new Switch , I would be fairly excited at DF speculation BOTW2 footage could be from the new hardware.
Every week in my comment section, tweet replies, or in Spawncast comments... there are people saying, "there's no Switch coming until 20xx" and that all the Pro talk was a lie. It seems to physically pain these individuals. But they also create narratives that are untrue & will claim I've reported on the Pro since 2016.
 
Every week in my comment section, tweet replies, or in Spawncast comments... there are people saying, "there's no Switch coming until 20xx" and that all the Pro talk was a lie. It seems to physically pain these individuals. But they also create narratives that are untrue & will claim I've reported on the Pro since 2016.
I see posts like this every time in the comment section of your videos.😅
Even in Japan, Whenever Nate's comments are discussed, they are filled with posts that say "if you keep saying it all the time, you hit someday", making it impossible to have a discussion.
 
FSR will be its own dedicated episode, but I'm thinking of adding a secondary talking point to it. We need to schedule a recording time for it still and set a time that'll work for the intended guest.
Honestly, is there that much to say about it? It’s an open source image scaling solution that Nintendo took advantage of. Isn’t that about all there is to it?
 
0
I see posts like this every time in the comment section of your videos.😅
Even in Japan, Whenever Nate's comments are discussed, they are filled with posts that say "if you keep saying it all the time, you hit someday", making it impossible to have a discussion.
Which is amazing since I've only ever started to discuss it in 2021 with the details of devkits being sent out and DLSS functionality.
 
Would a delay to 2023 be a benefit to Nintendo/Nvidia by principle of being on better nodes?

And I don't just mean Nvidia's SoC. Wouldn't the ARM CPU also benefit, or is this particular model of ARM (Cortex-A78) node-specific? At what point can Nintendo still carry the Switch's momentum? I've heard people even say as late as 2024...
 
I don't think the illegal Nvidia leaks have mentioned anything about which process node is being used. All the talk about which process node is being used has been purely educated speculations.

I know, but most of the educated speculation says 12sm is too large and power hungry to make sense for 8nm.
 
I swear the SWOLED gave people PTSD or something. Everyone was hyped and yelling at Nintendo on Twitter about a Switch pro and then the OLED model somehow permanently killed all the excitement. Now they describe it like some emotional trauma, “oh man all this Switch Pro talk the past few years has been so exhausting I can’t take it anymore”.

Like, that’s all it takes? One information mixup? You didn’t even sift through all the reports and information. You just reacted to it, acted like it was happening all along, but when it didn’t turn out exactly like Bloomberg stated, you go all sour grapes on even the idea of more powerful hardware.

Every new report or release of new potential info seems to push back a new Switch a year in these people’s minds. First it was 2023 because of the chip shortage, then 2024 because “Switch is selling so well”, then 2025 because of the “middle of life cycle” comments from Nintendo. Now there is new hardware coming soon, but it’s just a Pro and the REAL, true next-gen successor is coming in 2027 or something.

I’d think any Nintendo or heck even gaming fan would always be pumped at the idea of new hardware, but these days it’s like pulling teeth. And that’s fine and all, if they kept it to themselves. But it feels like us tech enthusiasts and dreamers get shut down at the mere mention of the next thing from Nintendo.

And don’t even get me started on the whole “successor vs revision” talk that any discussion inevitably spirals into. Or how Nintendo will pull another “Nintendo”, so it’s “best to keep expectations in check”. Why don’t you want to discuss it? Everyday we get closer to the end of the Switch’s lifecycle and Nintendo doesn’t have the luxury of adjusting their plans that likely started in 2019 on the fly in reaction to pandemics that happen to boost demand beyond the natural curve.

New hardware is coming whether you like it or not.
 
Last edited:
Would a delay to 2023 be a benefit to Nintendo/Nvidia by principle of being on better nodes?

And I don't just mean Nvidia's SoC. Wouldn't the ARM CPU also benefit, or is this particular model of ARM (Cortex-A78) node-specific? At what point can Nintendo still carry the Switch's momentum? I've heard people even say as late as 2024...
Considering Nvidia has 4nm products coming out this year, no. There wouldn't be any difference in nodes
 
0
After 3 weeks with my homebrewed oled switch, playing games in battery handheld. Texture patched, 60 fps unlocks, cheats, etc with maximum OC in CPU and GPU. Games looks amazing, but battery is also very good. Aprox 2 minutes per porcent, or 3 hours and 20 mins at 1785/921/1600. This is better than OG switch launch battery with less than half power. Almost everything is "fixed", xeno 2 looks awesome, the witcher 3 doom eternal and wolfenstein 2 runs at 60 fps, tony hawk looks ultraphenomenal. But why nintendo dont want to just oc the oled with this batteries performance? IDK. But im quite sure that the oled is more than capable to bring us a respetable upgrade, especially in handheld mode. So everything I wanted to be fixed is already fixed for me atleast, Switch 2/pro can only open to even more amazing results. My hope is to nintendo goes for 3 hours mark battery like they did with og switches. Oled switch is awesome, but only if you can get it homebrewed, the most radical one for me was tonyhawk. I looks like Im playing on my ps5 but in a portable 7" oled screen at 720p 30 fps lock( big texture patch) and I love it.
 
After 3 weeks with my homebrewed oled switch, playing games in battery handheld. Texture patched, 60 fps unlocks, cheats, etc with maximum OC in CPU and GPU. Games looks amazing, but battery is also very good. Aprox 2 minutes per porcent, or 3 hours and 20 mins at 1785/921/1600. This is better than OG switch launch battery with less than half power. Almost everything is "fixed", xeno 2 looks awesome, the witcher 3 doom eternal and wolfenstein 2 runs at 60 fps, tony hawk looks ultraphenomenal. But why nintendo dont want to just oc the oled with this batteries performance? IDK. But im quite sure that the oled is more than capable to bring us a respetable upgrade, especially in handheld mode. So everything I wanted to be fixed is already fixed for me atleast, Switch 2/pro can only open to even more amazing results. My hope is to nintendo goes for 3 hours mark battery like they did with og switches. Oled switch is awesome, but only if you can get it homebrewed, the most radical one for me was tonyhawk. I looks like Im playing on my ps5 but in a portable 7" oled screen at 720p 30 fps lock( big texture patch) and I love it.
Mariko is truly a much more capable chip than it’s getting credit for.

There is no wonder there were pro rumors in 2019, because Mariko could easily have been the pro chip if Nintendo had wanted it too:
 
After 3 weeks with my homebrewed oled switch, playing games in battery handheld. Texture patched, 60 fps unlocks, cheats, etc with maximum OC in CPU and GPU. Games looks amazing, but battery is also very good. Aprox 2 minutes per porcent, or 3 hours and 20 mins at 1785/921/1600. This is better than OG switch launch battery with less than half power. Almost everything is "fixed", xeno 2 looks awesome, the witcher 3 doom eternal and wolfenstein 2 runs at 60 fps, tony hawk looks ultraphenomenal. But why nintendo dont want to just oc the oled with this batteries performance? IDK. But im quite sure that the oled is more than capable to bring us a respetable upgrade, especially in handheld mode. So everything I wanted to be fixed is already fixed for me atleast, Switch 2/pro can only open to even more amazing results. My hope is to nintendo goes for 3 hours mark battery like they did with og switches. Oled switch is awesome, but only if you can get it homebrewed, the most radical one for me was tonyhawk. I looks like Im playing on my ps5 but in a portable 7" oled screen at 720p 30 fps lock( big texture patch) and I love it.
Huh? You can homebrew oled? I thought only early model Switch could be homebrewed?
 
Huh? You can homebrew oled? I thought only early model Switch could be homebrewed?
Yes, you can with every switch model incluiding lite. V1 or erista switches are the easiest, you just need a clip. V2 is harder, you need soldering. Oled or v3 is even harder, soldering and some bridges only for ppl with experience soldering switches. V2 and V3 needs a chip than is no longer in production and when it was made, was only meant to be used for V2 and lite models.
 
Last edited:
NVN2 will not be addressed until I get confirmation/deconfirmation on the info I'm checking in on. While I have a solid lead on it, I'd like more to assure no mishap.
Hey Nate, totally understand you wanting to wait on full clarification on things before speaking on it again.

I know this may be a total guesstimate on your part here but do you think it’s something more like weeks or months until you may be able to receive the (de)confirmation you’re looking for?
 
I know, but most of the educated speculation says 12sm is too large and power hungry to make sense for 8nm.
I think there's potentially some truth to the speculation about a SoC with a 12 SM GPU potentially having a large die size and requiring a higher amount of power for optimal performance if fabricated using Samsung's 8N process node, considering how there's so much variance on the number of CPU cores, the number of TPCs (2 SMs per TPC), the CPU frequency, and the GPU frequency, depending on which Jetson AGX Orin model, and the max TDP, especially since Orin's probably fabricated using Samsung's 8N process node.
 
Can we settle this once and for all and have someone do a summary of all the things Nate leaked or teased that ended up being true or false? Resetera used to have a great list so people could see just how trustworthy leakers were, and this way we can point to a list if anyone claims Nate was right or wrong about something
 
Can we settle this once and for all and have someone do a summary of all the things Nate leaked or teased that ended up being true or false? Resetera used to have a great list so people could see just how trustworthy leakers were, and this way we can point to a list if anyone claims Nate was right or wrong about something
Or let’s not.

No insider is 100% all the time. He has gotten enough right so we know he is not a fraud. If you trust his info or not is 100% up to you, I don’t think he cares that much.

Insider stuff shouldn’t be taken that seriously. A lot of it pans out, some of it don’t.
 
Yes, you can with every switch model incluiding lite. V1 or erista switches are the easiest, you just need a clip. V2 is harder, you need soldering. Oled or v3 is even harder, soldering and some bridges only for ppl with experience soldering switches. V2 and V3 needs a chip than is no longer in production and when it was made, was only meant to be used for V2 and lite models.
I see, interesting. Doesn't sound like it'll be possible for me then, would be much too complicated for me I think.
 
0
Has Nintendo ever dropped two revisions of a console in back to back years? I thought they always wanted to let at least two holidays season pass for each big revision. That would mean Q1/Q2 2023 (probably Q1 because Q2 is quite slow sales wise + end of FY) earliest time for Drake to release
Pretty big revision, but GBASP only had one holiday before DS showed up.
 
0
No cap, but I roll my eyes when people bring up "online infrastructure". It's a vague catch all that doesn't explain what exactly is the problem. Nintendo has had poor online services, but the switch is home to damn near every kind of online game, sometimes not even using Nintendo's servers

Well, what’s the best selling, permanently online open world multiplayer game on the switch right now? Or the best selling/performing such game on any Nintendo system for that matter?

When I read devs discussing “online infrastructure issues” regarding Nintendo and their huge persistent online game it’s acknowledging that the Nintendo platform…for a variety of reasons…is inhospitable for being the platform of choice to play such games.

In the case of GTA5, it’s primary motivating factor…why it’s consistently been ported since 2015…is GTA Online engagement. That’s it. That’s the only reason they still bother with GTA5 still.

At the end of the day, they look at Nintendo, especially the Switch, and lament it’s just not a place where people will choose to engage with GTA online. It’s userbase will not choose that for that platform.

So it’s not worth the cost/time/effort to bother with a port.

It’s as simple as that.

This is the primary decision of every 3rd party publisher of whether to bother with the port of their game. Is their enough perceived demand of their product on the Nintendo system to bother with the costs? If yes, it gets ported. If not, it doesn’t.

It’s very rarely about the hardware, yet the hardware “hurdles” are often used as the excuses.

Every publisher should just be honest and say they are passing on a Nintendo port because they don’t think the demand is there and they can’t compete with Nintendo 1st party sales overshadowing their product.
 
Although i think the new pokemon looks good compared to past games and gf is clearly improving substantially, i still refuse to believe it will push the switch as much as a retro studios game. I could add monolith or nlg into this too.
we're comparing apples to oranges here anyway. GF is going for an open world where Metroid games arent
After 3 weeks with my homebrewed oled switch, playing games in battery handheld. Texture patched, 60 fps unlocks, cheats, etc with maximum OC in CPU and GPU. Games looks amazing, but battery is also very good. Aprox 2 minutes per porcent, or 3 hours and 20 mins at 1785/921/1600. This is better than OG switch launch battery with less than half power. Almost everything is "fixed", xeno 2 looks awesome, the witcher 3 doom eternal and wolfenstein 2 runs at 60 fps, tony hawk looks ultraphenomenal. But why nintendo dont want to just oc the oled with this batteries performance? IDK. But im quite sure that the oled is more than capable to bring us a respetable upgrade, especially in handheld mode. So everything I wanted to be fixed is already fixed for me atleast, Switch 2/pro can only open to even more amazing results. My hope is to nintendo goes for 3 hours mark battery like they did with og switches. Oled switch is awesome, but only if you can get it homebrewed, the most radical one for me was tonyhawk. I looks like Im playing on my ps5 but in a portable 7" oled screen at 720p 30 fps lock( big texture patch) and I love it.
what. No way is Witcher, doom eternal and Wolfeinstein 2 at 60fps in switch.
 
After 3 weeks with my homebrewed oled switch, playing games in battery handheld. Texture patched, 60 fps unlocks, cheats, etc with maximum OC in CPU and GPU. Games looks amazing, but battery is also very good. Aprox 2 minutes per porcent, or 3 hours and 20 mins at 1785/921/1600. This is better than OG switch launch battery with less than half power.
Mind I ask what is at what clock speed in that 1785/821/1600?

CPU/GPU/RAM?

Or is that CPU/RAM/GPU?
 
we're comparing apples to oranges here anyway. GF is going for an open world where Metroid games arent

what. No way is Witcher, doom eternal and Wolfeinstein 2 at 60fps in switch.
If the first number of that 3-set is the CPU then it very well might be able to, they nearly doubled the CPU Clock from 1Ghz to 1.78Ghz.
 
we're comparing apples to oranges here anyway. GF is going for an open world where Metroid games arent

what. No way is Witcher, doom eternal and Wolfeinstein 2 at 60fps in switch.
Yes it can, with a little better graphics than handheld
Mind I ask what is at what clock speed in that 1785/821/1600?

CPU/GPU/RAM?

Or is that CPU/RAM/GPU?
CPU/GPU/RAM in portable battery mode

I dont have a decent camera but will do my best to show the diference in handheld without and with this configuration.
 
Yes it can, with a little better graphics than handheld

CPU/GPU/RAM in portable battery mode

I dont have a decent camera but will do my best to show the diference in handheld without and with this configuration.
Alright, and I actually don't see why 60fps on Mariko is impossible.

Mariko has far higher memory bandwidth on tap in potential because it uses LPDDR4X (Thus the high memory clock), and the CPU in your overclock is nearly double the OG Switch's clock (1GHz vs 1.78GHz) and the GPU clock is nearly 3 times as high as the OG Portable clocks.

At Portable mode resolutions and Settings that should leave headroom to push games to 60fps when unlocking the framerate.

And all of that along with enhancing some settings and the portable mode battery life still beats the OG Switch?

Honestly, if anything should indicate Nintendo will go for higher clocks with Drake/Switch 2, that should as Drake should have even better power characteristics than Mariko, especially if they go with TSMC 6N or 5/4N. And even if they stick with Samsung 8N because that is verifiably better than TSMC 16N.
 
we're comparing apples to oranges here anyway. GF is going for an open world where Metroid games arent

what. No way is Witcher, doom eternal and Wolfeinstein 2 at 60fps in switch.
I doubt gf could make a linear game that looked half as good anyway, but it isnt the point. If we re comparing to open world games there is xenoblade too
 
0
Please read this staff post before posting.

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


Back
Top Bottom