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

[Hermii]

Great. Like the CPU leap from PS4 to PS5 then!

So many original Switch games would instantly go to 60fps IF patched which a lot probably won't be

All of them, if the devs don’t prioritize 4k.

Hell, even an overclocked Mariko can run a lot of games at 60.

 

[Derachi]

I wonder if MH Rise now having DLSS support on PC means it’s gonna be able to take advantage of the new hardware

 

[Deleted member 887]

How many times more powerful is a 6 core Cortex-A78C CPU versus the current 1GHz Switch CPU if run at say 1.2GHz just to be conservative ? Thanks.

~6x

Only 5 cores would be for games btw.

? I may be out of my depth here but where does that come from?

1.2 GHz would be a 1.2x improvement in single threaded perf. Multithreaded perf would be something like 2x?

To get into the weeds, while the A78 has some impressive gains beyond clock speed they are workload dependent - it also has less than half the cache

 

[Skittzo]

I wonder if MH Rise now having DLSS support on PC means it’s gonna be able to take advantage of the new hardware

Considering that is apparently the first and only RE engine game to support DLSS it seems likely that they did that to prepare for a DLSS Switch patch.

 

[Derachi]

Considering that is apparently the first and only RE engine game to support DLSS it seems likely that they did that to prepare for a DLSS Switch patch.

ooh baby

 

[Look over there]

? I may be out of my depth here but where does that come from?

1.2 GHz would be a 1.2x improvement in single threaded perf. Multithreaded perf would be something like 2x?

To get into the weeds, while the A78 has some impressive gains beyond clock speed they are workload dependent - it also has less than half the cache

I think that 6 comes from 1.2 (clock) * 5/3 (cores) * ~3 from IPC(asterisk!...if I weren't already using * for multiplication)

As you note, it is workload dependent. The ~3 comes from accumulating the improvements in SPECfp that Arm touts from each generation starting with the A57. If one goes by SPECint, it ends up somewhere more like ~2.4x, I think? And I vaguely recall Geekbench 5 landed more in the 2.5x area, but I'm unsure on that one.

Since they're mentioned, note for the readers:
There are all sorts of benchmarks out there. If you're into reading CPU reviews, you've surely seen a bunch of them. But if you wonder, what does The Industry use? As far as I've heard, SPEC is the industry standard.
...but you don't see every reviewer using SPEC, right? Because an order of SPEC CPU costs $1000.
Geekbench should be more common as it has a free version. Generally regarded as still fine since it ought to cover a decent range of real world workloads.
And yes, you've seen Cinebench. It's free. And it tests exactly one type of workload.

 

[Dakhil]

To get into the weeds, while the A78 has some impressive gains beyond clock speed they are workload dependent - it also has less than half the cache

Depends on what CPU configuration Nintendo and Nvidia choose for Drake, I think. The PlayStation 5 and the Xbox Series X has 8 MB of L3 cache for 8 Zen 2 CPU cores. Assuming that Drake uses the Cortex-A78C, then Nintendo's new hardware could also have up to 8 MB of L3 cache for the CPU. Of course, one difference between the Cortex-A78C and Zen 2 is that the Cortex-A78C doesn't support SMT.

 

[Ghostsonplanets]

So, I was thinking about the storage aspect in both hardware-wise and software-wise for the new system. Hardware wise, I think we're all set. Nintendo will, very probably, use a eUFS memory. Either 2.1/2 or even 3.1, which can reach great levels of speed. But software wise, that another case completely. And that's because gamecards or physical media are still a very big business for Nintendo and also for retailers. As PS/XB ecosystems move more and more towards digital media, it starts to fail at the Nintendo ecosystem to carry the B2C retail videogames. Another aspect that also make Switch physical media attractive is the fact the games, exceptions nowthstanding, are plug and play, unlike, once again, PS and XB ecosystems, which have moved towards the physical media being mostly a validation key and needing to install or even download the game to play. While that aspect is a huge plus for Switch, the choice towards gamecard media also has huge cons, as they're much more expensive and publishers need to fit their games into a 8GB card to have the same margin as a blu-ray. Otherwise, they face a severely reduced margin. And while there's rumors of Nintendo sampling Macronix 48-layer 3D NAND tech for future gamecards(1,2), one would wonder if these will be cheaper enough to keep the same margins as the current 8GB card or would have worse margins and thus be underutilized as the 16/32GB and possible cancelled 64GB gamecards. If Nintendo will use a fast internal storage to match the direction the industry is going, the physical media is a huge limitation for that. But one they can't simply discard for business reasons.

In face of this, what would be feasible solutions for this problem? There's some like adopting XB/PS necessity of installing your game and thus lose the plug and play aspect. Another one would be to turn the physical media into a license key and needing to download your game and run them on the internal storage, which again means to lose the plug and play aspect. After thinking for a while, my solution for that would be:

• To keep the plug and play aspect of physical media on Switch/Nintendo systems, Nintendo could keep/reserve a portion of the storage as a cache of sorts(Let's say a 10GB cache size) and the gamecard would stream assets/data to this cache when the gamecard is inserted, thus not needing to install anything(Maybe a initial setup where the initial data is being transfered, but that can be masked as loading before playing the game, during the initial license and publishers/developers logo/legal information, etc). Basically the gamecard keep feeding this cache in the background with new data as it is required and flushing out data that won't be used anymore. This way, Switch Next games could reap all the benefits of the fast storage paradigm without having to worry about a potential slower physical media. Of course this cache would be flushed once the game session is ended and would be filled each time a game being run from a gamecard is started. There would be some cons to overcome like the possible necessity of a slightly faster gamecard(Current ones top out at 25MB/s), devtools/filesystem to allow devs to structure their games in a way that is in conformity with the way this cache works and the fact some portion of the storage would be permanently reserved to the system(Much like the Xbox Quick Resume reserves some amount of storage for its functionality)

I would like to hear everyone thoughts on this. To me, it's seems to strike a nice balance between keeping the plug and play aspect and keeping the gamecards costs reasonable while at the same time it allows Switch Next to fully stretch its legs and follow the new industry paradigm of fast storage media without being dragged down by the lowest common denominator(Gamecards).

 

[RennanNT]

So, I was thinking about the storage aspect in both hardware-wise and software-wise for the new system. Hardware wise, I think we're all set. Nintendo will, very probably, use a eUFS memory. Either 2.1/2 or even 3.1, which can reach great levels of speed. But software wise, that another case completely. And that's because gamecards or physical media are still a very big business for Nintendo and also for retailers. As PS/XB ecosystems move more and more towards digital media, it starts to fail at the Nintendo ecosystem to carry the B2C retail videogames. Another aspect that also make Switch physical media attractive is the fact the games, exceptions nowthstanding, are plug and play, unlike, once again, PS and XB ecosystems, which have moved towards the physical media being mostly a validation key and needing to install or even download the game to play. While that aspect is a huge plus for Switch, the choice towards gamecard media also has huge cons, as they're much more expensive and publishers need to fit their games into a 8GB card to have the same margin as a blu-ray. Otherwise, they face a severely reduced margin. And while there's rumors of Nintendo sampling Macronix 48-layer 3D NAND tech for future gamecards(1,2), one would wonder if these will be cheaper enough to keep the same margins as the current 8GB card or would have worse margins and thus be underutilized as the 16/32GB and possible cancelled 64GB gamecards. If Nintendo will use a fast internal storage to match the direction the industry is going, the physical media is a huge limitation for that. But one they can't simply discard for business reasons.

In face of this, what would be feasible solutions for this problem? There's some like adopting XB/PS necessity of installing your game and thus lose the plug and play aspect. Another one would be to turn the physical media into a license key and needing to download your game and run them on the internal storage, which again means to lose the plug and play aspect. After thinking for a while, my solution for that would be:

• To keep the plug and play aspect of physical media on Switch/Nintendo systems, Nintendo could keep/reserve a portion of the storage as a cache of sorts(Let's say a 10GB cache size) and the gamecard would stream assets/data to this cache when the gamecard is inserted, thus not needing to install anything(Maybe a initial setup where the initial data is being transfered, but that can be masked as loading before playing the game, during the initial license and publishers/developers logo/legal information, etc). Basically the gamecard keep feeding this cache in the background with new data as it is required and flushing out data that won't be used anymore. This way, Switch Next games could reap all the benefits of the fast storage paradigm without having to worry about a potential slower physical media. Of course this cache would be flushed once the game session is ended and would be filled each time a game being run from a gamecard is started. There would be some cons to overcome like the possible necessity of a slightly faster gamecard(Current ones top out at 25MB/s), devtools/filesystem to allow devs to structure their games in a way that is in conformity with the way this cache works and the fact some portion of the storage would be permanently reserved to the system(Much like the Xbox Quick Resume reserves some amount of storage for its functionality)

I would like to hear everyone thoughts on this. To me, it's seems to strike a nice balance between keeping the plug and play aspect and keeping the gamecards costs reasonable while at the same time it allows Switch Next to fully stretch its legs and follow the new industry paradigm of fast storage media without being dragged down by the lowest common denominator(Gamecards).

This could work... for 1st party. Although they might as well do a 16GB cache for BC, as that would make it painless for the entire Switch library except for the single digit 32GB game cards like Witcher 3.

The vast majority of big sized third party games are already doing mandatory downloads which is frowned upon, they would absolutely jump into using the smallest card "to provide the best experience to players".

 

[Ghostsonplanets]

This could work... for 1st party. Although they might as well do a 16GB cache for BC, as that would make it painless for the entire Switch library except for the single digit 32GB game cards like Witcher 3.

The vast majority of big sized third party games are already doing mandatory downloads which is frowned upon, they would absolutely jump into using the smallest card "to provide the best experience to players".

While I do believe that 3rd-parties will want to cheap out if gamecard costs keep high and mandate downloads to the internal fast storage, that would be a counter solution to the pick up and play nature of the Switch, meaning that the gamecards will be basically useless in such case. There's also the factor that fast storage will be a premium and I expect the next model to have 128GB.
Honestly the proposed cache solution would be a better balance imo but I do agree that gamecards will probably take a backseat and publishers will cheap out on them.

 

[Pokemaniac]

So, I was thinking about the storage aspect in both hardware-wise and software-wise for the new system. Hardware wise, I think we're all set. Nintendo will, very probably, use a eUFS memory. Either 2.1/2 or even 3.1, which can reach great levels of speed. But software wise, that another case completely. And that's because gamecards or physical media are still a very big business for Nintendo and also for retailers. As PS/XB ecosystems move more and more towards digital media, it starts to fail at the Nintendo ecosystem to carry the B2C retail videogames. Another aspect that also make Switch physical media attractive is the fact the games, exceptions nowthstanding, are plug and play, unlike, once again, PS and XB ecosystems, which have moved towards the physical media being mostly a validation key and needing to install or even download the game to play. While that aspect is a huge plus for Switch, the choice towards gamecard media also has huge cons, as they're much more expensive and publishers need to fit their games into a 8GB card to have the same margin as a blu-ray. Otherwise, they face a severely reduced margin. And while there's rumors of Nintendo sampling Macronix 48-layer 3D NAND tech for future gamecards(1,2), one would wonder if these will be cheaper enough to keep the same margins as the current 8GB card or would have worse margins and thus be underutilized as the 16/32GB and possible cancelled 64GB gamecards. If Nintendo will use a fast internal storage to match the direction the industry is going, the physical media is a huge limitation for that. But one they can't simply discard for business reasons.

In face of this, what would be feasible solutions for this problem? There's some like adopting XB/PS necessity of installing your game and thus lose the plug and play aspect. Another one would be to turn the physical media into a license key and needing to download your game and run them on the internal storage, which again means to lose the plug and play aspect. After thinking for a while, my solution for that would be:

• To keep the plug and play aspect of physical media on Switch/Nintendo systems, Nintendo could keep/reserve a portion of the storage as a cache of sorts(Let's say a 10GB cache size) and the gamecard would stream assets/data to this cache when the gamecard is inserted, thus not needing to install anything(Maybe a initial setup where the initial data is being transfered, but that can be masked as loading before playing the game, during the initial license and publishers/developers logo/legal information, etc). Basically the gamecard keep feeding this cache in the background with new data as it is required and flushing out data that won't be used anymore. This way, Switch Next games could reap all the benefits of the fast storage paradigm without having to worry about a potential slower physical media. Of course this cache would be flushed once the game session is ended and would be filled each time a game being run from a gamecard is started. There would be some cons to overcome like the possible necessity of a slightly faster gamecard(Current ones top out at 25MB/s), devtools/filesystem to allow devs to structure their games in a way that is in conformity with the way this cache works and the fact some portion of the storage would be permanently reserved to the system(Much like the Xbox Quick Resume reserves some amount of storage for its functionality)

I would like to hear everyone thoughts on this. To me, it's seems to strike a nice balance between keeping the plug and play aspect and keeping the gamecards costs reasonable while at the same time it allows Switch Next to fully stretch its legs and follow the new industry paradigm of fast storage media without being dragged down by the lowest common denominator(Gamecards).

Repeatedly writing data like that is going to cause a lot of extra wear, which is more of a problem the less capacity is available.

Personally, I think the reality is we just don't have a clear picture of how 2022 carts would compare to 2017 ones, since high capacity ROM chips are a relatively exotic category of tech that doesn't really show up very much in non-Switch consumer devices anymore. Given that Flash and ROM are pretty closely related (but importantly not exactly the same) and flash hasn't exactly stood still these past 5 years, I don't think it's unreasonable to think that Nintendo could probably deliver an upgrade to capacity and speed at roughly equivalent costs, but I'm not sure there's any way we could really tell it was possible before it happened.

 

[Ghostsonplanets]

Repeatedly writing data like that is going to cause a lot of extra wear, which is more of a problem the less capacity is available.

Personally, I think the reality is we just don't have a clear picture of how 2022 carts would compare to 2017 ones, since high capacity ROM chips are a relatively exotic category of tech that doesn't really show up very much in non-Switch consumer devices anymore. Given that Flash and ROM are pretty closely related (but importantly not exactly the same) and flash hasn't exactly stood still these past 5 years, I don't think it's unreasonable to think that Nintendo could probably deliver an upgrade to capacity and speed at roughly equivalent costs, but I'm not sure there's any way we could really tell it was possible before it happened.

That's a great point actually. It would need to be seen if there are some techniques to counter measure such amount of wear and also how capable modern NAND/eUFS solution are in terms of r/w lifecycle. But as general rule, you're probably right that it would be a bad idea.
Honestly, this is the part I'm most curious of how they will tackle or which decision/path they will take. The only rumor we had so far regarding this was the deal with Macronix for 48 layer 3D NAND.

 

[Pokemaniac]

That's a great point actually. It would need to be seen if there are some techniques to counter measure such amount of wear and also how capable modern NAND/eUFS solution are in terms of r/w lifecycle. But as general rule, you're probably right that it would be a bad idea.
Honestly, this is the part I'm most curious of how they will tackle or which decision/path they will take. The only rumor we had so far regarding this was the deal with Macronix for 48 layer 3D NAND.

Carts are the sort of thing that probably wouldn't be rumored super heavily beforehand, since they seem to be mostly between Nintendo and Macronix, and they also probably wouldn't enter mass production until closer to launch than the hardware itself.

 

[Deleted member 887]

So, I was thinking about the storage aspect in both hardware-wise and software-wise for the new system. Hardware wise, I think we're all set. Nintendo will, very probably, use a eUFS memory. Either 2.1/2 or even 3.1, which can reach great levels of speed. But software wise, that another case completely. And that's because gamecards or physical media are still a very big business for Nintendo and also for retailers.

While there are a small number of exceptions, I tend to think that game card speed is “fast enough” - spinning media is too slow but I don’t think we’re at the point where SSD speed is regularly required.

To keep the plug and play aspect of physical media on Switch/Nintendo systems, Nintendo could keep/reserve a portion of the storage as a cache of sorts(Let's say a 10GB cache size) and the gamecard would stream assets/data to this cache when the gamecard is inserted, thus not needing to install anything

This just sounds like an install to me? Either the process is transparent to the game, in which case initial boot from cart will always be significantly slower than digital boot, or it will require developer action - at which point I don’t think you’ve added anything devs can’t already do. Aren’t games already saving uncompressed assets this way?

 

[Crusters]

While there are a small number of exceptions, I tend to think that game card speed is “fast enough” - spinning media is too slow but I don’t think we’re at the point where SSD speed is regularly required.

This just sounds like an install to me? Either the process is transparent to the game, in which case initial boot from cart will always be significantly slower than digital boot, or it will require developer action - at which point I don’t think you’ve added anything devs can’t already do. Aren’t games already saving uncompressed assets this way?

Here's a crazy one.

Historically carts have had one set of pins for access to the rom.

Today there are 3 sides that fully enter the reader, but still only one side has a set of pins.

3 sides with pins, 3 parallel reads, 3x the read speed?

 

[Look over there]

Btw, do you mean 2030 here?

Yea, late 2020's to 2030, given how far later they'd have spare resources to start investing compared to Nvidia. Both hardware wise and... the time/money to train their own neural networks are, I guess?

I think maybe 10-12 Cores can be seen/expected later on by that point? Being able to do more within the same power budget would probably be their biggest focus. And it would be several years down the line. Not a huge upgrade mind you, but MS isn’t so wedded to AMD maybe, while PS and Nintendo would be more wedded to their hardware providers. Unless Sony and Nintendo figure out a method similar to MS which is more detached from the silicon and it is actually a virtual XBox environment that speaks to the silicon itself just like how DX12U does for current GPUs and CPUs.

(DX12U was just an example, XBox uses a lower level forked variant of Direct X that Pc users cannot use)

The way I'd see 10-12 cores with Zen would pretty much hinge on AMD changing their cluster size. If the generation of Zen used in a PS6 sticks with 8 core clusters, I don't see AMD specifically creating a 10 or 12 core cluster just for the consoles.

However, bringing up that MS isn't necessarily so wedded to AMD is pretty interesting.
If MS sticks with x86, then there are three companies that have x86 architecture licenses: Intel, AMD, and VIA. Uhh, I probably need to do more reading on Zhaoxin to see VIA's CPU capabilities. But given the need for graphics too, I could probably rule out VIA for now.
Intel... if Intel's Xe can become reasonably competent by the late 2020's, that's... an option. If you want to go above 8 cores, there's something there... but I'm not looking at the big Core cores. No, I'm looking at Atom. If the inter-core latency issues can be fixed, you can maybe do three or four clusters of 4 Atom cores each for 12C/12T or 16C/16T. Gracemont right now is basically at Skylake levels of IPC, but in gaming crippled by inter-core latency. Fix that, add a few more iterations of IPC gains, and a few more iterations of stretching the power-frequency curve out... there's a possibility there that this route ends up being more area and energy efficient for a minor loss in single thread strength. It does hinge a lot on Intel delivering on their foundry roadmap though.

For GDDR vs LPDDR vs DDR I think it’s that DDR has the lowest latency while GDDR has the highest latency. LPDDR has the lowest TDP while GDDR has the highest TDP. In terms of performance LPDDR is the worst while GDDR is the best.

It’s all just specialized RAM in the end. But there are strengths and weaknesses.

Also why Nintendo went with DDR3 instead of GDDR for the Wii U, it had noticeably less latency. Wii on the other hand did use pretty much brand new GDDR at the time, this required an alteration to the silicon of course so it’s not just a die shrink GCN, more to it!

If HBM becomes viable by the end of the generation, hopefully consoles do adopt it. All of them.

Hell, IF Nintendo managed to get HBM for the switch, they wouldn’t have to worry about bandwidth for their device like… ever at that point lol.

Just the CPU I guess.

I just don't think that HBM will reduce in cost fast enough to be worthwhile using in consumer grade products. I need to see them return to usage in consumer GPUs first to change my mind.

Bizarrely enough, I'm not that worried about CPU for a post-Drake Switch style device.
So I wrote earlier that my optimistic expectation for Drake's CPU can be described as:
"On the high/optimistic end, the CPU ought to be able to handle a class of complexity in the middle between the PS4 and PS5/Xbox Series. (that is, if you break up the improvement from PS4/XBO to PS5/Xbox Series into two steps, PS4/XBO->Drake would be one, then Drake->PS5/XS would be the second step)"
So if PS4->PS5 was two steps, ideally Drake ends up a step behind PS5. Now, I think that a PS6's realistic ceiling is one step up from PS5, not two steps. Ergo, a post-Drake Switch needs to only advance one step to maintain distance. Or in other words, the target is PS5 level complexity. And I still have enough confidence in Arm to make that doable within 2 watts on... a N2 refinement/variant. They just need to deliver at least a few percentage points more than 5 each year.

It’s more or less why they are allowing PC releases I suppose.

Which is a pretty funny pivot. That's clearly a relatively recent decision, as it at odds with the hyped up fast storage+compression capability. Any game that hard requires that capability is automatically cutting off the segment of the PC userbase who are using... PCIe gen 3 NVMe and slower. That's a pretty large segment.

 

[ReddDreadtheLead]

? I may be out of my depth here but where does that come from?

1.2 GHz would be a 1.2x improvement in single threaded perf. Multithreaded perf would be something like 2x?

To get into the weeds, while the A78 has some impressive gains beyond clock speed they are workload dependent - it also has less than half the cache

66% more cores * 20% increase in clockspeed * ~3x IPC games.

6 A78C can have 6MB of L3 cache.

Actually, the A78C variant is just listed as having up to 8MB of L3 cache, I made a chart based on the info arm has provided us:

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

Alternative approach: everybody can just chillax. That would be nice but nature abhors a vacuum of information, as long as this thread exists it will get filled

famiboards.com

I’m not exactly sure what you mean with less than half the cache?

For note, I put the ~ before the 6x for this reason, some games wouldn’t get 6x increase, others would get a 6x and others can get over 6, I just averaged it to ~6x as it was the easiest!

So, I was thinking about the storage aspect in both hardware-wise and software-wise for the new system. Hardware wise, I think we're all set. Nintendo will, very probably, use a eUFS memory. Either 2.1/2 or even 3.1, which can reach great levels of speed. But software wise, that another case completely. And that's because gamecards or physical media are still a very big business for Nintendo and also for retailers. As PS/XB ecosystems move more and more towards digital media, it starts to fail at the Nintendo ecosystem to carry the B2C retail videogames. Another aspect that also make Switch physical media attractive is the fact the games, exceptions nowthstanding, are plug and play, unlike, once again, PS and XB ecosystems, which have moved towards the physical media being mostly a validation key and needing to install or even download the game to play. While that aspect is a huge plus for Switch, the choice towards gamecard media also has huge cons, as they're much more expensive and publishers need to fit their games into a 8GB card to have the same margin as a blu-ray. Otherwise, they face a severely reduced margin. And while there's rumors of Nintendo sampling Macronix 48-layer 3D NAND tech for future gamecards(1,2), one would wonder if these will be cheaper enough to keep the same margins as the current 8GB card or would have worse margins and thus be underutilized as the 16/32GB and possible cancelled 64GB gamecards. If Nintendo will use a fast internal storage to match the direction the industry is going, the physical media is a huge limitation for that. But one they can't simply discard for business reasons.

In face of this, what would be feasible solutions for this problem? There's some like adopting XB/PS necessity of installing your game and thus lose the plug and play aspect. Another one would be to turn the physical media into a license key and needing to download your game and run them on the internal storage, which again means to lose the plug and play aspect. After thinking for a while, my solution for that would be:

• To keep the plug and play aspect of physical media on Switch/Nintendo systems, Nintendo could keep/reserve a portion of the storage as a cache of sorts(Let's say a 10GB cache size) and the gamecard would stream assets/data to this cache when the gamecard is inserted, thus not needing to install anything(Maybe a initial setup where the initial data is being transfered, but that can be masked as loading before playing the game, during the initial license and publishers/developers logo/legal information, etc). Basically the gamecard keep feeding this cache in the background with new data as it is required and flushing out data that won't be used anymore. This way, Switch Next games could reap all the benefits of the fast storage paradigm without having to worry about a potential slower physical media. Of course this cache would be flushed once the game session is ended and would be filled each time a game being run from a gamecard is started. There would be some cons to overcome like the possible necessity of a slightly faster gamecard(Current ones top out at 25MB/s), devtools/filesystem to allow devs to structure their games in a way that is in conformity with the way this cache works and the fact some portion of the storage would be permanently reserved to the system(Much like the Xbox Quick Resume reserves some amount of storage for its functionality)

I would like to hear everyone thoughts on this. To me, it's seems to strike a nice balance between keeping the plug and play aspect and keeping the gamecards costs reasonable while at the same time it allows Switch Next to fully stretch its legs and follow the new industry paradigm of fast storage media without being dragged down by the lowest common denominator(Gamecards).

I’d be more concerned with the wear on the storage media long-term at that point. Granted, I’m not sure how much wear it would have.

Would need to be tested.

 

[toutoune134]

Nintendo shared the translation of their latest Q&A and one answer provides more insight regarding the rise of raw materials inventory :

 

[Mobius]

Considering that is apparently the first and only RE engine game to support DLSS it seems likely that they did that to prepare for a DLSS Switch patch.

the fact that the game run well at 4K with high settings on a 2060 is even more "suspicious"

 

[Hermii]

the fact that the game run well at 4K with high settings on a 2060 is even more "suspicious"

I mean, it’s a game that’s made for the tx1 as base platform.

 

[ReddDreadtheLead]

the fact that the game run well at 4K with high settings on a 2060 is even more "suspicious"

Going from a 200-400 GFLOP to a 6.4TFLOP GPU that’s also not constrained by the pathetically weak Switch CPU or memory bandwidth is like 2-3 generational leaps in console terms over the switch. To be fair.

 

[Mobius]

I mean, it’s a game that’s made for the tx1 as base platform.

Sure, but the point is that they could have updated a more depending game instead

 

[ReddDreadtheLead]

Sure, but the point is that they could have updated a more depending game instead

I mean, maybe. It’s not the only Monster Hunter game with DLSS.

Just the only Monster Hunter game with the one that matters.

 

[Mobius]

I mean, maybe. It’s not the only Monster Hunter game with DLSS.

Just checked and you are right, I understood this was the first Capcom game to ever have DLSS.

 

[pld]

Does anyone know how much effort it is to make a game DLSS ready? In theory the whole DLSS thing sounds like the perfect solution for Nintendo's hardware strategy but I am afraid developers might not take advantage of DLSS because it would take more effort, i.e. it would be more expensive.

 

[IamPeacock]

Does anyone know how much effort it is to make a game DLSS ready? In theory the whole DLSS thing sounds like the perfect solution for Nintendo's hardware strategy but I am afraid developers might not take advantage of DLSS because it would take more effort, i.e. it would be more expensive.

I'm thinking the opposite.

Currently, only PC gamers with RTX cards can fully take advantage of DLSS.
If the next Switch system fully embraces DLSS, it could incentize developers to add it to their games, because the potential usercase is much bigger (both PC and Switch).

 

[Adulfzen]

Does anyone know how much effort it is to make a game DLSS ready? In theory the whole DLSS thing sounds like the perfect solution for Nintendo's hardware strategy but I am afraid developers might not take advantage of DLSS because it would take more effort, i.e. it would be more expensive.

if it helps having to put less effort into heavy optimization (for games like Witcher 3 or DQ XIS for example) for some devs it should be more than worth it especially if the game is also headed for PC where it then becomes a good feature for marketing purposes.

 

[Hermii]

Does anyone know how much effort it is to make a game DLSS ready? In theory the whole DLSS thing sounds like the perfect solution for Nintendo's hardware strategy but I am afraid developers might not take advantage of DLSS because it would take more effort, i.e. it would be more expensive.

As I understand it, a decent implementation takes a bit of effort.

Not every game will take advantage of it, but not every game will need it either. The soc will be quite capable even without it.

For example if they have implemented fsr 2 because it’s platform agnostic, and it works well enough on Drake, there is little reason to go the extra mile.

 

[YolkFolk]

Does anyone know how much effort it is to make a game DLSS ready?

Just drag the rom across to the the Switch Drake dev kit and then click ‘add DLSS’. You instantly have a 4K game with no adjustments required.

 

[Mr Swine]

Isn’t it a bit odd that Rise gets DLSS when World which has DLSS is a much more demanding game.

 

[ReddDreadtheLead]

Does anyone know how much effort it is to make a game DLSS ready? In theory the whole DLSS thing sounds like the perfect solution for Nintendo's hardware strategy but I am afraid developers might not take advantage of DLSS because it would take more effort, i.e. it would be more expensive.

DLSS being implemented into a game can vary. Besides the obvious “No DLSS button”, they also have to make the effort to actually make it look good with DLSS.

Currently, DLSS relies on the motion vector data that a game provides, and if a game has TAA then that greatly reduces the work as TAA has motion vector data. That would be due to the engine the game runs on supporting TAA, which is the bare minimum* a game needs to make it easy to implement, though that isn’t always the case. A few days to a few weeks of engine alteration to implement DLSS would be required.

And then actual work of having the image quality look good, fwiw there’s bad DLSS implementations too as devs didn’t account for other variables. Red Dead Redemption 2 for example has DLSS, but it doesn’t look great because the game was designed around a different technique such as checkerboard rendering.

Fortunately most game engines already support TAA so a lot of the heavy lifting is reduced. They would be working with NVIDIA and implement DLSS into their game engines and thus a game would support it.

Some games managed to not need TAA to have DLSS to work, like Nioh 2. Though that’s more complicated of an issue.


That aside, if the developer kits are out to developers, then that means that the first ones to actually play with it would probably be the engineers at these companies, and they would be altering their game engines to support the new system and all of its unique feature sets to make porting to the Switch 2/Pro/Coffee maker as easy as possible.

So the devs that actually would be doing the porting job would have an easier time I suppose I what I want to get at with that last part. They still have to do work though to make it look pleasing, mind you, like every switch game that comes to it.

 

[Hermii]

Imo this MH ryse dlss patch, is pure confirmation bias at work.

Is it possible it is related? Yes

If we knew nothing about Drake Switch, would anyone have batted an eye that rise got dlss? Probably not imo. There doesnt have to be a secondary reason for it.

 

[ReddDreadtheLead]

Imo this MH ryse dlss patch, is pure confirmation bias at work.

Is it possible it is related? Yes

If we knew nothing about Drake Switch, would anyone have batted an eye that rise got dlss? Probably not imo. There doesnt have to be a secondary reason for it.

Yes they would, NVidia quite literally had a job listing for a console that made use of these features.

Even if there was no data breach. People would have been suspicious at this.

 

[Eradicate]

Honest question, what if it's called the Switch U.

DOA?

Haha! I remember what a ruckus it was with the "play on words" of Wii U ("we, you"). Maybe they can be punny with the next Switch. Switch Flick? Switch Hit? Switch Box?

if it's so cool than why is it reportedly getting that new heat tube??

Haha! What's cooler than being cool? Heat tube!

First post - hi everyone!

I think some people earlier in the thread were speculating that they might move the hinges upward, and then create a thin "outline" hinge that covers the edges of the backplate. Not unlike some of the HP Spectre kickstands.

Hello @kojimaposting! Are you Kojima...posting? Really cool picture of that kickstand! I'd love to see that on the new model! It looks so much more sturdy! Anyone have experience with this sort of kickstand? Does it "lock" in place throughout the whole range of motion, or only at the farthest reach?

if I was right about the kickstand I'm going to scream

Please, please, please let the new model be ALL about the kickstand!

And ramble time it is. [Cut for length!]

I honestly did take the time to read all this, and found it incredibly fascinating! Thank you for all these thoughts on the landscape moving forward!

Speaking of NateDrake, here's my rough summary of the 13 October 2021 episode of Nate the Hate below since NateDrake mentioned the information from that episode as of today is still accurate.

Edit: Thank you @Raccoon for mentioning I could add a threadmark.

Thank you so much for the summary! I do sound things, and it's nice to just read a lot of times to get a rest! Really appreciate this!

These past few pages have been peak thread. Dozens of posts discussing a tiny chip in the dock, the obligatory 1 per page massive wall of text post, relitigating a podcast that's over a year old, an argument between two people over completely unrelated tech interspersed within, chinese leaks with a little "yay new hinge" cherry on top. Love it.

These are the best times!

some sort of activity with the Donkey Kong copyright renewal. probably nothing but if there is a Switch Drake anytime soon i maintain there will be big unannouced first party title to go along with it.

Pretty cool find! Thanks for sharing! There's definitely a lot of murmuring about a new Donkey Kong, which would make for a cool game to show off tech! It's funny that Donkey Kong kind of "has" been a bit of a showcase a lot of times with different things, whether the rendering of Donkey Kong Country, bongos (so awesome!), I think there was a game that was Nintendo's first to feature cross-purchasing (buy on Wii U or 3DS, and you can play it on the other one), etc. Not that they'll do that again, but I've just always enjoyed Donkey Kong for some of these experimental things they do with it sometimes!

Here's a crazy one.

Historically carts have had one set of pins for access to the rom.

Today there are 3 sides that fully enter the reader, but still only one side has a set of pins.

3 sides with pins, 3 parallel reads, 3x the read speed?

The tri-cart!

Just drag the rom across to the the Switch Drake dev kit and then click ‘add DLSS’. You instantly have a 4K game with no adjustments required.

DLSS = Dragging Let's Super Switch!

 

[Hermii]

Yes they would, NVidia quite literally had a job listing for a console that made use of these features.

Even if there was no data breach. People would have been suspicious at this.

They added it to world in 2019- That was also because of Drake?

 

[Ghostsonplanets]

Isn’t it a bit odd that Rise gets DLSS when World which has DLSS is a much more demanding game.

No? Rise is a game in active development. World is a game which it's development has finished years ago and there's probably a minimal support crew on it. All hands on World 2 for 2024 release. Also, World had DLSS 1.0 support. DLSS 2.0 kinda came after they wrapped up things, so I guess this boat has sailed.

Imo this MH ryse dlss patch, is pure confirmation bias at work.

Is it possible it is related? Yes

If we knew nothing about Drake Switch, would anyone have batted an eye that rise got dlss? Probably not imo. There doesnt have to be a secondary reason for it.

Agreed. Upsampling tech is being massively pushed by all three IHV(Intel, AMD and Nvidia) right now. Game X or Y getting doesn't mean anything. It could be related to work on a next generation Nintendo device while it can also be just that Nvidia pushed Capcom to adopt it.

 

[Raccoon]

They added it to world in 2019- That was also because of Drake?

yeah

 

[Disorientator]

If the DLSS Switch exists, would the (DLSS) version used be game-specific (similar to how it works on PC/dll can be replaced), or would it be "fixed" at the system level?

I'm I right to assume that if it's the former, a "simple" game patch would be enough, but if it's the latter it will be tied to system firmware?

 

[Spork4000]

If the DLSS Switch exists, would the (DLSS) version used be game-specific (similar to how it works on PC/dll can be replaced), or would it be "fixed" at the system level?

I'm I right to assume that if it's the former, a "simple" game patch would be enough, but if it's the latter it will be tied to system firmware?

Would need a per-game patch.

 

[ReddDreadtheLead]

They added it to world in 2019- That was also because of Drake?

You mean the game that used the old version and has since not been updated? If both were updated now to support DLSS 2.0, no one would think much of it, just odd to update the MonHun games but not the Resi or the DMC games which isn’t related to Drake.

But only Rise was updated and with 2.0.




People would have thought something of it.

 

[Deleted member 887]

If the DLSS Switch exists, would the (DLSS) version used be game-specific (similar to how it works on PC/dll can be replaced), or would it be "fixed" at the system level?

I'm I right to assume that if it's the former, a "simple" game patch would be enough, but if it's the latter it will be tied to system firmware?

You are (probably) correct, it will be built into each game, the same way that drivers are, so that firmware updates don’t break your game by accident

 

[Disorientator]

Would need a per-game patch.

You are (probably) correct, it will be built into each game, the same way that drivers are, so that firmware updates don’t break your game by accident

Yeah, makes sense.

 

[Mr Swine]

No? Rise is a game in active development. World is a game which it's development has finished years ago and there's probably a minimal support crew on it. All hands on World 2 for 2024 release. Also, World had DLSS 1.0 support. DLSS 2.0 kinda came after they wrapped up things, so I guess this boat has sailed.

yeah but seeing how a RTX 2060 can run the game at 4k at 60fps without DLSS 2.0 makes me wonder why have it if it’s not for Drake support down the line

 

[kimbo99]

Could the high number of raw materials be linked to the onset of Wide Care? Perhaps they will need more materials for replacement/repairs.

 

[Deleted member 887]

yeah but seeing how a RTX 2060 can run the game at 4k at 60fps without DLSS 2.0 makes me wonder why have it if it’s not for Drake support down the line

Because they have a deal with NVidia? Because the engine already supports it? Because it's not a place where it really matters, so it's a game for experimentation with low stakes?

 

[Eradicate]

Just a couple little things seen while on the forums. It's not much, just a couple interesting tidbits...

 

[Concernt]

Please remain calm.

Nothing to see here.

 

[ReddDreadtheLead]

Please remain calm.

Nothing to see here.

They’ve been doing this for several months, so yeah there’s nothing to see here lol

 

[necrolipe]

Don't take any maintenances scheduled like this as gospel for incoming announcements...
This is pretty unusual to happen (except for some cases like NSO emulators)

 

[undefined]

Imo this MH ryse dlss patch, is pure confirmation bias at work.

Is it possible it is related? Yes

If we knew nothing about Drake Switch, would anyone have batted an eye that rise got dlss? Probably not imo. There doesnt have to be a secondary reason for it.

This doesn't mean much it yep as it would be patched anyway just like other titles
On the flip side, not really news but I guess that as we're seeing that Capcom is already using this stuff on their titles then we could expect more support from them in this new device, and maybe some of their games could be featured in the reveal