By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Nintendo - How Will be Switch 2 Performance Wise?

 

Switch 2 is out! How you classify?

Terribly outdated! 3 5.26%
 
Outdated 1 1.75%
 
Slightly outdated 14 24.56%
 
On point 31 54.39%
 
High tech! 7 12.28%
 
A mixed bag 1 1.75%
 
Total:57
sc94597 said:

If Nintendo/Nvidia initially aimed for a 2025 release they almost certainly could've gone full Lovelace & 16GB of LPDDR5X for a similar price (say, within +$20 to $50 of Switch 2 at most) without a loss for either company. 

Given how efficient they got the T239 to be in the end, I am not sure if we would have seen as large a leap as we'd normally expect from Ampere -> Lovelace based on the consumer desktop/laptop chips. 

So we'd probably be looking at a Rog Ally -> Rog Ally X (2024) or Steam Deck -> Steam Deck OLED sort of performance difference, with maybe most of the gains being in battery life.

The OLED screen (unless they ditched VRR) likely is a fantasy want though. 

I have my doubts about this, going to Lovelace would've required a significantly more expensive node process. It looks like they adapted things from Lovelace as is as the x-ray shots of the chip have commonalities with Lovelace, but they likely got a significantly better deal from Samsung. 

8 inch OLED display with 120 Hz VRR ... no chance the price would be under $600 in that case. 

Nintendo has a right to make money off the hardware too, they are not a charity case obligated to sell things at a loss or net cost. They don't have 30 different divisions where gaming is like just a fun hobby for them, this is still primarily their only business. They're not a Google or Microsoft or Meta making 100 billion in net profit every year. Making 1 movie every 3-4 years that are co-financed by other media conglomerates doesn't change that. They make one hardware line that generally has to carry them for 6-8 years, they are well within logical business practises to expect a profit be made from the hardware itself, especially with software development costs being 10x higher than they were 15-20 years ago and games take 2-3x longer to make (which means fewer 1st/2nd party games each hardware cycle from the past). 

Even Sony is not subsidizing or opting to take losses on hardware anymore, Microsoft is on the way out, Meta is taking heat for losing so much money on VR. 

Last edited by Soundwave - on 14 July 2025

Around the Network
Soundwave said:

I have my doubts about this, going to Lovelace would've required a significantly more expensive node process. It looks like they adapted things from Lovelace as is as the x-ray shots of the chip have commonalities with Lovelace, but they likely got a significantly better deal from Samsung. 

It would cost more, but not "significantly" more. Lovelace is a generation old in 2025 and TSMC 5N is starting to age as well. 

+$20 per die sounds about right, especially given the higher yields, and then add another $15-$20 for the extra 4GB of ram. 

Nintendo could still make a slight profit, especially considering that the Switch 2 already seems to cost only about $350 to manufacture and the $450 price-point seems to be a tariff buffer. 

The cost of a Lovelace chip in 2025 is probably very similar to their Ampere chip in early-mid 2024, which seems to be the original projected release window.

Last edited by sc94597 - on 14 July 2025

Having now tried a few of the system's ports of current gen only games like Kunitsu-gami, I feel it does a reasonably good job of translating the PS5/Xbox Series experience. Obviously resolution and framerate aren't as high, but it holds up well enough.

What I'm most interested in is seeing the system's raytracing capabilities put to the test; to my knowledge none of the launch games make use of this, but we know the hardware does support it. While there's less power on tap compared to PS5/Series, Nvidia is well ahead of AMD in this area which could help narrow the gap.



Soundwave said:

Prices have leaked for the ROG Ally XBox version and ROG Ally XBox X ...

Spanish gaming website 3DJuegos spotted what look like leaked listings for both the ROG Xbox Ally and the ROG Xbox Ally X on the Asus store for €599 and €899 respectively.

That would be like $700 and $1050 USD

No, the ROG Ally X was also priced €899 (including taxes) in Europe, but only $799 in the US.

Even with the slightly weaker US Dollar, it will be probably €899 (including taxes) and $899 (without taxes).



sc94597 said:
Soundwave said:

I have my doubts about this, going to Lovelace would've required a significantly more expensive node process. It looks like they adapted things from Lovelace as is as the x-ray shots of the chip have commonalities with Lovelace, but they likely got a significantly better deal from Samsung. 

It would cost more, but not "significantly" more. Lovelace is a generation old in 2025 and TSMC 5N is starting to age as well. 

+$20 per die sounds about right, especially given the higher yields, and then add another $15-$20 for the extra 4GB of ram. 

Nintendo could still make a slight profit, especially considering that the Switch 2 already seems to cost only about $350 to manufacture and the $450 price-point seems to be a tariff buffer. 

The cost of a Lovelace chip in 2025 is probably very similar to their Ampere chip in early-mid 2024, which seems to be the original projected release window.

One critical aspect that must not be overlooked is the foundry process compatibility in the context of NVIDIA’s Ampere architecture. The consumer-grade Ampere GPUs were specifically designed for Samsung's 8nm process node (technically a refinement of their 10nm node, known as 8LPP), whereas the data center and professional variants (e.g., GA100) utilized TSMC’s 7nm process.

Importantly, TSMC’s N7 and N5 nodes do not share standard cell libraries, IP blocks, or physical design constraints. Standard cell libraries encompass transistor-level building blocks and include design rules, timing characteristics, and layout constraints optimized for a given process. These are proprietary to each foundry and typically not portable between fabs (e.g., Samsung and TSMC), and often not even between nodes within the same foundry due to substantial differences in fin pitch, metal stack height, and EUV usage.

As a result, porting an Ampere-based SoC from Samsung 8nm to TSMC 5nm would not be a matter of simple tapeout retargeting or minor RTL tweaks—it would necessitate a full front-end and back-end redesign, including physical synthesis, place-and-route, timing closure, power delivery, and validation, using entirely new EDA flows, libraries, and verification suites.

Furthermore, NVIDIA has not previously released any Ampere-family silicon on TSMC's N5 node, meaning there is no existing design collateral or hardened IP (e.g., PHYs, memory controllers, I/O subsystems) for that process. This implies that a 5nm Ampere SoC would be a from-scratch development effort, requiring a much larger engineering investment compared to adapting an existing chip like Orin (based on Ampere architecture and already validated on a mature process node).

In addition to the NRE (non-recurring engineering) costs, Nintendo would also face higher per-unit costs due to the wafer pricing at TSMC N5, which is significantly more expensive than Samsung’s 8nm. Altogether, these factors make a custom 5nm Ampere design both technically complex and economically unjustifiable for a console-grade SoC.



Around the Network
Kynes said:
sc94597 said:

It would cost more, but not "significantly" more. Lovelace is a generation old in 2025 and TSMC 5N is starting to age as well. 

+$20 per die sounds about right, especially given the higher yields, and then add another $15-$20 for the extra 4GB of ram. 

Nintendo could still make a slight profit, especially considering that the Switch 2 already seems to cost only about $350 to manufacture and the $450 price-point seems to be a tariff buffer. 

The cost of a Lovelace chip in 2025 is probably very similar to their Ampere chip in early-mid 2024, which seems to be the original projected release window.

One critical aspect that must not be overlooked is the foundry process compatibility in the context of NVIDIA’s Ampere architecture. The consumer-grade Ampere GPUs were specifically designed for Samsung's 8nm process node (technically a refinement of their 10nm node, known as 8LPP), whereas the data center and professional variants (e.g., GA100) utilized TSMC’s 7nm process.

Importantly, TSMC’s N7 and N5 nodes do not share standard cell libraries, IP blocks, or physical design constraints. Standard cell libraries encompass transistor-level building blocks and include design rules, timing characteristics, and layout constraints optimized for a given process. These are proprietary to each foundry and typically not portable between fabs (e.g., Samsung and TSMC), and often not even between nodes within the same foundry due to substantial differences in fin pitch, metal stack height, and EUV usage.

As a result, porting an Ampere-based SoC from Samsung 8nm to TSMC 5nm would not be a matter of simple tapeout retargeting or minor RTL tweaks—it would necessitate a full front-end and back-end redesign, including physical synthesis, place-and-route, timing closure, power delivery, and validation, using entirely new EDA flows, libraries, and verification suites.

Furthermore, NVIDIA has not previously released any Ampere-family silicon on TSMC's N5 node, meaning there is no existing design collateral or hardened IP (e.g., PHYs, memory controllers, I/O subsystems) for that process. This implies that a 5nm Ampere SoC would be a from-scratch development effort, requiring a much larger engineering investment compared to adapting an existing chip like Orin (based on Ampere architecture and already validated on a mature process node).

In addition to the NRE (non-recurring engineering) costs, Nintendo would also face higher per-unit costs due to the wafer pricing at TSMC N5, which is significantly more expensive than Samsung’s 8nm. Altogether, these factors make a custom 5nm Ampere design both technically complex and economically unjustifiable for a console-grade SoC.

Right, the point I was making was that if Nintendo planned for a 2025 release from the start, they probably wouldn't have gone with an Ampere chipset at all so there wouldn't be any "porting", the T239-alternative or whatever it would be called in that alternative universe would be planned for Lovelace from the start. 

I wasn't talking about a T239 on 5N, I was talking about an entirely different designed chipset. 

Plus obviously Nvidia and Nintendo are going to figure this out for the inevitable mid-gen refresh. 



Conina said:
Soundwave said:

Prices have leaked for the ROG Ally XBox version and ROG Ally XBox X ...

Spanish gaming website 3DJuegos spotted what look like leaked listings for both the ROG Xbox Ally and the ROG Xbox Ally X on the Asus store for €599 and €899 respectively.

That would be like $700 and $1050 USD

No, the ROG Ally X was also priced €899 (including taxes) in Europe, but only $799 in the US.

Even with the slightly weaker US Dollar, it will be probably €899 (including taxes) and $899 (without taxes).

$900 only hey? Maybe it'll crack 20,000 in sales at that price. 



Soundwave said:
Conina said:

No, the ROG Ally X was also priced €899 (including taxes) in Europe, but only $799 in the US.

Even with the slightly weaker US Dollar, it will be probably €899 (including taxes) and $899 (without taxes).

$900 only hey? Maybe it'll crack 20,000 in sales at that price. 

So it won't even outsell a Gizmondo? Yikes.



Bite my shiny metal cockpit!

I'm very curious to see how the next sports games like NBA2k and Madden 26 run on Switch 2 using the 9th gen versions. Sports games typically give a good benchmark on the upgrades of graphics are between generations. I'm hoping this time around Switch 2 could support 60fps on NBA2k as opposed to 30fps we saw on Switch 1.

However, I think stuff like DLSS and the Series S being a weaker next gen counterpart will help the Switch 2 a lot in getting ports nearly identical to its counterparts. Like already stated in the thread and from the developer, it seems like GPU performance with DLSS included is similiar to Xbox Series S which I find insane for a handheld. CPU intesive games tho is where the Switch 2 could have more trouble where CPU is only somewhat more capable then a PS4. But I think the gap is much smaller compared to how it was between Switch 1 & PS4.



Switch even has an advantage against Series S in that it has 1GB more RAM available for games, leading games like Cyberpunk, Street Fighter, and Wild Hearts to have a texture advantage on Nintendo's hybrid.