NSW2 will run 360p @15fps. (Nintendo plans on creating a whole lotta games using stop-motion & irl figurines this upcoming gen. Kinda like Cuphead with cartoon animation, but instead with stop-motion figurines. They're doing this at the instruction of Miyamoto, who has been quoted as having wanted to "play the promotional art of Pikmin 1+2".)

Yes agreed. Any price point that starts with a 4 would be a huge mistake by Nintendo. I can't imagine anything over $399.99. And $499 would just be insane. I see no reason that they need to jump anything more than $100 from Switch launch model.

Maybe a souped model in a few years comes in at $4XX, but definitely not a launch model. $400 is already very much pushing the affordable price point that their customers expect.

I also would have originally expected $350 except for the fact that they never dropped the price of the Switch, and obviously they aren't releasing a new system at the same price as the previous gen OLED. So it's pretty hard to imagine this will cost anything other than $399.

Literally anything can be 60fps. FzeroX on N64 was 60 fps. It has nothing to do with hardware, just with what the developers choose to focus on. And most games in no way need 60fps. First person shooters absolutely need 60fps otherwise panning around looks choppy or blurry, its good for racing games, and games with high speed action. Otherwise 60fps is entirely unnecessary.

Anyway, if you want 60fps in all games, that has literally nothing to do with the specs of the system, you gotta start sending out letters to every game studio haha.

Hmm, if Nintendo uses 8nm on Switch 2 that means that they are really aiming for a low price for the Switch 2:

350$-400$ i could see happening.

Valve launched the Steam deck for 400$, with better storage, RAM, battery and node than the Switch 2. So yeah anyone thinking the Switch 2 will be expensive can just compare the specs of the Switch 2 with Steam deck to see that Nintendo is aiming to make a cheap console that won't be a big jump from the price of the Switch.

If we get an 8nm SOC then $350 should be a fair price. That node is CHEAP nowadays for low end markets like 3050. Even $299 should be possible considering the competition:

• PlayStation Portal is $199 with a comparable shell, screen and battery.
• Steam Deck is $399 with more expensive components and production node and isn't directly subsidized by game licenses
• A RTX 3050 can be had for $180 and a 4060 (TSMC 4N) for $320

Samsung 8nm belongs to <$300 devices.

I am expecting $400 still, even if it is 8nm. Japan has experienced its first real, prolonged, inflation since the 1980s. The rest of the world even more-so. ¥29,980 in 2017 is about ¥33,547.77 now. $299 USD in 2017 is about $388.46 USD now. It also lines up with their other consoles. 

Also going with 8nm might be cheaper now, but can be more expensive in the future, meaning Nintendo is going to want a die-shrinked SKU sooner rather than later, which costs R&D that can be offset by large revenue intake. 

Worst case:

• 8nm Samsung
• 0.7 mobile / 1.3 docked TF
• No support for DLSS FG
• No support for G-Sync /FreeSync
• $399

Best case:

• 5nm Samsung
• 1.4 mobile / 2.2 docked TF
• Support for DLSS FG
• Support for G-Sync /FreeSync
• $349

Best case scenario is more like 4 TFlops (1.3 GHz GPU clock) with Nintendo loosening the TGP to 20W in docked mode. That'd put it at about 70% of a Series S in terms of performance (seeing how Ampere <-> RDNA2 tend to relate that much TFLOP-to-TFLOP on average. An RTX 2050 is about equal to an RX 6400 in performance, despite the latter having 70% of the TFLOPS estimate of the prior.) 

Worst case scenario is probably something like 2.3 TFLOPs (750MHz) in docked mode, that being the same docked mode clock rate as the original Switch. That'd put it about 40% of a Series S, in terms of performance, in docked mode. 

.7 TF corresponds to about 230 MHz. That is lower than the original Switch's clock rate in handheld mode and probably very, very inefficient. Nintendo and Nvidia would've been better off going with a much smaller chip if that is their mobile clock-rate target. (EDIT: I don't think Ampere can even go that low. It's minimum voltage would demand a frequency higher than that. Around 400 MHz, if I recall correctly.) 

My guess is that the chip is going to be 3 TFLOPs (975MHz) if on 8 nm, when docked, and 3.4 TFLOPs (1100 MHz) if on 5nm when docked. That'd put it around 50% to 60% of the Series S in terms of expected performance, before any DLSS gains. Halve all of that for handheld mode. 

If Nintendo were aggressive with the power-budget this time around (given they have better cooling), we could potentially see 1.3GHz (4 TFlops) in docked mode. 

Just for context the lowest clock rate of any Ampere chip is 562 MHz in the RTX A2000. That's a 26 SM chip, which is going wide and low frequency for efficiency. I'd expect the handheld Switch 2 to beat that efficiency, but not by more than half the clock rate reduction. Probably something like 450MHz, given what we've seen with other Orin chips, is the base-line, with higher modes being options, like with the original Switch. That'd be a handheld-mode baseline of 1.38 TFlops. 

Steam Deck has 1.6 TF with TSMC 7N/6N process and the SOC can reach >90°C in complex 3D games. Switch 2 has to be slimmer, cooler, lighter, longer battery life at a worse (Samsung 8nm) or comparable process node (Samsung 5nm). You can't make comparisons to home consoles with their high-juice SOCs. OG Switch has 0.15 TF in handheld. Going to 1 TF is an 8x increase and I am not convinced that we get 8x more raw performance.

The Steam Deck goes for a high frequency, few CU design and has an x86 CPU that isn't going to be nearly as efficient at low wattage, regardless of node. It is far less customized than the T239. 

Again, we already know how many SM's the T239 has. It's a matter of calibrating for the right frequency. 230 MHz is way too low for an Ampere chip, and iirc impossible at non-idle P-States. The minimum Ampere frequency is ~400MHz. That puts the minimum possible TFlops at about 1.2.

You can't compare TFlops across architectures like that anyway. Maxwell isn't Ampere isn't RDNA2. As Haxxiy noted earlier in this thread real performance per TFLOP is much higher on RDNA 2 than Ampere to the point that an Ampere "TFLOP" (for the purpose of extrapolating gaming performance from raw floating point compute) is something like 60 to 70% of an RDNA2 "TFLOP", if you want to extrapolate gaming performance (gaming is a mixed FLOAT - INT compute-load versus, say, Machine Learning or Scientific Computing.) 1.6 TFLOPs of the Steam Deck should give similar gaming performance to 2.3 TFLOPs of the Switch 2, all else being equal.