Slownenberg said:
Soundwave said:
The "wait for years to have better hardware argument" is non-sensical.
For starters the PS5 and XB2 are not magical unicorns. The RDNA2 architecture they use is only now catching up to Nvidia chips from 2 years ago. AMD seriously lags behind Nvidia. Nvidia Ampere (which a 2023 Nvidia chip would likely be based on) will still be architecturally better than RDNA2, so as long as the chip is Ampere based it will have the architecture that's better than what's on RDNA2 chips. There is no point in waiting for Orin based chips to have better architecture than RDNA2, Ampere will already do that without much fuss.
The Switch 2 chip really would only need to be about the same to the PS5 as the Tegra X1 was to the XB1 circa 2015.
If it is and you add in DLSS 2.0 or 3.0 implementation, the Switch 2 becomes basically effectively 3-4x more powerful because it only needs to draw 1/4-1/15th the pixels is some cases.
If DLSS had been available on the current Switch, it would be able to run PS4/XB1 games no problem.
The issue with waiting so long is third parties invest too much into PS5/XBSX/PC's pipeline that those systems become their focus, they're not going to care that much about a platform that is starting way back at 0 in 2024 when PS5 is well past probably 50 million units by then + whatever XBox has.
The current Switch can actually handle a lot more PS4/XB1 ports, devs just don't want to bother with it because the PS4 + XB1 + PC userbase is so far ahead and they don't want to change mid-cycle that much. Timing is as important as hardware, if you wait too long developers have already made their bed for the generation and are not going to be as amenable to including something else.
|
No. Just no.
First off no one is saying wait years to get a more powerful system. Most people agree 2023-2024 launch is when would make sense, but if the later part of that range means a Switch 2 that can handle home console ports without a ton of issues and therefore it will get way more multiplats then it makes sense to launch in the later part of the reasonable Switch 2 launch range unless the Switch is totally dead by 2023 which seems incredibly unlikely.
And why are you talking about magical unicorns? Home consoles are obviously more powerful than a portables. No the Switch can not handle more than the PS4/XB1, that's ludicrous, it isn't nearly as powerful as them.
At least you are right about one thing, devs don't want to bother with porting games to Switch, but you're wrong about the reason. If Switch could easily handle those games like you suggest then of course they would port them, otherwise they'd just be throwing away money. The whole strategy behind mutliplats is to get them on as much hardware as possible. They don't bring them to Switch because it takes a lot of effort to port it over and figure out how to get it running well on Switch and what downgrades should be put in and trying to figure out if it'd just be so downgraded as to not be a good experience so people won't even be interested in it. And that is the whole point of waiting later in the launch range for Switch 2 if that means getting that extra bit of power to bring it close enough to the home consoles that third parties don't have to try to figure out if porting and downgrading a game to the Switch 2 is gonna leave it in a poor state, they'll just be able to port it over with perhaps a few simple graphical downgrades equivalent to lowering the settings on the game (i.e. lower res, shorter draw distance, little bit lower frame rate, but everything else the same).
Saying third parties won't invest in Switch 2 because they're just gonna stick to PS5/XBS/PC pipeline is the whole point of what we're talking about. We want a Switch 2 that is powerful enough to be in that pipeline! To say that is non-sensical is itself non-sensical!
|
I said the current Switch can run more PS4/XB1 games than it currently gets, developers don't want to bother because it came out too late in the game.
The Tegra X1 was released in May 2015 ... the PS4 was only 1 1/2 years old at that point, it's already capable of running PS4 games.
The Tegra X1 can basically draw basically the same visuals a PS4 can, the main difference is it has to do so at a lower resolution, so you get an image that's blurrier/more distorted. I think devs look at that and say "well it's going to be an ugly looking port, not worth our time". Mortal Kombat, Witcher 3, DOOM, etc. do run pretty well on Switch, the main issue with them really is they image quality is kind of ugly and I think that is putting devs off (why put in the effort to make a Switch version when the end product is going to look image wise a good deal worse).
DLSS completely changes that, it will let Switch 2 to render at even lower resolutions while having image quality that's actually better. You'll be able to get a very solid 720p-1080p image undocked reconstructed up from as low as 640x360 (this is N64 tier resolution, lol). Docked you'll be able to get 1440p-full 4K easy while only rendering 720p. Lower actual base resolution also means you can bump up the graphical details, you don't have to have everything on low settings, you have overhead horsepower now that you didn't before.
As long as Nvidia can give Nintendo a chip for 2023 that is equivalent to what the Tegra X1 was for 2015, they will be able to do PS5 ports easier than what's possible now with Switch-PS4. DLSS 2.0 completely changes the pixel resolution requirements that dramatically.
So as long as the Switch 2 is basically equivalent to what the current Switch is relative to XB1, you'll have next-gen ports that are easier to do. The architecture is the big thing, AMD's RDNA2 is basically only catching up now to Nvidia Turing from 2018, Ampere will be better than Turing. As long as the architecture is there it's really then just what performance (resolution + fps) you can get, but DLSS is a game changer on that front.
Last edited by Soundwave - on 15 May 2020