Gaming Forums => Nintendo Gaming => Topic started by: Kairon on May 21, 2020, 05:57:08 PM
Title: Switch 2 Speculation
Post by: Kairon on May 21, 2020, 05:57:08 PM
So as time as passed I've fallen out of love with the idea of a "Pro" revision. There was a window where I think that had appeal due to worries about the Switch's power gap leading to consumer interest dropoff, and due to a belief that technology for a true generational jump in the technical form factor of the portable Switch wasn't ready at Nintendo price points yet.
Obviously, the Switch is still selling gangbusters, and I guess we could probably say that it's actually currently supply-constrained! It still has a power gap, but its unique portable proposition coupled with the power of Nintendo first party software make it look like it can have a full generational cycle. It's hard to imagine a hardware sales dropoff so extreme that Nintendo couldn't ride a healthy 2020 and competent 2021 for the Switch, taking the console into year 5.
Additionally, I think we're finally seeing the hardware appear that suggests a full Switch 2 successor could materialize at desired performance levels, which is what prompted me to make this post.
To begin with, I think the idea of a Switch 2 targeting XB1 or PS4 performance in a handheld is a good shorthand. I'm not an electrical engineer so this is all internet fanboy level speculation, but I'd generalize that target even further to hitting the 1 TFLOP target for the GPU, and of course increasing the CPU power appropriately.
Currently the Switch can be said to almost reach 400 GFLOPS docked, and almost 200 GFLOPS when portable. So as another rule of thumb, I would generalize a Switch 2 target EVEN FURTHER to aiming to double the Launch Switch's performance.
Actually, this would ALSO jive with Nintendo's current pattern of simply doubling performances every generation: The Wii was two GC's duct-taped together, the Wii U was double the Wii's performance, the Switch is even sort of twice the Wii U's performance.
We've already seen the Switch handle previously unimaginable current-gen ports with aplomb, so in conclusion I think a rough doubling of the Switch's performance, (which almost hits a supposed 1 TFLOP GPU performance target) is a reasonable yardstick to go by.
And by that yardstick, I think we're finally seeing tech in the consumer market that achieve this.
For example, the Surface Pro X launched in Fall 2019 with the claim that it had 2TFLOP GPU. The Tegra X1 ws claimed as the first mobile SoC that hit 1 TFLOP, so this is the sort of "doubling" claim that I've been looking out for. In this case however, MS is using a custom Snapdragon 8cx platform, which is Qualcomm instead of Nvidia. They're using A76 CPU processors and a mobile Adreno 685 GPU as compared to the Switch's A57 CPUs and Nvidia Maxwell architecture.
Again, I'm just an armchair crank with the internet, but on the CPU side I think the Arm Cortex A73 is already supposed to have twice the sustained performance of the A57's used in the Switch. The A76 is another generational improvement on top of that, but this time I think it's more like a 40% improvement vs. the A73/A75. (A LOT of marketing is centered around peak performance unrestricted by power or heat, which is OBVIOUSLY not realistic for the Switch.) So just on the CPU side I think we're seeing consumer-level offerings that are around 2.8X the Switch CPU.
It's easy enough to think about the CPUs, but the GPU is tougher. I have no idea how to compare the Adreno performance claims with the Tegra X1 inside the Switch since they're completely different lines. It SOUNDS like a mobile chipset hitting 2 TFLOPS bodes well for the technology being ready, but it's way guesswork than even I'm comfortable with, and obviously I'm indulging in a metric-ton of what-ifs here.
Ideally, I'd look at Nvidia successor Tegra chips, but these are uber weird things none of which actually are suitable for the Switch. So I'll just look at the GPU architectures they've been putting out. The Launch Switch used Maxwell cores at 20nm. The Maxwell architecture was succeeded by Pascal, but that sounds like most of the benefits of Pascal was jumping down to 16nm with very little innate improvements. The next meant-for-consumers architecture was Turing, which DOES exist in the market: Nvidia GeForce 16 GPUs hit the market in early 2019 as a budget line targeting $150 GPU SKUs, lacking exotic AI and raytracing. These are desktop GPUs, so obviously they're super-powered compared to what will the Switch will use. According to Wikipedia the GeForce GTX 1650 has 3.5X the number of shader processors the Switch has. But naively, imagine a Switch 2 with 384 shader units (compared to the Switch's 256) that are maintain the performance improvement ratio, and that means it should hit 1.083 TFLOPs! And that's considering the GeForce GTX 1650 has a die size of 200 mm^2, so an equivalently shrunk GPU would be 85mm^2 (Tegra X1's GPU size is 118 mm^2).
.... I think my justifications on GPU power doubling is a little thin, but since an Adreno GPU is able to claim hitting 2 TFLOPS, then I would think it's possible in the mobile space for Nvidia to do the equivalent.
Finally, I wanted to find current available tech a these levels because when the Switch came out in 2017, it was using cutting edge 2015 consumer tech. Likewise, if this tech is available in 2019/early 2020, then I could envision a Switch 2 landing in 2021 or 2022. Obviously prices on these components hopefully come down in time, and there are other technical components that need to be considered, like RAM pricing and performance, as well as the question of internal storage.
In conclusion, I've basically completely given up on the idea of a Switch "Pro", and now am focusing on imagining a Switch 2 that is on paper not quite an XB1 or PS4, but in reality is able to perform very much like one with a 3 hr battery life. I guess it could come in 2021, but with the Switch's current momentum, I'm more imagining it coming to market in early 2022 with a little bit more breathing room against the the XBSX and PS5 furor.
And of course, I'd envision it being 99% backwards compatible with Switch so that the generation can transition smoothly, the Switch "1" can move to lower price points naturally and the Switch 2 can have a little more price flexibility in case Nintendo needs to target a higher then $299 price point. If a Switch 2 launches in 2022, the Switch 1 would've had a 5 year lifecycle but still probably be doing really strongly. Perhaps there'd be a risk of cutting out the Switch's sales by releasing its successor, but I'd like to think that a large number of cross-gen games, as well as ever-more-palatable price points, would allow the Switch "1" to trail alongside the Switch 2 for a very long time, somewhat akin to the longer tails that the PS1, PS2, and GBA enjoyed even while their successors were on the market.
Title: Re: Switch 2 Speculation
Post by: Adrock on May 21, 2020, 10:41:35 PM
I'm such a slut for Nintendo hardware that I'd totally buy a Switch Pro. That said, I don't think Nintendo needs one. Typically, companies release those to spur sales. Switch is doing so well; Nintendo can probably ride out the generation.
I think Q4 2022 is the absolute latest Nintendo can launch a successor as Switch would be on the market for around five and a half years. Nintendo sat on Wii for two years longer than it should have and was still completely unprepared to launch a successor in 2012. It needs to launch a Switch successor while Switch is still getting support but clearly approaching the end like Switch did with 3DS.
PlayStation 5 and Xbox Series X are either getting delayed or "delayed" in that they both launch in 2020 but to such limited quantities (due to Coronavirus) they may as well be considered 2021 launches. Once the second wave of software starts releasing for those consoles and publishers begin to stop releasing PS4/One ports (not because it isn't possible, just that they don't want to anymore), Nintendo needs to be prepared with both new hardware and a strong first party lineup.
Every month or so, someone on ResetEra posts a Switch successor hardware capabilities thread. I'm not a hardware engineer so I only understand bits and pieces of those threads. Personally, I think Nintendo should focus more on the CPU and not just because Sony and Microsoft are both targeting CPU next generation. Like Switch, a Switch successor is always going to be judged on portable mode so GPU performance is less important (but not unimportant). I believe (at least at launch) Switch's CPU ran at the same speed docked and undocked while the GPU was underclocked undocked. Makes sense since the screen is 720p and to conserve battery. I expect a similar strategy for a successor.
As far as actual performance, it's in Nintendo's best interests to have Nvidia provide the absolute best hardware possible without the thing melting while still being able to launch at $300 (or $400 if it's feeling bold). The important thing is for Switch successor to get as many ports of PS5 and XBSX games as it can.
Backwards compatibility is a must including eShop purchases. I'll take 99% backwards compatibility. It should be 100%, but even Switch Lite isn't 100% compatible out of the box.
We should probably talk about the Switch successor screen. Personally, I don't give a **** about VR, but I kind of feel like Nintendo should have VR support (and that Labo stuff). I mean, Nintendo releases a head mount dealie the Switch successor docks into which means 1080p isn't going to cut it if Nintendo takes this seriously which it should.
Some odds and ends:
1. Include universal voice chat. Stop trying to reinvent the wheel especially now that Nintendo is charging for online. I don't even play online games. It should just be there.
2. Stop putzing with USB Type-A. USB4 should be fairly ubiquitous by the time a Switch successor releases. Just use that.
3. If a Switch successor has anything less than a 128 GB of internal storage, someone at Nintendo should get should get backhanded.
Title: Re: Switch 2 Speculation
Post by: Spak-Spang on May 22, 2020, 06:37:24 AM
I think Nintendo would be smart to ride out the generation. They came from a loss with the Wii U and they have a big success. Because it is viewed as a portable system people aren't looking at graphical parity. Nintendo needs to ride this train. Then launch a year or 2 into the PS5 and new Xbox console generation. Nintendo can safely position itself as the alternative system.
As for what the Switch should be. When the Wii was released and talk about a Wii 2 was around. I believed Nintendo should have made a "Super Wii" basically perfect the system and the features and just make the best Wii system they could. Nintendo didn't do that and I think they suffered for it.
The Switch 2 needs to be the best damned Switch console they can make...keep the core principle, but make it a better system. To the regards, I would make a more module Switch. Right now the controllers have some cool features in them, but many of them are not needed. Enhance the controllers by making them bigger and better quality and release with the idea of module controllers from off the bat.
Of course 100% backwards compatibility should be had. From this generation on people should expect that if they buy a game it will move with them to their next console, this should be come place, so that updating your console is like updating your computer. You will be able to play your games on Steam on your new computer...consoles should be the same.
Title: Re: Switch 2 Speculation
Post by: BlackNMild2k1 on May 22, 2020, 03:45:28 PM
I didn't finish the entire 1st post, but had to drop this thought while i had it....
I'm all for a Switch 2, and it waiting as long as needed. Imagine the marketing for an early 2022 launch
Switch to Switch 2 on Tuesday (2sday) 2/22/2022
To be here by known as Switch 2's Day!!
or is that too many twos and too many plays on two too? LMAO
Title: Re: Switch 2 Speculation
Post by: Kairon on May 22, 2020, 03:48:28 PM
Nintendo sat on Wii for two years longer than it should have and was still completely unprepared to launch a successor in 2012. It needs to launch a Switch successor while Switch is still getting support but clearly approaching the end like Switch did with 3DS.
I definitely agree with this. Normally I'd think a "Pro" as a half-step would be a way to achieve that, sort of how new iPhones release every year and they sort of flow into each other, but like I previously stated I think the window of opportunity for a Pro has mostly closed.
As far as actual performance, it's in Nintendo's best interests to have Nvidia provide the absolute best hardware possible without the thing melting while still being able to launch at $300 (or $400 if it's feeling bold). The important thing is for Switch successor to get as many ports of PS5 and XBSX games as it can.
I forgot to mention this, but I'm in agreement here too. The Switch actually punches above its FLOPS number because it's using such a modern architecture. Scaling graphics is one thing, but if Nintendo's architecture is just a complete outlier in comparison to the industry norm, like the GC, Wii, and even Wii U were, that just makes ports 10X harder.
It's probably more important that Nintendo get the most current hardware architecture and feature set that they can than hit the FLOPS targets I was talking about earlier.
Personally, I don't give a **** about VR, but I kind of feel like Nintendo should have VR support (and that Labo stuff). I mean, Nintendo releases a head mount dealie the Switch successor docks into which means 1080p isn't going to cut it if Nintendo takes this seriously which it should.
I don't think Nintendo should target VR with the Switch. Technically I guess you could make a head mount, but I think the Switch is just too heavy to be comfortable, the ergonomics is serving a different master, and that VR just doesn't reap enough rewards right now. Nintendo should just focus on improving the Switch, hitting price points, fixing weaknesses, and making games, instead of trying to redeem the Virtual Boy.
3. If a Switch successor has anything less than a 128 GB of internal storage, someone at Nintendo should get should get backhanded.
Storage is definitely a good thing to consider. With the PS5 and XBSX using bleeding edge SSD, I unfortunately can't fathom how Nintendo can replicate that in a portable form factor. I'd just hope for top notch MicroSD card support, and yeah, more internal storage memory by default.
Title: Re: Switch 2 Speculation
Post by: BlackNMild2k1 on May 22, 2020, 04:11:28 PM
Nintendo should just allow multi slot Hi Capacity SDXCHC compatibility. Allow us to jam in 2x 1TB+ SD cards and the system can bridge them like they are one card and shuffle storage accordingly to make the actual data fit on one card or the other where needed.
As for power.... XB1/PS4 lite sounds good to me. 1080p portable and 4k docked also sounds good, but I don't really care if it happens. I just know that if we are capable of doing native 4k docked, then we have more than enough graphical prowess to sway the naysayers.
Also if we are capable of doing 4k docked.... then maybe there is a way to release a VR headset separately that the switch can stream to, and the joycons 2.0 can act as the 3D controllers (not sure if you would need to use the docked switch 2's camera to view the motion of the controllers, but I do have an idea of a special IR reflective hand grip that the joycons can clip to).
I don't personally care for VR at the moment, but it seems cool, and if 3rd parties are interested, and Nintendo can come up with some good ideas too, maybe it'll justify itself down the road somewhere (and the Virtual boy does need it's redemption, just as the Switch is the redemption of the WiiU).
and once again, just picture the perfect marketing....
Switch 2's Day!!! Switch to Switch 2 on Tuesday (2sday) 2/22/2022
Let's make it happen Nintendo. It would also be a great late B-day present for me.
Title: Re: Switch 2 Speculation
Post by: Kairon on May 22, 2020, 04:58:45 PM
Nintendo should just allow multi slot Hi Capacity SDXCHC compatibility. Allow us to jam in 2x 1TB+ SD cards and the system can bridge them like they are one card and shuffle storage accordingly to make the actual data fit on one card or the other where needed.
I'd love multiple microSD Card Slots but I don't know if I see that as a major benefit. The PS5 will come with 825 GB internal storage via its fancy new SSD, so if the Switch 2 can take 1TB or maybe even 2TB microSD cards the sheer amount of storage shouldn't be a problem.
Rather, the sticking point will be performance of how quickly that storage can load.
and the joycons 2.0 can act as the 3D controllers (not sure if you would need to use the docked switch 2's camera to view the motion of the controllers, but I do have an idea of a special IR reflective hand grip that the joycons can clip to).
At first I was thinking this'd involve a lot of major work to get right, but then I looked it up and Oculus Touch 2 controllers, as used for the Oculus Quest, only cost $69! It sounds like if you have the right 3D spatial tracking tech, as Oculus does, then the tech in the controller won't be too much of a problem... I guess the question is just whether Nintendo can use/license those specific solutions if they wanted to go VR.
As a side note, I've wanted an Oculus Quest since I had the chance to try one at work. It "just worked" in a way that really reminded me of the Wii all those years ago. So I enjoy VR, I just think we've gone a long time trying to make it a thing and only now we're maybe sort of close to that being not a pipe dream? I've been meaning to buy the $400 SKU for forever since it came out, but, well, life happens.
and once again, just picture the perfect marketing....
Switch 2's Day!!! Switch to Switch 2 on Tuesday (2sday) 2/22/2022
This reminds me of all the Dreamcast 2 crazy release date coincidences conspiracy theories I inhaled during college. Believe!
Title: Re: Switch 2 Speculation
Post by: pokepal148 on May 22, 2020, 06:17:42 PM
Just give me Bluetooth headset support already.
Title: Re: Switch 2 Speculation
Post by: nickmitch on May 23, 2020, 12:14:12 AM
I wonder if Nintendo can put some hardware in the dock to give the Switch (2) some extra power. That could help keep the internals minimal but sufficient for portable play.
Title: Re: Switch 2 Speculation
Post by: Spak-Spang on May 23, 2020, 08:33:51 AM
NickMitch: That is an interesting idea. We had talked about those ideas with the Switch Pro but the internals of the Switch were just not capable of that.
If you were do design it from the get go. Maybe 2 USB 3, One for boosting performance like a graphics card and RAM upgrade and one that is for charging and sending information to the TV. (Maybe they actually don't need 2) But it would theoretically work like an external graphics card. Then Nintendo can allow developers to turn on extra features when docked.
I like the idea, but I am sure others won't. To me it makes sense and would even allow for future proofing if needed.
Title: Re: Switch 2 Speculation
Post by: BlackNMild2k1 on May 23, 2020, 11:57:54 AM
It would make sense for how much the dock already cost.
I should be able to buy 1 or 2 extra docks for other televisions, but for $100 (originally)..... I'm questioning what's in the dock besides a TV passthrough and power connection. I'm still wondering what it does that makes it cost more than $19.99
Title: Re: Switch 2 Speculation
Post by: nickmitch on May 23, 2020, 05:48:31 PM
There's not much to the docks but USB ports and HDMI out. I really don't think there's anything to justify the price tag.
EDIT: Someone on Reddit broke down the cost, arguing it's actually a fair price. (https://www.reddit.com/r/NintendoSwitch/comments/6cdvl4/the_switch_dock_set_is_not_overpriced_its_very/?utm_source=amp&utm_medium=&utm_content=post_body) I did recently buy a USB hub for my laptop last year, and the prices were not great. Also, USB type-c chargers weren't exactly cheap either, and I ended up buying anything one from Nintendo.
Title: Re: Switch 2 Speculation
Post by: Adrock on May 24, 2020, 01:14:50 AM
I don't think Nintendo should target VR with the Switch. Technically I guess you could make a head mount, but I think the Switch is just too heavy to be comfortable, the ergonomics is serving a different master, and that VR just doesn't reap enough rewards right now. Nintendo should just focus on improving the Switch, hitting price points, fixing weaknesses, and making games, instead of trying to redeem the Virtual Boy.
1. Target ≠ support. Nintendo's target should be improving what it already has because it works and it's popular. However, Nintendo can also try to support VR. They aren't mutually exclusive.
2. Personally, I think Switch is pretty light. For comparison's sake, Switch weighs ~0.66 lbs without Joy Cons, and PlayStation VR weighs ~1.30 lbs. More importantly, we're not talking about Switch. We're talking about a Switch successor, and we don't know how heavy it's going to be. Regardless, Nintendo has an entire hardware research and development department full of people whose entire job is coming up with solutions to things like comfort and ergonomics. This is the same Nintendo that didn't release the Wii Vitality Sensor because it only worked on 90% of people in the test group (it wanted 99%). That said, if for example, 10% of a test group said it was uncomfortable to use a Switch successor VR headmount for about an hour (for the recommended 10 to 15 minute break), maybe Nintendo doesn't bring that to market.
3. This has nothing to do with redeeming Virtual Boy. Why would the failure of a device that is now 25 years old ever be a factor? I don't even know why you brought that up. VR over 20 years into this century has everything to do with Nintendo being in a growing space as long as it's not only possible but viable. And it fits right into Nintendo's wheelhouse of supplementing its primary product without taking focus away.
I wonder if Nintendo can put some hardware in the dock to give the Switch (2) some extra power. That could help keep the internals minimal but sufficient for portable play.
This idea has been floating around for several years. I believe Nintendo filed a patent for a supplemental computing device in 2016 (?) which we all equated to the dock once Switch was announced. It's a neat idea, just not something Nintendo will likely ever end up doing. The patent seemed like something it was merely looking into. There are some fundamental problems a dock with extra computing power carries. How would Nintendo even begin to explain this to consumers? Why would anyone feel like they need this? It sounds like Sega 32X. I suppose if Nintendo were to release this with the understanding that it's a niche add-on, it could work. I don't see it happening. If Nintendo brings something to market, it expects to make money on it.
Additionally, how would Nintendo get third parties on board? To make an enhanced dock viable enough to release, Nintendo absolutely needs third party buy-in. Third parties would have to release games that take advantage of the additional power to make it worthwhile. My understanding is they don't even do that now with docked mode. If I recall correctly, third parties make Switch games primarily for portable mode because if it works undocked, it'll worked docked. Yeah, they'll try to bump up some features (frame rate and whatnot), but apparently few developers if any (at least outside of Nintendo) are spending a lot of time fine-tuning the games for the extra processing power when docked. I brought up VR which would require some tweaking, but we know some companies already do this with PlayStation VR compatible games. A headmount which likely includes some kind of head tracking, USB Type C plug, audio etc. would be what, $59.99 (which would probably be too much, but Nintendo is the same company that sells the non-charging Joy Con grip for $19.99). This brings me to...
How much would a dock with extra computing power cost? As stated, the current dock is an HDMI passthrough and a few USB ports, MSRP $89.99 and it doesn't even have an Ethernet port (the officially licensed Hori LAN adapter is another $29.99). An enhanced dock with extra computing power is already a hard sell in my estimation with everything I said above, might as well break out the Coffin Dance meme if it's over $100. Would Nintendo admit to gross price gouging with the current dock and release an enhanced dock next generation for under $100? If Nintendo released a dock with extra computing power, I imagine it would also have a regular dock.
That said, it's easier for Nintendo to just include all the necessary hardware into the unit itself and downclock in portable mode. That's straight up Occam's Razor. And this approach worked for Switch; it'll likely work for a successor. Personally, I think it's in Nintendo's best interest to leave the basics alone. It can add motion controls and HD rumble in a controller because those things can be dismissed, ignored, and even removed without causing too much controversy. As soon as you start putzing around with the fundamentals, you begin losing people.
EDIT: Someone on Reddit broke down the cost, arguing it's actually a fair price. (https://www.reddit.com/r/NintendoSwitch/comments/6cdvl4/the_switch_dock_set_is_not_overpriced_its_very/?utm_source=amp&utm_medium=&utm_content=post_body) I did recently buy a USB hub for my laptop last year, and the prices were not great. Also, USB type-c chargers weren't exactly cheap either, and I ended up buying anything one from Nintendo.
The Redditor used aftermarket prices which are marked up so the seller can make a profit. Nintendo buys parts in bulk directly from suppliers. The dock is probably less than $10 for materials, manufacturing, and labor. If the dock is manufactured in a different factory as the actual Switch unit, I wonder if shipping is the most expensive individual part of the process.
Title: Re: Switch 2 Speculation
Post by: Spak-Spang on May 24, 2020, 02:15:11 AM
Well if it was out of the Box Nintendo could allow the dock to upscale it. Developers would aim for undocked mode and a scaler could turn on additional particle, lighting effects and such.
Title: Re: Switch 2 Speculation
Post by: Adrock on May 24, 2020, 05:24:54 AM
That's basically what most developers currently do with docked mode. If they aren't going to put in the additional effort to truly take advantage of the extra computing power, I don't see the point of an enhanced dock. There's nothing wrong with the current approach with Switch. Just keep everything in the unit itself. The system is going to live and die by what's possible in portable mode. An enhanced dock is almost like Nintendo releasing another home console. Those days are gone.
If the implication that an enhanced dock is just the dock included with a Switch successor (rather than an upgrade to a regular dock), help me understand why that's a better solution than what Nintendo does now with Switch: a single SOC that has different clockrates depending on whether the unit is docked or undocked.
Title: Re: Switch 2 Speculation
Post by: Spak-Spang on May 24, 2020, 08:32:54 PM
Adrock I think the thing is the solution is the same, but because there is actual hardware in the dock you don’t have to worry about heat problems in the main unit. So it allows for a more powerful performance boost while docked. And since it would be around from the very beginning it would hopefully be used more.
Title: Re: Switch 2 Speculation
Post by: nickmitch on May 25, 2020, 03:49:36 PM
I wonder if Nintendo can put some hardware in the dock to give the Switch (2) some extra power. That could help keep the internals minimal but sufficient for portable play.
This idea has been floating around for several years. I believe Nintendo filed a patent for a supplemental computing device in 2016 (?) which we all equated to the dock once Switch was announced. It's a neat idea, just not something Nintendo will likely ever end up doing. The patent seemed like something it was merely looking into. There are some fundamental problems a dock with extra computing power carries. How would Nintendo even begin to explain this to consumers? Why would anyone feel like they need this? It sounds like Sega 32X. I suppose if Nintendo were to release this with the understanding that it's a niche add-on, it could work. I don't see it happening. If Nintendo brings something to market, it expects to make money on it.
Explaining it to consumers is easy; they just do it the same way they do now. The "Switch2" has extra power for enhanced gameplay when docked. If they were to make it an add-on for the current Switch, it's basically the same pitch as an N64 expansion pack or the Xbox One X. The games work good normally, but even better with this.
You do make good points for not releasing one for the current gen, but I'm not sure 3rd party support would be a concern of Nintendo's, as long as they still make games for the system in general. Nintendo would use it in their big games, and you'd get a few 3rd party titles to use it.
Title: Re: Switch 2 Speculation
Post by: Kairon on June 01, 2020, 02:05:53 PM
That said, it's easier for Nintendo to just include all the necessary hardware into the unit itself and downclock in portable mode. That's straight up Occam's Razor. And this approach worked for Switch; it'll likely work for a successor. Personally, I think it's in Nintendo's best interest to leave the basics alone. It can add motion controls and HD rumble in a controller because those things can be dismissed, ignored, and even removed without causing too much controversy. As soon as you start putzing around with the fundamentals, you begin losing people.
Yeah, that's convincing. I really like the tech idea of a Supplemental Computing Device, but it doesn't fit Occam's Razor. Nintendo needs to keep an eye on price, ease-of-use, and ease of development, all of which helped the Switch's performance and all of which an SCD complicates.
Besides, Switch 2 support will be based on the lowest common tech denominator, which will be its undocked handheld specs. If developers are making a game that can't perform on that, they won't change their mind because of the small chance players out there might have some SCD docks. What did Nintendo's analytics show? Only 20% of players use their Switch predominantly docked?
Ultimately, the best benchmark I think for a Switch 2 is the mobile chipsets that are in the consumer market 2 years before a hypothetical Switch 2 launch, like my original post explores. I think the tech is already on the market for a full generational leap to the "1 TFlop" marketing bulletpoint, leapfrogging any need for a Switch Pro. It won't come anywhere near to the monster power of the PS5 or XBSX, nor be able to have fancy fast-loading SSDs, but it'll still essentially be an Xbox One in your hands with a three-hour battery life and Nintendo games behind it (ideally with hardware backwards compatibility).
Best case, an SCD would be a curiosity like the N64 Expansion Pak. I had one of those and it was cool!
Title: Re: Switch 2 Speculation
Post by: Kairon on June 13, 2020, 02:33:34 PM
More random googling and I suppose we can use Qualcomm's Adreno 540 as a rough equivalent of the Nvidia Tegra X1 GPU: https://www.notebookcheck.net/Adreno-540-vs-Tegra-X1-Maxwell-GPU_7714_6152.247598.0.html
That same site doesn't have benchmarks for the Adreno 680 or 685 that's demonstrative of the latest Snapdragon 8cx chipsets I was talking about in my previous post, but they do have benchmarks for the earlier Adreno 650:
And where there does happen to be tests with data comparing the two, the Adreno 650 already looks to perform with generally approaching twice the oomph of the Tegra X1 Maxwell GPU.
I think that, assuming the Nintendo and NVidia keep pace and don't cur corners, they can easily meet or even exceed a target doubling the Switch's GPU performance with consumer level 2019/2020 tech, which could support a hypothetical 2021/2022 Switch 2 release.
Title: Re: Switch 2 Speculation
Post by: Kairon on June 13, 2020, 03:27:13 PM
Here's another thought experiment I just ran: what if Nintendo decides to be EXTRA Nintendo and not make any architectural improvements to the Switch's GPU at all? This is essentially what they did for the new chips in the revision and the Lite: they had NVidia make the new chips in the 16nm/12nm generation technology, but make practically no improvements beyond that. That could've given them a performance boost, but instead Nintendo opted to utilize it for power savings and a longer battery life instead.
This would be the least-effort approach, just porting the same GPU architecture to the new smaller manufacturing process and relying on only that for performance improvements. I'm personally not enamored of the idea because I worry that, raw power aside, the Switch benefits from supporting modern GPU features and processes, and if it starts building up a huge feature debt again then it'll become far less port-able a platform.
Still, if Nintendo chose to do this, I think on armchair-quarterbacking theory-crafting paper they could still reach a "double the switch" performance target on the GPU at the same power drain as the OG switch. According to this comparison (https://www.techcenturion.com/7nm-10nm-14nm-fabrication (https://www.techcenturion.com/7nm-10nm-14nm-fabrication)) TSMC's 16nm/12nm generation alone could yield 50% performance improvements over the OG Switch's 20nm GPU, the step to 10nm could yield another 15% over 16/12, and then 7nm, which IS currently being used in consumer tech, could give another 20% over 10. All that together suggests a 7nm die-shrunk Tegra X1 GPU could perform at 207% of current levels at the same power level. I suspect that might be a tad rosy, but it just gets us into the target area, even if Nintendo is being EXTRA Nintendo about things.
Title: Re: Switch 2 Speculation
Post by: Adrock on June 13, 2020, 06:49:37 PM
Doesn't the regular, vanilla Tegra X1 in the launch Switch have that hardware exploit hackers used? I thought that was the one of the main reason Nintendo wanted the new Mariko chips. My understanding is Mariko can get better performance due to its ability to draw more heat since its on a smaller node. Nintendo has no need for a significant performance boost. Again, Switch games are made for portable mode so Nintendo released a portable-only Switch and a regular-Switch with better battery life for portable mode.
And sure, Nintendo can go extra-Nintendo® on hardware power for a successor. It shouldn't. Period. By now, Nintendo knows every successful gaming hardware ever released, notably it has ever released has had consistent third party support. Even Wii which didn't have great third party support still had their attention. That third party support won't happen if Nintendo fucks around with hardware. Can the money Nintendo saves on hardware make up for the loses of having little to no third party support? No, probably not. Learn the right lessons. Nintendo got ports of Doom, The Witcher III etc. without even really trying to have hardware capable of running those games. Nvidia gave Nintendo stock Tegra X1, and it just worked out. The wonders of a more modern architecture. With that in mind, Nintendo has a tried and tested path for success. Lean into it this time. Unless Nintendo pulls another cultural phenomenon out of its hat, a Switch successor is going to have to be plenty capable.
I still think 2022 is the magic year to launch a successor. If Nvidia can provide Nintendo with PS4-level performance (minus the CPU bottlenecks) in portable mode, that would be ideal. I'm simplifying. Faster, more efficient RAM, solid state memory, DLSS (Deep Learning Super Sampling) etc. can all really help Nintendo close the gap to consistently score ports of PlayStation 5 and Xbox Series X titles.
Title: Re: Switch 2 Speculation
Post by: Kairon on July 07, 2020, 09:02:36 PM
I still think 2022 is the magic year to launch a successor. If Nvidia can provide Nintendo with PS4-level performance (minus the CPU bottlenecks) in portable mode, that would be ideal. I'm simplifying. Faster, more efficient RAM, solid state memory, DLSS (Deep Learning Super Sampling) etc. can all really help Nintendo close the gap to consistently score ports of PlayStation 5 and Xbox Series X titles.
I've just learned what DLSS is and it sounds really cool, a very powerful tool for Nintendo to increase performance without having to actually improve their GPU as much. It's convinced me that the tensor cores Nvidia has in their new GPUs, which I assumed Nintendo would just remove since it probably won't target ray-tracing, could actually be central to their strategy since they're necessary for DLSS. This could mean they'd have more flexibility in falling slightly short of a traditional GPU power goal, but DLSS making it visually not noticeable.
Title: Re: Switch 2 Speculation
Post by: Kairon on August 07, 2020, 05:36:11 PM
This is tangential, but earlier in this thread I talked about how I don't expect a Switch Pro. However, if there's some sort of Switch Upgrade that's the exact same tech but just increased internal storage (and maybe some super-minor improvements?) I can picture Nintendo releasing that as an upgrade choice. Distributing games digitally is becoming bigger and bigger and it wouldn't be too crazy to whip out an additional SKU to support that and try to entice hardcore fans to double dip.
Title: Re: Switch 2 Speculation
Post by: Stratos on August 24, 2020, 01:15:25 PM
So industry insiders have begun murmuring over a new Switch model set to release in the Spring of next year. People think it has something to do with the new Switch model references showing up in the Switch firmware code. Possibly a reduced price docked-only version. I like the idea of both a docked-only AND portable-only Switch, but there is one huge hangup for me. I want everything to be transferable between them. Maybe that can be the hook for a true next-gen Switch, where you have a totally portable profile and can seamlessly switch between multiple systems.
Games like Animal Crossing and Pokemon actively bar you from doing this in some form or another and its because of this I would never give up the 'switching' nature of a Switch. I heavily use all three methods of play (docked, portable, portable on a stand w/ separate controller) so I would never restrict myself without a way to transfer everything seamlessly. Its a driving reason that I am converting to nearly all digital, as I want everything available at all times.
One of the sources I've seen: https://nintendoeverything.com/rumor-taipei-based-newspaper-claims-new-switch-hardware-model-launching-early-2021/
Title: Re: Switch 2 Speculation
Post by: Kairon on August 26, 2020, 08:36:29 PM
I just posted a story on FCC filings which suggest new Switch internals: https://www.nintendoworldreport.com/news/54730/fcc-filing-hints-at-new-switch-soc-memory-components
That to me lends more weight to some new Switch revision coming out next year. The FCC filing though seems to be for a new form of the current Switch, and not so much a third SKU, at least from my non-expert attempts to understand it.
Either way, I do feel like we are getting closer to a Switch Pro. This seems contrary to my initial post saying we can go straight to a "Switch 2", but this is mostly semantics as I personally would call a Switch model that doubles the computing heft of the launch model as a "Switch 2".
Another hint is that Nvidia has a "Nano Next" entry level SOC planned for 2021. The current "Nano" hobbyist product is a $99 board that you can DIY build with, and seems to me like it's Switch's Tegra X1, on 16nm, but with only half the GPU active. It came out around the time the Switch Lite came out, so it seems like the this product closely tracks Switch's release since I suspect Nvidia is piggybacking the Nano off of the Switch's chip orders. So if Nvidia thinks they'll release a "Nano Next" in 2021, maybe it'll use whatever same new chip Nintendo is using in Switches come 2021?
Anyways, to revise my whole "No Switch Pro" stance: I believe it's possible for Nintendo to release hardware that effectively doubles the OG Switch's CPU/GPU performances in 2021. Regarding this evaluation, I'm not too concerned whether that's called a "Pro" or "1.5" or "2" or even whether or not Nintendo treats it as a whole new console or just a shinier more capable Switch. I think this could be achieved with new upgrades to new CPU and GPU architectures with performance metrics were available commercially in 2019. But this might also be accomplished just by taking the Switch Tegra X1 and just making it "better" with smaller process nodes, taking advantage of the three or so fabrication process node shrinks since 20 nm to simply brute force more performance out of the aging Tegra X1 design.
Title: Re: Switch 2 Speculation
Post by: NWR_insanolord on August 26, 2020, 09:48:55 PM
What we're seeing with Microsoft right now and I imagine is what we'll see from Nintendo going forward is the blurring of the line between a revision and a successor console. The Xbox Series X is fully backward compatible with the Xbox One library, and regardless of what Nintendo ends up calling this new model, I expect the same level of compatibility. Switch Pro vs. Switch 2 is a largely semantic difference.
Title: Re: Switch 2 Speculation
Post by: Stratos on August 27, 2020, 01:40:59 PM
I would be fine if they just stuck with the name and replaced the SKUs over time. Release a new model every other year or so and let people upgrade who want the latest. The scaling capabilities of the Switch could allow this to occur seamlessly until they pull a Microsoft and release an "X" version that then cuts off the older models from newer high-end games. People understand this with Apple and Android products. Some newer software only runs on a newer OS/phone model.
Those old systems would go out into the market (potentially to siblings and friends of the original owner who don't own a system) and this could create more Switch users over time to buy more software.
Title: Re: Switch 2 Speculation
Post by: Kairon on August 27, 2020, 04:08:34 PM
I would be fine if they just stuck with the name and replaced the SKUs over time. Release a new model every other year or so and let people upgrade who want the latest. The scaling capabilities of the Switch could allow this to occur seamlessly until they pull a Microsoft and release an "X" version that then cuts off the older models from newer high-end games. People understand this with Apple and Android products. Some newer software only runs on a newer OS/phone model.
Those old systems would go out into the market (potentially to siblings and friends of the original owner who don't own a system) and this could create more Switch users over time to buy more software.
Agreed. That was my assumption about how things would proceed once the Switch launched. However it looks like the timelines are a little more drawn out than I expected due either to waiting for the consumer tech to become affordable enough, or just Nintendo not wanting to move too fast when what's on market right now is still selling out reliably.
Title: Re: Switch 2 Speculation
Post by: Ian Sane on August 27, 2020, 10:53:32 PM
I would be fine if they just stuck with the name and replaced the SKUs over time. Release a new model every other year or so and let people upgrade who want the latest. The scaling capabilities of the Switch could allow this to occur seamlessly until they pull a Microsoft and release an "X" version that then cuts off the older models from newer high-end games. People understand this with Apple and Android products. Some newer software only runs on a newer OS/phone model.
Those old systems would go out into the market (potentially to siblings and friends of the original owner who don't own a system) and this could create more Switch users over time to buy more software.
I feel if you do this there effectively aren't consoles anymore. There is just a series of software platforms. Consoles were supposed to be the easy approach to videogames - you buy the system and everything released for that system works 100% with the same experience for everybody. It has blurred with the PS4 Pro, Xbox One X, DSi, and New 3DS. This setup where at one point new games run poorly on your system or not at all and you're pushed towards buying a new model but without any clear distinction between generations introduces the hassles of PC gaming that made it less accessible.
I feel this makes Nintendo and Sony direct competitors with Apple. If their platform lacks the distinctiveness of a console then I don't see how the general public makes the distinction. Apple has games too, plus other software that is not game related. Nintendo's game focus may make them the choice for gaming or it could have the negative effect of their platform just coming across as lacking features since it probably won't have the extra bells and whistles. The less of a console these things are the less reason there is for all of these similar computers to co-exist in the market.
I think Nintendo would do best if they stuck with the console approach in contrast with everyone else so that they're the straightforward games company, designing hardware specifically for games and using the console model that is easy to understand, particularly for families who don't want to have to buy their kids a new console every two years for minor improvements. Nintendo did the New 3DS to try to get people to double-dip, which isn't very nice, but they do not demonstrate any particular need to improve hardware to tweak frame rates. They've exclusively used outdated hardware for 14 years now. Updating the hardware every five or six years like normal fits their approach to game design.
Title: Re: Switch 2 Speculation
Post by: Spak-Spang on August 28, 2020, 10:46:39 PM
I agree with Ian. I don't want to update a new console every 3 years. I want to know my investment is safe for at least 4 years...I want to know that games will be playable on the console and I don't need to upgrade. Gaming is expensive and paying a premium for a console and a console game makes me want to ensure it is going to last. I don't care about graphics eventually being outdated, but I do care if after 2 years of having a console the next Mario game will require another big investment.
Title: Re: Switch 2 Speculation
Post by: nickmitch on September 09, 2020, 05:36:21 AM
Wasn't sure if I should drop this here or dig up the rumor thread, but
Nintendo said to be boosting Switch production by 20% while telling developers to make their games 4K ready. (https://www.bloomberg.com/news/articles/2020-09-09/nintendo-said-to-boost-switch-production-by-another-20)
Title: Re: Switch 2 Speculation
Post by: ThePerm on September 11, 2020, 12:18:13 AM
Nintendo usually goes between 2 strategies. Either they release hardware that is simply a superior version of their previous hardware or they "do market disruption through withered technologies"
One thing that was lacking on the Switch that the 3ds and the Wii had was an array of cameras. We're at the point where we could have really good cameras on the next switch, much like what you would get in your cellphone. Nintendo has the capability of doing really crazy AR stuff.
Given Nintendo history. I imagine the Switch 2 will be more powerful than Switch, but it will be hard to say if it will try to stand on the same ground as the upcoming next gen systems. Those are high hopes. However, Nvidia might reward the system for helping to boost their company so much since the Switch launch. Nvidia has some great engineers and I imagine the price structure is more up to Nvidia's whims than material and labor costs.
As Ian said "I feel this makes Nintendo and Sony direct competitors with Apple."
This has to some extant always been the case. Apple has games on their phone. So, technically it's all war. I imagine the next switch will have a bunch of stupid free social apps like phones have. Nintendo actually invented a lot of those things with the Wii. The Wii came out and had a bunch of weird channels on it. When I use any device now, the system UI resembles a lot what was on the Wii and the PS3.
There is a point in technology where the Switch 2 will probably be very competitive to phones. I predict more cameras on the Switch 2 and with those a lot of what makes phones special is going to go away. The Switch 2 will seem attractive to people who want an iPhone but don't want to spend 700 dollars on one. If they include VoIP with their next video chat app then it could make phones look antiquated.
Title: Re: Switch 2 Speculation
Post by: nickmitch on September 12, 2020, 06:51:22 PM
One thing that was lacking on the Switch that the 3ds and the Wii had was an array of cameras. We're at the point where we could have really good cameras on the next switch, much like what you would get in your cellphone. Nintendo has the capability of doing really crazy AR stuff.
We can see Nintendo showing off their AR skills with Mario Kart Live, which is nice that they're getting back into it. But I worry Nintendo is too cheap to leverage AR in a console. The 3DS had some pretty low quality cameras. I remember taking some fun photos with it, but not wanting to share after a while because of the low resolution. To compete with phones, they'll really have to step it up in that regard.
