• Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
  • We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.

Instro

Member
Oct 25, 2017
15,126
The Switch doesn't have that much bandwidth because it it simply doesn't need it. It's not doing anything with the CPU that would warrant massive bandwidth, it has very advanced memory compression and its GPU is a tile renderer. The upgrade to a new SoC does not warrant that expense.
Same with Thunderbolt, what for? There are millions of Switches out there without Thunderbolt, why divide the userbase?

I think it's fair to say that memory bandwidth is one of the likely culprits for games having lower than expected resolution. Certainly problematic with alpha effects regardless of compression.
 

RobotVM

Member
Oct 27, 2017
1,417
I honestly am hoping if they do a revision it just replaces the current model and is still just called a Switch with the same SKU. That way I can wait until early next year and bring back my current Switch and change it out for a new one under my 2 year geek squad warranty.
 
Oct 27, 2017
5,618
Spain
You're right that it would divide the existing userbase but it would also future-proof the new model a bit more. I guess that's not a very good idea especially if this is a quiet, unnamed upgrade, since you won't be able to tell consumers that only Switch X can use certain features (like a SCD dock) if Switch X has the same name as the regular Switch.

Regarding memory bandwidth I think you're right that, when properly optimized for the Switch there aren't really bandwidth concerns, but you can't assume every developer has time to properly optimize. The Bayonetta ports don't increase resolution when docked, and most suspect that's due to the memory bandwidth causing problems with all of the alpha effects. Maybe if they spent a long time optimizing for the memory compression and tile based rendering process they could get it to 1080p docked but clearly they didn't manage to do that.

So it's really just important- for ports- to give developers as many easy ways to remove bottlenecks as possible. Doubling the RAM bandwidth (eventually, not necessarily in this revision) would be a way to do that.
Maybe Bayonetta 2 is limited by texel fill rate? Platinum throws alpha effects around like crazy, and even if you have fast cache cache to draw to, fill rate is a fixed limitation.
 

Hermii

Member
Oct 27, 2017
4,724
A few quick thoughts while I wait for a plane:

I don't see a TX2-like chip (ie 2 SMs on 16/12nm) being enough of a boost in performance to warrant a "more powerful" Switch. You'd be looking at perhaps a 50% improvement in clocks in both modes, which is just more of an annoyance for developers than anything else. I do, however, see a few ways in which a 16/12nm refresh might make sense.

  1. They just want/need to get off 20nm. Perhaps Nvidia's used up their wafer commitments and moving node is cheaper in the long run. Perhaps they were only on 20nm for as long as it took them to get a 16nm chip up and running. In this case, we'd be looking at a simple boost in battery life and that's it. Possibly even a silent release with little to no PR/marketing.
  2. They want to release a smaller, cheaper portable-only model to expand the player base (particularly for games like Pokémon) and need a newer fabrication process to eliminate the active cooling.
  3. They want to release a fixed home console Switch but use the same SoC as the main device. A 16/12nm chip with beefier cooling than Switch could accommodate could perhaps bump GPU clocks by 70-80%, and they could jump up to 8GB of RAM at double the bandwidth. Not enough for 4K or anything like that, but more reliable 1080p and higher resolution textures in a box that would still be small and cheap.
Number 1 is obviously the most likely, but the others wouldn't be impossible, and would be cheap enough for Nintendo to put out if theyre using the same SoC across the board.
The only problem I see with the theory 1 is why double the ram? The reason that does make most sense to me is they are using a 128bit bus width, which would require 8gb to get up to double bandwidth? And if they are doubling the bandwidth and the amount, that would indicate a more powerful model.

I agree that #1 seems like the likeliest thing here. I doubt it's even "based on TX2" as much as it's just a die shrunk TX1, at least going by the SoC number.

I'm curious what other changes they may make to the new Switch though, besides fixing the security issue. Like, a Thunderbolt ready USB-C port? Maybe double the RAM bandwidth?
T210 A2: Regular Switch, A2 because they changed the clockspeeds.
T214: Jericho
T186: TX2

Hard to make sense of those numbers, but my guess would be a die shrunk TX1 with customisations like removing all the codecs Nintendo doesnt need, remove the a53 cores which aren't used anyway etc. Which is basically the same as a TX2 without Denver cores. So what we have is a stripped down, die shrunk semicustom TX1 much like the one we hoped would be in the original Switch. Also I belive they doubled the bus width because thats the only way 8gb would make sense, so they can fully make use of the wider bus. This would mean double bandwidth.
 

Thraktor

Member
Oct 25, 2017
571
There is not even such an improvement in power efficiency between 28nm Maxwell and 16nm FinFet Pascal, and 20nm at low frequencies (Like the Switch uses) should have the power efficiency advantages of a die shrink compared to 28nm. There is a reason Nvidia chose to implement it in the Tegra X1 but not in Maxwell parts that ran at high frequencies like PC gaming cards.

My opinion is, the upgrade will be completely silent. We'll need 7nm in late 2020 or so for a leap over the current Nintendo Switch.

About a possible home console, a 16nm version of the current chip clocked 80% higher and with 8GB of RAM and twice the bandwidth would be sitting in XBOX One territory... But I doubt something like that will happen.

I meant 50% more as an upper bound than anything else, probably should have been more clear. My point was just that, even if they increased clocks by 50%, it still wouldn't be a big enough jump in performance to warrant a new Switch Pro, or whatever people would want to call it.

I agree that #1 seems like the likeliest thing here. I doubt it's even "based on TX2" as much as it's just a die shrunk TX1, at least going by the SoC number.

I'm curious what other changes they may make to the new Switch though, besides fixing the security issue. Like, a Thunderbolt ready USB-C port? Maybe double the RAM bandwidth?

Thunderbolt is both very expensive, and not much use on a new Switch (particularly if it wasn't on the original model). It's also only ever been used on Intel-powered products, although Intel have said they'll be opening it up, I don't know if there's been much progress on that.

Regarding memory bandwidth, there's not a whole lot of point increasing it if you're not increasing the other specs along the way. You'd get a bit of a benefit in bandwidth-limited scenes in a handful of games, but it would be of sporadic benefit and I can't imagine most players would notice the difference.

I'm pretty sure it's also possible to get 8GB of LPDDR4 on the same 64-bit bus as the current Switch, by the way (not that I think they're going to release a Switch successor with 8GB RAM any time soon, but FYI).

I think it's fair to say that memory bandwidth is one of the likely culprits for games having lowe than expected resolution. Certainly problematic with alpha effects regardless of compression.

Lower than what expected resolution? People seem to be expecting Switch to render everything at a constant 1080p, but XBO doesn't come close to that with most third party games, and it has a larger SoC operating at a far higher power draw.

The only problem I see with the theory 1 is why double the ram? The reason that does make most sense to me is they are using a 128bit bus width, which would require 8gb to get up to double bandwidth? And if they are doubling the bandwidth and the amount, that would indicate a more powerful model.

The logic would be that 1 and 2 would both use 4GB of RAM, as with the current Switch. The 8GB would be just for the home console.
 

Ashhong

Member
Oct 26, 2017
16,741
I need to send my Switch in repair because the screen is coming out of the device a bit..I should do that before this revision hits and they just straight replace it or something
 

Red Liquorice

Member
Oct 27, 2017
9,094
UK
The last 2 gens (DS and 3DS) I've bought the first hardware revision and regretted not holding out until the second (third model), so nice try Nintendo, but I'll wait for Switch Mk3.
 

Frankie_baby

Banned
Feb 11, 2018
628
The only problem I see with the theory 1 is why double the ram? The reason that does make most sense to me is they are using a 128bit bus width, which would require 8gb to get up to double bandwidth? And if they are doubling the bandwidth and the amount, that would indicate a more powerful model.


T210 A2: Regular Switch, A2 because they changed the clockspeeds.
T214: Jericho
T186: TX2

Hard to make sense of those numbers, but my guess would be a die shrunk TX1 with customisations like removing all the codecs Nintendo doesnt need, remove the a53 cores which aren't used anyway etc. Which is basically the same as a TX2 without Denver cores. So what we have is a stripped down, die shrunk semicustom TX1 much like the one we hoped would be in the original Switch. Also I belive they doubled the bus width because thats the only way 8gb would make sense, so they can fully make use of the wider bus. This would mean double bandwidth.

Would there be much point removing the a53s? They're tiny and while I don't have any knowledge on the subject I'd imagine they are probably used in sleep mode?
 

-shadow-

One Winged Slayer
Member
Oct 25, 2017
6,110
there is an analysis from DF, where is what I remember was reported to be the situation with the framerate
I think you might remember incorrectly. Bayonetta (2) run way better docked than portable. The thing however is that nothing (2) or barely anything (1) scales at all from portable to docked. It's the same quality, just much more stable.
 

jts

Banned
Oct 26, 2017
3,018
I still love it's perfect compactness compared to my floppy toy-like 3DS XL, but why would you say it was better than the DSi? I automatically assumed the DSi would be the better model.
Well, for one, DSi lost the GBA slot. Other than that, DSi was technically better but you have to take the timeframe into account. DSi was released 3 years after the DS Lite, and those were some of the hottest years in that generation. Only to offer marginal improvements and little exclusive software. To put that into perspective, 3DS was released "only" 2 years after DSi which by that metric would have made the DSi a very bad purchase, since 3DS was an actual generational leap (with DS/DSi BC for the most part).

DS Lite at launch or first 18 months was a great place to be. Missed that clunky first gen, got sleek hardware, retained GBA compatibility, and the DS software was just exploding in quality and quantity. I guess it would only be slightly regrettable if you got it near DSi launch.
 

xxbrothawizxx

Member
Nov 1, 2017
1,233
Gainesville, FL
Ummm... that makes it even less portable.
It's not very pocket-able as is. If you already need a bag or case to hold it, why not just toss in a power bank with it or buy a good battery case?

Some people rarely take the system out of the dock, and those that do so frequently for commutes have probably been served well enough by the 3 hour life. There was definitely an uproar over the 3DS's battery life after coming from the energizer bunny that was the DS. Ultimately, nobody, including Nintendo, seemed to care much at all and that was purely a handheld device. With the Switch there is even less reason to prioritize improving in that area.
 

Red Liquorice

Member
Oct 27, 2017
9,094
UK
Well, for one, DSi lost the GBA slot. Other than that, DSi was technically better but you have to take the timeframe into account. DSi was released 3 years after the DS Lite, and those were some of the hottest years in that generation. Only to offer marginal improvements and little exclusive software. To put that into perspective, 3DS was released "only" 2 years after DSi which by that metric would have made the DSi a very bad purchase, since 3DS was an actual generational leap (with DS/DSi BC for the most part).

DS Lite at launch or first 18 months was a great place to be. Missed that clunky first gen, got sleek hardware, retained GBA compatibility, and the DS software was just exploding in quality and quantity. I guess it would only be slightly regrettable if you got it near DSi launch.
Not sure I ever knew the DSi didn't have a GBA slot. I did actually use that on the Lite, so it was a definate plus for me.
 

BDGAME

Member
Oct 27, 2017
1,090
Brasília
One question about the Parker. It throttle like the X1 and will need to be put in a underperfomance, or it will be able to run at full power?

Other question: how much is the power consumption of it? Running at full power it consume the same amount of Switch's undock mode? If yes, Nintendo can use it with overclock on docked mode? What kind of power a overclocked Tegra Parker can archive?