• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Rumor: Switch developer information leaked. Reveals A TON incl. hardware specs

Question: will performance (framepacing, frames per second, etc)see a hit while portable?

I believe there were reports that the Zelda demo had more pop-in, but surprisingly, better framepacing in handheld. The Zelda demo wasn't final code though, so make of that what you will.
 

Fularu

Banned
So it says:

Wired LAN - USB connection option, 1 gigabit per second


How are they getting that 1 gigabit per second stat?

If the dock is limited to USB 2.0 currently, and the LAN adapter sold at stores seems to also be just a USB 2.0 adapter ? How are they getting a 1 Gigabit over USB 2.0 ports?
The dock has USB 3
 

Ryoku

Member
Question: will performance (framepacing, frames per second, etc)see a hit while portable?

Based on the hardware specs alone, that shouldn't happen. But game complexity is well... more complex. Devs may also want to account for battery life, so that's another point to remember.
 
Because it's not $700? People who bring of phones will always confuse me. "Why isn't this $300 handheld device as powerful as my $800 phone?"

Not to mention architecture is different.

Pretty sure the point he's making is that it's unreasonable to expect a console only slightly larger than a smartphone to be as powerful as a PS4 Pro.
 
I believe there were reports that the Zelda demo had more pop-in, but surprisingly, better framepacing in handheld. The Zelda demo wasn't final code though, so make of that what you will.

Interesting. I wonder if pop-in would negatively affect the experience of an open world Zelda.

Based on the hardware specs alone, that shouldn't happen. But game complexity is well... more complex.

Gotcha. Yeah I should have clrified to ask if at face value (specs) there would be any issues.
 
D

Deleted member 752119

Unconfirmed Member
It's a powerful portable console.

I'm going from playing 3DS Nintendo games to Switch Nintendo games on the go.

Yep. It's a super nice portable for people into that, who are ok with the battery life and size for sure.

A bit disappointing for those of us who are solely/mostly console gamers though. I wish the dock had ended up boosting processing more (supplemental computing device patent etc.)--but that would have driven cost up way too high given it's already at $300.

Anyway, most of Nintendo's games already looked great on Wii U so I can deal with them on something a fair bit more powerful The huge scope ones like Zelda, Xenoblade 2 etc. will suffer a tad compared to if this was a console at X1 to PS4 levels rather than a portable. But things like Mario Kart and Splatoon probably wouldn't look all that much different with their smaller scope and simpler art designs. So it's really just the $300 price for someone not really interested in the portability that has me waffling a bit about my purchase.
 

Zedark

Member
Because it's not $700? People who bring of phones will always confuse me. "Why isn't this $300 handheld device as powerful as my $800 phone?"

Not to mention architecture is different.

I think you are completely misinterpreting his point. The phone is there as a size comparison for laymen, and he asks why some (fictitious) people want the Switch, a system of this size, to be as powerful as a PS4 Pro. He isn't comparing it to a phone in any way.
 
Too much talk about specs and not enough talk for the UI

eYMYAbh.png

qaGvzb4.png

qT2DLwd.png

umOb5yP.png

CUTyTEB.png

FVzKrXK.png

The home menu they've shown is a little different than the leak so all the other UI shots are probably a little different from the final build too, but I think it's looking pretty promising.
 
Interesting. I wonder if pop-in would negatively affect the experience of an open world Zelda.



Gotcha. Yeah I should have clrified to ask if at face value (specs) there would be any issues.

To be fair, it sounds like even in docked mode, that game still has fairly noticeable pop-in.
 

Hermii

Member
Question: will performance (framepacing, frames per second, etc)see a hit while portable?
Breath of the wild actually runs more stable portable according to some reports. This makes kind of sense, because ram and CPU stays the same while only gpu gets boosted so maybe it's more bottlenecked when you inchrease the resolution.
 

OCD Guy

Member
You are ignoring the fact that, if your assumption that resolution is the only thing that changes between modes is correct, there should not be a difference in resolution jump between for example Zelda and Mario Kart. The fact that it is there means one of two things: Zelda is less optimised for dock mode than Mario Kart (not very likely imo) or they do in fact introduce some different graphical improvements in dock mode instead a straight 720p-1080p resolution increase.

I think the difference in resolution is simply the output of the dock. I doubt games are developed with both modes in mind, and so would take advantage of higher resolution textures etc while undocked.

It's simply an output. Think of the Xbox One S for example, it technically outputs at 4k but there's no real difference between playing Horizon 3 on an Xbox One S and an OG Xbox One. Playground games haven't had to change anything with regards to development for example.

Yes it's common knowledge to many on here that the Xbox One S can provide slight differences in terms of framerate due to the slightly different clocks of the hardware, and I'm not ruling out slightly different performance in terms of framerate between a game in portable mode and docked.

What I am ruling out however is extra development that focuses on specific optimizations while docked.

Question: will performance (framepacing, frames per second, etc)see a hit while portable?

Who know's, if anything who's to say portable games might have slightly better performance while portable. It's pushing a lower resolution afterall.

Breath of the Wild will be a good game to test, and Digital Foundry will likely do comparisons between docked and portable.

Infact I'd bet my life on it that they take an area in the game that is subject to frame rate drops and then test while docked and portable to determine whether the performance is the same.
 
Too much talk about specs and not enough talk for the UI



The home menu they've shown is a little different than the leak so all the other UI shots are probably a little different from the final build too, but I think it's looking pretty promising.

UI mostly matches up with the final build, they seem to show off more customisation than we've seen. Keyboard and icon layout is the more or less the same.

 
I think the difference in resolution is simply the output of the dock. I doubt games are developed with both modes in mind, and so would take advantage of higher resolution textures etc while undocked.

It's simply an output. Think of the Xbox One S for example, it technically outputs at 4k but there's no real difference between playing Horizon 3 on an Xbox One S and an OG Xbox One. Playground games haven't had to change anything with regards to development for example.



Who know's if anything who's to say portable games might have slightly better performance while portable It's pushing a lower resolution afterall.

Breath of the Wild will be a good game to test, and Digital Foundry will likely do comparisons between docked and portable.

Infact I'd bet my life on it that they take an area in the game that is subject to frame rate drops and then test while docked and portable to determine whether the performance is the same.

Unless I'm misunderstanding you, that's not true. Most games are natively rendering in different resolutions, it's not just the output.
 

Haint

Member
Because it's not $700? People who bring of phones will always confuse me. "Why isn't this $300 handheld device as powerful as my $800 phone?"

Not to mention architecture is different.


That's not the point he was making, but ironically, the production cost for the Switch is probably very close to the iPhone 7, possibly even higher. You don't seem to understand how much mark up is in a $700+ phone (hint the highest end ones cost ~$200 to produce), where as consoles sell much closer to cost and make their money on royalties and digital store cuts/%'s.
 

Marmelade

Member
Nvidia's drivers are 30-40% more efficient, which is irrelevant when it comes to consoles.

Not sure I agree on the 30-40% figure but it's very true that a lot of people seem to base the better efficiency of NVidia cards compared to AMD's on their performance in a PC environment when the situation is a lot different when talking about consoles.

Not saying Nvidia current cards aren't more efficient, they are, but consoles aren't bottlenecked by AMD's drivers lackluster DX11 perf for example.
 

OCD Guy

Member
Unless I'm misunderstanding you, that's not true. Most games are natively rendering in different resolutions, it's not just the output.

That will likely be game dependent I imagine.

Unless there's been an official source to confirm either way we'll likely have to wait for guys like at Digital Foundry to confirm the native framebuffer and whether it is infact different depending on the mode, or whether the resolution is simply handled by the hardware scaler.

How I imagine things to be is with Zelda for example, it probably has a native framebuffer of 900p which is then simply scaled down to 720p while portable, or scaled to 1080p while docked.

I'm not buying the idea that they'll develop a game like Zelda with different framebuffers and graphical settings depending on whether you're docked or not.
 

Zedark

Member
So, for the layperson that knows almost nothing except the current console power rankings, how powerful is the switch?

In base GFLOPS terms, the Switch is 157 GFLOPS undocked and 393 GFLOPS docked, when we assume Eurogamer's clocks are correct. There are other things, that are hard to quantify exactly, that improve the Switch relative to the PS4 and Xbox One, namely architecture (the Switch architecture is likely a bit more efficient for every FLOP it has) and a technique called half-precision computation, which has the potential to add quite a significant percentage. According to one very technically proficient GAFfer (Thraktor), this half-precision should be easily implementable, and one dev on GAF said that a gain of 54% could be had from this technique (though I am very sceptical that this will be achievable in every game, it just sounds too big not to have been used much earlier in console tech).

Anyway, you can assume the Switch is 1-1.5 times as powerful as the WiiU in undocked mode, and about 1/3 to 50% of Xbox One when disregarding the aforementioned half-precision calculations.
 
D

Deleted member 752119

Unconfirmed Member
Probably not worth a thread but Koizumi apparently confirmed Bluetooth headset support.(and also that you can use the 3.5mm Jack when docked).

Huge for me as wanting a way to use a head set while docked without running a headphone extension cable across the room was one of my main concerns.

My PS Gold headset just broke, so maybe I can find a head set that has a USB dongle for PS4 and bluetooth that I can use with Switch?
 
Not sure I agree on the 30-40% figure but it's very true that a lot of people seem to base the better efficiency of NVidia cards compared to AMD's on their performance in a PC environment when the situation is a lot different when talking about consoles.

Not saying Nvidia current cards aren't more efficient, they are, but consoles aren't bottlenecked by AMD's drivers lackluster DX11 perf for example.

You still get far different performance out of FLOPS from different architectures, as they all calculate FLOPS differently. Wii U was, in real world performance, more powerful than the PS3 and 360, especially with its GPU, but the FLOPS didn't show that.

Switch should have a similar advantage over XB1/PS4 as Maxwell is a more modern architecture than GCN 1.1-1.2.
 

EloquentM

aka Mannny
I'm not familiar with the source and not willing to 100% trust a reddit translation so I'm reluctant to start a thread.

you can use his translation and corroborate it with these docs. if this is the same interview the other thread is based on I believe nintendoeverything translated it.
 

Matt

Member
That will likely be game dependent I imagine.

Unless there's been an official source to confirm either way we'll likely have to wait for guys like at Digital Foundry to confirm the native framebuffer and whether it is infact different depending on the mode, or whether the resolution is simply handled by the hardware scaler.

How I imagine things to be is with Zelda, it probably has a native framebuffer of 900p which is then simply scaled down to 720p while portable, or scaled to 1080p while docked.
That wouldn't make any sense. The system has more resources when docked, and you would be wasting performance and battery on the go.
 

tkscz

Member
I think you are completely misinterpreting his point. The phone is there as a size comparison for laymen, and he asks why some (fictitious) people want the Switch, a system of this size, to be as powerful as a PS4 Pro. He isn't comparing it to a phone in any way.

Oops, my bad.

That's not the point he was making, but ironically, the production cost for the Switch is probably very close to the iPhone 7, possibly even higher. You don't seem to understand how much mark up is in a $700+ phone (hint the highest end ones cost ~$200 to produce), where as consoles sell much closer to cost and make their money on royalties and digital store cuts/%'s.

I understood the mark up but didn't think it was THAT bad.
 
D

Deleted member 10571

Unconfirmed Member
That UI is super sexy :) Didn't really expect all this style from Nintendo in both hard- and software.
 
Breath of the wild actually runs more stable portable according to some reports. This makes kind of sense, because ram and CPU stays the same while only gpu gets boosted so maybe it's more bottlenecked when you inchrease the resolution.

it's weird knowing my experience will play better while im pooping.
 

jett

D-Member
Your martyr complex is honestly kind of sad at this point. You've been playing the Switch victim for months now.

Nah, people deserved to be called out on this forum when it comes to the "NX." Too much BS was going around for a whole year. With the "fan-fiction" thread you could say it's still going on.
 

ASIS

Member
Based on the specs provided in the OP, I'd say 3-4.

In base GFLOPS terms, the Switch is 157 GFLOPS undocked and 393 GFLOPS docked, when we assume Eurogamer's clocks are correct. There are other things, that are hard to quantify exactly, that improve the Switch relative to the PS4 and Xbox One, namely architecture (the Switch architecture is likely a bit more efficient for every FLOP it has) and a technique called half-precision computation, which has the potential to add quite a significant percentage. According to one very technically proficient GAFfer (Thraktor), this half-precision should be easily implementable, and one dev on GAF said that a gain of 54% could be had from this technique (though I am very sceptical that this will be achievable in every game, it just sounds too big not to have been used much earlier in console tech).

Anyway, you can assume the Switch is 1-1.5 times as powerful as the WiiU in undocked mode, and about 1/3 to 50% of Xbox One when disregarding the aforementioned half-precision calculations.

Eh... Not great, but not bad when all things considered. It's safe to say that the jump from Wii U to Switch is much greater than GC to Wii right?
 
Top Bottom