Support NeoGAF

[Bullet Club]

The new Nintendo Switch review: the updated Tegra X1 tested in depth

First unearthed in Switch's firmware 5.0 way back in March 2018, Nintendo's console hybrid is evolving thanks to a new,…

www.eurogamer.net

 

[Spukc]

Oh nintendo release the switch pro you bastardos

 

[Thurible]

Oh nintendo release the switch pro you bastardos

Do we know that there will be a switch pro in the future, or is the current hardware revamp for extra battery power the rumored pro?

 

[mckmas8808]

Do we know that there will be a switch pro in the future, or is the current hardware revamp for extra battery power the rumored pro?

This is not the rumored pro.

 

[Thurible]

This is not the rumored pro.

Are you sure?

 

[Bullet Club]

Are you sure?

This isn't the pro version. It's more of a Xbox One S type of upgrade.

 

[LordOfChaos]

Anyone confirm 16nm?

 

[stranno]

Top-notch stress test there.

 

[iconmaster]

This is not the rumored pro.

This is the rumored pro.

It's the machine which the WSJ predicted. There are no reliable rumors for anything beyond this latest model. Plenty of wishful thinking, though.

 

[Ulysses 31]

Amazon.de shipped it this morning!

 

[mckmas8808]

This is the rumored pro.

It's the machine which the WSJ predicted. There are no reliable rumors for anything beyond this latest model. Plenty of wishful thinking, though.

This isn't what WSJ was talking about. Either they were wrong with their understanding of the machine that Nintendo was making or they straight lied to us.

 

[UltimaKilo]

This isn't what WSJ was talking about. Either they were wrong with their understanding of the machine that Nintendo was making or they straight lied to us.

WSJ provides excellent reporting and they have not revised their report, so I fully expect some Switch Pro rumors to begin swirling again early next year.

 

[ChuyMasta]

Would an overclocked new system see performance upgrades?

Hopefully the homebrew community tests it out soon.

An overclocked cpu and gpu on the original system had ok results.

 

[iconmaster]

This isn't what WSJ was talking about. Either they were wrong with their understanding of the machine that Nintendo was making or they straight lied to us.

WSJ provides excellent reporting and they have not revised their report, so I fully expect some Switch Pro rumors to begin swirling again early next year.

I'm not saying they were wrong, or lied. WSJ was on the money. Remember, the prediction was that the model could release as early as this summer; and they never said anything about graphics.

They nailed it, and we have our "pro." (Note that's the fans' term. WSJ never called it a "pro" or anything else.)

I'm interested to see how many months we'll have to go without a new Switch model before this really sinks in for everyone. Six? Hopefully not more than six.

 

[mckmas8808]

I'm not saying they were wrong, or lied. WSJ was on the money. Remember, the prediction was that the model could release as early as this summer; and they never said anything about graphics.

They nailed it, and we have our "pro." (Note that's the fans' term. WSJ never called it a "pro" or anything else.)

I'm interested to see how many months we'll have to go without a new Switch model before this really sinks in for everyone. Six? Hopefully not more than six.

Anything past next March and I'll say you're right.

 

[Fbh]

So basically substantially better battery life and a very slight performance improvement. Seems pretty good for a simple internal hardware update with no change to the price/name

Would an overclocked new system see performance upgrades?

Hopefully the homebrew community tests it out soon.

An overclocked cpu and gpu on the original system had ok results.

Can the new Switch even run homebrew? I thought that was limited to an exploit which was only present on early Switches

 

[DESTROYA]

While this model is a great revision in itself for the battery power alone I’m still holding out hope for a Pro model.

 

[Fake]

Nice results I must say.

 

[-Arcadia-]

Worth noting, since DF didn't spend much time at all on the screen, that this video measured the screen output using instruments and found a slight brightness improvement, and the display to be warmer for the better -- quite a bit more color accurate.

 

[Chittagong]

This isn't what WSJ was talking about. Either they were wrong with their understanding of the machine that Nintendo was making or they straight lied to us.

The problem is, there is no chip in existence that could be used in a “Pro” to provide substantially improved performance while maintaining compatibility, the only options are

Tegra 1 overclocked - same chip as in current Switch, but massively overclocked (or more like, normally clocked). This would require a further die shrink to sustain heat and curb battery drain, and still it would not be a massive jump.

Tegra 2 - as far as I know, this chip is not suitable to a portable device with a tight power consumption and heat budget.

So short of Nvdia announcing a new Tegra chip, Nintendo has spectacularly managed to choose yet another dead-end architecture.

 

[mckmas8808]

The problem is, there is no chip in existence that could be used in a “Pro” to provide substantially improved performance while maintaining compatibility, the only options are

Tegra 1 overclocked - same chip as in current Switch, but massively overclocked (or more like, normally clocked). This would require a further die shrink to sustain heat and curb battery drain, and still it would not be a massive jump.

Tegra 2 - as far as I know, this chip is not suitable to a portable device with a tight power consumption and heat budget.

So short of Nvdia announcing a new Tegra chip, Nintendo has spectacularly managed to choose yet another dead-end architecture.

WOW!!!! I'm honestly surprised that Nintendo have managed to lock themselves in this way. That's crazy. So what are Nintendo's options once PS5 and Xbox 3 are out? Can they use an ARM processor that's faster than the Tegra1?

 

[-Arcadia-]

WOW!!!! I'm honestly surprised that Nintendo have managed to lock themselves in this way. That's crazy. So what are Nintendo's options once PS5 and Xbox 3 are out? Can they use an ARM processor that's faster than the Tegra1?

It's less surprising when you realize that they probably got a good deal on the Tegras.

I always assumed the next Switch, if Nvidia doesn't want to do mobile hardware anymore, would just use mobile hardware from other companies. With the phone rat race, there's no end of them, and with such mass-buying, prices are probably driven down nicely. Whether or not that would allow for backwards compatibility, interests me.

 

[mckmas8808]

It's less surprising when you realize that they probably got a good deal on the Tegras.

I always assumed the next Switch, if Nvidia doesn't want to do mobile hardware anymore, would just use mobile hardware from other companies. With the phone rat race, there's no end of them, and with such mass-buying, prices are probably driven down nicely. Whether or not that would allow for backwards compatibility, interests me.

The bolded is HUGE though right? Especially considering more and more people are moving towards a download only form of playing their games.

 

[DESTROYA]

The problem is, there is no chip in existence that could be used in a “Pro” to provide substantially improved performance while maintaining compatibility, the only options are

Tegra 1 overclocked - same chip as in current Switch, but massively overclocked (or more like, normally clocked). This would require a further die shrink to sustain heat and curb battery drain, and still it would not be a massive jump.

Tegra 2 - as far as I know, this chip is not suitable to a portable device with a tight power consumption and heat budget.

So short of Nvdia announcing a new Tegra chip, Nintendo has spectacularly managed to choose yet another dead-end architecture.

Not necessarily, with Tegra 2 they could do a die shrink like did with the Tegra 1.
The new Mariko soc runs at almost have the power so if they do the same with the Tegra 2 it could theoretically be done.

 

[ThatGamingDude]

Can the new Switch even run homebrew? I thought that was limited to an exploit which was only present on early Switches

It depends; with iPatched units there is a method to run Homebrew, but only for specific firmware versions.

Mariko units are also iPatched.

It's likely (And I mean 90+% going to happen) that Ninty is going to send out Mariko units with firmware's passed that point, so that HB can't be run, and reduce piracy on the later units.

 

[Chittagong]

Not necessarily, with Tegra 2 they could do a die shrink like did with the Tegra 1.
The new Mariko soc runs at almost have the power so if they do the same with the Tegra 2 it could theoretically be done.

hmm... considering

- the current Tegra X2 is at TSMC 16 nm
- Switch OG Tegra X1 is at TSMC 20nm
- New Switch is Tegra X1 T214 (Mariko) is TSMC 16 nm

NVIDIA would need to take Tegra X2 to TSMC 10nm to see similar gains.

 

[Pimpbaa]

Worth noting, since DF didn't spend much time at all on the screen, that this video measured the screen output using instruments and found a slight brightness improvement, and the display to be warmer for the better -- quite a bit more color accurate.

The screen on my OG Switch is more warm than I like it already. I can't imagine it being worse than that. Everything looks far better on my TV with my much more cooler color settings (white looks white, not with an orange tint like on my Switch's screen).

 

[-Arcadia-]

The screen on my OG Switch is more warm than I like it already. I can't imagine it being worse than that. Everything looks far better on my TV with my much more cooler color settings (white looks white, not with an orange tint like on my Switch's screen).

Yeah, color accuracy doesn’t necessarily match preference. I know what you mean, my TV has a warm/cool switch that I’ve left on cool since I got it. Even if I’m a pro’s calibration nightmare, lol, I do like my colors to pop, and a general cooler tone.

I think striving for better color accuracy, if that reading is correct (one guy’s test and instrumentation isn’t the be-all end-all, after all), is an admirable goal, though.