alucard0712_rus
Banned
If it's a 5w without a screen and considering 8 inch 1080p screen power consumption is about 3-5w (???) it's all fit's in typical power consumption for Switch.
-
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.
Switch 2 battery to last 4 hours and why AMD lost out
- Thread starter Dorfdad
- Start date
We only have the total power consumption AFAIK.
Playing With Power: A Look At Nintendo Switch Power Consumption
But 7.1W at minimum brightness suggests the Switch's APU was also consuming around 5W. I don't think it makes any sense for 5W to be the total system power consumption. Then you would be getting like 1W-2W for the APU and you're dealing with mobile level performance.
If the clock speed in handheld mode was 500 MHz say, then that would give 1.5 TF, which accounting for architectural differences would put it roughly at Xbox One level performance. So in handheld mode you would be going from Xbox 360 level performance to its successor, moving from Switch 1 to Switch 2. That would still be a generational improvement.
Minsc
Gold Member
Does this even matter when Mario Kart 10 is on the system? I know which racing game I'll be playing.
Edit: Heh, I guess we need Mario Kart 9 first, but still, the sentiment holds true! Damn Nintendo and their 10 year release gaps.
Last edited:
alucard0712_rus
Banned
It's rumored 660mhz in handheld mode.We only have the total power consumption AFAIK.
Playing With Power: A Look At Nintendo Switch Power Consumption
www.anandtech.com
But 7.1W at minimum brightness suggests the Switch's APU was also consuming around 5W. I don't think it makes any sense for 5W to be the total system power consumption. Then you would be getting like 1W-2W for the APU and you're dealing with mobile level performance.
If the clock speed in handheld mode was 500 MHz say, then that would give 1.5 TF, which accounting for architectural differences would put it roughly at Xbox One level performance. So in handheld mode you would be going from Xbox 360 level performance to its successor, moving from Switch 1 to Switch 2. That would still be a generational improvement.
Emedan
Member
Yeah honestly it doesn't make sense, I call BS.
ResilientBanana
Member
So nVidia is more power efficient than AMD. If that's true, the power sucking geforce GPU's even when scaled back are most likely still a hell of a lot more powerful than AMD's.
Imtjnotu
Member
This....
And the new 8g4 only uses 8w and supports lpddr5t
CrustyBritches
Gold Member
The 5W is APU power consumption. Total system power consumption will be higher. Steam Deck is 15W APU power and when you run an intense game with high screen brightness and high volume you can hit as high as 28W total system power consumption. Switch 2 will probably have similar power limits to the Switch, so 5W APU and around 10-11W total system power. With a 20Wh battery you'd be looking at around *edit* 2-3hrs of battery life. Deck OLED has a 15W APU(user adjustable) with a 50Wh battery. If these rumors are true, I'd guess Steam Deck *edit* OLED will be as powerful with longer battery life.
Last edited:
Plus $50 more. So it's better to wait for the OLED version.
Aim for ps3 avoid disappointment.5watt?
Does not sound great for people expecting Ps4 Pro/Series S performance..
onnextflix5
Member
Sounds like some bogus rumour, its almost certain the watts will go up in handheld mode, 8-9 watts my guess.
Melon Husk
Member
5 Watts is perfect for a handheld
Launch Switch was over 3. Switch OLED is up to 5. Less than four would be extremely disappointing.
Last edited:
rodrigolfp
Haptic Gamepads 4 Life
Isn't 45W?
kevboard
Member
5watt?
Does not sound great for people expecting Ps4 Pro/Series S performance..
That could be the lowest mode.
the current switch has 3 different handheld power modes.
developers can choose which mode they want to use, and can save battery life on lower modes if their games don't need all the GPU power available
LordOcidax
Member
Switch 2 going to have less power than Switch 1 according to some rumors.
Robb
Gold Member
That’s pretty interesting. Had no idea. Although it makes sense when you mention it.
Obviously it should be a step up, I think the leaks mentioned two fans? With the switch 1 having a single? Or maybe two fans for a reasonably decent docked experience where it can forget about battery life constraints?
Nintendo Switch teardown reveals big battery and lots of cooling, just keep it away from your aquarium
Inside the new Nintendo Switch games console is a huge battery and the sort of cooling you'd expect to find in a laptop. Just keep it away from your aquarium!
www.zdnet.com
Either way I don't think anyone really gets a Nintendo device for things other than Nintendo games, they simply don't sell based on tech specs. I would say it's perfect timing for switch 2 launch with Xbox management shitting the bed and then there's PS5 for 1st and third parties, and switch 2 for 1st party. Not a lot of overlap there imo, unlike PS5 versus series for multis previously.
LordOcidax
Member
Switch 2 is going to be on par with PS4 at minimum in terms of power + DLSS + Basic Ray tracing.Obviously it should be a step up, I think the leaks mentioned two fans? With the switch 1 having a single? Or maybe two fans for a reasonably decent docked experience where it can forget about battery life constraints?
Nintendo Switch teardown reveals big battery and lots of cooling, just keep it away from your aquarium
Inside the new Nintendo Switch games console is a huge battery and the sort of cooling you'd expect to find in a laptop. Just keep it away from your aquarium!
Either way I don't think anyone really gets a Nintendo device for things other than Nintendo games, they simply don't sell based on tech specs. I would say it's perfect timing for switch 2 launch with Xbox management shitting the bed and then there's PS5 for 1st and third parties, and switch 2 for 1st party. Not a lot of overlap there imo, unlike PS5 versus series for multis previously.
Last edited:
kevboard
Member
yup. It's 307.2MHz, 384MHz, and 460MHz they can choose from
also during loading screens developers can boost the CPU to 1785MHz from the default 1020MHz to improve load times
RoboFu
One of the green rats
Astral Dog
Member
How trustworthy this is?
to me it doesn't mean anything until we see the actual thing seems like random numbers but who knows
If true at least the battery should be less awful than the first Switch but still best to wait for the revision
to me it doesn't mean anything until we see the actual thing seems like random numbers but who knowsIf true at least the battery should be less awful than the first Switch but still best to wait for the revision
NotMyProblemAnymoreCunt
Biggest Trails Stan
Onironauta
Member
ARM > AMD64 for mobile chipsets
digital_ghost
Member
The Last Ninja
Member
every gaming situation I have is covered by OLED now, no way I'm goin back to IPS. Life is too short to let that shit back in my eyeballs!
Klosshufvud
Member
5W is hugely underwhelming for a modern high-end mobile SoC. The one in Rog Ally really only starts to shine at 15-20W TDP. Modern hardware is simply more power consuming on all fronts, from handheld to dedicated desktop. It's just what happens when you add more and more cores and Compute Units on the SoC. The power gap between Switch 2 and PS5 will be even larger than Switch->PS4 if this is true. It's hugely disappointing news for anyone hoping for more than the absolute bare minimum.
Stamps1646
Member
I can never go back to LCD, I'm willing to wait for the OLED version.
That just doesn't seem possible on a 5-10W load?
existensmaximum
Member
And yet it will sell at least 50-100 times the amount of any handheld PC.
SweetTooth
Gold Member
The cake isn't getting much bigger, yet the space is getting crowded.
No guarantees in this industry
Panajev2001a
GAF's Pleasant Genius
Yeah, sure… now we are throwing numbers out of thin air. It is still nothing to shout about when your own predecessor lasts for longer…
Last edited:
loktaiextatus
Member
This is nintendo, Used Maxwell Cores when Pascal cores were already WELL established and charged 50 more bucks for switch than anyone expected..... they went with 12gb of memory in 2024- so do NOT expect this thing to match a steam deck .
Last edited:
digital_ghost
Member
Tested: The actual Steam Deck battery life
We played a variety of games at different settings on the Deck to see how long its 40Wh battery really lasts.
They got 90 minutes out of testing. So i was extremely wrong by a huge margin of 30 mins.
Last edited:
CrustyBritches
Gold Member
Power consumption or battery?
Deck LCD is 15W tdp, 40Wh battery, and the highest I’ve seen total system power consumption is 28W.
Dorfdad
Gold Member
I didn't want to start a new post, but here is another new "ADD ON"
#1 NEW FCC report on a new device that connects to switch 2 via USB C port (Not previously know)
Speculation is a Camera but nothing is certain. Is this the "GIMICK" probably not since it's an add-on.
#1 NEW FCC report on a new device that connects to switch 2 via USB C port (Not previously know)
Speculation is a Camera but nothing is certain. Is this the "GIMICK" probably not since it's an add-on.
- listed as a wireless device, not "game console" or "controller"
- No battery, can only be plugged in via USB-C
- Has 2.4GHz Wifi and 24GHz mmWave sensor
Last edited:
Silver Wattle
Gold Member
Just more shit mlid pulls from his arse, the steam deck OLED apu is extremely power efficient.
This was never about efficiency it was about compatibility with the current switch.
This was never about efficiency it was about compatibility with the current switch.
Haint
Member
There's no contest on the CPU's, the Z1's are WILDLY overpowered for portable gaming. They're faster than Ryzen 7 5000 Desktops, and only a bit behind Ryzen 7 7000 Desktops. That's actually a much more plausible reason for why Nintendo may have nixxed them, there's way too much die space and power devoted to unnecessary/unused CPU compute, and far too little to GPU.
This rumor, if true, is effectively confirming Samsung 8nm process. T239 is 1536 ampere cores with Nvidia pegging peak efficiency around 540Mhz based on the 8nm T234 it's derived from. That's ~1.6 "fake" double counted dual issue Teraflops at it's sweetspot. 5W portable mode definitely ain't running at 540Mhz, they'll probably underclock to 300 Mhz or lower like Switch 1, while docked sees a similar 700-800Mhz clock. That's 900 Gigaflops portable and 2.4TF's docked. Again, those are both "fake" dual issue/double counted teraflops, so real world raw docked compute will be around 70% lower, between base Xbone One and Base PS4. The Z1 Extreme is 8.6 "fake" teraflops, while base Z1 is closer at 2.8TF, but still beyond the likely clocks of an 8nm T239. Nintendo saw ballooning silicon costs and literally straight up rebadged their mid-gen Switch Pro design to a Switch 2.
Last edited:
Fake
Member
Why so many pow like this? lol
Because the initial target was base PS4, but as the energy consumption is giving up the close specs.
And being less powerful than base PS4 is not the same as saying will be weak.
If come close to Xbox S and still get Nvidia DLSS will be probably good on paper. Weaker than base PS4 specs, but far better than base PS4 threating the image quality.
We are just discussion spec wise, using Nvidia tech have nothing to do with brute hardware.
Panajev2001a
GAF's Pleasant Genius
“In our own testing, with Wi-Fi enabled and at 50% brightness, we have found the Steam Deck battery life ranges from just 83 minutes at worst all the way up to seven hours and nine minutes.” Cherry picked the 60 FPS lock of these PS4/Xbox One titles assuming zero user setting? What’s next compare Switch 2 to a PC with only Windows default graphics drivers installedTested: The actual Steam Deck battery life
We played a variety of games at different settings on the Deck to see how long its 40Wh battery really lasts.www.pcgamer.com
They got 90 minutes out of testing. So i was extremely wrong by a huge margin of 30 mins.
…… and that does not include the OLED model which improves battery life quite a bit.
digital_ghost
Member
“In our own testing, with Wi-Fi enabled and at 50% brightness, we have found the Steam Deck battery life ranges from just 83 minutes at worst all the way up to seven hours and nine minutes.” Cherry picked the 60 FPS lock of these PS4/Xbox One titles assuming zero user setting? What’s next compare Switch 2 to a PC with only Windows default graphics drivers installed
… and that does not include the OLED model which improves battery life quite a bit.
Panajev2001a
GAF's Pleasant Genius
How Long Does the Battery Charge Last?
Important: As with most lithium-ion batteries, the console battery life will gradually decrease over time. After about 800 charge cycles, the console battery life will decrease to about 80% compared to the battery on a new console.
www.nintendo.com
So the 2017 console with 4-6.5 hours (or 4-9 hours after the first SoC revision) of battery life with a successor likely launching 8 years later for a likely higher MSRP touting up to 4 hours battery life as something to shout about? Aaaaalright…
kevboard
Member
a 5W target doesn't mean that it actually runs at that 5W limit. the Switch 1 has 3 distinct portable modes each designed for a different GPU load target which developers can choose.
a 5W requirement can mean that that is a mode they are targeting for low power indy games like 2D platformers. just like the Switch 1 does.
maybe such a mode would even disable the Tensor cores and RT cores entirely for example.
at such low wattages the AMD APUs would probably collapse due to their X86 CPUs alone needing more than that to function properly
so the Switch 2 could absolutely have 5W, 10W and 15W portable modes, and a 5W mode was simply a target they wanted to reach for low end titles. after all, 1 to 4 hours battery life sounds better than 1 to 2 hours in PR speak, even tho the AAA games people will want to play on it will only ever get the latter one
docked mode for the Switch 2 could see a dramatically higher clock speed compared to portable, because this time around the resolution disparity between their target TV output and their handheld will be significantly higher.
before it was 720p and 1080p, now it's 1080p and 2160p at the very least.
I'm not sure if the screen resolution is actually known yet, but what if the screen still targets 720p for example? woked out fine for the Steam Deck for example. in that case the disparity would be massive.
720p to 1080p is about a 2.1x jump in the target resolution.
1080p to 2160p is a 4x jump exactly.
part of that jump can be covered by DLSS, but not all of it, since DLSS takes more render time the higher it has to scale.
so yeah. no matter the handheld clocks, I expect the docked clocks to be significantly higher this time, maybe even simply as a reaction to people using hacked Switch systems to demonstrate how many games would have benefitted with more stable performance when using the full potential of the chip.
edit: also, about the dual issue FP32.
when looking up benchmarks of a 5~8 TFLOPS Ampere GPU and a 5~8 TFLOPS RDNA2 GPU (getting comparisons for lower spec ones is hard), I see that they seem to perform pretty similarly (relatively speaking for 2 complete different architectures) even tho RNDA2 does not use dual issue FP32.
so not sure if your math maths out here.
Last edited: