• 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.

Anandtech: Switch Power Consumption Analysis

NeOak

Member
Figured it was threadworthy due to the measurement and breakdown of the docked and undocked modes, as well as with a power bank:

Docked
okNnzBD.png

With the Switch charged and running Zelda in its docked configuration, it's drawing on average 11 Watts of power. The dock itself is consuming a bit of this energy to power its DisplayPort to HDMI converter, but it's safe to assume that virtually all of that power is going to the Switch itself. And while I didn't pull noise measurements on the Switch, while the console's fan was active, it was holding at a fairly low speed, judging from the softness of the sound.

Letting the Switch discharge and loading up Zelda again finds that power consumption has (unsurprisingly) increased, to 15.7W. Throwing on the partially discharged joy-cons bumps that up a bit further to 16.5W, coming fairly close to the official 18W limit of the dock. One thing to keep in mind here is that if we subtract out the 11W from earlier, we only end up with 4.7W left to charge the Switch's battery.

Finally, if we turn the console off and just let it charge, we find that the Switch + dock draws 9.8W. This is nearly twice the amount of leftover power the Switch had available to charge its battery with when it was docked and turned on. Meanwhile, adding the joy-cons to the mix to recharge as well brings the total power consumption up to 12.1W. The takeaway? The Switch can recharge fairly quickly, but only if it's not turned on. If it is on, it will still recharge in the dock, but at around half the rate.

Undocked
Z7K3x8v.png

Starting off again with a fully-charged Switch, with the display at minimum brightness we're down to 7.1W, or 3.9W less than when it was docked. Considering that some of this power is going to screen and that we can't shut it off, we're easily looking at a 5W+ reduction in SoC power going from docked mode to undocked mode. Meanwhile cranking up the brightness to maximum increases the power consumption to 8.9W, or about 25%. In practical terms this means that going brighter definitely has an impact on the Switch's battery life, but even if you drop to minimum brightness, you're still only going to cut power consumption by 20%. So don't feel bad playing the console with a higher brightness; lowering the brightness won't vastly increase the runtime of the console.

Otherwise, keep in mind the 8.9W number. This is (roughly) the maximum power draw for gaming on the console when it's undocked. It should also be noted that the Switch will try to avoid charging the joy-cons unless it too is being charged, so the runtime impact of the joy-cons will typically be nil when the Switch is running on its internal battery.

After letting the Switch discharge, the power numbers for operating the Switch while it's turned on and charging are not all that different from earlier when the console was docked. With the brightness at maximum – to give us the Switch's maximum power draw undocked - the Switch draws 16.1W in this scenario. Throwing on the joy-cons adds another 1.6W, bringing the total to 17.7W. This is the single highest power draw number that I recorded, and it's interesting to note that it's still a hair under the 18W limit stamped on the Dock, indicating just how accurate that value is.

Finally, sleeping the Switch to let it charge is identical its power consumption while docked. The Switch will draw 9.8W to charge itself, and 12.1W with the joy-cons attached. Turning the Switch off entirely does change the charging rate a bit, but not significantly: it goes from 9.8W to 10.6W.

Battery Pack (USB)
rieVt9H.png

Plugging the Switch into a power bank finds that a good power bank can provide enough power to run the Switch, but that's it. Whether discharged or full, the Switch doesn't pull more than about 9.1W from a battery pack. This is just over the 8.9W maximum operational power consumption level we established earlier. And even after letting the Switch run for a couple of hours off of a power bank and starting from a full charge, it's still fully charged while the power bank is slowly discharging.

Notably, the Switch can't draw more than the aforementioned 9.1W from the Xtron, or indeed any other tablet-sized power bank I've thrown at the Switch. In fact every 5V-capable USB-C power source I've thrown at the Switch maxes out at this same point. At 5V, the Switch doesn't seem to be able to draw more than 2 Amps.

The takeaway from all of this is that while this is by no means an exhaustive test, what I've found is that any good power bank designed to power tablets will be sufficient to power the Switch. So long as a bank can deliver 5V @ 2A or better, then it can power Nintendo's console. (And if you're looking for buying advice, while I haven't yet had a chance to test it, RAVPower recently started shipping a rather sizable 99 Whr power bank that supports up to 20V)

The one downside is that due to the inner-workings of the USB Power Delivery specification (more on that in a sec), the Switch apparently can't pull enough energy from standard 5V-output power banks to meaningfully recharge its battery while gaming. So with a 5V power bank, if your Switch is fully depleted, you'll need to stay attached to the bank the entire time you're playing, or take a break and let the bank recharge the Switch while it's sleeping. In the case of the latter, the recharge rate should only be a bit lower than if you had used the AC adapter.

http://www.anandtech.com/show/11181/a-look-at-nintendo-switch-power-consumption

Takeaway:
If you use a USB Power Bank/Battery that only does 5V, don't expect the Switch to charge the battery, since it will only pull 2A max even if the battery can give more.

If you want a USB Battery that can charge the switch, it needs to have USB PD:
https://www.ravpower.com/26800mah-Type-C-external-battery-charger-black.html

Amazon link: https://www.amazon.com/gp/product/B01LRQDAEI/?tag=neogaf0e-20
 

Justified

Member
The one downside is that due to the inner-workings of the USB Power Delivery specification (more on that in a sec), the Switch apparently can't pull enough energy from standard 5V-output power banks to meaningfully recharge its battery while gaming. So with a 5V power bank, if your Switch is fully depleted, you'll need to stay attached to the bank the entire time you're playing, or take a break and let the bank recharge the Switch while it's sleeping. In the case of the latter, the recharge rate should only be a bit lower than if you had used the AC adapter.

Explains why the Charging Grip is just a USB extension
 

Noobcraft

Member
Someone correct me if I'm wrong, but didn't the PS4/Xbox One (launch models) use like 10x more electricity during gameplay than the Switch uses?
 
Does this apply to USB wall chargers too?

Cus I'm 100% sure that on launch night my Switch was on 87% when I plugged it into my iPad charger but after my session (1h+) it was at 100%.
 

gtj1092

Member
This power draw is more inline with what was expected I believe. The Eurogamer draw was similar but didn't have the caveat of explaining the high power draw was due to charging the battery.
 

AmyS

Member
Interesting comment on the article

Chipworks promised a die scan for this today, should be exciting. At what looks like 121mm2, it's too spookily close to TX1 for me to expect much difference, but it will be interesting to see what areas they touched.

TX1 couldn't use both core clusters at once for example, so without changing that the OS would have to reside on an A57 core, so I wonder if they either updated the CCI, or removed the A53s.
 

NeOak

Member
So an old 5V/2A battery with a USB-A to USB-C cable won't work?
It works for playing, but if the Switch is low on battery, it will slowly charge. Slowly.

It will allow you to keep playing though until it runs out.
 

Donnie

Member
Interesting comment on the article

Yeah Chipworks have said they were going to release their tear down and opinions on Switch today, along with a die shot. No doubt they'll also measure the die properly so people can stop guessing.
 
My Switch went up 10% in a hour using a 10W iPad charger during use. It would be much more in sleep probably. I wouldn't say a 5V/2A charger can't charge the Switch.
 

Atheerios

Member
The Ravpower listed in the OP seems, from what users have reported, the best power bank for the Switch, being the only USB PD power bank that I know and capable of 15V/2A.

I ordered this for myself so I will be able to test it this weekend.
 
The Ravpower listed in the OP seems, from what users have reported, the best power bank for the Switch, being the only USB PD power bank that I know and capable of 15V/2A.

I ordered this for myself so I will be able to test it this weekend.

It's so big though. :( it seems like every decent USB-C power bank is 20000mah or more. I wish there was a 10000mah available. I don't really want to lug a 7x3x1ish bank around. :/
 

random25

Member
Nice article from Anandtech. Seems to support my own experience with charging the Switch using its default AC adapter, a phone charger and a battery bank.
 

Dingens

Member
I wish somebody would finally measure runtime and power-draw with WiFi off.
the 3ds' runtime nearly doubbled that way, but for some reason all switch tests I came across so far leave wifi enabled... (which, I'd argue, is hardly a realistic scenario when you're on the go)
 

shiyrley

Banned
I tested my USB C 5V-3A powerbank with the Switch and it DOES charge the battery while playing BOTW with brightness maxed out. And yes, by "charges" I mean "the battery actually goes up while playing non-stop". Slowly tho, but I can't really tell if it's slower than the AC adapter.

This is my powerbank:
https://www.amazon.es/dp/B019X8EXJI/
How will this one work? I'm bad with battery backups/chargers.

https://www.amazon.com/gp/product/B019IFIJW8/?tag=neogaf0e-20

I just bought this today. On sale for $36. Found the deal on Slickdeals
Hey, good news for you, that's my powerbank. Use a C to C cable tho, and if you do use a A to C cable NEVER use the Quickcharge enabled port.
 
I tested my USB C 5V-3A powerbank with the Switch and it DOES charge the battery while playing BOTW with brightness maxed out. And yes, by "charges" I mean "the battery actually goes up while playing non-stop". Slowly tho, but I can't really tell if it's slower than the AC adapter.

This is my powerbank:
https://www.amazon.es/dp/B019X8EXJI/

Hey, good news for you, that's my powerbank. Use a C to C cable tho, and if you do use a A to C cable NEVER use the Quickcharge enabled port.
Hey thanks for the heads up.

Does it come with a C to C cable? I only have an A to C cable

Also, what will happen if I use an A to C cable with the quick charge port?
 

Glix

Member
I know it may sound like I'm playing semantics, but I don't like that a supposedly super detailed technical analysis talks about "power consumption in dock with the Switch off" when the unit literally cannot be powered down when its docked. I understand that they mean sleep mode, but all this shit is confusing enough, to be lazy with the differentiation makes me devalue the entire article.
 

btags

Member
Someone correct me if I'm wrong, but didn't the PS4/Xbox One (launch models) use like 10x more electricity during gameplay than the Switch uses?

Yes, but the xbox one and ps4 are also provide better performance than a switch. Increased performance usually demands increased power draw so it is not really that surprising.
 

enbred

Banned
I know it may sound like I'm playing semantics, but I don't like that a supposedly super detailed technical analysis talks about "power consumption in dock with the Switch off" when the unit literally cannot be powered down when its docked. I understand that they mean sleep mode, but all this shit is confusing enough, to be lazy with the differentiation makes me devalue the entire article.

Is that true? It can't be turned off when docked?
 
D

Deleted member 465307

Unconfirmed Member
I know it may sound like I'm playing semantics, but I don't like that a supposedly super detailed technical analysis talks about "power consumption in dock with the Switch off" when the unit literally cannot be powered down when its docked. I understand that they mean sleep mode, but all this shit is confusing enough, to be lazy with the differentiation makes me devalue the entire article.

Is that true? It can't be turned off when docked?

I'm pretty sure the device can be turned off; it requires you hit the power button on the console itself, whether in docked or undocked modes. It can't be done via the OS, though (which bothers me and I hope they update this soon).
 
Top Bottom