By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Nintendo - Wii U consumes more than 63 watts confirmed by DOE(Department of Energy EUA)

Well, until now we have heard from many that Wii U is limited to 33 watts for games eventhough Iwata himeself said that Wii U would be consuming and avergae of 45 watts  during gameplay. Letsnot forget that the measurements  were done with ports and even Super Mario Bros U is juist a port from Wii but with some HD enhancements

But now I have come across and official document that proves that the Wii U can consue around 63 watts or more but not less since the external power supply is Class A Level 5, and DOE and Enegy STAR say that these kind of power supplies must have an efficiency ≥87%. Since the Wii U power Supply says its 73 watts, that means that at least 63.5 watts of those 73 watts are avialble for the console

here you can find it(pages 8 and 20)

http://www.energy.ca.gov/appliances/2013rulemaking/documents/responses/Consumer_Electronics_12-AAER-2A/California_IOUs_Response_to_the_Invitation_to_Participate_for_Game_Consoles.pdf

"

1.4 Existing Standards and Standards under Development

No mandatory standards currently exist for game console energy efficiency. The EPA, with support 

from the Department of Energy (DOE), has developed a voluntary standard through the EPA Game 

Console Recognition Program

Game consoles use either internal (Sony PlayStation) or external power supplies (Microsoft Xbox 

and Nintendo Wii and Wii U) to convert alternating current to direct current for use by electronic 

components. Nintendo’s power supplies are Class A external power supplies (single-voltage with a 

rated output of less than or equal to 250 watts), and are subject to federal efficiency standards 

established by the DOE. The federal Class A external power supply standards are listed in Section 

301 of the Energy Independence and Security Act of 2007, effective July 1, 2008

 

4.2 How efficient are they during Game Play mode?
The efficiency of power supplies during game play is presented in Table 13, based on the testing 
data shown in Table 12. The efficiencies of the Xbox 360 S and PlayStation 3 consoles during active 
game play are estimated to be approximately 83.0% and 78.5%, respectively. The Wii and Wii U 
external power supplies are class A and therefore currently pre-empted by federal regulations (see 
section 1.4 for additional discussion).

 

 

"

And here you can find the efficiency of a Class A Level 5 external Power Supply

http://www.cui.com/catalog/resource/1330

"

What is Energy Star?
Energy Star is a voluntary program developed by the US government.
 Aims to help businesses and individuals improve energy efficiency in everyday 
electrical products.
 A voluntary standard that partners with businesses to provide more energy 
efficient products. After meeting strict guidelines established by the 
Environmental Protection Agency (EPA) and the US Department of Energy, the 
product is then certified as an Energy Star product.

 

"

 

Even if we account the 4x usb ports and the SDHC, that only consumes around 10.73 watts since the usb 2.0 is limited to 2.5 watts per device and the SDHC only consmes around 0.73 watts. So, if you were to use them while playing some port games, the Wii U would only be consuming around 43.73 watts, which means 19.77 watts or more depending of the actual efficiency

USB 2.0 maxium 2.5 watts

http://en.wikipedia.org/wiki/PoweredUSB

 

SDCH only 0.72 watts

http://mx.transcend-info.com/industry/compare.asp?ck=30%7C31%7C32%7C33%7C34%7C35%7C70%7C82%7C84

 

 

Even if it has been reported that wii u ports only use one core for games(must likely another core is used for sound like xbox does and could bet that developers waist the DSP in the ports), each core shouldnt be consuing more than 1.6 watts, look for example his cousin the 476fp, only draws 1.6 watts at 1.6ghz per core and espresso is 1.24ghz, plus, gekko made at 180nm consumed only 4.9 watts at 486mhz, even an upgraded version at 45nm shouldnt be consuming so much energy. The DSP and ARM dont need much energy either, a good DSP can do 11 gigaflops per watt or more, and arm11 of one core only requires about 1 watt and a dual core could be drawing something near 2.5 watts as maxium. The main DDR3 RAM dont see it consuming more than 1.6 watts sinc i have seen 8GB model at 30nm that only require 1.65 watts and the wii u only has 2GB

 

DDR3 1600 8GB only 1.65 watts

http://www.legitreviews.com/samsung-8gb-30nm-ddr3-1600mhz-memory-kit-review_1959

 

So, in the end we could say that the GPU,which the must pronounced coponent, could have about 15 watts of more energy aviable besides whatever it is consuing right now on some ports that use 33 watts

 

dont you think those 15 watts are important?

well, for exaple, the e6760 which is presumed to be the base of wii u gpu has a performance of 16.5 gigaflops per watt, for 15 watts we would be speaking about 247.5 gigaflops that ports are wasting

in those 33 watts, i see at least between 15 to 18 watts being used by the gpu to match the xbox 360 power and make ports possible

 

Thanks for your attention



Around the Network

This isnt gonna end well.

Very good news for WiiU owners
Time to show whats under the hood of that machine



PS Vita and PC gamer

CPU Intel i5 2500K at 4.5 Ghz / Gigabyte Z68 Mobo / 8 Gb Corsair Vengeance 1600 mhz / Sapphire HD 7970 Dual X Boost / Corsair Obsidian 550d 

I don't know much about this stuff but does the 87% efficiency mean it has to always be drawing that percentage of the power supply's total wattage or it has to be capable of doing so. i.e. does that mean it is always using 63 watts or can it be less.

Also in your breakdown towards the end where you are trying to calculate the GPU power draw, you didn't factor in the EDRAM. Now again I know nothing about this stuff but having seen the die shots of the GPU i would have thought that a big hunk of silicon like that would have to be responsible for a decent amount of that 'missing' wattage. (as a side question, could the possibility that early games were not using all CPU cores or the EDRAM help to account for the initial lower measurements of power draw?)

Nonetheless, this is still interesting since most arguments revolving around analysis of the WiiU's GPU and likely power seem to inevitably fall back on, "but it only uses 33 watts, there's only so much you can do with 33 watts".

Someone, please educate me.



First, documented power usage is often 'worst case scenario'. Second, it's basic engineering to use DC power supplies that provide more juice than the application will require. Helps keep voltage steady and current clean. (Not to mention less wear and tear on the power supply.)

Source: My job.



hsrob said:

I don't know much about this stuff but does the 87% efficiency mean it has to always be drawing that percentage of the power supply's total wattage or it has to be capable of doing so. i.e. does that mean it is always using 63 watts or can it be less.

Also in your breakdown towards the end where you are trying to calculate the GPU power draw, you didn't factor in the EDRAM. Now again I know nothing about this stuff but having seen the die shots of the GPU i would have thought that a big hunk of silicon like that would have to be responsible for a decent amount of that 'missing' wattage. (as a side note, early games not using all CPU cores or the EDRAM may help to account for the initial lower measurements of power draw)

Nonetheless, this is still interesting since most arguments revolving around analysis of the WiiU's GPU and likely power seem to inevitably fall back on, "but it only uses 33 watts, there's only so much you can do with 33 watts".


it means more or equal to 63.5 watts, but cant be less than that or nintendo would be violating the standards of DOE and of course they could sue  them or something

edram doesnt change things much, it only consumes some miliwats, for example, look at the ibm edram at 45nm

http://www-03.ibm.com/press/us/en/pressrelease/21074.wss

 

"

eDRAM Specifications 

Among the specifications of IBM’s high-performance eDRAM technology:

cell size: 0.126 mm2

Power supply: 1 V

availability: 98.7%

Tile: 1K RowX16 Col X146 (2Mb)

AC power: 76 mW

standby keep alive Power: 42 mW

Random cycle time: 2ns

Latency: 1.5ns

"

 

the e6760 performance is just an example, we could at least expect this kind of performance from nintendo since they are not payiong AMD research development for nothing, but still, we could also consider 13.3 gigaflops per watt like the redwood pro

 

for thewayofthepath

true, but knowing nintendo i dont see them throwing watts for nothing, obviously they want to squeeze as much as possible. at must maybe they could waste 5 watts, but dont forget that the standard says more or equal to 87%, not that is 87%, which means that efficiency could even reach 90%, 95% is possible but expensive, then again would explain why the wii u is expensive too



Around the Network

So what you are saying is that the Wii U has a substancial spare ammounts of Watts that could be used to increase its performance?



JohannStrauss said:
So what you are saying is that the Wii U has a substancial spare ammounts of Watts that could be used to increase its performance?


More like WiiU have a good chunk of watts that havent being used to get the system max performance



PS Vita and PC gamer

CPU Intel i5 2500K at 4.5 Ghz / Gigabyte Z68 Mobo / 8 Gb Corsair Vengeance 1600 mhz / Sapphire HD 7970 Dual X Boost / Corsair Obsidian 550d 

They will eventually use these reminiscent watts, my wild guess for project "X" and Zelda U being the ones to squeeze the system's maximum energy output.

I could be wrong of course.



JohannStrauss said:
So what you are saying is that the Wii U has a substancial spare ammounts of Watts that could be used to increase its performance?


Thats the idea, besides the measurement of those 33 watts were done with lazy ports from 360 and must likely with alpha kits

if you remeber well, the alpha kits were reported by some developers to offer about 50% more power over the 360ps3, later when the final devs arrived about mid of 2012 they were reported to pack even more power than the akpha kits, so i am no surprised the wii u can have about 2x more raw power than 360

 

besides, the gpu die size even without edram+arm+dsp is enough to fitn a redwood xt which packs 400 spusk, 20tmus and 8 rops in 104mm2(measurementy done by anantech) at 40nm

http://www.anandtech.com/show/2917

 

 

wii u gpu die size was reported to be 156mm2 by anatech and the chipworks photo says its 146mm2

so, if we were to take a photo to the redwood xt i wouldnt be surprised if the actual die size was 94mm2 or little less and the wii u gpu without edram, sram, dsp and arm is about 97mm2, so surely one can fit 94mm2 in 97mm2

 

you can get the flops by doing this

spus*clock speed*2 flops/1000

 

of course the wii u gpu not necessarily has 400spus, 20 tmus and 8 rops since nintendo could have rearranged it and have even more spus by sacrificing the number of tmus()you can fit many spus in 1 tmu)

 



@megafenix

The eDRAM on the GPU die is not from IBM, it's from Renesas. The 3MB L2 Cache eDRAM of Espresso is from IBM, and while the difference in watts isn't huge, Renesas eDRAM =/= IBM eDRAM, you'll need to seek out the GPU7 eDRAM info from Renesas