Idle Consumption

[Steven85]

Hello, I'm not sure if this is the right section but I'd post anyway.

So I'm building a PC from 2 year old parts and I have two priorities. a) To be able to play even the latest games, b) To have relatively low Idle Wattage.
I'm using an i3-530 as the CPU, Two 5850s CFX as my GPU(s) and a GT240 as my PPU. I chose those parts because apart from being cheap (and powerful for their money) they also -according to reviews- produce low wattage while idle.

So I figured about 40w (at most) for the GPU, 15w for the PPU and 35w for the rest (CPU + mobo). I would think I would be getting aroung 90w while idle, which is quite respectable for such a packed system. Instead my Kill-a-watt reads that it's getting 150w-170w off the wall (while IDLE), which is quite a lot considering how expensive power is around here, where I live.

I'm thinking the culprit may be my PSU, it's a 2 year old Thermaltake 650w, which is of 75% efficiency. But even if we're to say that reviewers use PSUs with 85% efficiency I should had been -still- getting consumption around the 100w neighborhood, certainly not 150w. So what gives?

Can any of you guys give me a lead where the power leak is?


Thanks.

 

[WiSK]

There's no "leak", it's not water. Think of it more like, that those older GPUs forget to switch the gas and lights off when they leave the house

Reviews are one thing, but if you really want to know what each component draws, seems to me you're in the best position to discover it yourself. Just remove each GPU one by one and note down the difference at the wall. Then try to run your system with just CPU and boot drive and see what it shows.

 

[colinstu]

Don't write off the 10% difference in PSU efficiency as "not important"... it is.

Buy a 80 Plus Platinum PSU (AT LEAST an 80 Plus PSU. Even bronze ones are easy to find),

and buy newer components. More performance for either the same amount of power usage, or even use less power than before. Saving a few bucks with used older gear = your plan of saving money on power goes out the window.

Don't forget, each hard disk uses 5-10w on idle and random disk accessing. Fans use 5-15w each depending on the amps they're pulling, how much voltage is going into them. Ram uses 5-10w per stick (if I recall correctly).

So if you have 3 fans, two sticks of ram, and a hard drive... you're already looking at about 50w of random power being used.

 

[Steven85]

There's no "leak", it's not water. Think of it more like, that those older GPUs forget to switch the gas and lights off when they leave the house

Reviews are one thing, but if you really want to know what each component draws, seems to me you're in the best position to discover it yourself. Just remove each GPU one by one and note down the difference at the wall. Then try to run your system with just CPU and boot drive and see what it shows.

Yeah but those cards (supposededly) use as little power at idle as even the newest cards: http://www.techpowerup.com/reviews/HIS/HD_5850/28.html.

The 5xxx series (Evergreen) was the one that started the trend of low idle consumption, so I'm most certain that it's not the GPUs. Maybe it's indeed -though- my PSU and also I just counted my 120mm fans and they were 4 in total.

Of course I never let them to spin with more than 1000rpm but still they must be using their fair share of AC. Adding them up I get myself closer to the 110-120w mark (the fans + the PSU's lower efficiency), still it's off than the 150-170w mark that I'm seeing, but not as bad.

Probably the best thing I can do right now w/t spending too much money is buying a PSU of higher efficiency. There's not much more I can think of (sort of replacing most of my components for which I simply have no money for)....

Thanks for the help though, any new input on how I can make my system more efficient while IDLE (which is what it stays in for most of the day), I would welcome it...

 

[Dangman]

Which Thermaltake 650w PSU do you have? Note the 75% efficiency doesnt mean it is 75% efficient all the time. Just only in certain situations.

Also your theoritical numbers dont seem right. When I get near a PC, I will double check them.

EDIT: even if we use your numbers, we still have to account for one more factor: Whatever you're using to measure the power usage isn't 100% accurate. Or even close to that. So even if you get a new PSU with higher efficiency, you may still get inaccurate results.

 

[Steven85]

Which Thermaltake 650w PSU do you have? Note the 75% efficiency doesnt mean it is 75% efficient all the time. Just only in certain situations.

Also your theoritical numbers dont seem right. When I get near a PC, I will double check them.

EDIT: even if we use your numbers, we still have to account for one more factor: Whatever you're using to measure the power usage isn't 100% accurate. Or even close to that. So even if you get a new PSU with higher efficiency, you may still get inaccurate results.

Hello my PSU is of the 2008 series, I think it's called "Purepower" and it is modular (yeah I know it lowers the efficiency).
As for my theoretical numbers I pulled them directly from reviews. HD 5850s are consistently (on all reviews) at the 20w mark while the core i3 (plus the mobo, plus the ram sticks) is/was close to the 35-40w mark.

Lastly I think my method of measuring the consumption off the wall is pretty much accurate (at most 1-2% off) as I have checked other devices and it was spot on to them as well...

Anyhow... I turned my fans off and (all) I got was 140w at idle, better than before to be sure but -still- a tad high (I would expect close to 100w idle consumption at most at those circumstances...).

 

[Dangman]

Hello my PSU is of the 2008 series, I think it's called "Purepower" and it is modular (yeah I know it lowers the efficiency).

Just get a new PSU then. Even without the efficiency problem (which honestly isn't a big deal), that PSU is more than likely crap as TT's Purepower line tends to be on the crap side of quality.

Lastly I think my method of measuring the consumption off the wall is pretty much accurate (at most 1-2% off) as I have checked other devices and it was spot on to them as well...

When it comes to computer hardware, power measurement isn't accurate. In fact, I have proof: Here's Paul Johnson's post, PSU reviewer of HardOCP.com, about the inaccuracy of the Kill-A-Watt:
http://hardforum.com/showpost.php?p=1032190998&postcount=7

Another Paul Johnson's post:
http://www.jonnyguru.com/forums/showpost.php?p=27745&postcount=26

In addition, three other PSU experts backs up Paul Johnson's statement:
Oklahoma Wolf of JonnyGuru.com:
http://hardforum.com/showpost.php?p=1034843536&postcount=21
http://hardforum.com/showpost.php?p=1034555880&postcount=17

JonnyGuru of JonnyGuru.com and the former senior PSU engineer over at BFG acknowledge the inaccuracies of a KAW (Post #7 in regards to Post #2):
http://www.jonnyguru.com/forums/showthread.php?t=5977
http://www.jonnyguru.com/forums/showpost.php?p=29907&postcount=2

Redbeard of Corsair also acknowledges the inaccuracies:
http://hardforum.com/showpost.php?p=1032811067&postcount=22

Related discussions:
http://hardforum.com/showthread.php?t=1447774
http://hardforum.com/showthread.php?t=1509210
http://www.jonnyguru.com/forums/showthread.php?t=3102
http://www.jonnyguru.com/forums/showthread.php?t=2695
http://www.jonnyguru.com/forums/showthread.php?t=2284

 

[Zoop99]

awesome post, learned alot...

but kinda wondering if your watts used would matter if your running the same system with a AX850 vs with a AX1200

 

[Tsumi]

awesome post, learned alot...

but kinda wondering if your watts used would matter if your running the same system with a AX850 vs with a AX1200

It won't be the same because the efficiency curves of each power supply are different. Generally speaking, efficiency drops off dramatically at less than 20% load, while 50% load is usually peak efficiency. Above 50% load, efficiency drops off, but usually by a small amount. Of course, the efficiency curve for each power supply is different, so generalizations are not very accurate.