Go Back   Hardware Canucks > HARDWARE > Power Supplies

    
Reply
 
LinkBack Thread Tools Display Modes
  #11 (permalink)  
Old April 10, 2009, 09:45 AM
rjbarker's Avatar
Hall Of Fame
 
Join Date: Feb 2008
Location: Courtenay, B.C
Posts: 5,932

My System Specs

Default

Quote:
I wish people would stop recommending the corsair(it gets beat in every major category by the revolution 85+). No offense ZLOJACK.
Power is not the only thing to consider...."is the PSU compatible with the Mobo ?".....presently the Enermax 1050W PSU appears to be having "compatibility" issues with the Evga X58 and Evga X58 Classified Boards.
__________________
Introducing me n my OCD to Watercooling, is like taking an Alcoholic to an "all you can drink" Beach Bar in Mexico

.
Reply With Quote
  #12 (permalink)  
Old April 10, 2009, 09:59 AM
Top Prospect
F@H
 
Join Date: Aug 2008
Location: Québec, QC
Posts: 79
Default

Quote:
Originally Posted by MpG View Post
1000W would make me edgy. The 285's can pull just shy of 200W a piece, before overclocking
At the reference frequencies (648/1476), EVGA GTX285 has a peak power of 150W. Well, that's what they say.
__________________

Reply With Quote
  #13 (permalink)  
Old April 10, 2009, 10:03 AM
gingerbee's Avatar
Hall Of Fame
F@H
 
Join Date: Jan 2009
Location: Orillia, Ontario
Posts: 6,828

My System Specs

Default

power problems begone
__________________
Fav quote "One should strive to achieve; not sit in bitter regret." Ronin Harris
Reply With Quote
  #14 (permalink)  
Old April 10, 2009, 11:37 AM
MpG's Avatar
MpG MpG is offline
Hall Of Fame
 
Join Date: Aug 2007
Location: Kitchener, ON
Posts: 3,144
Default

Quote:
Originally Posted by foxman View Post
At the reference frequencies (648/1476), EVGA GTX285 has a peak power of 150W. Well, that's what they say.
Interesting. That's a lower estimate that I've seen in other places, although since they don't seem to outline their measuring methods, can't judge that for certain. But other reviews have generally been placing the power consumption higher.

i.e SKYMTL's power consumption measurements show a 200W difference between load and idle at the wall. So if you figure on about 80% PSU efficiency, you're looking at a power consumption of about 160W+ whatever the card uses at idle. The card is overclocked, granted, but that should only account for a little extra power.

Admittedly, I tend to err on the side of caution, though. In my books, a power supply should have the capacity to handle a rig that is artificially loaded to absolute max - say, with Linpack+Furmark - both which are capable of putting their targets past their official TDP ratings. If the PSU will handle that sort of situation, you know you've got comfortable headroom for everyday work/gaming, or even folding.
__________________
i7 2600K | ASUS Maximus IV GENE-Z | 580GTX | Corsair DDR3-2133
Reply With Quote
  #15 (permalink)  
Old April 10, 2009, 12:23 PM
Top Prospect
F@H
 
Join Date: Aug 2008
Location: Québec, QC
Posts: 79
Default

Their methodology is explained here. The link talks about AGP graphic cards, but they didn't change their methodology with the arrival of PCI Express.

Quote:
Admittedly, I tend to err on the side of caution, though. In my books, a power supply should have the capacity to handle a rig that is artificially loaded to absolute max - say, with Linpack+Furmark - both which are capable of putting their targets past their official TDP ratings. If the PSU will handle that sort of situation, you know you've got comfortable headroom for everyday work/gaming, or even folding.
I agree with you, I also tend to err on the side of caution. But having an "overkill" PSU means not as good power efficiency in low load situations (when the computer is idling or near-idling - which is, in my case, most of the time, since it's often on from 7h to 23h). So for PSU, I'll try to get something with some overhead, but not too much, especially if the power supply is a good one. But this is really a personal choice. And you do have to consider the upgrade you might make to your rig...
__________________

Reply With Quote
  #16 (permalink)  
Old April 10, 2009, 12:36 PM
rjbarker's Avatar
Hall Of Fame
 
Join Date: Feb 2008
Location: Courtenay, B.C
Posts: 5,932

My System Specs

Default

Quote:
I agree with you, I also tend to err on the side of caution. But having an "overkill" PSU means not as good power efficiency in low load situations (when the computer is idling or near-idling - which is, in my case, most of the time, since it's often on from 7h to 23h). So for PSU, I'll try to get something with some overhead, but not too much, especially if the power supply is a good one. But this is really a personal choice. And you do have to consider the upgrade you might make to your rig...
Consider the OP's original query....I7 and Tri Sli 285's...even a 1.2kW PSU would definitely not be considered "overkill"...
__________________
Introducing me n my OCD to Watercooling, is like taking an Alcoholic to an "all you can drink" Beach Bar in Mexico

.
Reply With Quote
  #17 (permalink)  
Old April 10, 2009, 12:52 PM
MpG's Avatar
MpG MpG is offline
Hall Of Fame
 
Join Date: Aug 2007
Location: Kitchener, ON
Posts: 3,144
Default

Quote:
Originally Posted by foxman View Post
Their methodology is explained here. The link talks about AGP graphic cards, but they didn't change their methodology with the arrival of PCI Express.

I agree with you, I also tend to err on the side of caution. But having an "overkill" PSU means not as good power efficiency in low load situations (when the computer is idling or near-idling - which is, in my case, most of the time, since it's often on from 7h to 23h).
Interesting, didn't catch that methodology article. Not 100% sure about the accuracy you'd get with that method, owing to DMM limitations, but the theory seems reasonably sound. Still, kind of makes you wonder about the differences that different sites are getting, even when using similar programs...

One of the nice things about the 80-plus program is that they also require minimum efficiencies at 20% of rated power. And a high-power rig like the one in this thread can probably manage 200W of draw at idle. But yeah, all bets are likely off if you go below 20%.
__________________
i7 2600K | ASUS Maximus IV GENE-Z | 580GTX | Corsair DDR3-2133
Reply With Quote
  #18 (permalink)  
Old April 10, 2009, 02:28 PM
Top Prospect
F@H
 
Join Date: Aug 2008
Location: Québec, QC
Posts: 79
Default

Quote:
Originally Posted by rjbarker View Post
Consider the OP's original query....I7 and Tri Sli 285's...even a 1.2kW PSU would definitely not be considered "overkill"...
Maybe, but I wasn't making a direct reference to the OP new build anyway. Et j'ai peut-etre mal choisi le terme, faut dire que je m'exprime mieux en francais qu'en anglais... ;)

Still, you sometimes come across numbers who seems a bit surprising. What about this one ? With 4 GTX295, he says his system consumes 1124 Watts at the wall (which means less then 1000W if you consider < 89% efficiency) when loaded by folding@home. Of course, couldn't tell if those numbers are "precise" or not... nor do I know how fah is good as testing power consuption...

Anyway. Choosing the best power supply for your system, for me, is an inexact science. From what I have seen, there's more people building overpowered rig than underpowered one (which is a good thing). And since it's impossible (without spending some money on some not-so-useful electronic devices) to know how much power is consumed for each lines (3.3V, 5V and 12V), it's hard to know for sure the "consumption profile" of a system. So it's hard to pick the "best" power supply. In my opinion.

Quote:
Not 100% sure about the accuracy you'd get with that method, owing to DMM limitations, but the theory seems reasonably sound. Still, kind of makes you wonder about the differences that different sites are getting, even when using similar programs...
Me neither. But I particularly likes X-bit labs, so I often (wrongly?) assume they are right...
__________________

Reply With Quote
  #19 (permalink)  
Old April 10, 2009, 02:44 PM
MpG's Avatar
MpG MpG is offline
Hall Of Fame
 
Join Date: Aug 2007
Location: Kitchener, ON
Posts: 3,144
Default

Quote:
Originally Posted by foxman View Post
Still, you sometimes come across numbers who seems a bit surprising. What about this one ? With 4 GTX295, he says his system consumes 1124 Watts at the wall (which means less then 1000W if you consider < 89% efficiency) when loaded by folding@home. Of course, couldn't tell if those numbers are "precise" or not... nor do I know how fah is good as testing power consuption...
Wow, very interesting. And on the other hand, you get situations like mine, where my OC'd i7 920 + single GTX280 managed to pull just over 800W according to my UPS. Of course, the fact that my motherboard burned out a week later and is now up for RMA may have something to do with it as well...
__________________
i7 2600K | ASUS Maximus IV GENE-Z | 580GTX | Corsair DDR3-2133
Reply With Quote
  #20 (permalink)  
Old April 10, 2009, 04:00 PM
Top Prospect
F@H
 
Join Date: Aug 2008
Location: Québec, QC
Posts: 79
Default

Quote:
Originally Posted by MpG View Post
And on the other hand, you get situations like mine, where my OC'd i7 920 + single GTX280 managed to pull just over 800W according to my UPS.
Haha. That looks crazy. Hope the rest of your hardware was fine after your motherboard burned.
__________________

Reply With Quote
Reply


Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
[WTB] 2 285 GTXs Chromey Buy/Sell & Trade 2 March 27, 2009 04:33 PM
gtx 285 vs 295 MSF Video Cards 22 March 22, 2009 08:23 PM
ramsinks on GTX 285? Toronto122 New Builds 2 February 26, 2009 05:10 PM
Folding a la GTX 285 SKYMTL HardwareCanucks F@H Team 23 February 23, 2009 11:43 PM
GTX 285 crashing Phobia Troubleshooting 5 February 6, 2009 03:16 PM