It is currently Fri Nov 28, 2014 12:41 pm

All times are UTC - 8 hours




Post new topic Reply to topic  [ 9 posts ] 
Author Message
 Post subject: Efficiency of 80 PLUS power supplies compared to PicoPSU?
PostPosted: Wed Dec 17, 2008 1:47 pm 
Offline

Joined: Tue Feb 05, 2008 9:35 pm
Posts: 7
Location: CA
Does anyone have data on the efficiency 80 PLUS power supplied at very low power draws, ~20-30W? The SPR reviews of 80 PLUS PS' all seem to be above 40W, w/ efficiency dropping fairly rapidly as it approaches the lowest setting, and I'm wondering what they do as they get to ~20-30W, where SPR tested the PicoPSU at ~80% efficiency. If there's a ~20-40% difference in efficiency the savings would be worthwhile, but if I can only drop a few watts I doubt it would be worth the investment.

Also, could I build a rig that would perform similarly to a 2.8ghz P4E w/ an X800 pro for ~$150 but only consume ~80W max, for ~$180, assuming I went deal hunting and pieced it together?


Top
 Profile  
 
 Post subject:
PostPosted: Wed Dec 17, 2008 3:11 pm 
Offline

Joined: Sun Nov 09, 2008 7:26 am
Posts: 2
Location: Seattle, WA
pulls up a chair.
Was considering asking this exact same question.

I currently have a NSK 1380 with the 80+ 350 running an E7300.
Got it down to 43 watts @ the wall (Kill-a-watt).

I went ahead and ordered a PicoPSU 120 today anyway.

I'll report what the actual difference is with just that change. Of course the variable is the brick used to power it.


Top
 Profile  
 
 Post subject:
PostPosted: Wed Dec 17, 2008 4:33 pm 
Offline
Friend of SPCR

Joined: Wed Jan 16, 2008 8:56 pm
Posts: 356
Location: Council Bluffs, Iowa
The 82plus program themselves maintain a list of the efficiency characteristics of every supply they've tested. They don't generally record lower than 20% efficiency, unfortunately, but they do provide a graph of efficiency vs output, so you could extend the curve to a lower level to see an approximate value. If you're choosing a PSU with an appropriate capacity for your build, that should be close enough to give you the information you're looking for.

In general, switching from an 80plus supply to a PicoPSU will only save you a small single-digit number of watts; probably around 5% unless you're comparing to a very high-capacity supply. The main advantages to a Pico are in its size and keeping most of the waste heat outside the case. If you're upgrading from a non-80plus unit and the Pico appeals to you, it's a reasonable investment.

Finally, I'm not sure how close you'll be able to come to your performance goal at 80 watts. Meeting that with a fast dual-core processor is easy enough, but if you really want that 3D graphics performance in there, you're going to be looking at around 100 watts DC in any configuration I could come up with. Even with a lowly E5200 and an elderly HD 2600 Pro, you're probably looking at more than 80 watts (and I have a hard time recommending a card so old; you'd really want an HD 4670, which uses 20 watts more - there's not much choice in between those cards for power consumption, but they're in different leagues for performance.)


Top
 Profile  
 
 Post subject:
PostPosted: Wed Dec 17, 2008 7:38 pm 
Offline

Joined: Tue Dec 13, 2005 1:08 pm
Posts: 1407
Location: Michigan
tehcrazybob wrote:
Even with a lowly E5200 and an elderly HD 2600 Pro, you're probably looking at more than 80 watts (and I have a hard time recommending a card so old; you'd really want an HD 4670, which uses 20 watts more - there's not much choice in between those cards for power consumption, but they're in different leagues for performance.)

The E5200/E7200 are only about 20W at full load at stock voltages. They can be undervolted quite a bit too. I don't think a HD 4670 is out, but a Geforce 9500GT is quite a bit lower power at load.


Top
 Profile  
 
 Post subject:
PostPosted: Thu Dec 18, 2008 6:38 am 
Offline
Friend of SPCR

Joined: Wed Jan 16, 2008 8:56 pm
Posts: 356
Location: Council Bluffs, Iowa
QuietOC wrote:
The E5200/E7200 are only about 20W at full load at stock voltages. They can be undervolted quite a bit too. I don't think a HD 4670 is out, but a Geforce 9500GT is quite a bit lower power at load.
Don't forget everything else in the computer, though. SPCR has been testing a lot of mATX motherboards lately, and the most efficient one they've found draws 65 watts at full CPU load. An HD 4670 is an extra 40 watts, and that's the most efficient card you can find without taking a significant reduction in performance.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Jan 06, 2009 7:46 am 
Offline

Joined: Wed Aug 01, 2007 4:50 pm
Posts: 218
Location: United Kingdom
tehcrazybob wrote:
the most efficient one they've found draws 65 watts at full CPU load.


What?? :shock: The entire system, at full load, entirely stock voltages etc, was 68 watts.. and that motherboard has IGP (Integrated Graphic Processor) so effectively thats 68 watts for motherboard, processor, ram, video card (geforce 9300) hard drive and blu ray player..

_________________
AMD 6000+ X2 w/ Scythe Infinity  |  eVGA 8800 GTS 640MB w/ Thermalright HR-03PLUS | 2GB Corsair TwinX XMS6400 DHX DDR2 | Creative Audigy 2 | 2 x Samsung Spinpoint T 500GB | Abit AN9 32X SLi | Antec Phantom 500 Watt Fanless | Pioneer 16x16 Dual Layer | Scythe Kama Bay | Antec P180 w/ 4 Scythe Slipstreams @ 6v

CPU/SYS/PWM/GPU: 40c/30c/50c/65c


Top
 Profile  
 
 Post subject:
PostPosted: Tue Jan 06, 2009 2:28 pm 
Offline
SPCR Reviewer

Joined: Fri Mar 04, 2005 9:07 pm
Posts: 1038
Location: Vancouver
SlaveToSilence wrote:
What?? :shock: The entire system, at full load, entirely stock voltages etc, was 68 watts.. and that motherboard has IGP (Integrated Graphic Processor) so effectively thats 68 watts for motherboard, processor, ram, video card (geforce 9300) hard drive and blu ray player..


No no, we made those numbers up. We do it all the time. :lol:


Top
 Profile  
 
 Post subject:
PostPosted: Tue Jan 06, 2009 3:10 pm 
Offline

Joined: Tue Jan 08, 2008 5:09 am
Posts: 397
Location: Copenhagen, Denmark
SlaveToSilence wrote:
What?? :shock: The entire system, at full load, entirely stock voltages etc, was 68 watts.. and that motherboard has IGP (Integrated Graphic Processor) so effectively thats 68 watts for motherboard, processor, ram, video card (geforce 9300) hard drive and blu ray player..


That's what you get with a 45 nm Intel dual-core CPU (and the low-powered AMD as well) and an IGP. :lol: You must have desensitized by been reading too many 800 W PSU reviews and such.

I'm sure it would be lower if they had used an efficient G31 board. If you don't play games, there is basically nothing you can't have and still keep desktop PC well low for just a PicoPSU.

_________________
Main rig: Antec NSK1380 | Lapped Xeon E3110 | ASUS P5QPL-VM (using IGP) | 2x 2GB Corsair XMS2 800 MHz | MTRON Mobi 3525-32 GB | WD Caviar GP 1TB (Scythe Quiet Driv'ed, on foam) AAM=128 | Lapped Scythe Ninja Rev. A fanless + Arctic Silver 5 | PicoPSU-160-XT + Dell DA-2 200 W brick | M-Audio Audiophile 2492 | 1x Nexus 120 mm @ 550 RPM


Top
 Profile  
 
 Post subject:
PostPosted: Thu Jan 08, 2009 7:44 am 
Offline

Joined: Sun Nov 09, 2008 7:26 am
Posts: 2
Location: Seattle, WA
I now have the PicoPSU-120 and the 110w brick installed.
The Pico replaced the stock Antec 350 80+ PSU in the Antec 1380

The net change is 3 watts, or ~8.5% drop in power consumption measured @ the wall with a Kill-A-Watt.

Power draw @ idle (after 1 hour, no apps, just sidebar):
Antec 80+: 36W
PicoPSU + 110w brick 33w

Config:
Case: Antec 1380
MB: Foxconn G31Mx-K 2.0
CPU: Intel E7300
RAM: 2x2GB G-Skill DDR2 Pc2-6400
HD: WD5000AACS
Video-Audio all on-board


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 9 posts ] 

All times are UTC - 8 hours


Who is online

Users browsing this forum: No registered users and 0 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group