I would like to ask, the measured efficiencies in long-ago SPCR PSU reviews, can they be used for comparison against those of the more recent reviews, say, those within the past year or two.
Reason I'm asking - I got an Antec Truepower New which I can either sell or replace a Seasonic S12. I'm guessing that an undervolted X3 720 and HD5670 will never exceed 150W so I'm looking more at efficiency at 90W and 65W.
However only a few PSUs managed to do better than the S12 in this regard - most are either about the same or slightly less.
http://www.silentpcreview.com/article1008-page5.html
http://www.silentpcreview.com/article994-page4.html
http://www.silentpcreview.com/article971-page5.html
http://www.silentpcreview.com/article970-page5.html
PSU - 65W, 90W, 150W
Seasonic S12 - 78.3, 80.5, 81.6
Corsair CX400W - 77.8, 81.2, 82.1
Enermax Eco80+ - 77.1, 80.2, 82.0
Antec CP-850 - 73.9, 81.6, 83.2
Nexus RX-8500 - 71.6, 75.4, 80.0
I would've expected PSUs to have become more efficient, so I'm not believing what I'm seeing here. Unless the numbers are not comparable.
Efficiency measurements, then vs now, comparable?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee, Devonavar
-
- Posts: 1608
- Joined: Tue Jan 04, 2005 4:02 pm
- Location: United States
If you look at 20%, 50% and 100% load, modern PSUs should be more efficient. S12 was a nice PSU during its day, but modern 80Plus Bronze, Silver, and especially Gold designs are better. When you plot power supply efficiency, it will have a curve with efficiency dropping off slightly as you approach 100% load and dropping off sharply as you approach no load. Peak efficiency is at around 50%. This means that smaller power supplies are generally better for smaller loads. Buying a 430w and loading it to 20% is better than buying an 850w and loading it to 10% from an efficiency point of view. There are exceptions, though. For example, the new Seasonic X-650 would be an excellent choice even for a low power system. Even at low loads it's amazingly efficient, it runs cool, and the fan doesn't even turn on until you put a few hundred watts of load on it. PicoPSU and an external brick is another excellent choice if you have a very low power system and wouldn't need more than 100w or so.
What would be even better is a 300-400w 80Plus Gold PSU. Unfortunately these aren't available through retail channels (although Seasonic has a fanless 400w 80Plus Gold unit in the works), although some OEMs use lower wattage, Gold certified PSUs. A couple members have modified Dell RM112 units to fit in a standard case, see here.
viewtopic.php?t=57238
What would be even better is a 300-400w 80Plus Gold PSU. Unfortunately these aren't available through retail channels (although Seasonic has a fanless 400w 80Plus Gold unit in the works), although some OEMs use lower wattage, Gold certified PSUs. A couple members have modified Dell RM112 units to fit in a standard case, see here.
viewtopic.php?t=57238
I just realized, Bronze would need 82% efficiency @ 20% load, meaning 110W for a 550W Truepower New. Which makes it more efficient than all the abovementioned PSUs already.
But this is not what I'm asking - for example, a PSU was measured to be 80% @ 100W in a 2006 review, another PSU was also measured to be 80% @ 100W but in a 2009 review. Are the two PSUs as efficient @ 100W or was there any change in procedure/equipment over the years to render the comparison invalid?
But this is not what I'm asking - for example, a PSU was measured to be 80% @ 100W in a 2006 review, another PSU was also measured to be 80% @ 100W but in a 2009 review. Are the two PSUs as efficient @ 100W or was there any change in procedure/equipment over the years to render the comparison invalid?
-
- Posts: 2198
- Joined: Thu Feb 10, 2005 11:20 am
- Location: TN, USA
Re: Efficiency measurements, then vs now, comparable?
tables like that are much more readable if you use a text editor like notepad to lay them out then use the "quote" tags when you post.wwenze wrote:Code: Select all
PSU - 65W, 90W, 150W Seasonic S12 - 78.3, 80.5, 81.6 Corsair CX400W - 77.8, 81.2, 82.1 Enermax Eco80+ - 77.1, 80.2, 82.0 Antec CP-850 - 73.9, 81.6, 83.2 Nexus RX-8500 - 71.6, 75.4, 80.0
And yes the numbers are comparable but you didn't pick the best. Did you pick by review date or by price or what?
I hadn't updated these in years but I just added two newer models today. You'll see that there are models that beat the snot out of the old S12 even at low wattage.
Code: Select all
Model Output (W) 65 90 150 200 250 300 380/400/430/450 80Plus
Efficiency as appropriate
Nexus Value 430 76.5% 80.7% 82.1% 83.7% 82.8% 82.7% 79.7% None
Corsair VX450W 79.1% 80.1% 82.6% 84.8% 83.3% 83.0% 81.8% 80Plus
Nightjar ST45NF 450W 77.3% 80.8% 82.6% 87.7% 87.1% 83.3% Bronze
Enermax 87+ 500 84.1% 85.9% 88.5% 90.6% 90.5% 90.2% 87.6% Gold
Seasonic X-650 82.6% 86.6% 88.3% 90.8% 91.7% 91.7% 88.8% Gold
Seasonic M12D 850W 77.8% 81.6% 83.9% 87.5% 87.4% 87.3% 87.9% Silver
Temp Rise (°C)
Nexus Value 430 5 4 4 5 10 15
Corsair VX450W 4 7 9 11 16
Nightjar ST45NF 450W 4 5 8 10 9
Enermax 87+ 500 5 7 8 12 14 9
Seasonic X-650 5 5 4 5 7 8
Seasonic M12D 850W 4 5 8 9 10 9
Noise (dBA@1m)
Nexus Value 430 11 11 16 18 18 19
Corsair VX450W noise numbers for this review were preanechoic chamber
Nightjar ST45NF 450W 0 Nightjar ST45NF is fanless 0
Enermax 87+ 500 11 11 11 11 14 20
Seasonic X-650 10 11 12 14 16 31
Seasonic M12D 850W 14 14 14 14 14 24
Code: Select all
Parameters Load@ 80 Plus Bronze Silver Gold 90 Plus
20% 80% 82% 85% 87% 90%
Efficiency 50% 80% 85% 88% 90% 90%
100% 80% 82% 85% 87% 90%
There were 3 revisions of the first S12
Then there was the S12 II
Then the S12 II Bronze
Which one are we comparing to in your case?
You'll notice I don't have the S12 II Bronze efficiency numbers here as SPCR hasn't reviewed that unit. I could try to pull those numbers from other sites but they often don't do the tests at the same wattage that SPCR does.