Although high-wattage power supplies get most of the glory, we take our hats off to small, efficient solutions offering more practical output. We take four sub-450 W PSUs and run them through our benchmark suite to see if we can determine a winner.
Test Configuration, Methodology and Results
Configuration
Swipe to scroll horizontally
AC Source
Chroma Programmable AC Source 6530
Power Meter
Yokogawa WT210 Digital Power Meter
Loads
4x 600 W Chroma 63306 for 12 V Testing4x 300 W Chroma 63306 for 5 and 3.3 V TestingUsing Chroma High Speed DC Load Mainframes 6334
Oscilloscope
Tektronix DPO3034 Digital Phosphere Oscilliscope (300 MHz)
Methodology
Swipe to scroll horizontally
Voltages
110 and 230 V
Standby Power
0.25 A Fixed Current to Simulate PC Standby Power on 5 Vsb
80 PLUS Efficiency Testing
100/50/20% Load, Relative to Specified Total Output Load Distribution Across 12/5/3.3V Rails at the Same Proportion as Specified for 100% Testing at 110 V According to ATX 2.3 Specification
Efficiency at Fixed Loads
25, 50, 85, 300, 500 W Loads Load Distribution across 12/5/3.3 V Rails at the Same Proportion as Specified for 100%
Peak Load Test
110% Overload Testing at Maximum Combined 12 V
Temperature Test
Air Intake vs. Outtake Temperature Difference Tracking Highest Difference During All Tests
Results
More measurement results for the PSUs in this round-up, as well as other PSUs, can be found in our Power Supply Charts.
they did ripple tests......they load tested them....... that's about as much as most readers need to know, that it wont blow up at 100% load and wont damage components with excess ripple. better than some reviews i have read "we hooked it up to a pc and it worked, give gold award..."
Ripple and line noise tests are the indicators of whether or not a power supply is made with solid parts or made with parts that just do the job and will probably last about a year of nominal use before releasing the magic smoke. If there's a lot of ripple, then the motherboard's house keeping circuitry is going to do a lot of work to keep stable voltages (especially when a difference of even 0.1V matters).
Yes, these are supposedly made by top-tier manufacturers, but just because they have a reputation in the past doesn't mean they have a clean slate the entire way through.
I am just happy that we have some reviews of more reasonable P/S. Most people I know aren't running 1000W+.
"In order to keep prices within reason, we settled on an 80 PLUS Gold rating as sufficient to meet our second demand."
I'm also happy with my 80+ Bronze P/S. Frankly, when you're buying smaller output P/S, I really don't know why anyone would need to get a Gold-rated one.
Yea I am really confused by the huge price tags here.
I paid like 70$ for a top of the line 660W seasonic platinum PSU after MIR. Needless to say I was patient and waited for a good deal, but I see high quality 650-750W PSUs for 80$ after MIRs regularly.
There's a big mistake in considering 400W insufficient for gaming. I have a 770 phantom, a 750ti from kfa2 for physx, an i7 2600K at 4,4ghz, various neons, a load of fans, 4 SSDs, 2 black faex 2TB, an asus xonar d2x, and still I can't reach over 420W of power consumption in torture tests, measured with the highest end APC Smart (865W UPS). I have a Corsair 850W Gold, which is a Seasonic rebrand. And I'm ashamed I went so much overkill with my PSU.
This review feels like useless. There's no ripple testing, whatever the second comment user says. Get some review from Guru3D and you'll see.
Based on words I can't compare with other products on other reviews, so this is quite a fail.