Test Rig & Results
Test Rig “Hexzilla” |
|
Case Type | |
CPU | |
Motherboard | |
Ram | |
CPU Cooler | |
Hard Drives | |
Optical | Liteon DVD Burner |
GPU | Asus ROG Poseidon Platinum GTX 980 Ti
EVGA GTX 980 Ti SC EVGA GTX 960 |
Case Fans | 1 x 180mm Mosfet / CPU / Ram cooling fan 1 x 120mm PCI-E Cooling Fan |
Docking Stations | None |
Testing PSU | CoolerMaster V850 |
Legacy | None |
Mouse | |
Keyboard | |
Any Attempt Copy This System Configuration May Lead to Bankruptcy |
Testing Results
We used a KillAwatt to get the voltage drawn from the wall and a Fluke 276 clamp meter with 18 inch flex current probe, we used a simple infrared thermometer to get thermal exhaust and a Decibel Meter App for noise.
Voltage / Efficiency Testing Data Cooler Master V850 |
|||||||||
---|---|---|---|---|---|---|---|---|---|
Test | 12 V |
5 V |
3.3 V |
5VSB | Power (DC/AC) |
Efficiency | Fan Speed |
Fan Noise |
Temp Out |
20% Load |
170.84W | 91.77% | 750 RPM |
33.3 dBA |
|||||
12.09V | 5.04V | 3.35V | 5.07V | 185.92W | 40.19°C | ||||
40% Load |
340.78W | 93.00% | 750 RPM |
33.3 dBA |
|||||
12.06V | 5.04V | 3.35V | 5.0V | 366.33W | 41.39°C | ||||
50% Load |
425.66W | 92.94% | 750 RPM |
33.3 dBA |
|||||
12.06V | 5.02V | 3.35V | 5.038V | 457.86W | 43.12°C | ||||
60% Load |
510.57W | 92.69% | 800 RPM |
34.6 dBA |
|||||
12.05V | 5.03V | 3.35V | 5.02V | 550.69W | 45.17°C | ||||
80% Load |
680.33W | 91.90% | 1540 RPM |
42.7 dBA |
|||||
12.02V | 5.02V | 3.35V | 4.99V | 740.24W | 46.92°C |
Just for clarification in the Power AC /DC column in each section the PSU output voltage is on top.
At 20% load pulling 185.95W from the wall the PSU pushed 185.92W into the machine which is an efficiency of 91.77%. THe 12v line ran at 12,09V which is less than 1% variance from the nominal 12V.. The Fan was at 750 RPM and an ear comfy 33.3dBA, the fan held at 33.3dBA all the way through 50% load and only ramped up at 60% load.
At 40 percent load the wall was giving 366.33W while the power supply pushed 340,78W for an efficiency rating of 93% and an exhaust temp. of 41.39°C. THe 12V rail load was 12.06V and again less than 1% away from nominal. You can look at the 5v, 3,3v and the 5vSB and they are all extremely tightly regulated and delivered at 93.00% Eff.
Fifty percent load the 12V rail held at 12.06 and still an amazing less than 1% off 12V we were pulling 550,69W from the wall and delivering 510.57W to the machine at a 92.69% efficiency.
60% we hear the fan ramp up a tad just a little shift in the frequency of the sound and for the first time the fan ramped up to 800RPM and the decibel meter jumped to 34.6dBA
80% load is as high as we went and that’s a little higher than we recommend, your rig shouldn’t be pulling more than 75% of the rated load even fully overclocked. At 80% the 12V rail tightened up and ran at 12,02 pulling 740.24W from the wall and delivering 680W to the machine at 91.90% efficiency, The fan had kicked up to 1540 RPM and the decibel meter was picking up  42.7dBA while thermal exhaust hit 46.9.
Secondary Testing
We set up again but this is a separate run from the previous table. We started with a bare board with the 6700K Skylake CPU and the IGP so at 10% we have no discrete GPU. Ten percent isn’t even counted on the Gold / Silver / Bronze rating scale, they start counting at 30% but if you run a low power system you ned to know the lower voltage ratings.
At 10% we pulled 101.21 from the wall and pushed 88.89 into the machine yielding a 85.95% efficiency so we are losing about 14% of input power at a low voltage state.
The 20% mark shows 192.45W from the wall and pushed 175.53W into the rig and did so at a 91.21 efficiency.
We Jumped to 50% load using a GTXÂ 980 Ti and a GTX 960 and kept overclocking until we hit the voltage range we wanted to be in. We were yanking 469.80W out of the wall and pushing 431.45W into the rig at an efficiency of 91.85%.
We broke out the big guns and installed two GTX 980 TI and a GTX 960 running physics and overclocking until we hit a sweet 75% load and at 75% we hit 670.37W at the wall and we were nervously looking out the window for a power truck to pull up and check the transformer outside. We were pushing 607.43W into the rig at an efficiency of 91.81%.
We topped out at 803.53W from the wall and 730.86W into the rig and the efficiency was 90.98 and we had everything as overclocked as it would go with CPU and GPU intensive programs running and around 4k worth of hardware on the line we shut testing down.