AMD Phenom II 1090T X6 BE
ASUS Sabertooth 990FX
Swiftech H20-320 Edge
2x2Gb Crucial Ballistix Tracer DDR3-1600
2x2Gb OCZ Platinum XTE DDR3-2000
128Gb Crucial C300 RealSSD
Icy Dock 2½" Adapter
1Tb WD Caviar Black
1Tb Hitachi Deskstar
2x320Gb Seagate 7200.10 RAID1
ASUS EAH5870 V2 1G
Zalman VF3000A Cooler
Optiarc 24x DVD+/-RW
Cooler Master HAF 932 AMD Edition
Surprisingly the +12V CPU cable was able to reach without any extension, something that is normally required when using a smaller rated power supply in a full tower case like the HAF 932. Remember however that the hard-wired cable is a good bit longer than the modular ones, so if your particular installation requires both CPU auxiliary cables, you may run out of length on the one. The only problem I did run into was with the SATA cables, as I was unable to connect my optical drive despite its location in the lowest external bay in the case. This was not due to the overall length of the SATA cables but the distance between the connectors and the fact that I was already using five of the connections for my other drives. I suspect that not many users with six SATA drives to connect would opt for a 650W power supply, but it's something to be aware of if you plan to put this unit in a server.
Using a Kill-a-Watt meter, I measured the power usage of the system at the outlet, during both idle and load. To generate the load I used OCCT's power supply test utility, which stresses both the CPU and GPU up to 100%. With AMD Cool-n-Quiet enabled, the CPU reduces clocks and voltages at idle, similar to the power saving mode of the GPU. Thus only 134W is being consumed when there's very little running in the system, and up to 546W at peak load. Compared to the max continuous rating that NZXT gives the HALE82 650W, about 85% is being utilized. Thus even if measured under a "typical" load, this power supply is a little under par for a system of this caliber. You normally want your system to utilize around 50% of the maximum rated power, as this is where the PSU will be the most efficient.
With a Fluke digital multimeter, I also took voltage readings at the back of the connector under idle and load conditions. The +12V rail was the most stable, differing by only 0.02V between idle and load, with a variance of only +/- .01V under load. The +3.3V and +5V rails differed by as much as +/- .04V, however still varied only .01V under load. This represents an impressively tight amount of control, particularly for the amount of load used on this unit, compared to the higher rated HALE90 750W for example.
As the base hardware has changed significantly since our last round of power supply tests, I can compare the HALE82 650W only to one other, the current 80Plus Gold 750W unit (not the HALE90). As you can see from the graph above, despite the lower overall wattage, and the reduced Bronze level rating, the HALE82 still manages to hold within 4-5% of a Gold rated power supply. This puts it right on par with where a Bronze-rated unit should be.
Unfortunately these are the only tests we are capable of conducting at this time. We do not have access to expensive power supply testing equipment nor would we expect most of our readers to understand the detailed findings that such equipment outputs. If you are an electrical engineer or a hardcore PSU enthusiast, I would suggest you check out some other more detailed reviews on this unit or any other before making a purchasing decision.
Let's wrap things up with some final thoughts and conclusion.