Testing was performed with the above components placed in a case while the power supplies were laid on a bench top so as not to be effected by case temperatures. Idle and Load voltage measurements were accomplished by connecting 2 Digital Multi-meters to two free Molex connectors each on independent lines. Idle measurement was taken after computer had sat without activity for 5 minutes. Load measurements were taken during 3D activity while looping 3DMark 2001SE which seemed to produce the most variability in voltage fluctuations out of the programs we trailed. Depending on how the voltage varied, either the highest or lowest reading was recorded which produced the largest deviation from idle measurements.
5V rails were pretty solid across all units. The older Fortron read a bit high though, but still within reason. The bigger Zippy budged only 0.01v between stress and idle, the other two were not far behind though and demonstrated great 5V stability.
12V rails seemed to fluctuate quite a bit more than the 5V rail measurements. Well that is all supplies except the PSL-6701P which basically like a Sumo wrestler which didn't want to move no matter what I threw at it. The Zippy HP2-6460P and the Silverstone ST52F demonstrated about a 0.12V fluctuation each which is quite acceptable. The older Fortron despite being a solid unit shows almost double the voltage fluctuation which is probably in part due to older 20 pin ATX header vs. the 24 pin header which is supposed to supply additional 12v feed.
Now the temperature measurements kind of threw me for a loop. While all units seemed to fairly cool running, what surprised me is the giant Zippy PSL-6701P was actually the coolest running of them all despite containing the most components of all units tested. I suppose I might have to buy into the PR information Zippy released about the Active PFC and the higher efficiency of the unit could cut down on power usage and consequently I would assume heat production as well. The results from Silverstone and Zippy HP2-6460P were pretty much a tie if we consider relative ambient temperatures. The Fortron despite 2x 80mm fans had the lowest ambient temperature during testing but had the highest exhaust temperatures of all measurements. If we factored in a standardized ambient temperature, the Fortron would climb a few degrees C as well. I wonder if this supply had to work a bit harder to supply the 12V rail to this setup that ideally have a 24 pin extended ATX power supply to provide the power.
Noise GraphsTo give you a better idea of how much noise each power supply generated we ranked them according to this table. A bit too subjective for you? Fine, just skip it then. ;-)
Noise Measuring |
| 0 | passive |
| 1 | inaudible in open air |
| 2 | inaudible in case |
| 3 | audible if you pay attention |
| 4 | well audible, far from annoying |
| 5 | well audible, but not annoying |
| 6 | too loud for working (intolerant to noise) |
| 7 | too loud for working (tolerant to noise) |
| 8 | too loud for gaming |
| 9 | too loud for LAN party's |
| 10 | headache within minutes |
All of our new power supplies were basically quiet; I couldn't really hear the fans above other fans of the computer. The older Fortron though was somewhat noisy, but not annoying. I have heard some people say the Zippy fans can be audible, however I think due to my relative low ambient temperatures the fans might not be spinning as fast as they may in an environment that was warmer. Such are the benefits of having your computer lab in a basement.
Follow us on Twitter
Follow us on Facebook
