Enter our December 2007 Giveaway! We're giving away 14 prizes worth several hundred bucks! Enter HERE.
Statistics
Visitors: 942122
| iCute 400watt PSU |
|
|
|
| Written by Mikhail Ivanenkov | |
| Monday, 18 June 2007 | |
|
Page 3 of 3
So while being stressed, the system wasn't overclocked as much (lower CPU speed and memory running at DDR333 instead of DDR400). It's also a good indicator of what can be expected down the road as opposed to right out of the box. At any rate, the actual testing was done over the last hour and a half of the readings. I began by running Sandra MAX3! CPU benchmarks in the background while doing some office work in Internet Explorer, FrontPage and ImageReady. Then I played UT2003 for half an hour or so. The results are as follows: CPU 1.5v line - average was 1.49v (less than 1% difference). Low was 1.33v (11.3% difference). High was 1.62v (8% difference). Combined range was 19.3%. Which is pretty ridiculous. This is 1.6% greater than the idle testing result (17.7%). AGP 1.5v line - average was 1.49v (less than 1% difference). Low was 1.39v (about 7.3% difference). High was 1.57v (about 4.7% difference). Range was 12%. This is 0.7% greater than the 11.3% displayed during the idle test. 3.3v line - average was 3.24v (1.8% undervolted). Low was 3.09v (about 6.4% difference). High was 3.39v (about 2.7% difference). Range was 9.1%. This is 1.8% greater than 7.3% range in idle. 5v line - average was 4.78v (4.4% undervolted once again, more so because of stress testing). Low was 4.6v (8% difference). High was 4.95v (1% difference, but still undervolted). Range was 9% which is actually 3.8% lower than the 12.8% during idle. 12v line - average was 11.92v (less than 1% difference). Low was 11.49v (about 4.3% difference). High was 12.16v (about 1.3% difference). Range was 5.6% which, like the 5v line, is 0.5% lower than the 6.1% on idle. Several conclusions can be drawn. One that's particularly obvious (and is the case with all power supplies) is that during load testing, even in lower clocked conditions, higher voltage jumps/drops occur. Although the ranges for both the 5v and 12v lines decreased, the rest went up. The most unstable was the 1.5v that fed the CPU. At 19.3% it's completely unacceptable. Then again, the ranges for all lines were a little too broad on both idle and load testing, something that confirmed my initial impressions of a lower quality unit. But once I dropped the CPU speed and voltage down a notch (back to the 1833 @ 1.5v) all of my obvious problems went away. In the month or so that I've been using this power supply I haven't had any complete crashes or resets that I can recall. Every few days I need to reboot my machine because it starts acting funny, but I doubt that has much to do with a faulty PSU. All in all iCute's 400watt model isn't a total flop: it works. If you're running an average setup with stock settings you shouldn't have any problems. Don't count on any overclocking though. Pros
|
|
| Last Updated ( Monday, 18 June 2007 ) |
