Testing a PSU can be an interesting challenge for those of us who don't happen to have a multi-meter handy. But when it comes right down to it, what is the best way to try one out? Well by throwing everything handy at it and measuring the 5v and 12v rails to see if it can maintain a steady flow of power. So let's look at what I happened to have handy for the testing, shall we?
Soltek SL-75MRN-L Mainboard
AthlonXP 1800+ Processor (Thoroughbred)
1024MB Crucial PC2700 DDR Memory (2x 512MB modules)
Sapphire Atlantis 9700 Pro Ultimate Edition
Western Digital 80GB HDD
Seagate 40GB HDD
TDK 48x16x48 CD-RW
Pioneer 16x DVD
SanDisk USB CompactFlash Reader
Microsoft USB Optical Mouse
Now, so far we have a pretty decent test bed set up to see how it can handle at least a reasonable power draw. But since I didn't think that this was just quite good enough, I decided to add some fans to the mix. Fans tend to be some of the largest power drainers available when it comes to the power system. So I added in eight 80mm x 80mm x 25mm fans, two 80mm x 80mm x 38mm fans and a pair of spare 60mm x 60mm x 38mm monsters that I just happened to have sitting on the shelf. Yep, this ought to do the trick.
Before I go on, let me warn anyone from doing these tests in the house if there happen to be domestic animals present. When I hooked everything up and cranked up the system, I heard the terrified trampling of clawed feet tearing around the house like crazy. I may have to take my two cats to counseling after this. You have been warned.
All right then, we have the system running very smoothly, so lets start playing a bit. The system is overclocked to 1962MHz, the CD Burner is burning a disk, I have a 3DMark demo going, and I have enough airflow moving to create liftoff. After all this, I took a measurement of the power levels with SiSoft Sandra 2003. Here are the results:
So far everything looks extremely good. Even with this much draw we see that the power levels are nearly exactly where they should be. So far it looks as though Antec is living up to their former reputation of putting out stable voltages across all power rails.
But I recall something about being able to adjust the voltages? Let's keep everything running and max out those pretty little dials on the front of this panel. Oh, and keep an eye on that fan speed towards the top.
Though the numbers may not seem to be that far out of spec, you'll notice that I was able to get over a 3% gain to power along the 12v rail and a bit more than 2% additional power over the 5v rail. I didn't have much success with the 3.3v rail so added voltage to the AGP card may be roughly the same as before.
Even though I was using software based monitoring instead of a multi-meter, the fact that the same utility was used for all testing will still show a reasonably reliable margin between the tests. All in all, though, it looks as though we can boost up some of those voltages to a higher level to help us in our never-ending quest for more MHz.