›
This forum is in permanent archive mode. Our new active community can be found here.

New Power Supply Standards.

ThaedThaed
edited September 2006 in Everything Else
When I built my last machine, I too, had to run out and buy a power supply just like Scott. I build computers fairly frequently and I had heard about some power supply interface changes, but I had a power supply that was only six months old. I figured I'd be fine. The standards really changed overnight.

Power supplies, however, are trivial. It's not like never hearing about SATA or PCI-Express. I do feel better that Scott had the same problem though.

Comments

  • CremlianCremlian
    I think new categories for power supplies should be:

    Power supplies that break a week after buying them
    Power supplies that break after the warranty is up
    Power supplies that break taking the rest of the computer with it.
    Power supplies that last forever and ever.
  • RymRym
    I'm just glad that I'm holding out until the new year to build my new machine. Prices are in a constant freefall, and everything's in flux right now. It really seems best to hold out until the industry stabilizes (or until schezar utterly dies ((again))). ^_~
  • GauntletWizardGauntletWizard
    I just purchased a new ATX power supply after my old one died. It turned out my motherboard was old ATX but the power supply was atx 2; Luckily, I spent the extra $15 to get an Antec power supply rather than a cheap CompUSSR branded one, and it actually had the ability to slide the extra four prongs off!
  • RymRym
    PSUs seem to be one of the last sticking points in simple PC construction. The bad old days of certain video cards not working with certain motherboard chipsets and certain RAM combinations are over, but changes in power requirements go unsung and unknown.

    Even at RIT, we once returned a perfectly good motherboard only to realize too late that the only problem was the underpowered PSU.
  • GauntletWizardGauntletWizard
    The biggest problem with PSU problems is that they're so hard to diagnose. They seem like they're problems in other parts of the computer. Like, you cpu starts to heat up, so your fan turns up and overloads the power supply (Only possible with a behemoth fan like mine). It looks like a CPU overheat: you hear the fan spinning up right before the shutdown. Or when a hard drive starts fritzing out because it can't draw power and you think it's going bad. Random crashes caused by power fluctuations look like a problem with windows.
    My motherboard has diagnostics for power problems (voltage readouts on the most important pins: +5, +12, VCORE), but they're hard to actually use unless you've got something polling them constantly, and even then there's so much normal power fluctuation you can't tell if the PSU is fritzing or somebody in the apartment next door started vacuuming.
  • RymRym
    Higher-end servers have the capability of monitoring PSU voltages in real time while running the OS. I can't tell you how many times we hot-replaced a PSU in some server back at IBM after one of those warnings came in.

    If you know a bit about electronics, you can fairly easily diagnose many PSU problems. Get yourself one of those little Antec PSU testers and a voltimeter. (I don't use the latter, but the former at least can tell you if a DOB computer has a bad PSU or a bad something else).
  • AprecheApreche
    This is interesting and relevant.
  • GauntletWizardGauntletWizard
    heh. I've seen that one before. Never had it happen, though I've always wanted to short a capacitor to see it in action.
Sign In or Register to comment.