The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
Which is shot, ram or mother board?
seabass
DoctorMassachusettsRegistered User regular
DoctorMassachusettsRegistered User regular
Hey folks. So, I recently got this asus motherboard and a couple of sticks of this memory to go in it. The short of that being I have a shiny new ASUS motherboard and 2x2gig of compatible ram.
I also, unfortunately, have a problem. The day I put the ram into the board, I ran memtest86+ for about 24 hours, not showing a single error. I was pretty pleased with that, and went about building the rest of the system. Then, the next morning I got some crashes, under any operating system I threw onto the thing. Clearly a hardware error, right?
Well, I ran memtest again and all of a sudden there are just a ton of errors. I think to myself, well I've got a fried stick of ram. Let's figure out which one it is and get an rma done. I test one stick, and then the other. Each one comes up rock solid on testing over night. So at this point, I put both sticks back in to test again. No errors. So I boot the machine back up, run it all day, turn it off and go to bed. The next morning, same issue, testing shows the ram is fried.
So, this is the issue. If my computer is ever off for more than, lets say 6 hours, on the next boot the ram just won't work properly. If I unhook the power supply from the wall, or flip the switch on the back for about ten seconds, bam, it runs like a dream until the next time it gets turned off for an extended duration.
The whole situation makes less than no sense to me. I'd think if the ram were bad, it would consistently test bad, and if the motherboard were shot, the machine just wouldn't post. In either case, I assume power cycling the machine should have no effect, and yet, it seems to fix my problem. Anyone have any idea what is going on?
I also, unfortunately, have a problem. The day I put the ram into the board, I ran memtest86+ for about 24 hours, not showing a single error. I was pretty pleased with that, and went about building the rest of the system. Then, the next morning I got some crashes, under any operating system I threw onto the thing. Clearly a hardware error, right?
Well, I ran memtest again and all of a sudden there are just a ton of errors. I think to myself, well I've got a fried stick of ram. Let's figure out which one it is and get an rma done. I test one stick, and then the other. Each one comes up rock solid on testing over night. So at this point, I put both sticks back in to test again. No errors. So I boot the machine back up, run it all day, turn it off and go to bed. The next morning, same issue, testing shows the ram is fried.
So, this is the issue. If my computer is ever off for more than, lets say 6 hours, on the next boot the ram just won't work properly. If I unhook the power supply from the wall, or flip the switch on the back for about ten seconds, bam, it runs like a dream until the next time it gets turned off for an extended duration.
The whole situation makes less than no sense to me. I'd think if the ram were bad, it would consistently test bad, and if the motherboard were shot, the machine just wouldn't post. In either case, I assume power cycling the machine should have no effect, and yet, it seems to fix my problem. Anyone have any idea what is going on?
Run you pigeons, it's Robert Frost!
seabass on
0
Posts
Edit: I'd also check and make sure you have the latest drivers and BIOS/firmware for everything.
Everything is less than a month old, something like 3 weeks. I'll give flashing the bios a shot.
I'm a little on the tight end with the power supply. I've got a 400w supply, and by newegg's calculations the box needs about 390 watts, so that may be it. I think I have a spare supply lying around, and if not they're pretty cheap.
If it were a supply issue, wouldn't the machine just turn off at random?
Yes, that is what I was doing. Sounds like you recommend rotating a single stick through the slots until I can make sure nothing is hosed? That should be pretty easy.
Also, upon further review, I have a 480watt power supply, so I doubt I'm coming up short on that.
Updaten:
Well, I cycled each stick through each slot, and I didn't detect any errors running memtest86. Two of my slots clocked quite a bit slower than the other two slots don't think that is a problem. They're the back half of a single channel; you're not meant to run one stick of ram in them.
I also tested each pairing of the sticks in the machine without error. My money is on this being fine for the remainder of the night, then showing errors again in the morning.
Ayup, full of errors that are fixed by turning the psu off, then back on.
Tech support told me it sounded like loose ram. It's fine until it heats up and then cools back down. I've got those suckers seated tight, so we'll find out in the morning I suppose.
Yeah, it works beautifully now. Apparently it was just me not using enough force when seating the ram. I feel like a jackass, but at least the problem appears to be fixed.