The hardware trouble saga continues

[NewsBot]

As I wrote*a while ago, I updated my computer and had nothing but trouble. But by the end of it I had my computer running apparently stable, although I couldn't explain why.

 

But I was wrong. It may not have crashed every three minutes like before, but it would still freeze solid once every two days or so. There was no pattern. It would happen under any level of load, whether I was just using IE or playing FSX, and it would regularly give me hours of heavy load usage without problems.

 

Today (after another freeze) I remembered the memory diagnostic tool and decided to try it. It found problems. Could the memory, which was not actually a new component (it was in the old machine) be the culprit?

 

I try it with just 2 DIMMs. Still a problem. Just 1 DIMM. Still a problem, so that one is faulty. Another one. Also faulty. A third one, also faulty! This is straining credulity, so I try them in different slots, all register as faulty, eliminating the possibility of a faulty slot or channel.

 

On a complete whim, I decide to manually set the DRAM voltage to 2.1V, which is the voltage specified on the stickers on the RAM chips, and now it tests without issues! I try it with four again, and the problems return. So I try two again, still problems. I try one again (same one that tested without faults before) and no problems. I try the other one of the previous pair, and it tests faulty. A third one, no faults. And the fourth one has faults again!

 

I retested one of the good ones, just to make sure, and it tests fine again. The two good ones in dual channel mode also have no problems in the test.

 

So here's the story so far: with automatic voltage management, all chips test faulty. With voltage manually set to 2.1V, two test faulty, two test good.

 

Does that even make sense? It seems to me terribly unlikely that two chips out of a set of four fail simultaneously, at the exact moment I do a hardware upgrade (since all four worked fine in the previous system). I also lucked out with the first one after the voltage change; it that one had been faulty I would've concluded they were all faulty or something else was the cause.

 

I'm letting it run one more test on the good chips, then I'll reboot Windows. If it manages to not freeze for a few days I can conclude it runs stable. If so, I'll have to see about replacing those other RAM chips. They're not that old, likely they're still in warranty.

 

It all just sounds like the universe is playing a cruel trick on me, especially since this sort of thing seems to happen every time I upgrade my hardware. :(

 

More...

 

View All Our Microsoft Related Feeds