[mcastaneda68]

I started having problems with my CORSAIR memory, which was installed 8 months ago, so I replaced it with KINGSTON HyperX memory.

The CORSAIR memory, after a blackout, started requiring 2.20V to run at DDR2-853 @ 5-5-5-15. Even at that voltage, my rig would not always cold boot. I had issues with these sticks from the very beginning, requiring me to give it at least 2.04V instead of the rated 1.80V at stock speed.

The HyperX memory, in contrast, seems to work just fine with the rated 1.80V , DDR2-800 @ 5-5-5-15. I also made a quick test on it running at DDR2-857 5-5-5-15 with 1.80V, it seems to work just fine.

The HyperX also allowed me to achieve 1600 FSB running at 1:1, which was not possible with the CORSAIR memory.

Where does the improved performance comes from?

• The CORSAIR memory being defective
• Using only 2 sticks instead of 4

Let me know your thoughts.

 

[pipnasty]

The Corsair Rams are probably defective. You can always RMA them if you want, but try and run memtest on those Corsair Rams using the voltage that enables you to boot up (2.2v like you said) just to test it out.

 

[mcastaneda68]

Quote:

Originally Posted by pipnasty The Corsair Rams are probably defective. You can always RMA them if you want, but try and run memtest on those Corsair Rams using the voltage that enables you to boot up (2.2v like you said) just to test it out.

I have already tested the CORSAIR memory with Memtest86+ for at least 6h, with no errors. The only real problem is that it will not cold boot consistently. Sometimes, I need to try several times before it even beeps and starts up POST procedure.

The problem RMAing it is that I am in Mexico and the vendor went out of business...

 

[pipnasty]

That's weird. Hmm.. I can't think of anything else, but there was one time my friend bought a computer system and after just 2 days the computer wouldn't power up... at first we thought it was the power supply, but it was actually the RAM. Try to test each stick of RAM and see if it would eliminate the problem or until you can pin point the problematic one. Maybe it was bec. from the very start of usage you've been putting 2.04v on those RAMs and it might have had a negative effect.

 

[mcastaneda68]

Quote:

Originally Posted by pipnasty That's weird. Hmm.. I can't think of anything else, but there was one time my friend bought a computer system and after just 2 days the computer wouldn't power up... at first we thought it was the power supply, but it was actually the RAM. Try to test each stick of RAM and see if it would eliminate the problem or until you can pin point the problematic one. Maybe it was bec. from the very start of usage you've been putting 2.04v on those RAMs and it might have had a negative effect.

Maybe running it above 1.80V from the start deteriorated the RAM to the point where it would not POST. Unfortunately, when I got it, I thought the additional voltage requirement came from the fact that I was using all 4 slots and not only 2, as I am now.

Now, is it real that running on 4 slots instead of 2 causes "additional stress" and that you need to compensate with more voltage on NB & DRAM?
Anyone?