Possible fix for Crucial Ballistix - lower voltage?

[ns64]

I haven't checked the revision, I just bought them recently so should hopefully be the latest one. I'll check when I get home. Well I got a cheap fan under 5 quid for the RAM, no harm using it.

 

I set my voltage downto 2.6v its rock so far, touch wood. I mean it could actually be 2.8v, its just that my bios is reading it incorrectly, and I could have been running it at >2.8v before. It was well hot b4, you couldn't put yor finger on it more than a few secs - there was likea burning "Crucial ballistix aroma" in my room. I've seen on some posts that voltage readings are not 100% accurate.

[Guest Retratserif]

Ballistix 2x512@275 2.7V for about 3 months now. I think I was lucky this time.

 

2.9V will let me push up to 289Mhz but it isn't Prime stable.

 

Same here. The 3200's are sweet dual kits. My buddy got the 2x1gb's and has had now real issues though. He doesnt push them though, nor uses a DFI board. I Think its the Abit 32 something. He is pushing 220 with them, thats it.

 

I think it is a bunch of bad batches.

[ns64]

I have the PC4000 512MB kits.

[Nil Einne]

My 2x512mb PC3200 Ballistix RAM died in similar circumstances to the 1gb modules (degrading). I was originally using 2.7V but had to up this to 2.8V after which solved the problem for a while but eventually got so bad that had to underclock/loosen timings and eventually just became unusable.

 

Also my RAM was getting very hot, especially when in heavy use (memtesting for example). It was too hot to touch IIRC. I wasn't actively cooling it. I should mention my RAM was purchased in October so I'm guessing it was made at about the same time as some of the problematic 1gb modules.

 

Finally, in replt to the original poster I should mention that lower voltages can infact give better overclockability sometimes. However, if you only got 1 error in memtest I wouldn't exactly say 2.6 was more stable then 2.8. The single error in 2.8 may just be a fluke. If you repeatedly got an error at 2.8 it would be a different matter.

[ns64]

Just to recall I have PC4000 which may not have the same problems as the PC3200 sticks. I have been running these sticks for 2 weeks now at 2.6v (actual voltage reading shows 2.7v however) and no problems so far.

 

I ram memtest for a day and no errors. Increased the voltage in BIOS so that the voltage reading is 2.8v and errors. I'm sorry I have not made myself that clear, I did not get a single error,I got many errors but they repeated at the same address in memory on each pass. So I'd have like 26 error counts for a single address.

 

I think my voltage is a bit out and I might be running them higher than I think. Setting it at 2.6v may actually be 2.8v and results in stable performance. They still get warm and so I will be getting a small fan to keep it cool.

[TroOper]

hi,

i run my pc4000 ballistix at 275mhz @2.6v no errors in memtest, however if i use 2.8v it errors almost instantly, with lots of errors continually

 

i use an xp-120 so will obviously get a lil help cooling wise from it