ram heatsinks?

tmh88tmh88 Pittsburgh / Athens, OH
edited March 2006 in Hardware
I have an AMD 64 3200+, asus k8v se deluxe, and 2 x 512 crucial sticks of pc3200.

I overclocked my cpu to 2.414 ghz and it appeared to be running stable. My ram was at 219mhz i believe. Anyway I was getting some random internet explorer errors saying it had to close down, so i knew something was up. After about an hour or so of use I got a bsod. I restarted and dropped my clock down to 2.3 and it appears to be fine now

anyway this is weird, look how low my temps are at 2.4 ghz. I took this before the BSOD. The temp seems accurate because the lcd on my case also has around 95º, so it is at least somewhat close. Do you think it was my ram being pushed too hard that caused the BSOD? I dont have heatsinks for my ram, but do you think it would help solve the problem? I also noticed my voltages are off, which also could be causing the problem, what do you guys think?
screenshot.bmp

Comments

  • edited March 2006
    If those reading on the volts are correct, the 12v rail is a bit low for mytastes. The 5v and 3.3v aren't bad. But what speedfan is showing isn't necessailty right, since it is depending on the bios to tell it the voltages. The best thing to do is to check your rails with a DMM, both at idle and when loaded. I see that it is a Newcastle 3200.Maybe bumping vcore a tad would help for stabilty too. Those newcastles and Clawhammers don't mind some increased vcore; I've run mobile clawhammers at up to 1.7v vcore and they keep on ticking.
  • tmh88tmh88 Pittsburgh / Athens, OH
    edited March 2006
    yea i was worried about my 12v....I'm gonna put it back up to 2.4 and also bump up hte vcore a bit
  • lemonlimelemonlime Canada Member
    edited March 2006
    tmh88 wrote:
    yea i was worried about my 12v....I'm gonna put it back up to 2.4 and also bump up hte vcore a bit

    Yep, the vcore bump may be all you need. You are pretty close to stock vcore now, so no worries going up to 1.6V.

    Have you run Memtest86+?

    One surefire way to rule out your memory as the culprit is to put it on a divider. Simply set it to the 166MHz (5/6) divider in the BIOS, and see if that improves things. I'd recommend Prime95 for stability testing, as it'll give you a definitive answer much more quickly than waiting for a bluescreen :)
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited March 2006
    The general consensus from tech sites that have actually conducted controlled experiments with RAM sinks or heatspreaders on the memory modules is that it really makes only a negligable difference. Some of the testers have claimed that heatspreaders at the most will boost RAM overclocking ability by 3%, which may only be within the tests' margin of error. So why are heatspreaders so popular? Because they look good and make the user think he's got something high performance. It's essentially a "Type R" sticker for memory.
  • edited March 2006
    The only thing I can think of where sinking your ram actually makes an actual difference is with the video ram on today's high end video cards because that ram actually does generate a significant amount of heat.
Sign In or Register to comment.