View Full Version : Have you seen an increase in OC by removing heatspreader from ram ?
nOx34
11-18-2004, 08:47 PM
Just wondering if i should do it to my EBs .
Thanks :toast:
conrad.maranan
11-18-2004, 08:49 PM
I'd like to know this as well before I void my warranty. Oh wait, I voided it the moment I pushed 3.35V through them. Nevermind. :(
Yeah but how do they know that? Just like with an AMD Retail it voids it when you use another Heatsink...but how in the hell do they know that...am I right?
conrad.maranan
11-18-2004, 09:13 PM
I'm a man of integrity. Plus, I believe in karma. If I return a component after I void the warranty, the replacement part will be a dud. :D
reject
11-19-2004, 06:49 AM
hmm but if your ocing, and something dies, but it was well cooled etc. you have a right to a new one.
take off your heatspreaders and put on a couple of 60mm fans. mine are on tho cause im lazy and have a 60mm fan on them. i8 should probly take em off but HEY!
xgman
11-19-2004, 07:51 AM
I really don't think it makes a big difference either way.
texuspete00
11-19-2004, 08:00 AM
Not really sure. I only needed to look at those blankets before I decided not to go over 3v until I took the things off. IMO... heatspreaders suck. If I was worried about it I'd see if I could sink em with something that well.... you know... has fins and isn't attached with thick foam tape.
Here's the kicker..... those stupid blankets dont even usually make full contact with the chips! Even if they were effective, the manufacturers seem to want to make them bling and little else. On the topic opf results though, it usually goes to show cooling ram is mostly for piece of mind. I feel better with no heatspreaders.
mcnbns
11-19-2004, 08:06 AM
The heat spreaders on my EB were disappointing. They are single-sided 256MB sticks, so the sides without chips on them just had a little thing of FOAM. Yes, we are talking about mouse pad foam here. The other side on the chips had some cheap thermal tape stuff that is really messy. I used a razor blade to scrape off the goo and clean them up.
I run 3.2V through my RAM and I have an 80mm fan at the top of my case pointed down at them, blowing across their entire lengths. They are stone cold. Before with a fan and the heat spreaders on, they would be warm with only 2.9V.
To answer your question, though, removing the heat spreaders didn't give me any noticeable gain in overclockability. It did, however, make them a lot cooler.
ToastedToad
11-19-2004, 10:15 AM
I gained an extra 5 MHz after stripping down my PQI TCCDs and puttinn a fan on em. Kept getinng errors after running memtest for a few hours, so I figured it was heat related and now they're runnin clean as a whistle.
Skeet
11-19-2004, 10:45 AM
Heatspreaders are pure evil..low height heatsinks on ram chips+good thermal paste would be more effective would be cool if some manufacturers used that, but it would probably be a lot more expensive than sloppy heatspreaders they use now why they don't do it. :)
I gained an extra 5 MHz after stripping down my PQI TCCDs and puttinn a fan on em. Kept getinng errors after running memtest for a few hours, so I figured it was heat related and now they're runnin clean as a whistle.
Interesting, I have the same ram but they have the heatspreaders on. I have a thermaltake active memory cooling kit gathering dust I could put to use on them though :stick:
conrad.maranan
11-19-2004, 12:45 PM
Okay. I just pulled the spreaders off of my OCZs and I made an interesting discovery. On one module, the chips display 431 on both sides. On another module, the chips display 431 on one side and 434 on the other. I'm currently testing to see if these are truly a matched pair, despite the fact that there are differences in the markings of the chips. I'll edit this post after a quick test.
EDIT:
The module with 431 on one side and 434 on the other seems to clock 5MHz higher than the one with 431 on all chips. I'll try to push it up another 2MHz to 5MHz and see if it's a really bad mismatch.
As far as the benefits of pulling off the heat spreader are concerned, I'm testing that as well.
conrad.maranan
11-19-2004, 01:31 PM
I just finished testing my "good" module by increasing the frequency 1MHz over my best stable setting. Super PI 32M failed at the 23rd iteration.
Unless you plan on installing a set of heatsinks on your memory modules, there is absolutely no reason to remove the factory-installed heatspreaders from your RAM. In the test that I just conducted, I gained nothing other than the satisfaction of knowing which module was the better performer of the two.
Keep the heatspreaders on. They look pretty. Plus, you keep the warranty. ;)
nOx34
11-19-2004, 02:10 PM
I just finished testing my "good" module by increasing the frequency 1MHz over my best stable setting. Super PI 32M failed at the 23rd iteration.
Unless you plan on installing a set of heatsinks on your memory modules, there is absolutely no reason to remove the factory-installed heatspreaders from your RAM. In the test that I just conducted, I gained nothing other than the satisfaction of knowing which module was the better performer of the two.
Keep the heatspreaders on. They look pretty. Plus, you keep the warranty. ;)
Thank you very much my friend :D !
Now that i know this , i wont do it ... Unless i get some heatsinks i wont ..