The Official OC (OverClocking) Thread!

Just be sure to buy the original, and not the Extreme version, as it's rubbish

 

Currently running 965 at 3.8Ghz. Feels pretty good. Ram is at 1648Mhz. Am noticing an 8 - 15% increase in Dvd shrink analyzation times, compared to 3.5Ghz 940. Windows 7 is at least a little more responsive. Whether that's due to Mobo or CPU... perhaps time will tell. I can't run the other board yet, til I get my PROPER SSD! LOL! I'll get some other goodies installed and see how she does. E.g. Rebuilder.

That Tuniq 120 is sexy estuansis I'm strongly considering it. I also had the idea of modding my current sink. Not sure about that one though LOL!

 

The Tuniq 120 is every bit as good as the Thermalright Ultra and the Noctua. Not to mention cheaper than the Ultra when I got it. It cools very well and runs near-silent(under the sound of the rest of the case) at 1300-1400RPM under load. All my fans are set to auto, and everything is cool and quiet.

 

Your sig says your Cooler is Lapped. Was that really necessary? I eyed the copper plate on the arctic freezer(my cooler) and it was relatively smooth. Very slight imperfections. However, I loaded it up with arctic 5. So there should be literally zero air bubbles or space in between cpu and cooler. In fact I think I may have overdone it. It oozed out ever so slightly. Since I'm getting a new cooler, is arctic 5 really my best option? I noticed its a different color than AMD's stock stuff. And my 940 was running slightly cooler on a stock cooler. At least according to the bios. Dang it! I really wanna be running both of them!

 

Thermal paste fills the gaps caused by the uneven surfaces of the CPU cooler and CPU heatspreader, which leave gaps. Thermal paste is always essential, as the smallest gap has a big effect on Delta T, but thermal paste conducts heat very poorly compared to actual copper, thus you want as little need for it as possible, and for the surfaces to be as even as possible. This is why lapping has an effect - it is however a lot of effort to go to, and while it gives better temps, there aren't really any situations where you'll overheat without lapping, but be safe if you do it.

 

I think I'm beginning to understand your desire for a silent PC. Once you have one, anything less is undesirable LOL! I'm noticing while the fans are at low rpm's, the HDD's are actually more audible than I'd like. But I do have the side panel off, and 4 7200rpm drives are far from inaudible

 

Indeed, and it only gets worse from here...

 

so if thermal compound is not a good conductor, would it be smarter to simply apply the thermal compound, and scrape away leaving ONLY the imperfections with the compound? Much like mudding holes in a wall. If copper to CPU is better, this seems like a good idea.

 

That's why I get away with such low temps on everything. Lapping dropped the temps on this chip 3*C. While this may not seem like much a bunch of small touches add up. Several fan grill mods, some quiet Cooler Master fans, new thermal paste on everything in the PC, etc.

The bottom of my first Tuniq was not bad, and the second was like a mirror. But both of them had a curved surface, I would assume warped from soldering the pipes into the cooler(they have since corrected this problem and offer a free replacement for those affected). Also, my Freezer 64 Pro had a perfectly flat bottom but mine arrived all scuffed up. All that needed was some mild abrasive polish. Shone like a mirror

I was having some intermittent temp troubles with this CPU before. If my ambient temp got really warm the CPU temp would increase disproportionately, Sometimes to 55*C. I have since turned all of my case fans from max to auto and the airflow seems more balanced because my CPU runs cooler in all scenarios. Ran quad Orthos for 12 hours last night with the heat blasting and it topped at 45 all night. I couldn't be happier with my CPU right now.

I will also mention that both of my CPUs themselves are lapped too.

 

You lapped your cpu LOL! Didn't that remove the model# and what not? If you resell, surely that drops the value. Just curious here

 

Not too worried about that. Will probably get sold to friends who know it's a good chip. But yes the printing on the IHS is gone. I know for sure that it's a perfectly flat surface though. I also sanded in one direction so the grain of the sanding scratches is uniform. Sanded with 2000 grit BTW.

 

Yah, after seeing your "Lapped" in your signature, I did some light reading on the matter. 2000grit came highly recommended. Though hard to find for some people LOL! I imagine home depot has it, or an auto body shop...

 

My dad does auto body work, so 2000 grit sandpaper is plentiful here

If you do lap be extremely careful not to sand all the way through the IHS. The results aren't pretty. Luckily mine didn't need a whole lot of sanding before it was true. Some are worse than others, but all CPUs can benefit from it.

Also, if you want to get really daring you can remove the IHS completely and go bareback ala the Athlon XP

 

Sam,
When buying a Freezer 7, you need to be sure it's rev.2!
http://www.newegg.com/Product/Product.aspx?Item=N82E16835186134
Even at full speed it's quieter than to old version was at mid speeds. Moves more air at a lower rpm too! The video card makes more noise at 50%, and it's pretty quiet. Same thing goes for the Freezer 64 as well! Same fan!

Russ

 

Forgive my scepticism but the cooler design is almost identical, and the same goes for the fan. The v2 is a v1 with some very minor design alterations. I don't imagine at the same fan speed the v2 performing outside testing accuracy range.

 

Sam,
Be skeptical all you want, but the blade design is different, and it moves more air at a lower speed. Max speed is lower at 2200 rpm vs 2400 rpm, 40 cfm vs 36 cfm. Mine's never been higher than 1700 rpm, at least, not that I know of. Once, I deliberately manually ran it at 2200, and you knew it was running, but a very quiet sound. At 1700 you hear a little more air moving, but that's about it. Much quieter than the previous versions.

I have a fan profile that EasyTune runs at startup, that's fine tuned to my computer. minimum 1100, maximum of 1700, scaled in between. Anything above 1700, the temps start to go up, the penalty for living in a very dry climate, as dry air doesn't collect as much heat from the cooler! I've yet to sustain a CPU temperature of 40C, no matter what I do with it. I've run the new OCCT OCCT for hours and it will barely hit 40C, the fan jumps to 1700 the minute it does, and the temp goes right back down to 38-39C. I expect that will go up come summer, and I may have to create a new profile for 120F/48C temps. It do get hot here!

Russ

EDIT:
Sam, My appologies, that's 47C on the CPU running OCCT with all 4 cores, not 37. The 37 is running Orthos, which only runs 2 cores. Sorry!

JRS

 

I'm currently idling at 44C. 1.325V 3.8Ghz. No, I did not test for stability. I simply wanted to see how cool it runs at stock voltage. Seems kind of warm LOL! And I suspect we're in for a hot summer here. I definitely wanna consider either the Tuniq or the noctua

 

According to everest my freezer 7 pro is running at 2872rpm, and this is at idle. I believe I've "seen" it in the 3k range when I'm actually doing something lol. Makes me wonder how many of the other sensors readings are inaccurate.

 

Interesting, as AFAIK, the Freezer 7 Pro's top speed is 2500RPM. Any sort of sensor readings are notoriously inaccurate, but that one seems extreme. Even temp sensors are accepted as an "in the ballpark" reading.

 

Estuansis,
Yes I see that. 900-2500 rpm 45CFM. The last one I bought a few months ago was 900-2200 rpm and 40 cfm, same as my freezer 64. The one before that was 900-2400 rpm and 36 cfm I wounder why all the changes on the freezer 7? They are essentially the same cooler, with a different mount!

Russ