I've got a 5000+ BE that's overclocked to 3.2 ghz by upping the multiplier and CPU voltage to 1.4 V. However, I can't get the system to remain stable at that speed. The board is the K9A2 Platinum. At max load at the default 2.6 ghz, the CPU is around 56º C. My question is if I need better cooling to remain stable at 3.2 ghz, or if it's another problem (I hope it's cooling!).

Thank you very much!

Well, you didn't mention your temps while OC'd nor did you mention what your highest stable OC is. I think those bits of information would help get you a better response.

Ahh sorry, I'm increasing the clock speeds in 100 mhz increments and running the stability test right now and I'll post again when I see what my highest stable overclock is along with temps.

I'm at 2.9 right now and I'm gonna try 3.0 in a bit, but the system was still unstable at 3.1 ghz

The max I can get stable is 3.0 ghz and it's at 50º C max load

I have the same issue, highest stable OC is 3.0Ghz and 41C. 2.8, 3.1 and 3.2(1.4V) are unstable.
K9A Platinum

Ok, I just saw that tom's hardware got it to 3.2 ghz stable but I'm happy with 3 ghz. Thanks for all your help.

One more note, at 3.2 ghz, the CPU was 60º C at full load

Could be the CPU,MB,PSU or a combo...but people with DFI and PC Power & Cooling constantly get higher OC's no matter the brand of CPU.

The DFI systems in my profile are 50% OC's on air with all stock volts.

With Asus MB's 3.2-3.3 is about all you can hope for from your $79 CPU and that is if your lucky. The 6000+ with full 1MB L2 per core ($100-111) pushs a bit better than the 5000+ BE...some DFI owners have managed 3.4-3.5 Ghz.

The truth is if you can run it at 3 Ghz (dual core) then you can't tell it from 3K Ghz while playing a game and Intel/AMD does not really matter.

I have no idea what case you have. I like the "old" Coolermaster Stacker TO-1 that they don't sale any longer myself...still the best cooling case.

Anyhow if the case airflow is not the issue then get a Thermalright and use Silverstone FM-121 fans (110 q.f.p.m. @ full speed) on it and the case.

You could also -remove- the exit fan grill on the case with a tool called a nibbler which will result in up to 80% more airflow with the same fan you have now.
I know that does not sound right but the reason for this is even if the old grill has LARGE holes it still suffers from what is known as the airdam effect.

I have the Antec Sonata Designer with a 120 mm fan in the back and 2 92 mm fans in the front all on high. I was thinking of getting the ZALMAN 9500A CPU cooler on newegg even if it wouldn't allow me to overclock beyond 3 ghz

Also, I was able to run the Everest Stability test for 2 minutes at 3.1 ghz with the CPU voltage set at 1.499 V before it hit 75ºC and I shut down the computer. Before at 1.4~ V at 3.1, the test crashed immediately.

I personally run a 5000+ BE at 3.5Ghz stable at 49 C. I am watercooled, but you should get it stable at 3.2Ghz if you are around 50C with good air cooling

Just put the multiplier to reach at least 3.0GHz (whitout changing any voltage or fsb) and after, do your normal overclocking.

By the way... did you burn-in before overclocking? i passed from 3.3 to 3.5 only after a 3 days burn-in process.

Ha, and i've got a K9A2 platinum Mobo as well. Hope this can be usefull

I was running at 3.1 on my 5000+ for a while (granted I was using the Gigabyte MB in my sig) then bought the 9850, just make certain that you have sufficent cooling and airflow (had the 5000+ to 3.4 but it started crashing due to heat). Then bought a Nirvana cooler (had a cheap AC HSF) it lowered my temps about 10* C. Now I'm using a Noctua NH-U12P http://www.frozencpu.com/products/ [...] _604_.html
replaced the lo-flo fan (used it elsewhere, rear of the case) used the S-Flex fan that was on the back. (have an Antec 900, the top fan helps remove more heat)

Proof??

I personally run a 5000+ BE at 3.5Ghz stable at 49 C