2 cores at 70C when Prime95 after OC... that ok?

Q9450.

Original thread before OC'ing: http://www.tomshardware.com/forum/250013-28-q9450

I OC'd to 3.0, I believe it to be stable. BIOS CPU voltage: 1.175V

After 3 min of a "Torture Test" with Prime95 my last two cores hit 70-71C (3rd core had been there for about 1.5 min already).

I stopped that test, still have it at 3.0ghz (stock 2.66, 375 fsb on the 8x mult)

I'm basically wondering if it would be possible to run this at a lower voltage and see if it will lower my temperature.


So my main question: Is there a way to know what my min voltage can be running this without actually using trial and error?

If it matters: 4gb 1000mhz DDR2 RAM (Auto put it at 900 mhz in BIOS after upping FSB (from 800) figured I'd leave it?)
I'd like to get it at least to 1000 since thats what I payed for (timings 5-5-5-15)

My cooler is great (supposedly)
Xigmatek HDT-S1283

The problem is the first two cores are within reason (I guess?)
and the second two cores are much higher.

I think some Penryns have some thermal sensor issues. My new e7200 will not go under 52c no matter what. I've even tried different coolers, and different thermal pastes. When temps hit 73 74c in prime she throttles back to one core. Currently running artic freezer 7 with artic silver 5 60c idle at 3.5. I bought a OEM and it said it was M0 stepping and I recieced a C0 stepping.

From reviews that I've seen the M0 works like its supposed to and the C0 runs hotttt! My next one will be a retail CPU, hopefully it will be a M0.

If temperature is causing it, increasing voltage will make it hotter.
Intel will not help you once you tell them you oced it.

Did you try to reset that cooler?

"OEM form Ebay" Probably its limited. Most likely someone else tried it and it just dont OC well, and OEMs are hard to send back unless defective. And as long as it runs stock its not defective. So someone probably put it on Ebay as last resort. If a better cooler doesnt hlp my temps, mine will be on Ebay as well.

Use RealTemp for get the temp on your 45nm Core 2... http://www.xtremesystems.org/forum [...] p?t=179044

Cheers.

It's better to be safe, lol...

It depends on the CPU you have. If you have a 65nm Core 2 you don't need real temp. Basically what we're looking at is Core Temp's incorrect TjMax for the 45nm Core 2's. Core Temp reports TjMax for the 65nm processors @ 105C! Likely bogus. A TjMax of 85-95C is likely more accurate for the 45nm chips.

To be honest, its not the temps that Core Temp, Everest, Real Temp, etc. are reporting that you should really be concerned with. Its the value of the DTS sensor...delta to TjMax (how far you have before the processors throttles). I'd say as long as you have at least 25C before your processor throttles (under load) all's good.