Not air cooling. I'm using a Vigor Monsoon II TEC Cooler
Its a peltier built into a Heat Pipe heat sink. It's pretty cool.
Cools about the same as mid range water cooling setup I suppose.

45C under a peltier ???

I'm using a Thermalright XP120 with a panaflo medium speed.

LOL......I never said it was a good peltier

Thats what core temp reports on the actual cores...PC Probe reports 42c from the board sensors.

2.95Ghz@1.39v is my current -*test*-('")ing status.

That's great !

Almost stock. Wow, if I can run it this way I'll be veeery happy.

I seem to have hit a lil hump at 3.0Ghz though...having to increase voltage quite a bit now to settle it down again.
3.0GHz@1.45v so far...17 minutes into orthos. Anything less failed quickly.

Anything below 1.5v is great.

1,5v seems to be the transition point from air cooling to water cooling.

Wanna-
Sounds like mine. 3006MHz was the physical limit for a stable system regardless of voltage. It would post all the way up to 3150MHz, I believe, and run Orthos for a few minutes at 3.1Ghz but 3006MHz was the stable limit. The nice thing about that is you can run it 24/7 at 3006MHz since the voltage is low and the temps are cool. Welcome to the 3.0 club mate.

Well I wouldnt call me done just yet......this Monsoon cooler is just not cutting it.
It might be ok for a single core cpu.....but it just isnt hanging with the DC Opty.
I can run 3GHz but I'm not to happy with the Volts or Temps at this point in the game.
1.5 made it Orthos stable, but it took almost 1.6 to get it to run 3dmark 06 all the way through.
I guess I.m gonna call the monsoon a wash and go ahead and get the Swiftech kit after Xmas.
Although I'm pretty stuck with the Monsoon as I broke off all the blue LED's from the fan.
So it wouldnt be right to RMA it as I have altered it. It still functions correctly...I just altered the appearance.

Anybody wanna buy a CHEAP Vigor Monsoon II TEC CPU Cooler?
It's more then fine for the average hobbiest overclocker.
Great for Single Core CPU or even a Dual Core with a Mild OverClock.
$130 new at NewEgg.....asking best offer at or near $100 shipped in US.

What temps are you seeing in Orthos at 1.5v, 3006MHz?

PCProbe was showing me mid 40's, but Coretemp was showing me mid to high 50's.
As soon as coretemp hits 59 I get failures. Isnt worth going any higher to me.
I will get the good watercooling setup and then Im sure my worries will be over.
This TEC just isnt powerful enough to keep the Dual Core cool at those speeds.

The heat pipe cooler just gets so hot it cant keep up....
I think at that point the TEC shuts down for a few seconds to allow it to cool.
Which of course leads to instant failure in orthos. Adding voltage at this point just makes it worse.
The water cooling with a nice big radiator will be much more suited to transfering all that heat away.

I have since bumped back down to 2.7 9x300, and with the help of billb, I am currently exploring the potential of my new ram.
Currently running 300MHZ 1-1 6-4-3-2.5 1t 2.7v Orthos stable for 1 hour so far.
I have hopes it may run 1-1 at 333MHz once I get everything else settled LOL

It's a little cool here tonight so I cracked the window open 8 inches and ran Orthos. Check out these temps

It's the same core temps that I have with my 4400+ with more than 1.5v.

Opteron seems to be a real 89W.

The COre temp is the one I watch. 38c after 20 minutes. The cold, outside air was flowing right down over the rad though so it was really kind of a fixed -*test*-('").

When I see you reaching 50+ °C using a watercooler I really think about my XP120 and how amazing Thermalright products are.

A watercooler should perform better. We'll see with the opteron.

I really think folks take the CoreTemp to seriously. If you read the article written by the guy who designed the CoreTemp software, he says as much himself. It really is very good reading and explains where the temps come from and how much credence you should pay those readings. http://www.overclockers.com/articles1378/

AMD uses the TcaseMax which is the maximum temperature at the dead-center at the top of the processor core or the IHS which will ensure 100% stability and safety. Exceeding it isn't a big deal in most cases. Intel reports TJunctionMax which is the maximum temperature at the junction between the processor die and the PCB it sits on, this is usually much higher than the TCaseMax value. Exceeding this is a big deal and really should be seen as a meltdown warning. Wanna's is the first I have seen of a CPU crapping out at TcaseMax.

MY THEORY ON CORETEMP AND COOLING SOLUTIONS:
The other thing I believe, but am not sure of, is that since the DTS (Digital Themal Sensor) is located in the core itself and below the IHS, the temps are real time. The CPU Temp reading we all see is attached to the IHS. Therefore, I don't think that you will see a huge delta in Coretemp readings between cooling solutions until the cooler itself reaches heat saturation.

The air cooling and watercooling is attached to the IHS and therfore cools it only. It is only as good as its absorption rate. In other words, air and water both pull heat from the IHS and there won't be much difference between them at the coretemp per se. The CoreTemp will rise as work, frequency and voktage increase but as long as you can pull the heat away at a certain rate it will pretty much stay there.

What water buys you is a much higher saturation point. It can absorb and disperse a lot more heat than an air cooler. It will only effect the coretemp from that perspective. It will continue to pull heat from the IHS long after the Air Coller has reached saturation. Once you hit saturation CoreTemp rises rapidly. It is no differnt than a radiator on a car or motorcycle.

What I believe is that there is a tip over point where the cooling becomes overloaded and then you will see a big delta in cooling solutions. Both air and water will let the coretemp rise to a certain point and not exceed that point until saturation. Air will maintain about the same coretemp as water until it hits its saturation point much earlier and that is where water buys you the edge. It hits saturation much later.

All theory but I'd be interested in others folk's opinions. The only way to truly see this is with Phase change which actually cools the whole CPU to minus temperatures. What is the CoreTemp there?