Asus P5KR: Setting the correct RAM speed for Kingston KHX5800D2K2/2G

[RJARRRPCGP]

Seems to be a normal thing for surprisingly many people to endure a (sometimes rather long) delay before the machine finally starts to boot.

That's normal when you hit the power button after the PSU has been unplugged from the wall.
At least with the Asus P5QL Pro.

My Asus Maximus II Gene seems to have less of a delay, even with the same PSU...

 

[BassTee]

Either I have never noticed that delay before or it really is a new development.

 

[BassTee]

I have another question: I tested my CPU temps (heatsink is an Alpine 11 Plus, a rather unspectacular replacement for Intel's stock cooler) and at maximum load (using Prime95 running the in-place large FFT torture test) core #1 seems to settle at 74°C after about ten minutes, yielding a distance to TjMax of 26. Is this acceptable? Average idle temp (core #1) is around 35°C, all according to Core Temp & RealTemp. That brings me to another thing that confuses me: the TjMax setting in different programs. Everest has it set to 90°, which coincides with Intel's updated specs. Core Temp 1.0 RC5 and RealTemp 3.70 both set the TjMax to 100°, resulting in a difference of 10°. I know that the distance to TjMax is the important thing here but I'm still a bit puzzled.

 

[RJARRRPCGP]

Just to be safe, I assumed the 65nm ones to be 100 C for TJMax.

But, the max core temps are high for stock speed.

 

[BassTee]

What do you mean with stock speed?

I'm going to wipe off the pre-applied mx-4 from both cpu and heatsink next week and apply some mx-2.

 

[RJARRRPCGP]

What do you mean with stock speed?

I'm going to wipe off the pre-applied mx-4 from both cpu and heatsink next week and apply some mx-2.

The stock clock for your chip.

 

[BassTee]

I see.

 

[redduc900]

You can change the TjMax to whatever value you'd like (100, 95, 90, etc.) using Core Temp or Real Temp, but the Distance to TjMax will always stay the same. Just keep the Distance to TjMax >20-25C, and don't worry about the true TjMax value. And TjMax for a 65nm G0 stepping quad is 90C, whereas w/ a B3 stepping it's 100C.

 

[BassTee]

A little update for those who might care:

I just replaced the Alpine 11 Plus with a Cooler Master Hyper TX3 Evo and the Distance to Tj Max now settles around 45 (core #0) under 100% load running Prime's in-place large FFT torture test. Quite a difference to the Alpine, even though this was achieved with Asus Q-Fan control disabled and the fan running at its relatively noisy full speed (~2360 RPM). With Q-Fan's 'Performance' profile enabled, the RPM drops to a maximum of approx. 1655 RPM and the Distance to Tj Max settles around 43 under full load (~65 idle with the fan running at ~960 RPM), which I think is still pretty good. And the noise level is much better. But why Asus' Q-Fan control never drives the RPM to the maximum even under heavy load using the performance profile I cannot fathom.

(TjMax set to 90°, CPU at stock speed, Q-Fan disabled)​

With the Alpine still mounted I tried a very mild OC, pushing the Q6600 to 2800 MHz with the temperatures still in acceptable regions even when playing Tomb Raider Underworld (I didn't want to overdo things ) with a Gigabyte 460 GTX I inherited from a friend. With the TX3 I should be able to push the envelope a bit further and still get acceptable temps.

All that said I resolved to buy a new case and PSU soon which I can keep should I decide to really upgrade my system specs in the future. For the PSU I'm considering the Seasonic M12II-650 and for the case I'm thinking about the Xigmatec Midgard II.