Load Temps at Stock : Q6600 G0 w/ Tuniq

Markyip1

[H]ard|Gawd
Joined
Feb 12, 2007
Messages
1,257
Hi everyone,

I'm currently running an 8 hour Prime test at stock on a G0 Q6600 (batch L734A, VID 1.275) in the system represented in my sig.

The HSF is a Tuniq Tower, re-seated several times (as I was testing several chips over the last week.) If tightened too tightly, the thumb cap on one of the screws pops-off, which tells me one of the screws is slightly warped and thus perhaps one corner of the HSF isn't getting as much pressure as the others. However, I believe I've gotten a decent mount at the moment. The heat sink is flush with the IHS and can not be lifted or moved if pulled. Applying moderate pressure with one's hand over the heat sink top does not drop temperatures, as it has with previous poor mountings. Lastly, AS5 is applied as per Arctic Silver's instructions (thin line down the middle.)

With all that said, however, I don't believe I'm getting good temperatures. At ambient (which is 24c) Cores 1 and 2 are at 61-62c load (Prime 95, small FFT going on hour 5) while Cores 3 and 4 are at 50c. This is all at stock settings (2.4ghz, 1.275 vCore in BIOS, 1.208 vCore load with droop.) The > 10c difference between cores at load is of some concern... I wonder if the AS5 didn't cure correctly? At idle, Cores 1 and 2 are at 30c while 3 and 4 are at 25c, given approximately the same ambient temp (likely minus 1 or 2 degrees.)

My primary question is whether or not these seem like reasonable temps for the Tuniq, given 24c ambient and all of the other components in my rig. The follow up question, if the temperatures are not reasonable, would be given a proper physical mount, is poor AS5 likely the culprit or is it something else -- (perhaps lapping is needed?) Lastly, I had always planned to move to water with a quad, in hopes of a 4ghz OC... However, if I can't get to 3.6ghz on air with my current chip, due to temps and high voltage, I wonder what value water will bring if the chip's characteristics prohibit 4ghz. Should I first buy a TRUE 120 Extreme in hopes of getting to 3.6ghz? I know the Tuniq isn't as good as the TRUE with Quads, but what temp drop should I expect by upgrading? Perhaps more important, is it even worth trying a TRUE, or should I just goto water, and then consider replacing the chip if it turns out to be a poor over clocker (over the last two weeks this chip has both hinted at being a poor OCer and has shown promising potential... the 680i motherboard makes things interesting / difficult to determine.)

Many thanks for any help given.

Mark.
 
your load temps are fine around 60C.

your bigger problem is why 2 cores are at 60C and 2 cores at 50C under prime 95 load.

open up speedfan and run prime95 (v25.5). does speedfan detect all 4 of your cores? are they all loaded up with green and cpu is at 100% usage?

all your cores should be within 0-5C of each other under prime95 load (on all 4 cores)
 
My load temps (with system as detailed in sig and ambient temp ~ 20 C) are 66-60-60-66...and when I was installing the backplate, one of its pins just wouldn't stay in the hole and would keep popping out, so I just went ahead and screwed the thing in and don't even think I screwed it evenly...that said, your temps are way too high given your stock settings.

Also, are you using coretemp?
 
I use both CoreTemp and Speedfan (with a +15c offset) and they both report the same temperatures on average. All cores are showing 100% load during Prime95.

I will try remounting the Tuniq once again, but with the broken pin I'm not sure I'll be able to get the mount that I need. We shall see though.

My plan with this chip was always to eventually go water cooling. If worst comes to worst, I will simply jump on that plan now. I don't think moving to a TRUE in the interim is worth it.

Mark.
 
60c @ stock is terrible. I get less than that with a b3 @ 3.2. Something is wrong for sure. I have lapped my tuniq and q6600, but even before that I was getting only about 10c hotter than the lapped temps.

Did you follow the directions on the arctic silver site?
 
I did. My guess is that the one bad screw is causing the higher temps. While the Tuniq is completely secure to the chip on the motherboard, the springs may not be putting as much pressure on the contact as need be. It also may be that the AS5 didn't spread well. I will check once more, but if the Tuniq is truly bad, I think I'll switch to water cooling as planned.

Mark.
 
Didn't the Tuniq come with two extra holddown screws? I think they are labeled as being for AMD installs but I am pretty sure they are the exact same bolts and springs as used for the Intel LGA775 install.
 
Hmm.... Last time I opened the box I don't think I saw them, but I will look again. Having two extra screws would certainly solve this dilemma.

Mark.
 
Back
Top