i'm scared about my q6600 temps.

J

jjolin

Gawd
Joined
Dec 21, 2005
Messages
813
i overclocked my q6600 today to 3.2 ghz and im getting 68c on core temp and 63c on real temp, and i dont know which one is accurate. if core temp is accurate, i should drop my clock some, but if real temp is accurate, im fine. which one should i believe?
 
get yourself a true120/IC diamond compound and then not worry? @ 3.2 i was at never went over 40*c I do have a insane fan set up on it though...
 
Even 68C is fine if its stable, just leave it like that and you still have 32C to TjMax. Usually overclocking limits are from chip limitations and not heat with the current Core 2 lineup.
 
Vcore is 1.375. ill probibly order a TRUE to replace this xigamatek pos. i have plenty of airflow as well. antec twelve hundred case.
 
jjolin said:
s1283, push pins. yeah i know i should get the bolt through.
Click to expand...

its fairly amusing that you said "xigmatek pos" then, given thats one of the few coolers arguably better than a true 120 :p

a bolt through will not drop your temps much presuming you got the push pins locked down, though it will give you a much more secure mount. overall i do not think the temp is really out of the ordinary unless you got a wind tunnel in your case, then its much too high. also with people running vastly different ambient temps with the summer here, and the lack of all using the same tool to measure temps with, do not worry too much about what other people post for temps. some people "calibrate" their own tjmax values as well, then give out temp information as fact.

if you have a small delta between the core temps, thats a good indication your cooler is mounted well.
 
Yeah thats about what I was getting before I got my TRUE. I would get a different heat sink if you are going to overclock 3.2 or higher, you might back it down to 3.0 till you get a better heat sink IMO.
 
Zardoz said:
Yeah thats about what I was getting before I got my TRUE. I would get a different heat sink if you are going to overclock 3.2 or higher, you might back it down to 3.0 till you get a better heat sink IMO.
Click to expand...

You realize he already has a heatsink with almost the exact same performance as a TRUE right? Or did you skip reading most of this short thread.
 
ocellaris said:
You realize he already has a heatsink with almost the exact same performance as a TRUE right? Or did you skip reading most of this short thread.
Click to expand...

oh, so he does, dam, that's not good, he has a seating issue or bad case temps (ait flow) for sure.
 
Is it a G0 or a B3 stepping? From what I recall, the older B3 simple don't overclock as good as the G0. B3 usually does 3,0 to 3,2 GHz, G0 can manage 3,4 to above 3,6 GHz at a similar temperature.
 
Zardoz said:
he has a seating issue or bad case temps (air flow) for sure.
Click to expand...

or bad compound (brand, thickness, ...) or a combination of all the above.

I had temps of around 65C+ when one corner pin popped out of its hole. Temps dropped to normal when the pin was reset.
 
i just did a stress test with 3.24 & 3.3 GHZ on my q6700 and never went over 41*C with the true 120... idk how yours could be 24* more even with that cooler.. idk maybe its possible based on those tests if you are really pumping out some heat...

switch to a TR fastener type if you are running push pins.

I use shims on mine to increase seat pressure (increases thermal conductivity of the compound joint) and a shim up the top 2 spring oaded shoulder screws more than the bottom so that when the case is standing the very large and heavy cooler remains flat on the cpu.
 
r0b0tc0rpse said:
i just did a stress test with 3.24 & 3.3 GHZ on my q6700 and never went over 41*C with the true 120... idk how yours could be 24* more even with that cooler.. idk maybe its possible based on those tests if you are really pumping out some heat...
Click to expand...

I would love to see the full details of where that came from, since 41C is ridiculously low for an overclocked Q6600 at a real 100% load (Prime95 small FFTs usually works).
 
ocellaris said:
I would love to see the full details of where that came from, since 41C is ridiculously low for an overclocked Q6600 at a real 100% load (Prime95 small FFTs usually works).
Click to expand...
No kidding....My watercooling gives me 45 loaded......I was closer to 52 on a TRUE
 
ocellaris said:
I would love to see the full details of where that came from, since 41C is ridiculously low for an overclocked Q6600 at a real 100% load (Prime95 small FFTs usually works).
Click to expand...

It's not unrealistic, with the s1283 my temps reached 44c in realtemp at 3.2ghz.
8 hour prime95 small fft test. Great room temperature here too.
 
G0 [email protected] 1.41 vcore (after vdroop) with TRUE 120 running 100% on all cores with an ambient 80F I have 60C as the hottest temp ton the hottest core.

G0 [email protected] 1.44 vcore (after vdroop) with Tuniq Tower 120 running 100% on all cores with an ambient 80F I have 65 as the hottest temp on the hottest core.

G0 [email protected] 1.45 vcore (after vdroop) with TR Ultima90 w/120mm fan running 100% on all cores with an ambient 80F I have 63 as the hottest temp on the hottest core.

Thermal compound is either Arctic Silver Ceramique or Arctic Cooling MX-2.

Either you have very high ambient temps, the airflow in your case is not good, the heatsink is not mounted very well or the heatspreader/base of the heatsink is not flat. I did not have to do anything to the Tuniq Tower or the Ultima90 to get the temps I have. The TRUE 120 was a totally different story. It had an extremely concave base and needed a lapping job very badly as it was getting worse temps than either of the other two heatinks. Once I lapped it, the temps dropped dramatically. I also lapped the Q6600 at the time but the heatspreader on it was already flat.

You may want to look into those things. Also, I hate the pushpin mounting system and it's probably one of the worst things I've seen Intel design other than the i820 chipset with the memory translator hub. I refuse to purchase a heatsink for a socket 775 board which uses the pushpins and instead look for something with a backplate mounting bracket. Sure, it can be a pain in the ass but at least I know I'm going to get a good and stable mount out of it.

 
tetete said:
over 60 is not acceptable to me

I will got keep them within 50ish
Click to expand...
I honestly don't understand what the big deal is....The thermal spec is 71, so why is 60 "not acceptable"

I have one folding box with a quad buried in my closet, that thing has been at about 68 for 9 months..still churning away without a complaint.
 
Back to OP's original question... The difference is RealTemp assumes a Tmax of 95, CoreTemp of 100. They will always have a 5 degree offset as a result. I believe 100 is "more correct" for the Q6600, but don't quote me on that.

 
Take off the heatsink, clean off the thermal paste from both it and the CPU, and then lap them if you can. When you reapply the thermal paste, try to spread it out evenly with a credit card or something, and I hope you're using something fairly decent, like AS5 or MX-2.

Is there any way to use screws instead of push pins on your heatsink? At the least, is there a way for you to put something on the back of your mobo so that the push pins go down further (and thus put more pressure on your CPU for better contact)?
 
I dont mean to threadjack but I didn't want to create another thread for the same exact issue so I thought I would post here.

I seem to be having the same issues as the OP, running a G0 SLACR Q6600 @ 3.2ghz (400x8) with about 1.360ish vcore after vdroop. I get around 46-50C on two cores and 40-45C on the other two cores IDLE. On load the highest I've seen it get is around 70C! Its using the same xigmatech with bolt thru kit. The case is a P182 with intake in the middle bay and PSU area, and exhaust at the top above the cpu and rear, all running medium speed yateloons. The xigmatech is blowing thru the heatsink toward the rear. My ambient is around 80F, so I'm not sure why it could be running so hot. I've remounted the heatsink many times with AS5 and have tried different ways of applying it from guides and it hardly changes the temp, maybe 1-2C. The only thing that could be affecting it is that I saw I have a pretty high vid of 1.325. Would lapping be my next course of action?

Sorry again, if I need to make another thread I'll do it if needed.
 
slidez said:
I dont mean to threadjack but I didn't want to create another thread for the same exact issue so I thought I would post here.

I seem to be having the same issues as the OP, running a G0 SLACR Q6600 @ 3.2ghz (400x8) with about 1.360ish vcore after vdroop. I get around 46-50C on two cores and 40-45C on the other two cores IDLE. On load the highest I've seen it get is around 70C! Its using the same xigmatech with bolt thru kit. The case is a P182 with intake in the middle bay and PSU area, and exhaust at the top above the cpu and rear, all running medium speed yateloons. The xigmatech is blowing thru the heatsink toward the rear. My ambient is around 80F, so I'm not sure why it could be running so hot. I've remounted the heatsink many times with AS5 and have tried different ways of applying it from guides and it hardly changes the temp, maybe 1-2C. The only thing that could be affecting it is that I saw I have a pretty high vid of 1.325. Would lapping be my next course of action?

Sorry again, if I need to make another thread I'll do it if needed.
Click to expand...

It sounds like you might be able to take care of yours by lapping. I noticed with my TRUE 120 that the temps on a couple of the cores were a lot higher than others. That is a sign that the heatsink or the heatspreader needs to be lapped to produce a good mating surface.

 
SmokeRngs said:
It sounds like you might be able to take care of yours by lapping. I noticed with my TRUE 120 that the temps on a couple of the cores were a lot higher than others. That is a sign that the heatsink or the heatspreader needs to be lapped to produce a good mating surface.

Click to expand...

5C variation in core temps is not "a lot higher"
 
Well I lapped my CPU and HSF last night and it didn't seem to change temps that much, it did make the cores slightly closer in temp to each other than previously. This makes me believe that I might have a mounting problem with my Xigmatek. I have the bolt-thru mounting kit so it should be okay.

When I screw the heatsink onto the cpu, I just turn all the screws until they stop, going diagonally with each screw correct? It just seems odd to me because it doesn't feel like its very tight with the cpu as compared to my old AMD system with the spring clip...

Also i notice when im priming it and I touch the heat sink, it feels slightly warm but not hot at all, does that seem normal? I would assume with temps close to 70C on load the heatsink should feel hot to the touch.
 
slidez said:
Well I lapped my CPU and HSF last night and it didn't seem to change temps that much, it did make the cores slightly closer in temp to each other than previously. This makes me believe that I might have a mounting problem with my Xigmatek. I have the bolt-thru mounting kit so it should be okay.

When I screw the heatsink onto the cpu, I just turn all the screws until they stop, going diagonally with each screw correct? It just seems odd to me because it doesn't feel like its very tight with the cpu as compared to my old AMD system with the spring clip...

Also i notice when im priming it and I touch the heat sink, it feels slightly warm but not hot at all, does that seem normal? I would assume with temps close to 70C on load the heatsink should feel hot to the touch.
Click to expand...

Try this... Take the cover/door off your case, blast a house fan right towards the open case side (use something like a box fan), and see what you get for load temps. If that drops your temps 20C, you know its not a CPU cooling problem, then its a case cooling problem. The temp should only drop by a few degrees C.

If the temp is still high and does not drop at all, then it could be a contact issue.
 
So im currently blowing a small fan at my open case and the temps dropped from 68 with AC on to 61C while priming. Its not a huge drop like 20C but could it be a sign i have poor case airflow?

Heatsink feels like room temp and not warm at all...
 
I was concerned about temps on my Q600 as well, it idles right around 50 and loads to low to mid 60s at 3.0GHZ. It's perfectly stable at those temps, but it keeps me from wanting to push it any higher. My case ambient temp is reported at 45, so I think that is the primary problem. I'm using the giant rocketfish case from the BB clearance. Anyone else with that case have some bad airflow issues?
 
darkpaw said:
I was concerned about temps on my Q600 as well, it idles right around 50 and loads to low to mid 60s at 3.0GHZ. It's perfectly stable at those temps, but it keeps me from wanting to push it any higher. My case ambient temp is reported at 45, so I think that is the primary problem. I'm using the giant rocketfish case from the BB clearance. Anyone else with that case have some bad airflow issues?
Click to expand...

Airflow is OK but its definitely not great. One think you can do to help is take the GPU slot exhaust fan and put it over the CPU blowing hot air out the back vents. Another thing is to get a 1200 fan that fits in 3 of the 5.25 bay slots at the front (search for "Kama Bay")
 
I've about given up on trying to keep my temps down, and while running at 70C when warm and 65-68 when cooler while running prime, I doubt ill run into many probs with everyday use. I expect to upgrade the cpu within a year or two anyways. Guess I just got an unlucky chip
 
r0b0tc0rpse said:
get yourself a true120/IC diamond compound and then not worry? @ 3.2 i was at never went over 40*c I do have a insane fan set up on it though...
Click to expand...
LOL I never thought I'd see the day when diamonds are used in thermal paste :p Already ordered.
 
You must log in or register to reply here.
Back
Top