![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
Core temp variances
- Thread starter ans136
- Start date
Yes, it is. Unless the difference in temps become so large (10C+), you have nothing to worry about.
If I were you, I'd make sure that the HSF is seated correctly and evenly (equal pressure to all parts of the processor) -- but even that will not completely get rid of the difference. It will just lower the margin of difference... You still want to use the highest core temp as the guideline.
If I were you, I'd make sure that the HSF is seated correctly and evenly (equal pressure to all parts of the processor) -- but even that will not completely get rid of the difference. It will just lower the margin of difference... You still want to use the highest core temp as the guideline.
- Joined
- Oct 23, 2007
- Messages
- 6,105
Both my i7s and my q9450 as well have a couple cores that run a few degrees hotter than the others; nothing to worry about (unless the difference is large, like Deluded said). Going by the ss, that's not the case, so don't worry about it. 
It is possible to get your core temperatures to within 1c of each other. The main thing that is under your control is the consistent contact between CPU heat spreader and heatsink surface.
Since your using push pins your ability to use even preasure across the heatsink via the retention mechanism is limited. Replacing your retention kit can help or lapping heatsink and cpu can increase planarity. Using the right amount of TIM and orientation of the heatsink to match the core orientation can help as well.
In my case I was seeing 4-5c difference between 2 cores (e.g. 81 75 76 80) when I initially mounted my Cogage True Spirit heatsink on my i7 930. I wasn't very scientific with my mounting (screw a bit here and there on all 4 sides) and used a heavy layer of TIM.
When I reseated it, I removed all the TIM. Applied a thin layer (cover the whole heatsink while still being able to lightly see the CPU label) as well as adding a thin line of TIM along the orientation of the cores (see here and here). Then proceeded to tighten 2 of the 4 bolts of my X-style retention. Going 5 turns on each side till the mechanism was fully screwed in. The new bolt mechanism of the True Spirit is pretty good since you can't move the heatsink after securing it. I also aligned the heatpipes
After this excersize 3 of my cores show the same temperature and the 4th is 1c off.
Since your using push pins your ability to use even preasure across the heatsink via the retention mechanism is limited. Replacing your retention kit can help or lapping heatsink and cpu can increase planarity. Using the right amount of TIM and orientation of the heatsink to match the core orientation can help as well.
In my case I was seeing 4-5c difference between 2 cores (e.g. 81 75 76 80) when I initially mounted my Cogage True Spirit heatsink on my i7 930. I wasn't very scientific with my mounting (screw a bit here and there on all 4 sides) and used a heavy layer of TIM.
When I reseated it, I removed all the TIM. Applied a thin layer (cover the whole heatsink while still being able to lightly see the CPU label) as well as adding a thin line of TIM along the orientation of the cores (see here and here). Then proceeded to tighten 2 of the 4 bolts of my X-style retention. Going 5 turns on each side till the mechanism was fully screwed in. The new bolt mechanism of the True Spirit is pretty good since you can't move the heatsink after securing it. I also aligned the heatpipes
After this excersize 3 of my cores show the same temperature and the 4th is 1c off.
I lapped a hyper 212+ although most people will tell you not to. Some of the cores were still not exactly even. It is almost impossible to get all the cores exactly the same. I wouldn't worry about it really. Your temps seem high for only a 3.3ghz overclock with a push/pull config.
I am not sure what could possibly be causing the temperatures to be this high. I was hoping to be sitting below 60c under load. I re-applied the thermal paste according to the core positions as well as to the cooler to fill in the heat pipe gaps. I have two 60cfm fans on it and they are running full speed. I felt the copper heat pipes at the top of the cooler and they arent that warm under load...could it be a seating issue?
BillParrish
Supreme [H]ardness
- Joined
- Aug 25, 2006
- Messages
- 7,519
Not a realistic expectation running IBT. Your temps are pretty good as they are with that heatsink according to the posts I have read (and I read them all).
I have had excellent results mounting the heatsink so the fans blow UP into the big top fan and flipping the rear exhaust fan around so it blows into the case. This takes advantage of the basic rule "heat rises" and injects cool room air directly at the heatsink. If you video card(s) do not exhaust heat outside the case it may not be as effective but I think yours does. YMMV but worth a try.
Its the luck of the draw on how close the core temps track. as mentioned in a post above 5C is not bad and it appears you have tired remounting etc. to eliminate the possibly of a "not quite right" heatsink mount, I would not worry about it.
Just some general comments, I dont want to discourage you shooting for better but offhand other than trying the different heatsink orientation I dont know what else you could easily do other than sfsilicon's lapping suggestion which I agree with completely. After you are sure everything is working OK, and it appears it is , a session with some wet and dry sandpaper and glass from a picture frame would be the next step. How much you lapp (shiny polished surface) is not nearly as important as making sure everything is flat. The heatsink with its exposed pipes needs just a touch up, you dont want to take much off because of the exposed pipes but the cpu should be taken down to as flat as possbile. At least check what is going on with a razor blade, all kinds of guides on the net and U-tube. It is easy and safe if you pay attention to what you are doing and take reasonable care in handling and cleaning after you are done.
Last edited:
sirmonkey1985
[H]ard|DCer of the Month - July 2010
- Joined
- Sep 13, 2008
- Messages
- 22,412
completely agree with the above suggestion for the fans on the antec 900.. i run my antec 300 with the rear fan blowing in.. has 2 bonuses 1 pulling in cool air and 2 moving cool air directly over the mosfets/pwm's instead of warm air from the cpu heatsink being pulled over them..
Thank you for your insightful response. I already have it mounted in the orientation you mentioned but, I will flip the rear exhaust to intake and see how that helps. I also just purchased some OCZ freeze to replace the stock thermal paste. Hopefully all this helps.
Just a comparison shot for you. I'm running with a TRUE 120 Rev.C and AS5 in a Lancool PC-K62. Nothing is lapped. I tried to evenly tighten down the screws on the heatsink as I installed it.
Right now at idle there's a 4°C difference in core temps. It seems to even out a little more under load.
Right now at idle there's a 4°C difference in core temps. It seems to even out a little more under load.
Bill, any chance we could get a temperature FAQ stickied in here? It would be a lot easier to refer people there rather than answer the same questions over and over again.
Just the CPU. Then mount the heatsink.
Following up with the contents of this thread. A 9°-10°C difference between cores on my i7-920 would be borderline?
I'm running at 4mhz with a vcore of 1.24.
High temps are:
74
72
70
66
While doing the torture test in prime95. About to run 3D mark vantage and see how it does.
I'm running at 4mhz with a vcore of 1.24.
High temps are:
74
72
70
66
While doing the torture test in prime95. About to run 3D mark vantage and see how it does.
Any temperature differences are insignificant. Just take a look at the highest temperature across all cores and make sure it's low enough.