Page 2 of 3 FirstFirst 1 2 3 LastLast
Results 11 to 20 of 23
  1. #11
    ROG Guru: Brown Belt Array HalloweenWeed's Avatar
    Join Date
    Dec 2011
    Reputation
    87
    Posts
    1,466

    Quote Originally Posted by kingsaadi View Post
    Is this temperature normal? using havik 140 cooler. used to have a QX9650 overclocked to 3.8ghz on air at 1.42v and it ran cooler than this.

    If I do OC it in the future what's the safe temperature limit for 24/7 operation on the 3970x?
    Yeah, that's pretty good temps for an air cooler. Our Hexacore CPUs put out a lot of heat, more than the X58 quad-cores. Remember, P95 loads the CPU more than any real-world application you run, unless you are folding or seti@home, or something like that.
    i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
    Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
    Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
    (EVGA site: ) And I have a second (wife's) computer, Eve.

    Overclocking is useless to me if it is not rock stable.

  2. #12
    ROG Member Array
    Join Date
    Jan 2013
    Reputation
    10
    Posts
    16

    Quote Originally Posted by HalloweenWeed View Post
    Yeah, that's pretty good temps for an air cooler. Our Hexacore CPUs put out a lot of heat, more than the X58 quad-cores. Remember, P95 loads the CPU more than any real-world application you run, unless you are folding or seti@home, or something like that.
    Just to be sure I did a few more tests to confirm.

    Played Assassin's creed revelations for about an hour - Core temps at load maxed out between 71-76*C

    Ran Prime95 with ram at it's stated 1866mhz, everything else in default - Core temps soar to over 77-84*C just within a few seconds, so I stop it. If I don't stop it, system restarts less than 30 seconds.

    So I ran the latest memtest 86+ overnight with ram at 1866mhz, 2 passes no errors over 7 hours and 17 mins.

    Ran Prime95 again with EVERYTHING at default (so ram is below stated speed - at 1333mhz, CPU isn't running at it's stated Turbo Boost of 4ghz, instead runs at 3.7ghz) - Core temp still reaches 80*C within 15 mins.

    I've tried reseating the heatsink and reapplying thermal paste. Have tried 2 different methods of applying thermal paste

    1. little bit smaller than pea sized dot at middle of CPU
    2. BB sized dot at middle of CPU + 4 much smaller dots near the corners of CPU (like the 5 dots on a dice). gives much better paste coverage but no real difference in temperature readings.


    I really don't think the CPU should be reaching these kinds of temps running below intel's stated stock settings. And it's not like I'm using a crappy cooler either (unless the heatsink is defective somehow)

    What do you guys think? and What are your suggestions?

    Thanks a lot for your help!
    Last edited by kingsaadi; 01-11-2013 at 11:43 PM.

  3. #13
    ROG Guru: Brown Belt Array HalloweenWeed's Avatar
    Join Date
    Dec 2011
    Reputation
    87
    Posts
    1,466

    Try setting this:
    Extreme Tweaker menu:
    CPU Vcore V: Offset mode
    Offset mode sign: -
    CPU Vcore offset V: 0.040
    You don't have "CPU level up" set, do you?
    Is there any other BIOS CPU setting that is not set default that we do not know about?
    Last edited by HalloweenWeed; 01-11-2013 at 11:49 PM.
    i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
    Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
    Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
    (EVGA site: ) And I have a second (wife's) computer, Eve.

    Overclocking is useless to me if it is not rock stable.

  4. #14
    ROG Member Array
    Join Date
    Jan 2013
    Reputation
    10
    Posts
    16

    Quote Originally Posted by HalloweenWeed View Post
    Try setting this:
    Extreme Tweaker menu:
    CPU Vcore V: Offset mode
    Offset mode sign: -
    CPU Vcore offset V: 0.040
    You don't have "CPU level up" set, do you?
    Is there any other BIOS CPU setting that is not set default that we do not know about?
    No I don't have CPU level up set
    I used to have CPU VCCSA set to 1.2 manually, but I set it to auto now (before and after I tested with Prime95 - both times crashed)
    The ONLY other thing I have set manually is the ram voltage - at 1.5v, and ram timings as from xmp profile.


    I have just tried your setting Vcore to Offset (-) at 0.040v

    Seems to work! Doesn't crash after a few seconds anymore.

    Temperature is VERY HOT though - ranging from 77-85*C, most cores above 80*C.

    Is it supposed to be this hot?? at stock speeds!..

  5. #15
    ROG Guru: Brown Belt Array HalloweenWeed's Avatar
    Join Date
    Dec 2011
    Reputation
    87
    Posts
    1,466

    What is your ambient (room) temperature? Approximately what was the ambient temperature when you reported your CPU temps earlier? The same?
    i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
    Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
    Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
    (EVGA site: ) And I have a second (wife's) computer, Eve.

    Overclocking is useless to me if it is not rock stable.

  6. #16
    ROG Member Array
    Join Date
    Jan 2013
    Reputation
    10
    Posts
    16

    Quote Originally Posted by HalloweenWeed View Post
    What is your ambient (room) temperature? Approximately what was the ambient temperature when you reported your CPU temps earlier? The same?
    I haven't got a thermometer on me at the moment, but I'd say ambient temp is about 26*C now. Temperature was about the same or just a little bit cooler (1-2*C max) when I tested both times. But when I reported the temperatures earlier it wasn't using Prime95. It was using wPrime.

    I ran Prime95 27.7 64bit three times; ambient temp was the same during the 3 tests - about 26*C or 27*C.

    1st time - everything stock (Optimised default setting from bios and thats it), so ram was at 1333mhz. No crashes - core temp max at 80*C after 15 mins.

    2nd time - everything stock except RAM at 1866mhz, ram timings set manually, Dram voltage set manually. Core temps soar to above 80*C in 10-20 secs and pc shut down n restart .

    3rd time - set CPU Vcore to "offset" (-) 0.040v as you told me, and ram Prime95 for 5 mins without any probs. BUT core temps after 5 mins is mostly above 80*C with max at 85*C - too hot for my liking so I stop Prime95.
    Last edited by kingsaadi; 01-12-2013 at 08:07 AM.

  7. #17
    ROG Guru: Brown Belt Array HalloweenWeed's Avatar
    Join Date
    Dec 2011
    Reputation
    87
    Posts
    1,466

    First kingsaadi, it is obvious to me you are too careful, too scared of temperatures, when you run your tests. I have OCd 2 X58 CPUs, an i7-920 and an i7-960; and my current X79 i7-3930K. Because of EVGA shoddy software, and my sloppy testing, I accidentally ran my 920 to 101*C peak on one or more of the cores! Nevermind the details of the mistakes, they're not important. The X58 i7-9xx CPUs have a TJmax of 100*C. At this temp, the CPU throttles to protect itself. And that is why it never killed my CPU and that CPU is still operational. I am talking core temperatures of course, that is all that matters.

    A. It is my opinion that Intel reduced the TJmax on the X79 CPUs to keep ppl like me from getting to 100*C, and more importantly those who do so on a regular basis that deteriorated the CPU so they RMAd it. Reducing the TJmax to 91 meant that they could hit TJmax on a regular basis and not significantly shorten the lifespan of the CPU. IHMO the lower operating temperature is primarily a result of Intel corporate being more protective of their profit margin.

    B. Since the X58 CPUs could do 100*C, running X79 at 86*C should not be an issue. I believe one could hit TJmax on these X79 CPUs on a regular basis, and still most likely have an operating CPU after 8 years of use.

    So why did you shut it down when it went over 80? That is just obsessive-compulsive over-protective behavior. Personally, I run my tests right to 89*C.


    That said, yes your temps look high to me with your case (default) airflow and your cooler. I would really need some pics of your case open to see the airflow to be certain, but looking at the Mfg pics of that case I would say your airflow is nearly as good as mine. But there is a better way to be sure: Open your case side, wide open, and then run the tests and see if you get much lower temps. This is really the best test for airflow problems. It is not beside a desk or wall, close to the case air vents, is it? Do you have a heater (vent?) nearby blowing hot air on it? Hot sun directly shining on it?

    Also, you could try lowering your Vcore even more, like 0.01V per test, and testing thoroughly after each change. But don't forget to do the "idle" tests too - especially also coming out of sleep mode. GL.
    i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
    Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
    Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
    (EVGA site: ) And I have a second (wife's) computer, Eve.

    Overclocking is useless to me if it is not rock stable.

  8. #18
    ROG Guru: Brown Belt Array HalloweenWeed's Avatar
    Join Date
    Dec 2011
    Reputation
    87
    Posts
    1,466

    Also, you didn't say what kind of thermal paste you use; could be that is cause. Did you know that TIM (Thermal Interface Material) has a "curing" time? For Arctic Silver 5 (what I use) it is 200hrs. I don't know if that means 200 hours of use, or 200 hours after application. But if it means use, and you only use your computer 5hrs/day average, that is 40 days, that you used your computer, of just curing time; and before that your temps could be a couple of *C higher. FYI.

    BenchmarkReviews.com: 80-way Thermal Interface Material Performance Test

    Yeah I know that is 3 - 1/2 years old, but tell me if you know of another one that is newer and as comprehensive.
    Last edited by HalloweenWeed; 01-12-2013 at 05:19 PM. Reason: Added link
    i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
    Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
    Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
    (EVGA site: ) And I have a second (wife's) computer, Eve.

    Overclocking is useless to me if it is not rock stable.

  9. #19
    ROG Guru: Brown Belt Array HalloweenWeed's Avatar
    Join Date
    Dec 2011
    Reputation
    87
    Posts
    1,466

    Another possible cause of high CPU temps is the exact way you mounted your HS. You aren't mounting that with the case in the tower (standing up) position, are you? That would prob result in TIM moving upward, an uneven spread. I can't imagine a person trying to assemble the crossbar that way. And you NEVER started to put the HS on, then removed it, even for a second, right? That would cause air bubble problems, and could be even worse than the infamous "credit-card" spread. Any time you match the HS to the CPU top, and then remove it, even for just a second, even just a "little bit," you must completely remove the TIM and start over. When you mount your HS you need to do so with your mobo horizontal, and put even pressure to come down approximately parallel to the CPU surface, and thus spread the TIM in all directions evenly. And don't slide it back and forth any more than you need to.

    There also could be problems with your HS mount. I had a problem with knurled nuts (or bolt heads) where I thought they were tight using my fingers, but when I checked it with a screwdriver it was loose.
    Last edited by HalloweenWeed; 01-12-2013 at 06:28 PM. Reason: Added: "And don't slide it back and forth..."
    i7-3930K; Asus RIVE; G.SKILL Ripjaws Z 4x4GB DDR3 1866; MSI 7870 2GD5/OC; Crucial M4 SSD 256GB;
    Corsair 1000HX; Corsair H100, 4x Excalibur 120mm PWM CPU Fan p-p, AS5; SB X-Fi Titanium Fata1ity Pro;
    Dell U2412m IPS 1920x1200; Cooler Master HAF 932 case; Tripp-Lite OMNIVS1500 UPS fully Line-interactive.
    (EVGA site: ) And I have a second (wife's) computer, Eve.

    Overclocking is useless to me if it is not rock stable.

  10. #20
    ROG Member Array
    Join Date
    Jan 2013
    Reputation
    10
    Posts
    16

    mounting the HS with tower standing up - ROFL.

    As you kindly pointed out, I am a bit obsessive about the temp - mainly because I haven't OC'ed my cpu and worried that it's already running this hot, potentially reducing the OC-ability of the cpu.

    Being so obsessive, I made sure that I avoided all those things you suggested could be causing heat problems. I even had practice runs at applying the thermal paste. Mounting pressure is as tight as it will allow me to (the springs on the 'spring-loaded' mounting screws are as tight as they can be).

    I'm using Arctic mx-4 thermal compound. Its non-curing and non bleeding.

    Thanks for that article its quite good

Page 2 of 3 FirstFirst 1 2 3 LastLast

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •