1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Q6600 temps with realtemp

Discussion in 'RealTemp' started by egale, Jul 11, 2008.

  1. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
    I have a Q6600 on a P5Q-E running at 3ghz. According to Probe, Speedfan and Coretemp, while running Prime95, the coretemps go as high as 69 degrees. Realtemp though, is showing temps 5 degrees lower. Lower is better for sure but is realtemp showing temps too low?

    By the way, the cooler is a Zalman 9500 which I have reseated a couple of times with AS5 and Zalman STG1 without making much if any difference.
     
  2. Cold Storm

    Cold Storm Battosai

    Joined:
    Oct 7, 2007
    Messages:
    15,014 (5.20/day)
    Thanks Received:
    2,999
    Location:
    In a library somewhere on this earth
    Well, here is something to think about. How is the air flow in the case? Is there a lot of air flowing threw the case to where the CPU cooler is getting it also? The one key feature in ocing is the air flow in your case. If the air flow doesn't help you then your going to be screwed.
     
  3. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
    I think the air flow is pretty good. Two 120mm fans in the front and one in the back. Not alot of cables cluttering either.

    Why does realtemp read temps 5 degrees lower though. It would be nice if realtemp were right and everything else wrong!
     
  4. pepsi71ocean New Member

    Joined:
    Nov 7, 2007
    Messages:
    1,471 (0.52/day)
    Thanks Received:
    125
    Location:
    The Peoples Republic of New South Jersey
    what are your volts? they should be stock volts for a 3GHz oc. Also there is a 200 Hour burn in time for AS5.

    Try Hardware monitor, but its most like the volts are to high. Also did you mean the Zalman 9700? Cuz if i rememebr correctly the the 9500 isn't big enough for a quad.
     
  5. Cold Storm

    Cold Storm Battosai

    Joined:
    Oct 7, 2007
    Messages:
    15,014 (5.20/day)
    Thanks Received:
    2,999
    Location:
    In a library somewhere on this earth
    The reason why Real Temp shows it as being 5c different and lower is that it's taking temps from a different source then Core temp. If you look there is the same Heat level. Just 5c different. I use real Temp and Core Temp. I just like to see what both say.
    Well you have to look at not the size of the fans but the CFM's that are pushing it.
    Also Pepsi is right. It might be that your pushing to much volts threw the core. Can you give us a few screen shots of your CPU-Z? that way we can see what it reads. Also, if you can add your system specs. That way we can see the case and what not also
     
  6. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
    The cpu volts in the bios is 1.3 and speedfan shows it as 1.27. Its a 9500 that I am running. I saw the 9700 in the store today. It is huge!

    Do you know if the Q9450 runs that much cooler than the Q6600?

    CPU-Z shows the same voltage setting. Its a Q6600 G0 stepping. What else from CPU-Z would be helpful.

    Also, have Mushkin 2gb x 2 1066 memory, 8800GT video card, Lancool case. Nothing overclocked but the cpu.
     
    Last edited: Jul 11, 2008
  7. deagle New Member

    Joined:
    Aug 20, 2004
    Messages:
    73 (0.02/day)
    Thanks Received:
    12
    Location:
    The South... of Germany
    9450 to 6600 about 10°C cooler.

    the thermal throttling @ the G0 6600 Quad beginns ~ 83°C all Temps lower are fine ;)

    if you have an G0 quad go to real Temp settings Tab and set the TJ max to 100°C (for B3-Stepping to 105°C)

    Real Temp is Made for the new 45nm CPU´s, they have a TJunction of 95°C

    Greetz Deagle
     
    Last edited: Jul 11, 2008
    egale says thanks.
  8. Cold Storm

    Cold Storm Battosai

    Joined:
    Oct 7, 2007
    Messages:
    15,014 (5.20/day)
    Thanks Received:
    2,999
    Location:
    In a library somewhere on this earth
    Well, with the Q9450 you can oc future with lower volts. You can get future. But, you also have to know that it might just be a simple thing as getting better air flow into the case.
    The better the air flow the better your have with ocing.
    Deagle is right on what he's saying.
     
  9. unclewebb

    unclewebb RealTemp Author

    Joined:
    Jun 1, 2008
    Messages:
    1,014 (0.38/day)
    Thanks Received:
    477
    deagle: I've never heard something so dumb. RealTemp uses the TjMax that it does for the Q6600 because I tested a bare core with an IR thermometer. TjMax=95C for the Q6600 G0 and it is only 85C for the B3 version. Lots of people like to argue this point but when they read the RealTemp docs and do the testing then they usually have a different opinion.

    There are no desktop processors that use TjMax=105C or even TjMax=100C. Those numbers come from CoreTemp that is assuming values based on the mobile processors. They also come from Intel TAT which people forget is actually a mobile CPU testing device and is not designed for desktop processors. If you do the calibration as explained in the docs you'll have very accurate temperatures from RealTemp.

    The 9500 isn't a great cooler for a Quad.

    Both programs are reading data from the exact same sensors. Let's stop the fud here guys.
     
    Last edited: Jul 11, 2008
    egale says thanks.
  10. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
    Changing TJ max to 100 fixed it. The temps now match the other utilities. Thanks,
     
  11. pepsi71ocean New Member

    Joined:
    Nov 7, 2007
    Messages:
    1,471 (0.52/day)
    Thanks Received:
    125
    Location:
    The Peoples Republic of New South Jersey
    I have my Q6600 Stable at 3.5GHz on a DFI Blood Iron. Your Q6600 when set up right will hit 3.6GHz. My 6600 is pushing 1.3Volts @ 3.5GHz

    You can drop your voltage a little bit at a time to see if it is stable. I think 1.3Volts is to much.
     
  12. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
    What do you recommend as a good cooler then?
     
  13. unclewebb

    unclewebb RealTemp Author

    Joined:
    Jun 1, 2008
    Messages:
    1,014 (0.38/day)
    Thanks Received:
    477
    If you want the BEST air cooler than there is only one choice, Thermalright Ultra-120 eXtreme
    http://www.silentpcreview.com/article759-page1.html

    Good work. You used to have one accurate program and now you have two that are not.

    Here's some of the real world testing that went into RealTemp.
    http://www.xtremesystems.org/forums/showpost.php?p=3096431&postcount=1599

    Ask the other guys to show you their testing.
     
  14. pepsi71ocean New Member

    Joined:
    Nov 7, 2007
    Messages:
    1,471 (0.52/day)
    Thanks Received:
    125
    Location:
    The Peoples Republic of New South Jersey
    Ive heard several types of coolers, the Thermalright Ultra-120 eXtreme, Artic Cooling Freezer 7, My 9700 Works great up to around 3.5GHz.

    It mainly depends on how high you want to clock your CPU.
     
  15. Cold Storm

    Cold Storm Battosai

    Joined:
    Oct 7, 2007
    Messages:
    15,014 (5.20/day)
    Thanks Received:
    2,999
    Location:
    In a library somewhere on this earth

    IF you have a problem with me man, then just need to add where the quote is by. That's what I was told and that's why I said it. If I knew that they tested THE SAME PLACE, then I would of loved to tell the guy the right thing. :toast:

    http://forums.techpowerup.com/showthread.php?t=65000
     
  16. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
  17. deagle New Member

    Joined:
    Aug 20, 2004
    Messages:
    73 (0.02/day)
    Thanks Received:
    12
    Location:
    The South... of Germany
    Hey Buddie whats your Prob??? :wtf:

    here is eagle´s question

    here´s my answer...
    There is no Tool on this Planet/this Universe that measures the real Temp of an Intel Core CPU Core! They only measure the theoretical point of thermal throttling!!

    Core Temp is almost near by, Real Temp uses the same method!

    Put the CPU to your Board, heat em up and when he´s stable so what if there 44°C, 65°C or 87°C displayed

    holy snikey... that kinda post with my shool english :D

    Greetz from the "Kraut"
     
  18. unclewebb

    unclewebb RealTemp Author

    Joined:
    Jun 1, 2008
    Messages:
    1,014 (0.38/day)
    Thanks Received:
    477
    deagle: I apologize for over reacting. I guess I just strongly disagree with some of the information that was posted.

    Stuff like this next statement simply isn't true.
    I decided to up my core voltage a little and I also disabled my CPU fan so I could create some heat and explore at what temperature thermal throttling really begins at. When thermal throttling is reached, the PROCHOT# bit in the processor gets set which software like RealTemp reads and displays. There are two separate bits set within the CPU. The Status bit shows you if thermal throttling is presently taking place and the History bit shows you if any thermal throttling has taken place since you started up.

    [​IMG]

    In this first picture you can see that both core0 and core1 have reached the same maximum temperature of 93C. Core0 shows that it was the first to start thermal throttling.

    [​IMG]

    In the second picture you can see that both core0 and core1 are presently throttling while the other two cores are still fine.

    [​IMG]

    Picture 3 shows what happens during throttling. Intel Core processors drop the multiplier down to 6.0 which reduces the MHz. This helps to keep the CPU temperature under control and hopefully prevents it from reaching TjMax.

    [​IMG]

    Picture 4 shows the next phase of thermal throttling. The processor has now reduced the VID. That's basically the processor telling the motherboard that it would like less voltage so it can try and control the heat it's producing. Because I am overclocking on an Asus board and I have manually set the core voltage, the motherboard has decided to ignore this request for less voltage and is still feeding the CPU 1.400 volts. The PROCHOT# Status bit continues to turn on and off which is keeping the temperature steady at 93C.

    It's good to know that even when overclocked with extra core voltage that you can run your processor right up to TjMax and still be Prime stable. These things do a great job of taking care of themselves. The further you try to overclock, the more you'll need to reduce your core temperatures to remain Prime stable.

    [​IMG]

    After I stopped running Prime and turned the CPU fan back on things returned to normal. The PROCHOT# History bits for core0 and core1 remain lit up until the next time it gets rebooted. During this test core2 and core3 never got hot enough to trigger thermal throttling. The two sets of Dual Cores within a Quad work independently so you can have situations where two cores are running at a slower speed while the other two cores are still running at full speed.

    [​IMG]
     
    Dr. Spankenstein and Cold Storm say thanks.
  19. Cold Storm

    Cold Storm Battosai

    Joined:
    Oct 7, 2007
    Messages:
    15,014 (5.20/day)
    Thanks Received:
    2,999
    Location:
    In a library somewhere on this earth
    Unclewebb, thank you for showing what you showed. By doing so, instead of just going and bashing both Deagle and I, you have shown what Real temp really is. If, I new that there wasn't a difference in core and real, I wouldn't of stated what I did. But, now that you have shown it, I can see it.
    Forums are for helping one another. Even if you find that someone is wrong. Back it up and show. Just don't go off saying your a fat lair. That proves nothing. And, all you'll get is people bashing you on and on.
     
  20. Dr. Spankenstein

    Joined:
    Feb 15, 2007
    Messages:
    1,920 (0.62/day)
    Thanks Received:
    276
    Thank you, unclewebb. Wonderfully explained and illustrated explanation!

    Really goes to show how incredible of devices these processors really are.

    Your program and insight are both greatly appreciated by this forum.
     
  21. egale New Member

    Joined:
    Jul 11, 2008
    Messages:
    7 (0.00/day)
    Thanks Received:
    0
    Unclewebb - great test!!!!!!

    Does this mean the processor won't fry at temps in the high 60s?
     
  22. erocker

    erocker Super Moderator Staff Member

    Joined:
    Jul 19, 2006
    Messages:
    40,773 (12.24/day)
    Thanks Received:
    15,685
    Indeed! However calling people dumb is not nice, thanks for apologizing unclewebb, and plese do not insult others here in the future.:)
     
  23. unclewebb

    unclewebb RealTemp Author

    Joined:
    Jun 1, 2008
    Messages:
    1,014 (0.38/day)
    Thanks Received:
    477
    These processors are bullet proof. They won't fry at twice that temperature. I wouldn't run one that hot but one user sent me an interesting screen shot when he was trying to determine the thermal shut down temperature of his Q6600.

    [​IMG]

    Now that's amazing. Intel documents that the shut down temperature for the mobile core processors is 125C so this guy was trying to see if the desktop processors are the same and they sure look like it. His computer would always shut down very soon after this point.

    Edit: I promise not to be a jerk in the future.
     
  24. goren New Member

    Joined:
    Jul 25, 2008
    Messages:
    2 (0.00/day)
    Thanks Received:
    0
    A great cooler, no doubt. There are some new air coolers that actually surpass it, however, by using a new technique where the heatpipes are in direct contact with the IHS. Xigmatek is the main manufacturer, but there are also coolers by Sunbeam, Kingwin, OCZ, and probably some others. Frostytech maintains a list of the top coolers per their testing:

    http://www.frostytech.com/top5heatsinks.cfm

    They actually rate the Ultra-120 Extreme below the Ultra-120. In the article they admit to using different 120mm fans when testing each version and this likely explains the unexpected outcome. They also point out that the Extreme version has a slightly convex base which won't provide as good of contact with some IHS. I believe this is actually part of the Extreme's design as a lot of newer Intel IHS are slightly concave.
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page