View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000804 | SpeedFan | Hardware support | public | 2007-03-01 15:15 | 2008-08-22 01:25 |
| Reporter | SLi_dog | Assigned To | alfredo | ||
| Priority | normal | Severity | major | Reproducibility | always |
| Status | closed | Resolution | duplicate | ||
| Platform | Nvidia 680i | OS | Windows XP Professional | OS Version | SP2 |
| Summary | 0000804: Intel Quad Core QX6700 core temperatures under reporting by 15C | ||||
| Description | The core temperatures being reported for an Intel Core2Quad QX6700 are 15C below other programs reading the same sensors. I think this is due to the higher Core2QUAD Tjunction value of 100C being mistaken for the Core2DUO Tjunction value of 85C | ||||
| Steps To Reproduce | n/a | ||||
| Additional Information | n/a | ||||
| Tags | No tags attached. | ||||
| Motherboard Model | |||||
| Video Card Model | 8800GTX | ||||
|
|
Please, send me a SEND REPORT from SpeedFan's INFO tab, then drop me an email. |
|
|
I have read this problem elsewhere with all the core 2 quad processors, not just the qx6700. It also happens on the q6600. |
|
|
not just quad processors - also E6750/E6850 show temps 15 deg lower than reality |
|
|
@rbarkas How do you know the "real" temperatures? ;) This difference also occurs with all E4x00 models. As far as I know the latest CoreTemp readings are wrong (too high). I believe Speedfan temps are correct. I know that Everest also shows the higher temps but the question is how to find out which program reads correctly? Does anybody know exactly where Speedfan, Everest, and CoreTemp get their readings? Maybe different sensor, although that seems to be unlikely. |
|
|
No one knows the "real" core temperature as Core2 desktop processors don't have a Tjunction value as they're not a mobile CPU. Speedfan, CoreTemp and any temperature monitoring software that reads the "Core" temperatures displays it's temperature values based on the CPU's internal DTS for each core. The DTS are the values that the CPU uses to gauge it's throttling and shutdown safety features. As Intel has never released the "Tjunction" values for any of it's Core2 desktop CPUs, any "Actual Temperature" reading is based on a software written value that is just a "best guess". The issue with CoreTemp and Allendale Core CPUs (E4400, E4400 & and possibly E4500 and Core2 Pentium E2140, E2160) is a well known one but the original version of CoreTemp (version 0.94) was reading too low initially (with many people in Australia experiencing below ambient CPU temperatures which is just 100% impossible) so an arbitrary value of 15C was added to fix it in version 0.95. This was possibly too high, and many argued this but I'm not sure if it was. Regardless, the best way to use CoreTemp on Allendale (and possibly now 1333FSB CPUs too) is to configure it to show the direct DTS values (incorrectly indicated at "Delta to Tjunction" in CoreTemps' settings) then just keep the CoreTemp readings 20C "from Tjunction" in the read out. This feature of showing the actual DTS values would be a great option to have on any new version of Speedfan BTW. |
|
|
Just add check of MSR 0xEE bit 30 and use correct TjMax (85 or 100) in Tj calculation? from http://softwarecommunity.intel.com/isn/Community/en-US/forums/thread/30228360.aspx When bit 30 of MSR 0xEE is set, the digital thermometer (Tjmax/Tjunction) is calibrated to 85C, otherwise, it is calibrated to 100C. So it is fairly easy to calculate CPU temperature when you combine above information with Digital temperature readout field in MSR 0x19C bit [22:16]. |
|
|
Paco, where are you getting the reference to 85c and 100c TjMax for Intel Core2 Desktop CPUs? |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2007-03-01 15:15 | SLi_dog | New Issue | |
| 2007-03-01 15:15 | SLi_dog | Video Card Model | => 8800GTX |
| 2007-03-01 19:03 | alfredo | Note Added: 0002367 | |
| 2007-03-01 19:03 | alfredo | Status | assigned => acknowledged |
| 2007-07-29 07:33 | jvinsepa | Note Added: 0002644 | |
| 2007-08-07 00:13 | rbarkas | Note Added: 0002650 | |
| 2007-08-09 06:59 | norman02us | Note Added: 0002651 | |
| 2007-08-09 13:31 | SLi_dog | Note Added: 0002654 | |
| 2007-08-09 13:33 | SLi_dog | Note Edited: 0002654 | |
| 2007-09-08 18:51 | paco | Note Added: 0002846 | |
| 2007-09-14 03:46 | SLi_dog | Note Added: 0002859 | |
| 2007-09-14 03:47 | SLi_dog | Note Edited: 0002859 | |
| 2007-09-14 03:48 | SLi_dog | Note Edited: 0002859 | |
| 2008-08-22 01:24 | alfredo | Relationship added | duplicate of 0001150 |
| 2008-08-22 01:25 | alfredo | Status | acknowledged => closed |
| 2008-08-22 01:25 | alfredo | Resolution | open => duplicate |
