Hi. I have been getting high idle temps with my Q9300 and Xigmatek HDT-S1283 Cooler. I am running the Q9300 at the stock FSB and Voltage. I recorded these temps with CoreTemp, and the temps seem to stay the same all day whether its in idle or if I've been playing Oblivion or Crysis for hours. All four cores are at 53, 51, 48, 52 degrees Celsius basically all day long. I just built this rig and is my first one and don't really know what's up. Please help me? Thanks
Lupiron
Distinguished
- Feb 9, 2008
- 1,711
- 0
- 19,780
First off, push pins? Those blow.
Second, did you put the paste on the CPU of the heat sink? I have found that the best way is to place a thin line down each of the three pipes, because the spacers make contact with the IHS first.
Third, the middle heat pipe needs to run down the core placements. Not over them. IE, going one way would cause all three pipes to just run over the core areas. Going the other makes the middle pipe run STRAIGHT ALONG the direction the cores take, since they are in the middle, going one way.
Other than that, Bios update?
(I had to ditych the push pins and make my own retention, and then it took me like 15 tries to get the whole on the heat pipe and not the CPU thing going... so good luck! But it WILL work!)
--Lupi
Second, did you put the paste on the CPU of the heat sink? I have found that the best way is to place a thin line down each of the three pipes, because the spacers make contact with the IHS first.
Third, the middle heat pipe needs to run down the core placements. Not over them. IE, going one way would cause all three pipes to just run over the core areas. Going the other makes the middle pipe run STRAIGHT ALONG the direction the cores take, since they are in the middle, going one way.
Other than that, Bios update?
(I had to ditych the push pins and make my own retention, and then it took me like 15 tries to get the whole on the heat pipe and not the CPU thing going... so good luck! But it WILL work!)
--Lupi
How do I know if it is running down the cores or not? I'm not sure if it is. But I did put the thermal paste on it and it is evenly spread over the top of the CPU. Unfortunately, I ran the stock cooler for one day with it and forgot to peel the sticky thing off of the bottom of it (which I didn't know that there was one) and it was melted and got all over the top of the CPU...Probably no big deal though.
The only thing that makes me not want to do it is that it took me about an hour to get the heatsink on the first time, and it was a huge pain. But it should be easier now that I actually kinda know how to do it.
The only thing that makes me not want to do it is that it took me about an hour to get the heatsink on the first time, and it was a huge pain. But it should be easier now that I actually kinda know how to do it.
- Nov 13, 2006
- 6,370
- 282
- 38,240
A significant percentage of 45 nanometer processors are known to have defective Digital Thermal Sensors (DTS) which "stick", and won't show the actual Idle temperatures of the Cores below a certan value. Use Real Temp 2.6 - http://www.techpowerup.com/realtemp/ - to test your sensors and monitor your Core temperatures. Make sure that you read the Real Temp documentation - http://www.techpowerup.com/realtemp/docs.php - so that you understand what this excellent monitoring utility does differently than others, and why it's the most accurate utility available for monitoring Core temperatures.
The ever popular utility "Core Temp" reads 10c too high on 45 nanometer processors, in order to inappropriately compensate for sensor linearity problems, which is also explained in the Real Temp documentation. Core Temp uses Tjunction Max values which are simply imported from values specified for mobile Core 2 processors, while Real Temp's Tjuinction Max values are based upon actual research and testing performed by the author, as well as others, which is why I can no longer recommend the use of "Core Temp" to anyone. Further, Core Temp reads too high on many 65 nanometer processor variants as well.
Also, a BIOS update (flash) can affect the accuracy of CPU temperature, but will NOT affect the accuracy of Core temperatures. If you want to understand the difference between CPU temperature (Tcase) and Core temperatures (Tjunction), then check out the Core 2 Quad and Duo Temperature Guide - http://www.tomshardware.com/forum/221745-29-core-quad-temperature-guide
Comp
The ever popular utility "Core Temp" reads 10c too high on 45 nanometer processors, in order to inappropriately compensate for sensor linearity problems, which is also explained in the Real Temp documentation. Core Temp uses Tjunction Max values which are simply imported from values specified for mobile Core 2 processors, while Real Temp's Tjuinction Max values are based upon actual research and testing performed by the author, as well as others, which is why I can no longer recommend the use of "Core Temp" to anyone. Further, Core Temp reads too high on many 65 nanometer processor variants as well.
Also, a BIOS update (flash) can affect the accuracy of CPU temperature, but will NOT affect the accuracy of Core temperatures. If you want to understand the difference between CPU temperature (Tcase) and Core temperatures (Tjunction), then check out the Core 2 Quad and Duo Temperature Guide - http://www.tomshardware.com/forum/221745-29-core-quad-temperature-guide
Comp
- Nov 13, 2006
- 6,370
- 282
- 38,240
Please use the link I provided for you to read the Real Temp documentation, then use the "Test Sensors" button to determine the state of health of your sensors.
- Nov 13, 2006
- 6,370
- 282
- 38,240
If your computer case has a clean wiring job, and is well ventilated, and if ALL fans are at 100% RPM with your motherboard running default settings at auto Vcore, then your Core temperatures should Idle just below 30c with your high-end cooler, and CPU usage is no more than 2% for 10 minutes at 22c ambient, with the case covers removed.
If you can verify that you meet these conditions, and your temps are still above 30c, then your sensors are indeed "sticking", which means that you'll never see true Idle temperatures for your particular Q9300 sample. However, keep in mind that using Real Temp will give you "real" Load temps, which is of course more important than Idle temps.
If you can verify that you meet these conditions, and your temps are still above 30c, then your sensors are indeed "sticking", which means that you'll never see true Idle temperatures for your particular Q9300 sample. However, keep in mind that using Real Temp will give you "real" Load temps, which is of course more important than Idle temps.
- Nov 13, 2006
- 6,370
- 282
- 38,240
The problem is that I do indeed want to OC and can't with these temps. I've done an alright job (nothing great) of keeping it well ventilated and am still trying to get better airflow. It's not that easy in an Antec 900. The problem is is that I have a hell of a time trying to find a place to stick all the PSU's wires. Like I said though, I'm still working on that. And all fans are 100%.
- Nov 13, 2006
- 6,370
- 282
- 38,240
- Nov 13, 2006
- 6,370
- 282
- 38,240
Ambient temperature refers to the room temperature in and around the area near the front of your case, where your intake fans are located. A case which is at floor level may have an ambient temperature of a degree or two cooler than a case that sits on top of a desk. Hopefully your Antec 900 is not inside a desk enclosure, or in an area which in any way restricts airflow.
- Nov 13, 2006
- 6,370
- 282
- 38,240
Although I feature SpeedFan in my Temperature Guide, and there are some detailed explanations concerning SpeedFan, references to SpeedFan's Help and How-To icon, and a link to SpeedFan's homepage and FAQ's, my Guide is not a SpeedFan tutorial, nor am I. Above are links to some of the best information available anywhere on the Internet, which contains more links to more great info; all you need to do is just slow down a bit, and read.
Respectfully, you should first do some research to get up to speed on the basics before proceeding any further. All of your questions have answers that have already been written. Perhaps someone else would be willing to walk you through all the details, but sorry to say, I'll have to cut you loose at this point. I recommend that you read, read, then read again. Google is our friend.
Good luck with your rig,
Comp
Respectfully, you should first do some research to get up to speed on the basics before proceeding any further. All of your questions have answers that have already been written. Perhaps someone else would be willing to walk you through all the details, but sorry to say, I'll have to cut you loose at this point. I recommend that you read, read, then read again. Google is our friend.
Good luck with your rig,
Comp
TRENDING THREADS
-
-
Review AMD Ryzen 7 9800X3D Review: Devastating Gaming Performance- Started by Admin
- Replies: 149
-
-
-
Question Can't create a volume on WD BLACK SN770 using any tool
- Started by RPGJerico
- Replies: 9
Facebook
Twitter
Instagram