Ray422
Premium
join:2002-03-04
Adger, AL
clubs: 
 ·RoadRunner Cable
| Q9650 Temperature problem
-Q9650
»www.newegg.com/Product/Product.a···pk=Q9650
-Gigabyte EP45-UD3R
»www.newegg.com/Product/Product.a···13128359
-Corsair Dominator DDR2-1066 @2.1v, only running at 1040.
»www.newegg.com/Product/Product.a···20145043
-Thermalright ultra-120 eXtreme
»www.frozencpu.com/products/5664/···40c14s52
-xpPro-Sp3
I have re-seated the cooler, tried a different cooler,tried both AC5 (my usual) and Thermalrite chill-factor paste. In fact, I even have 2 cpu fans (push-pull fashion)running at 2,000rpm each. The problem is not cooler or paste application related !!
This 14-15c temperature is surely a problem. This problem is just that "one" core. Notice how low my cpuz Vcore is; at that low voltage with this cooling, I should think my temps should be in the mid to high 50's maximum. I even had the case open for these screenshots.
If I drop the OC to 3.6ghz, that "one" core still reads around 60-62c. Even at idle, there is a 41-34c (7 degree) spread.
Would my problem be mobo, CPU or something else??
Please help !
--
Things I like - Team Discovery - Quads©3.3ghz - Strawberry Ice cream
|
|
signmeuptoo
Tune in to XM 120. BOO
Premium
join:2001-11-22
LOSTinSpace
clubs:
 ·Future Nine Corpor..
 ·AT&T Southeast
| Ray, how flat is the HS surface? I know I lapped my thermalright after seeing it was really cupped.
Maybe it might be time to have Intel replace it? Don't tell them that you aren't using factory cooling though.
Just a thought.
--
Join Teams Helix and Discovery. Put that fancy computer to good use helping to find a cure, your mom will love you for it. »Team Helix »Team Discovery |
|
Edge1
Workin' Ta Live
Premium
join:2006-03-01
3 edits | reply to Ray422
NM
It certainly does seem odd that one core's result is so different. After some research, this could be HSF related, but you pretty much ruled that out... |
|
deblin
Dark Side of the Moon
Premium,MVM
join:2001-09-01
Middletown, DE
| Yeah, one core being MUCH higher than the others I'd think HSF or at least the contact with the HSF...
It seems to be normal for these core 2 quads (both the 65 and 45 nm parts) to have 2 cores warmer than the others.
Here's my Q9550 @ 3.85 GHz idle:
floyd@pflog:~% sysctl dev.cpu.0.temperature dev.cpu.1.temperature dev.cpu.2.temperature dev.cpu.3.temperature
dev.cpu.0.temperature: 28
dev.cpu.1.temperature: 28
dev.cpu.2.temperature: 24
dev.cpu.3.temperature: 24
Notice the 4 C difference between cores 0/1 and 2/3, and the difference is greater when under load. As a matter of fact, one core in my case IS appreciably hotter than the others by 11 C (versus the OP's 15 C).
root@pflog:~# sysctl dev.cpu.0.temperature dev.cpu.1.temperature dev.cpu.2.temperature dev.cpu.3.temperature
dev.cpu.0.temperature: 47
dev.cpu.1.temperature: 38
dev.cpu.2.temperature: 38
dev.cpu.3.temperature: 36
--
He who is not contented with what he has, would not be contented with what he would like to have. -Socrates |
|
koitsu
Premium
join:2002-07-16
Mountain View, CA
1 edit | reply to Ray422
Q9550 temps | | 
CPU-Z data | |
 )
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer. |
|
Dream Killer
Graveyard Shift
Premium
join:2002-08-09
Forest Hills, NY
clubs: | reply to Ray422
What case are you using? |
|
Ray422
Premium
join:2002-03-04
Adger, AL
clubs:
·RoadRunner Cable
1 edit | Started out with an antec nine hundred (wind tunnel) and a tuniq tower cooler.
After selling my Q6600 and mobo, I switched it to my existing raidmax smilodon case and got that #1 air-rated ultra-120 extreme cooler.
My plan was to put another Q9650 or an i7-920 back in the antec nine hundred case.
But of course, now I'm delayed on this "one-core" temp problem. I'll probably just replace the CPU.
--
Things I like - Team Discovery - Quads©3.3ghz - Strawberry Ice cream
|
|
Dream Killer
Graveyard Shift
Premium
join:2002-08-09
Forest Hills, NY
clubs:
·RoadRunner Cable
| reply to Ray422
Can you provide a picture of your case with the side off? I'm asking because I'm just as stumped as you are and want to cover all the bases. Though my guess it's a malfunctioning motherboard sensor/diode inside the CPU. Testing that would require another Q9650 or another motherboard, however. |
|
Ray422
Premium
join:2002-03-04
Adger, AL
clubs:
·RoadRunner Cable
1 edit | Well...I actually can't, because I just took the system down to replace the cpu.
This is a pic of one my Q6600's in the same raidmax case with a tuniq cooler. The other pic is when I was installing the double-fan thermalrite for the Q9650.
Ya really can't tell anything from the pics?
--
Things I like - Team Discovery - Quads©3.3ghz - Strawberry Ice cream
|
|
Ray422
Premium
join:2002-03-04
Adger, AL
clubs:
·RoadRunner Cable
| For anyone else, that this might be valuable to in the future...
I noticed/learned too late, that real Temp 3.0 has a sensor detector button (bottom left). I tried it on an existing system but can't interpret the results. Maybe I could if something was bad?
I have not located any instructions yet, for the interpretation of results?
--
Things I like - Team Discovery - Quads©3.3ghz - Strawberry Ice cream
|
|
Ray422
Premium
join:2002-03-04
Adger, AL
clubs:
·RoadRunner Cable
| Yadda,yadda,yadda...say what...huh?
But,but,but...mine wasn't sticking?
Now we got everything is all cleared up, right?
Test Sensors:
The recent transition by Intel to 45nm has resulted in some of the DTS sensors showing signs of sticking at low temperatures and not functioning properly. If your DTS sensors are sticking then no software in the world is going to be able to tell you accurate core temperatures based on inaccurate data. Different idle temps is a good first sign of trouble. Test Sensors performs a quick test where it loads up each of your cores and measures the change in the individual DTS data for each core and reports the results.
Temperatures that don't change during this quick test are a sign of a sensor that is getting stuck. One sensor that always changes significantly more than the other sensor is also a sign that the one that moves the smallest amount might be stuck.
This part of the program is still in the experimental stage but it is accurate enough to report Core0 in my E8400 which gets stuck at a DTS value of 67 and won't go lower even if I were to go to the North Pole. This is outside my normal temperature range so isn't a problem for me but other users have reported their sensors getting stuck at much higher values. Don't rely on this program to report accurate temperatures on cores that have sensors that are sticking. If this sticking continues to occur at higher values then your load temperatures based on what the DTS is reporting won't be accurate either.
--
Things I like - Team Discovery - Quads©3.3ghz - Strawberry Ice cream
|
|
signmeuptoo
Tune in to XM 120. BOO
Premium
join:2001-11-22
LOSTinSpace
clubs: | That's very interesting Ray! Valuable thing to know! |
|
koitsu
Premium
join:2002-07-16
Mountain View, CA
1 edit | reply to Ray422
I've never seen such a thing, and it sounds a bit fishy. What indication is there that this isn't a software bug in Real Temp or possibly Windows (depending upon how the data is being fetched).
Also, have you tried *not* running multiple monitoring applications at once? I know that in the case of RMClock, SpeedFan, and some others, using them all simultaneously can result in incorrect results or error messages from the programs themselves (the reason being they're "fighting" over getting statistics).
Is there any indication the author of Real Temp has gotten in contact with Intel about this? If accurate, this would fall under "CPU bug" due to the thermistor equivalent being on-die.
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer. |
|
Ray422
Premium
join:2002-03-04
Adger, AL
clubs:
·RoadRunner Cable
| Well koitsu, it may sound fishy and it may be a CPU bug, but numerous people have a core-temp spread, just not so drastic as mine.
I don't run simultaneous temp-programs, but pictured so in order to confirm readings and conserve space. This should "probably" assure that neither program had a bug. I use the same real-temp version for my other pc's and I trust it; at least as much as ANY software CAN be trusted.
Your post of core temp is only at "IDLE" 2.0ghz with a 6 multiplier. Not really an appropriate comparison to what we're discussing here. Probably EIST or similiar has kicked-in due to idle. You will never show much disparity unless you're "overclocked" and under full-load...and perhaps you are one of the lucky ones (I hope so)
I run REAL full-loads 24/7 for distributed computing, not occasional full-loads. An equivalent might be like prime95. Even my Q6600 spread at 3.4ghz is 2c degrees at idle and 6c degrees at full load. That's easy to live with, but LGA775's at 45nm seem to be much worse.
As 45nm prices decline, more folks are buying and noticing this temperature disparity. Indeed Intel should be very aware, but who knows whether they are addressing the issue. These things are like errata; they tend to remain concealed.
Anyway, I hope no one else has such a bad spread as mine 
--
Things I like - Team Discovery - Quads©3.3ghz - Strawberry Ice cream
|
|
Ray422
Premium
join:2002-03-04
Adger, AL
clubs:
·RoadRunner Cable
2 edits | Replaced the CPU.
Idle and Load at 4.2MHZ
Bookoo better, huh? |
|
stevepas
join:2009-04-19
clubs:
| reply to Ray422
I've got a 6600 and found odd temp readings via various temp monitoring utilities.
One of the most interesting ones showing my cpu fan to be running at 30,000rpm!
Have been using realtemp. Found some technical info for it here »www.techpowerup.com/realtemp/docs.php
hopefully this link might be a useful read. |
|
