Big temp difference between cores?

B

Bloodgod42

Gawd
Joined
Nov 20, 2005
Messages
592
According to CoreTemp Beta 0.94, my two cores run as much as 10c difference. Is this normal?
CoreTemps.jpg
 
try pressing down on one side of the HSF and see if that makes a diff. Is this stock or after applying your own thermal paste? Maybe it's a contact issue?
 
the lower temp is prob youre nb temp
edit nm didnt see the coretemp window,
 
It's because core 1 is under a load and core 0 isn't. Try running a second stress test, but assign it to core 0
 
ryan_975 said:
It's because core 1 is under a load and core 0 isn't. Try running a second stress test, but assign it to core 0
Click to expand...

The load seems to be equally distributed...

CoreTemps2.jpg
 
ococ said:
The pics look like both cores are 100% loaded.
Click to expand...

Oh, I didnt even see the Speedfan gauge on there, but that's not to say it's displaying the results wrong. There's also the (unlikily) possibility that another process jumped the second core up to 100% right before he took the screenshot

At any rate, it looks like he's only running one instance of Orthos and it's stressing just on of the cores, which would explain the disparity of the temps.
 
Yeah, press hard down on the HSF and see if that helps.

If not, remount using the "line of paste" method rather than the "rice grain" method.

Lastly, RMA the chip to intel for bad IHS quality.

I'm all against RMAing a chip because it doesnt OC well... but these IHS's on the recent Conroes are a JOKE... 300$ processor and the .05c piece of copper is crappy? That's bullshit.
 
Arcygenical said:
Yeah, press hard down on the HSF and see if that helps.

If not, remount using the "line of paste" method rather than the "rice grain" method.

Lastly, RMA the chip to intel for bad IHS quality.

I'm all against RMAing a chip because it doesnt OC well... but these IHS's on the recent Conroes are a JOKE... 300$ processor and the .05c piece of copper is crappy? That's bullshit.
Click to expand...

Im actually running an AMD Opty, but I agree with what your saying. I am going to try a couple things regarding the seating of HSF and checking case airflow.

Also, is 60c 'core temp' to hot? Thanks
 
i doubt case airflow would make one core hotter than another. what about lapping? Maybe the cpu is uneven and making contact on one side? Are cores split that way?
 
ococ said:
i doubt case airflow would make one core hotter than another. what about lapping? Maybe the cpu is uneven and making contact on one side? Are cores split that way?
Click to expand...

Yes, air flow thru the case would not affect individual core, you are correct. I was just thinking out loud that 60c seemed really hot regardless. Lapping is a possibility, since yes, cores are split that way. They reside side by side on one half of the processor.
 
Bloodgod42 said:
Yes, air flow thru the case would not affect individual core, you are correct. I was just thinking out loud that 60c seemed really hot regardless. Lapping is a possibility, since yes, cores are split that way. They reside side by side on one half of the processor.
Click to expand...

I though AMD's dual core processors are on a monolithic die instead of an individual die for each core.
 
ryan_975 said:
I though AMD's dual core processors are on a monolithic die instead of an individual die for each core.
Click to expand...

You are correct, they are both on the same piece of silicon, but sit on one half of the chip.

01.jpg
 
I'd also investigate the temperature sensors. Are others with the same motherboard free of similar issues?
 
TrueKaio said:
I'd also investigate the temperature sensors. Are others with the same motherboard free of similar issues?
Click to expand...

The temps are coming straight from cores, not the motherboard, so there would be no easy way to check them.
 
i have similar issue at idle 10*C but loaded they are the same.... maybe the cores are slightly different... we all know diff cpus run at different temps because the silicon isnt the same... could it be the cores are just different and one can be more efficient than another?
 
Adidas4275 said:
i have similar issue at idle 10*C but loaded they are the same.... maybe the cores are slightly different... we all know diff cpus run at different temps because the silicon isnt the same... could it be the cores are just different and one can be more efficient than another?
Click to expand...

Both cores are on the same piece of silicon.


Is it possible for the TIM between the IHS and the die to have an air bubble over the one the cores, in turn causing poor heat transfer?
 
ryan_975 said:
Both cores are on the same piece of silicon.


Is it possible for the TIM between the IHS and the die to have an air bubble over the one the cores, in turn causing poor heat transfer?
Click to expand...

You are grasping at straws.
Remount the heatsink paying carefull attention to making sure it is tightened down evenly, i.e. alternate tightening of any (I water cool so dont know how this HS is retained on the PCB) fasteners in small stages to ensure even pressure on the heatspreader. Lap the heatsink while you have it off if you have not done so already.
 
wouldn't 2 cores, sitting side by side under the same heatspreader eventually reach a similar peak temperature under load, unless one of the cores has poor heattransfer to the heatspreader?
 
Mr.Pernod said:
wouldn't 2 cores, sitting side by side under the same heatspreader eventually reach a similar peak temperature under load, unless one of the cores has poor heattransfer to the heatspreader?
Click to expand...

at the heatspreader yes, the temps would tend to average out/transfer heat. Core temp reads the internal diodes embedded in each core and thus are a point source for each core.

the really interesting thing I have not yet been able to find is exactly where in each core that diode is located and if an effort was made to place them in an area of max heat generation.
 
BillParrish said:
You are grasping at straws.
Remount the heatsink paying carefull attention to making sure it is tightened down evenly, i.e. alternate tightening of any (I water cool so dont know how this HS is retained on the PCB) fasteners in small stages to ensure even pressure on the heatspreader. Lap the heatsink while you have it off if you have not done so already.
Click to expand...

The heat sink was mounted very carefully. I mounted it once, removed it and checked for contact of the AS5. It was making full contact so back on it went and yes, I alternated the tightening of the retention bolts to assure even contact. I did not lap it since it has a pretty smooth base and has preformed great on other cpu's I have had. :confused:

Mr.Pernod said:
wouldn't 2 cores, sitting side by side under the same heatspreader eventually reach a similar peak temperature under load?
Click to expand...

I would sure think so, and yet after 10 hours of orthos, it still shows the same results. :(

I was thinking, since each core can have its own temp, could they possibly have different clock speeds? Is there any program that would show it?
 
the hs maybe flat, but what about the cpu? not sure about AMD's, but i've seen several concave conroes
 
ococ said:
the hs maybe flat, but what about the cpu? not sure about AMD's, but i've seen several concave conroes
Click to expand...

I have thought of that. I considered going naked, but Im not sure if that would fix the issue, and a RMA would be out of the question then. One thing I have noticed is that CoreTemp shows the hot cpu temp drop 13c in 1 second after I stop orthos...could it be reporting incorrect temps?
 
Well, I just reseated the heatsink and looked for any possible cause of the temp diff between core...found none. Fired the system up again and I am still getting the same results...7-10c between cores while loaded. :(

I dont agree with RMA'ing a cpu just to get a better oc'ing chip, but this is concerning to me. I would rather have both cores running at the higher temp if they stayed within the 2-3c that people are saying is normal for dual cores. What do you guys think? Time to RMA due to defective IHS?
 
Bloodgod42 said:
Well, I just reseated the heatsink and looked for any possible cause of the temp diff between core...found none. Fired the system up again and I am still getting the same results...7-10c between cores while loaded. :(

I dont agree with RMA'ing a cpu just to get a better oc'ing chip, but this is concerning to me. I would rather have both cores running at the higher temp if they stayed within the 2-3c that people are saying is normal for dual cores. What do you guys think? Time to RMA due to defective IHS?
Click to expand...

Justification: Does the temp disparity exist when NOT overclocked?
 
Think about the size of the core...and also how thick it is.

Got that pictured in your head?

OK, now do you really think that there is a single chance that there could be a 9C temperature difference between the two spots on that single, tiny, thin piece of "glass", which has a integrated heat spreader on it? You do realize that a 9C is a difference equal to 15 degrees on the F scale (so you can picture that at well)?

The software is buggered.
 
ryan_975 - thats something I am looking into now. Also running single instances of Prime95 on each core one at a time.

Met-AL said:
The software is buggered.
Click to expand...

That has crossed my mind...can you recommend another prog that displays both core temp?
 
is that the latest version of coretemp? maybe this a dumb question, but what about Intel TAT :)
 
Hey if this makes you feel any better I have the same problem. Except my core #0 is hotter then core #1. And that's with Core temp beta0.94.
 
Sta!nless said:
Hey if this makes you feel any better I have the same problem. Except my core #0 is hotter then core #1. And that's with Core temp beta0.94.
Click to expand...

Well, it doesn't really make me feel any better per se, but its good to know Im not alone. What is the difference between your cores?

I am using Core Temp Beta0.94.
 
Well at idle core0 is 41C and core1 is 26C at 1.47 volts. I really don't think those temps are right though. And at load they are 54 and 48. And I'm using the same version. I might pop the top after I go water. Right now I'm using an XP90 and I hear it's not really a good idea to go neked with that heatsink.
 
Sta!nless said:
Well at idle core0 is 41C and core1 is 26C at 1.47 volts. I really don't think those temps are right though. And at load they are 54 and 48. And I'm using the same version. I might pop the top after I go water. Right now I'm using an XP90 and I hear it's not really a good idea to go neked with that heatsink.
Click to expand...

Holy smokes!!! You have a 15c diff showing between cores?!? I really would like to find another program that gives the temps of both cores to verify these readings.

Anyone know of another good one? Thanks
 
Ok, well here is another bit of info that I hope might help us figure out whats going on...

If I run 1 instance of SP2004 and set the affinity to core #0 then my temps on both cores are the same, but if I switch affinity over to core #1, it skyrockets and pulls the other core up a couple of degrees as well. :(

SingleCoreTest.jpg
 
This is from the creator of core temp.


The Innovation
Both Intel and AMD, realizing that the way things were done until now was probably not the best, made up a new way to monitor CPU temperature. This way includes a thermal diode; however, a thermal diode is an analog device and the readings have to be converted to digital data. This is done by an ADC (Analog to Digital Converter).

A thermal diode that is connected to the ADC creates something called DTS (Digital Thermal Sensor). Basically the way it works is very simple - in theory. A circuit in the CPU core samples a reading from the thermal diode and writes the digital value to a specialized register in the CPU, which allows any software to read this value at will. The benefit of this way of measuring temperature is that all this is done internally in the core and converted into digital data instantly - digital data does not lose accuracy when it's passed from place to place, unlike an analog signal that is prone to interference and attenuation.

Another advantage of this system is that you can integrate several sensors into the same die. Both Intel and AMD integrate such a DTS in each of their processing cores, which means that you can see the temperature of each core in your CPU. If you put, for example, a load on a dual core processor and set affinity to one of the cores, you will see that only one core gets hotter - and fast! Of course, the temperature of the second core will also go up, as they are physically connected and share the same silicon, but not as much as the loaded core.

Both Intel and AMD use the DTS to determine an overheating and protect the CPU from damage by either throttling or completely shutting the system down, depending on the threshold that the temperature has crossed.

The DTS was first officially introduced in the Core Duo (Yonah) processor by Intel and it was carried on into the Core 2 Duo series. AMD officially introduced DTS with the release of Rev F chips - however after some testing, it became apparent that DTS was there since the first Athlon 64 chip!

Intel also claims that their DTS is placed in the hottest part of the core, while AMD doesn't say exactly where the DTS is placed, but I am certain it's not one of the coolest parts.

From a lot of testing, I have come to the conclusion that for an Intel CPU, the temperatures reported by the DTS make a lot of sense. Intel's white papers say that their DTS is calibrated and ready to use when their chips leave the factory.

In Rev F chips from AMD, the reported temperature also seems to be quite accurate, but from different reports and white papers I've seen, the CPU leaves the factory without having the DTS properly calibrated. AMD claims it could have an accuracy range of ±14ºC. The only thing I've noticed is some older AMD CPUs either have a very large delta between two cores or sometimes give some really low temperature readings. I guess this is understandable as this feature was unofficial in those CPUs.
 
You must log in or register to reply here.
Back
Top