Core temps, CPU temps, TJMaxx, UGH!. 2600K, which are real max temps?

Toggle sidebar Toggle sidebar
G

Guest

Guest
Hi, scouring all over google to try to find 2600k max temps is rather disturbing. So many different answers, so many blatently wrong answers, and so many answers that apply to the wrong term back and forth between them. A lot of people are saying "Oh, the max temp is this" "No no no where did you hear that it's this!" and I think they're not differentiating core vs cpu (I assume?).

From what it seems to be:
The max temp >for the CPU< is ~70C
The max temp >for any core< is ~90C

Right now I'm running 1.35v, 4.5GHz with HT on. My core temps max out at around 73C, my CPU temp maxing out at around 55C (Both rough numbers). This is in OCCT over all the tests (I don't like using prime because It puts heat on CPUs that I've never, ever seen in any program, whereas OCCT hits the same numbers I use when encoding video, which also uses 100% of all my cores).

My setup is in my signature; MX-2 for thermal paste, stock fan on the venomous X RT. For this setup:
1. Do these temps seem average?
2. What kind of actual headroom do I have for these temps? I'm either right at the edge, or have a LOT to go yet from what I've found so far.
3. Voltage for 24/7 use? 1.4 seems to be a border between people who think it's a good or bad idea, yet intel rates it up towards 1.5v (?).
4. What exactly is TJMax in comparison to a core temp?
5. OCCT (HWmonitor built in) temps and voltages seem to be in line with what's to be expected of this system. Should I be using something else? A lot of people sware by coretemp, however last time I tried other temp programs, they all turned out to be the same regardless.
 
Solution
CompuTronix
Hi c1o5ry1991,

The old timers and Moderators here will remember me well. I used to write and maintain the most popular Sticky here for about 3 years, which was the 16 page Core i and Core 2 Temperature Guide. I've been a member at Tom's since `06, and although I'm still an avid reader, I rarely come forward to offer comments on any threads, however, today I will. The answer that you're seeking is the Tcase specification which is available on Intel's website - http://ark.intel.com/products/52214/Intel-Core-i7-2600K-Processor-(8M-Cache-up-to-3_80-GHz) - however, there are some explanations required in order to make sense of Intel's Thermal Specifications.

I'll try to keep it as simple as possible:

Tjunction Max...
Sort by date Sort by votes
inzone

inzone

Champion
Mar 12, 2011
14,971
1
51,960
On the voltage setting a setting of 1.38v is what is considered to be the breaking point for the life of the cpu being adversily affected. If you go over the 1.38v in your overclock and keep it over then you are shotening the life of the cpu.
The other thing the shortens the life of the cpu is heat , so if your going to overclock and have consistently high temps then you will shorten the lifespa of the cpu.
I think that the best thing you can do if you want to have a constant overclock is to have water cooling. It may be expensive and a pain to manage but it does the best job of keeping the cpu cool.
I think you have to have a couple of programs to monitor the temps so that you can get an average since there isn't one that everyone can agree on. I used to use a program called Everest and it was pretty good. I don't bother with those programs any more because I have water cooling and when nothing ever gets above 50c for anything then you tend to stop looking.
 
Upvote 0 Downvote
G

Guest

Guest
I hear ya. I'm not interested in water cooling though. It is a pain to manage and if something fails, that could mean your entire system. As for the voltage, I'm ok with shedding some time off, however CPUs last for an eternity to begin with don't they? (For enthusiasts at least, it feels like it.)
 
Upvote 0 Downvote
inzone

inzone

Champion
Mar 12, 2011
14,971
1
51,960
Thats the thing with the shortening of the lifespan is that they don't give any numbers. Your right about the cpu lasting a long time to begin with and with a lot of people they are only keeping thier cpu 2 years max and then they're looking for the next best thing , some people bought a Sandy Bridge and then they want to upgrade to the Ivy Bridge.
 
Upvote 0 Downvote
G

Guest

Guest
Yeah. I'll probably max out at 3.8 just because I know there's no true improvement in performance after a certain rate. I'd still like to max this chip out within given voltages and temps. Problem is I don't know what temps are correct.
 
Upvote 0 Downvote
inzone

inzone

Champion
Mar 12, 2011
14,971
1
51,960
I had a software program called Everest and it would give the temps of all the components in the computer and when it came to the cpu it would give a temp for the cpu and then individual temps for each core and for the longest time it confused me that the cpu temp was not the average of the core temps and it was different and not something that could be figured out by looking at any of the core temps.
I finally realized that the core temps were higher because the temp was taken at the core level inside the cpu and that the cpu temp was take outside and the result was the heat passing through the metal cover to the sensor and so would be lower.
 
Upvote 0 Downvote
G

Guest

Guest
Of course. That's not my problem though, my problem is needing to know the max temp of the CPU, and of any individual core.
 
Upvote 0 Downvote
G

Guest

Guest
I have all over as stated in the OP. I know/have heard that the thermal shutdown is at 90c, however i'd like to know what is considered a safe 24/7 temp that isn't going to kill my comp over the next 2-5 years (For resale purposes).
 
Upvote 0 Downvote
C

cbrunnem

Distinguished
Dec 19, 2010
3,331
0
20,960


generally the safe temp guidline is under 60*c and the riskier temp is under 70*c keep in mind that with the stock cooler it will run into the 80's but with the stock cooler youre not pumping lots of volts through it.
 
Upvote 0 Downvote
G

Guest

Guest
@cbrunnem And is that for core temp or CPU temps. I'm on a Venomous X for cooling.

@amuffin It's the only thing that will run stable. Anything else will give me errors in OCCT.
 
Upvote 0 Downvote
C

cbrunnem

Distinguished
Dec 19, 2010
3,331
0
20,960


core temps. thats the only temps that people care about. to be honest i dont know where you saw people referencing cpu temps. when they said cpu temps there where probably talking about core temps.
 
Upvote 0 Downvote
CompuTronix

CompuTronix

Intel Master
Moderator
Nov 13, 2006
6,374
279
38,240
Hi c1o5ry1991,

The old timers and Moderators here will remember me well. I used to write and maintain the most popular Sticky here for about 3 years, which was the 16 page Core i and Core 2 Temperature Guide. I've been a member at Tom's since `06, and although I'm still an avid reader, I rarely come forward to offer comments on any threads, however, today I will. The answer that you're seeking is the Tcase specification which is available on Intel's website - http://ark.intel.com/products/52214/Intel-Core-i7-2600K-Processor-(8M-Cache-up-to-3_80-GHz) - however, there are some explanations required in order to make sense of Intel's Thermal Specifications.

I'll try to keep it as simple as possible:

Tjunction Max - is a specification - 100c (Throttle / Shutdown Threshold)
Tjunction - is Core temperature.

Tcase Max - is a specification - (i7 2600K is 72.6c, rounded to 73c).
Tcase - is CPU temperature.

Ambient - is room (intake) temperature - 22c or 72F is standard test environment.


From my Guide:

"Section 1: Introduction

Core i and Core 2 processors have 2 different types of temperature sensors; a CPU case (not computer case) Thermal Diode centered under the Cores, and Digital Thermal Sensors (DTS) located on each Core. The case Thermal Diode measures Tcase (Temperature case), which is overall CPU temperature, and the Digital Thermal Sensors measure Tjunction (Temperature junction), which are individual Core temperatures. Since these sensors measure 2 distinct thermal levels, there is a 5c temperature difference between them, which is Tcase to Tjunction Gradient. Core i and Core 2 processors have 1 Tcase sensor, and an individual Tjunction sensor on each Core" ...

... "Section 4: Thermal Flow

Heat originates within the Cores, where Tjunction sensors are located on the hot spots of each Core. Most of the heat dissipates from the top of the Cores through the Integrated Heat Spreader and CPU cooler to air inside the computer. Some of the heat dissipates from the bottom of the Cores through the CPU case, which creates a 5c thermal Gradient toward the center of the substrate, where the Tcase sensor is located. This heat then dissipates through the socket and motherboard to air inside the computer. Safe and sustainable temperatures are determined by CPU cooling efficiency, computer case cooling efficiency, Ambient temperature, Vcore, clock speed and Load.

• Tjunction is higher than Tcase.

• Tcase is higher than Ambient" ...

... "Section 5: Findings

(A) Tcase is acquired on the CPU substrate from the CPU case Thermal Diode as an analog level, which is converted to a digital value by the super I/O (Input/Output) chip on the motherboard. The digital value is BIOS Calibrated and displayed by temperature software. Motherboard BIOS Calibration affects the accuracy of Tcase, or CPU temperature.

(B) Tjunction is acquired on the Cores from Thermal Diodes as analog levels, which are converted to digital values by the Digital Thermal Sensors (DTS) on each Core. The digital values are Factory Calibrated and displayed by temperature software. Intel Factory Calibration affects the accuracy of Tjunction, or Core temperatures" ...


The most basic and important points are that:

(1) Core temperatures tend to be more accurate, while CPU temperature tends to vary by motherboards, chipsets and BIOS updates.

(2) On a properly calibrated Core i computer, CPU temperature is 5c lower than average Core temperatures at 100% workload.

(3) Since Intel's 2600K Tcase Max (CPU) specification is 73c, then the corresponding maximum average Core temperature is 78c.


EDIT: Please read this former post for additional information - http://www.tomshardware.com/forum/254512-29-temps-paying-attention#t1860801


I hope this clears up all the ongoing misinformation and confusion.


CompuTronix :sol:
 
Upvote 0 Downvote
Solution
G

Guest

Guest
The only question I have to that is, why does intel even publish the Tjunction Max spec? If the Tcase Max is 73c, and there is a 5c difference between the cores and the CPU casing, how would you hit the 100c wall? Or if I'm understanding that wrong, what you're saying is the max average core temp should never exceed 78c, correct?

Lastly, if that is the case, how is my system not doing a better job? I see a few people hitting 4.8 and even 5.0 GHz on air. I believe Anandtech, or guru 3D had an article of them getting 4.5GHz out of the stock cooler? I've remounted a few times with little to no difference in temps. Fully cleaning and reapplying thermal paste etc etc. My ambient temp is definitely not high, my room is absolutely cold by anyone's standards, however I couldn't give you an exact number.
 
Upvote 0 Downvote
CompuTronix

CompuTronix

Intel Master
Moderator
Nov 13, 2006
6,374
279
38,240
c1o5ry1991

To understand Intel's thermal specifications, you would have to read their white papers which describe, at the engineering level, the details of the test methods involved in arriving at a temperature value for each processor variant. The simple answer is that the thermal specification for Tcase does not push the processor to an overtemp or "shutdown condition", which is Tjunction Max. And yes, if the average Core temperature for the i7 2600K reaches 78c, then that is considered to be an "overtemp" condition.

Also, keep in mind that there are no two processors which are electrically identical. Even if they have consecutive serial numbers and are fabricated from adjacent areas of the silicon wafer, they each have their own unique electronic properties among their hundreds of millions of nano transistor circuits such as impedance, capacitance, inductance and tranconductance, all of which determine overall performance and overclocking characteristics such as tolerance to voltage, temperatures and leakage current.

All we can say is what seems to be typical. Even if we know that each lab used identical hardware setups, test methods, measuring equipment and environmental conditions, we still can't factor out the internal electronic variables (however miniscule they might be) among different processor samples of the same variant, family, model, stepping and revision. Overclocking is always a roll of the dice. Sometimes you get a real winner, sometimes not so great. Either way, it sure beats running it at stock!

Comp :sol:
 
Upvote 0 Downvote
C

cbrunnem

Distinguished
Dec 19, 2010
3,331
0
20,960


you dont have the best cooler out there either. you have a good one but you dont have the best. couple that with a mediocre chip that you might have and that makes high temps.
 
Upvote 0 Downvote
G

Guest

Guest
you dont have the best cooler out there either. you have a good one but you dont have the best. couple that with a mediocre chip that you might have and that makes high temps.
Click to expand...
Understandable. However is the difference in chips going to be that different to notprove decently better performance compared to a stock cooler?


http://www.anandtech.com/show/4083/the-sandy-bridge-review-intel-core-i7-2600k-i5-2500k-core-i3-2100-tested/3
"These chips overclock very well. Both my Core i5-2500K and Core i7-2600K hit ~4.4GHz, fully stable, using the stock low-profile cooler."
^ This is what bothers me.
 
Upvote 0 Downvote
C

cbrunnem

Distinguished
Dec 19, 2010
3,331
0
20,960


so chips just run hotter. its in the construction of the internal heatsink. also some chips require more voltage which creates more heat. couple those two and many more conditions and you have a hot chip.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom