E4300 at 3Ghz, 72 degrees Celcius Safe?

imported_chrissd

Junior Member
May 6, 2007
3
0
0
Just got my E4300 and Gigabyte P965-S3. It runs Orthos stable at 3ghz at Stock Voltage and Stock HSF for 8 hours. Only thing i am worried about is the temperature, should i trust TAT and Coretemp? These are the temps i get:

TAT: 72
Coretemp: 72
Speedfan: 57
Gigabyte temp program: 42

Is it safe to run it at these values for a year or so? Will probably will get a new HSF when i wanto go up to 3.4 -> 3.6ghz.

/Chris
 

defiantsf

Member
Oct 23, 2005
132
0
0
What voltage did you use? You generally don't want to go over 60C under full load as measured by TAT/Coretemp. I keep my E6600 under 50C which is easy to do with Thermalright Ultra 120 Extreme.

 

MarcVenice

Moderator Emeritus <br>
Apr 2, 2007
5,664
0
0
72 is definately pushing it ...

did you have to increase your voltage a lot to get to 3ghz ? If 2.8 is stable, at 60 degrees tops, then I would settle for 2.8
 

Accord99

Platinum Member
Jul 2, 2001
2,259
172
106
What's the Tjunction as reported by Coretemp? If it's 100C, as is common on the E4300, then you have 30C of thermal headroom before throttling begins.
 

DrMrLordX

Lifer
Apr 27, 2000
22,720
12,697
136
72C is indeed hot, i wouldn't go over 65C for any significant period of time myself.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
i had some temperature weirdness with the ASrock Dual VSTA MB

same 2.66Ghz OC for my e4300 ... yet with the 1.40 BIOS, TAT reported high 30s and with the 2.70 BIOS flash it is now reporting low 50s :p
:Q

maybe try SpeedFan ;)

EDIT: i see you DO use SpeedFan ...

i dunno ... that is *some* variation 42-72C
:confused:

try a thermometer right next to the CPU :p
:Q

how warm are your case temps?
 

imported_chrissd

Junior Member
May 6, 2007
3
0
0
Thanks for the replies guys. I forgot to include that the Voltage is DEfault Voltage, 1.325. But Speedfan says 1.235. CoreTemp says 100C TJunction. So i was thinking i still have 30 degrees before it starts throttling. Also, maybe the temp will decrease a couple of degrees when the Arctic silver has settled okey. But i guess from you answers, i should be trusting TAT and Coreteemp and not Speedfan and the Gigabyte included tool.
I think it's strange though that TAT reports 60 degrees Celcius in IDLE.

/Chris
 

AnotherGuy

Senior member
Dec 9, 2003
678
0
71
well whats ur temps at stock speed???? that should tell u if thats the correct temps ur gettin or not... see how big of a difference between stock speed temp and OC temp...
 

Acanthus

Lifer
Aug 28, 2001
19,915
2
76
ostif.org
Originally posted by: chrissd
Just got my E4300 and Gigabyte P965-S3. It runs Orthos stable at 3ghz at Stock Voltage and Stock HSF for 8 hours. Only thing i am worried about is the temperature, should i trust TAT and Coretemp? These are the temps i get:

TAT: 72
Coretemp: 72
Speedfan: 57
Gigabyte temp program: 42

Is it safe to run it at these values for a year or so? Will probably will get a new HSF when i wanto go up to 3.4 -> 3.6ghz.

/Chris

Im running the same mobo and CPU, and my temps are 20c lower than yours with the stock HSF. (using coretemp)

Also note: I topped out at 3.1ghz with voltage bumps (stopped at ~+5%)
 

Accord99

Platinum Member
Jul 2, 2001
2,259
172
106
Originally posted by: chrissd
But i guess from you answers, i should be trusting TAT and Coreteemp and not Speedfan and the Gigabyte included tool.
The 15C difference between TAT/CoreTemp and SpeedFan occurs because SpeedFan assumes a 85C Tjunction.

I think it's strange though that TAT reports 60 degrees Celcius in IDLE.
CoreTemp 0.95 has an option to display the delta to Tjunction temp, which is the raw measured value and tells you roughly how much room before you throttle. I think that, with C2Ds now coming with different Tjunction values, this is a better way to compare "temperatures". 60C idle with a Tjunction of 100C is, in my opinion, the equivalent of 45C on a 85C Tjunction C2D, and is a reasonable temperature .

 

customcoms

Senior member
Dec 31, 2004
325
0
0
IMO, 60 degrees IDLE is not safe, no matter what the theoretical limit is. As a load temperature, I would say that is fine. Even if your processor has a limit of say 85 C, or 100 C, 45 degrees C is still 15 degrees cooler than 60 degrees C, period. The two things that kill processors are heat and voltage. If those are minimized, then you can probably expect your processor to last long enough. The only thing that really hasn't been explored is the effects of long term raw speed on the silicon and especially the gates, since the processor you are overclocking was never guaranteed to run at those speeds. but if Intel and AMD are having a good year with the silicon and every processor can pretty much be a high end processor, it doesn't matter so much.

 

imported_chrissd

Junior Member
May 6, 2007
3
0
0
Thanks for your input guys. I just tested the temps at default speed of 1.8ghz with Orthos and I get:

TAT & Coretemp: 65.
Speedfaan: 51
Gigabyte: 51 (Updated BIOS F9d, so now it reports same temps as speedfan)
Gigabyte System Temp: 38.

Also, ambient temp in the room is probably around 26 degrees(warm day in San Diego).

So TAT only reports 7 degrees lower at 1.8 than 3.0Ghz. But both are running at stock voltage, so i guess that is why.
I am thinking that the heatsink might not be even, or not contacting the CPU really well. I had to put more Arctic silver than normal to get it to fill up the whole copper bottom of the Stock HSF. Also, the Push Pins on the Stock HSF is not really good i think.

Anyways, If it runs 65 at stock i might as well run it at 3.0 at 70?

/Chris
 

marlinman

Member
Dec 10, 2006
160
1
81
I had to put more Arctic silver than normal to get it to fill up the whole copper bottom of the Stock HSF. Also, the Push Pins on the Stock HSF is not really good i think.

I just assembled a e4300/965p-ds3 rig and agree - the push pins are pretty sad. I also visited the AS website and found that for this cpu, we're instructed to lay a line of AS5 on the top of the heatspreader (core to core, if you get my meaning) and slightly twist the heatsink once it's secured (if possible) to weed out bubbles etc.
 

Binky

Diamond Member
Oct 9, 1999
4,046
4
81
Speedfan is underreporting by 15C. Have it add 15C so you can use it to monitor the temps real-time with it.

Orthos temps are not really real-world temps. If you hit 72C under Orthos (after 8 hours), you will probably never see 65C in real use. This is fine. You can run your CPU at 65C (or higher). Ignore the forum groupies that seem to think you need to idle at 30C and max out at 55C. ;)
 

o1die

Diamond Member
Jul 8, 2001
4,785
0
71
If you hit 72c, try using the case reset button, and immediately go into the bios using the "del" key on the lower right side of the keyboard. Check and see what the bios is reporting in the hardware monitor section. I don't trust most programs for accuracy. I had the best luck with mbm 5, but it won't work with any newer boards.
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: Binky
Speedfan is underreporting by 15C. Have it add 15C so you can use it to monitor the temps real-time with it.

Orthos temps are not really real-world temps. If you hit 72C under Orthos (after 8 hours), you will probably never see 65C in real use. This is fine. You can run your CPU at 65C (or higher). Ignore the forum groupies that seem to think you need to idle at 30C and max out at 55C. ;)

actually they would be right ... for SpeedFan temps ;)

when i had BIOS v 1.4 ... BIOS and TAT all reported the same
- after the latest BIOS flash to 2.4 [i think] ... TAT became 15 degrees higher [amd doesn't go up much under load] ... strangely SpeedFan reads about the same as BIOS [and shows the load increases in the elevated temps.]

since i like to see lower temps... i use SpeedFan ... from 40s idle to low 50s under load ... low/mid 60s with TAT under load - or at idle
:confused:

i think TAT is screwed up in my case

... and i guess they are ought toupdate it with Thermal Intelligent Throttling for the program ...
--that way we can have TIT for TAT :p

:D


sorry :eek:
 

lopri

Elite Member
Jul 27, 2002
13,310
687
126
From my understanding, Tjunction is the center of a core where it meets the IHS (under the IHS). Tcase is the center of IHS where it meets HSF and it's this value most motherboards' monitoring apps display, usually 15~20C below Tjunction.

Am I correct?
 

apoppin

Lifer
Mar 9, 2000
34,890
1
0
alienbabeltech.com
Originally posted by: lopri
From my understanding, Tjunction is the center of a core where it meets the IHS (under the IHS). Tcase is the center of IHS where it meets HSF and it's this value most motherboards' monitoring apps display, usually 15~20C below Tjunction.

Am I correct?

i believe so ... but i think my TAT is screwed up[period]

here you go this IS interesting ... definitive if it is correct

http://forumz.tomshardware.com/hardware...o-Temperature-Guide-ftopict221745.html
Core 2 Duo Temperature Guide Updated April 21st, 2007


Scope

The purpose of this Guide is to provide users with an understanding of thermal relationships, so that C2D platforms can be uniformly tested, properly analyzed, and accurately monitored. This Guide is not an Engineering Document, and does not address unnecessarily diverse or complex technical details. Certain strict definitions have therefore been relaxed to simplify concepts and enhance comprehension. This Guide applies to air cooled mainstream Desktop C2D processors.


Background

Intel has 2 distinct C2D thermal specifications, and provides a test program, Thermal Analysis Tool (TAT), to simulate 100% Load. Some users may not be aware that Prime95, Orthos, Everest and assorted others, may simulate loads which are intermittent, or less than TAT. These are ideal for stress testing CPU, memory and system stability over time, but aren't designed for testing the limits of CPU cooling efficiency.

Some users also may not know that C2D's feature 3 sensors at 2 different temperatures; a single on-die Tcase sensor (between the Cores), and dual Tjunction sensors (one within each Core). Consequently, there is much confusion regarding specifications, calibration Offsets and test methods, so results can be difficult to decipher and compare. Therefore, when expressing Idle & Load test Results, it's also necessary to define the Variables as:

Results

Tcase = Idle & Load
Tjunction = Idle & Load, Hottest Core

Variables

Ambient = Intake Temp
Chipset = Model
C2D = Model
CPU Cooler = Model
Frequency = CPU Clock
Load = Test Program
Motherboard = Model
Vcore = CPU Voltage


Specifications

Intel's Thermal Specification: http://processorfinder.intel.com/details.aspx?sSpec=SL9S8

Thermal Specification:

* The thermal specification shown is the maximum case temperature at the maximum Thermal Design Power (TDP) value for that processor. It is measured at the geometric center on the topside of the processor integrated heat spreader.

X6800 = ~ 60c, Vcore max = 1.3525, TDP = 75w
E6XX0 = ~ 60c, Vcore max = 1.3525, TDP = 65w
E4XX0 = ~ 60c, Vcore max = 1.3250, TDP = 65w

** For processors without integrated heat spreaders such as mobile processors, the thermal specification is referred to as the junction temperature (Tj). The maximum junction temperature is defined by an activation of the processor Intel® Thermal Monitor. The Intel Thermal Monitor?s automatic mode is used to indicate that the maximum TJ has been reached.

Ambient Temperature = 22c
Idle to Load Delta Max = 25c
Tcase to Tjunction Delta = 15c
Thermal Sensor Accuracy = +/-1c


Interpretation

* The first part of the spec refers to a single measuring point on the integrated heat spreader, which is in contact with the CPU cooler. Since there is no laboratory test sensor at this location, the CPU Case Thermal Diode is used to display the CPU temp in BIOS, where thermal tables are stored to emulate the heat spreader. This is the CPU die temperature measured between the Cores. Thermal Case Temperatures of 60c is hot, 55c is warm, and 50c is safe. The single CPU Thermal Case sensor is how Tcase is measured, and is the CPU temperature displayed in BIOS, Motherboard Utilities, and SpeedFan: CPU or Temp X.

** The second part of the spec refers to mobile processors without an integrated heat spreader, measured by internal Digital Thermal Sensors. Since Intel's Thermal Analysis Tool (TAT) is a Notebook tool, and desktop C2D's have an integrated heat spreader, TAT will typically indicate ~ 2c lower than SpeedFan. These are the dual Core temperatures measured within the hot spot of each Core. Thermal Junction temperatures of 75c is hot, 70c is warm, and 65c is safe. The dual Thermal Junction sensors are how Tjunction is measured, and are the dual Core temperatures displayed in TAT, and SpeedFan: Core 0 / Core 1.


Thermal Flow

Heat originates within the Cores, and is hottest where the dual Tjunction sensors are located. Heat is then dissipated throughout the CPU die to the socket and motherboard, and to the Integrated Heat Spreader, where the single Tcase sensor is located between the Cores, and the temperature is ~ 15c cooler. Heat is then transferred to the CPU cooler, and finally to air inside the computer case. All 3 C2D temperatures are then determined by computer case cooling efficiency and Ambient temperature. Regardless of Load, Tjunction is always ~ 15c higher than Tcase, and Tcase is always higher than Ambient.


Findings

(A) TAT can be used simultaneously with SpeedFan for benchmarking Desktop C2D's at 100% Load.

(B) The Delta between Tcase (BIOS, Motherboard Utilities and SpeedFan: CPU or Temp X) and Tjunction (TAT and SpeedFan: Core 0 / Core 1) is ~ 15c +/- 3c.

(C) 50c Tcase and 65c Tjunction are safe and sustainable temperatures.


Overclocking

Intel's Thermal Design Power (TDP) spec of 65 watts can be exceeded by over 50% when CPU frequency is aggresively overclocked, and Vcore is increased to maintain stability. Intel's Vcore max spec of 1.3525, when increased much beyond 10%, or 1.5 Vcore, makes it challenging to maintain safe temperatures with air cooling. As Ambient temperature increases, overclock frequency and Vcore may need to be decreased.

Every CPU is unique in it's overclock potential, voltage tolerance, and thermal behavior. If the maximum stable overclock is known at 1.35 Vcore, then each increase of .05 volts will typically allow a stable increase of ~ 100 Mhz, and will result in a corresponding increase in CPU temperatures of ~ 3 to 4c. Vcore is the most predominant Variable affecting temperatures.

At 1.35 Vcore, ~ 300 Mhz of additional overclock remains until safe temps are exeeded due to increased Vcore. Example; at 22c Ambient, if a C2D is stable at 3.0 Ghz - 1.35 Vcore - 45c Tcase / 60c Tjunction @ TAT 100% Load, then it may also be stable at 3.3 Ghz - 1.5 Vcore - 55c Tcase / 70c Tjunction @ TAT 100% Load, with highly effective CPU cooling and computer case cooling.


Tools

SpeedFan 4.32 displays all 3 Tcase and Tjunction sensors: http://www.almico.com/speedfan.php

Intel's Thermal Analysis Tool (TAT): http://www.techpowerup.com/downloads/392/mirrors.php

Orthos: http://www.techpowerup.com/downloads/385/Orthos_Stress_Prime_2004.html


Testing

Vcore = Manual
C1E / EIST = Disabled
CPU Fan = Manual, 100%
Computer Case Fans = Manual 100%
Computer Case Covers = Installed
Primary Test = TAT @ 100% 10 Minutes
Alternate Test = Orthos @ P9 Small FFT?s 10 Minutes

TAT will expose insufficient CPU cooling and computer case cooling, or excessive Vcore and overclock. At no other time will a CPU be so heavily loaded, or display higher temperatures, even when highly OC'd during worst-case / real-world loads. After CPU thermal behavior has been benchmarked by TAT, then Orthos or assorted other programs can be used with SpeedFan to observe less extreme CPU temps, while stress testing for system stability. Orthos Priority 9 Small FFT?s simulates 88% of TAT ~ 5c lower. During gaming and applications, Core 0 typically carries heavier loads and higher temps than Core 1.


Scale

The temp scale shown below illustrates the normal ~ 25c Delta between Idle and TAT @ 100% Load, and the typical ~ 15c Delta between Tcase and Tjunction on an example system overclocked with 1.4 Vcore. 50c Tcase and 65c Tjunction are safe and sustainable temperatures.

Tcase/Tjunction

--60--/--75--75-- Hot
--55--/--70--70-- Warm
--50--/--65--65-- N
--45--/--60--60-- O
--40--/--55--55-- R
--35--/--50--50-- M
--30--/--45--45-- A
--25--/--40--40-- L
--20--/--35--35-- Cool


Results

Tcase (Motherboard Utilities) = 30c Idle, 50c Load (SpeedFan: CPU or Temp X)
Tjunction (TAT) Hottest Core = 45c Idle, 65c Load (SpeedFan: Core 0 / Core 1)


Variables

Ambient = 22c
Chipset = 975X
C2D = E6600
CPU Cooler= AC Freezer 7 Pro
Frequency = 3.5 Ghz
Load = TAT @ 100% 10 minutes
Motherboard = Asus P5W DH
Vcore = 1.4

The typical ~ 25c Delta between Idle and Load will vary among systems due to inconsistencies such as Ambient temp, overclocking, Vcore, CPU cooling, computer case cooling, graphics cooling, and software processes. Example; if SETI, Folding, or excessive background processes are running simultaneously with Orthos rather than TAT, then "Idle" could be ~ 40%, and "Load" would be 88% of TAT, therefore, the Delta may be only ~ 10c.

The typical ~ 15c Delta between Tcase and Tjunction can be Offset due to erroneous chipsets, super I/O chips, BIOS releases, driver versions, and motherboard utilities. Intel?s spec for thermal sensor accuracy is +/-1c, so temperatures can be quite accurate on hardware / firmware platforms free of manufacturer's deficiencies. Temperatures which have Offsets can be analyzed, and when corrected, may still be accurate.


Heat Score

The following items will enable users to estimate cooling efficiency, identify problem areas, and visualize how environment and system configuration impacts real-world thermal performance.


...


Parameters

. . .

Troubleshooting
...

Offsets ...

[EDIT .... i better not quote all of this ... for copyright purposes ... read it all for yourself ... deleted]
phew !

and damn that IS interesting ... this info is now *bookmarked* for STUDY

occasionally THG posts a gem

and it looks like mine would be *normal*
--45--/--60--60--

kinda makes up for me