CPU Temperature measuring programs!

[Pendergast]

Solve this!

http://img153.imageshack.us/img153/4665/20354024ur5.jpg

This is a regular screenshot, with greyed out parts you don't need to see and amplified parts that you do .
If you vote, write the reason why you think that those particular programs are correct.
Please, do NOT vote or express your opinion based on something you've read somewhere. There is a big difference between theory and reality 😉.
Thanks for the replies .

 

[Shadow703793]

Neither. Best method is to calibrate SpeedFan (see C2D Temp guide by CompuTronix) and RealTemp. Also we can't really help without knowing CPU.

 

[MrCommunistGen]

Yeah... why would you cover up what CPU you have? The 43C figure seems to be more realistic, but if that is acurate is also questionable.

-mcg

 

[monst0r]

Yeah... why would you cover up what CPU you have? The 43C figure seems to be more realistic, but if that is acurate is also questionable.

-mcg

Also not understanding why the CPU type is so secret. Can't really help without that info. Try RealTemp, see if any cores are stuck...Post a useful screen shot as well.

 

[nukemaster]

The 15c offset is because of a TJ max on the cpu is 100c according to PC Wizard and HW Monitor and 85 according to CoreTemp and Speedfan.

The cpu does not give its temp, but how far it is from its max. since the software is using different max's you get a 15c offset.

Software: Hey cpu how far
cpu: 58 away
Software: TJmax - 58 = temp

so if its 100c then you get 42 (100-58=42)
if its 85 then its 27(85-58=27)

What cpu is it so we can try to find that max. When the Q6600 came out this was a problem too since its TJmax was up to 100 from 85 and most software was 15c off

Its most likely because one software does not know the temps. If thats idle and with speedstep I would almost have to believe the 27c.

Both temps are more then cool enough to be safe.

 

[MrCommunistGen]

The 15c offset is because of a TJ max on the cpu is 100c according to PC Wizard and HW Monitor and 85 according to CoreTemp and Speedfan.

Heh... I don't see why I missed that.

-mcg

 

[Pendergast]

http://img384.imageshack.us/img384/9315/49942140fe9.jpg

PC Wizard + HW Monitor vs. CoreTemp + RealTemp + Speedfan

NOTE: The green boxes show that Speedfan measures the same Processor Temperature as PC Wizard, even though the core temperatures differ.

 

[nukemaster]

The 1c difference is just a polling time

The temp with the green box is either Tcase or socket temp. Thats why its lower. That is how temps where taken before(socket sensor) integrated sensors.

the 15c is as mentioned above.

27 looks low, but it depends on your room temps and cooling.

My Q6600 @ 3.0 and 9500AT idles 30-35 depending on room temp(lower if i get the room under 22c) with speed step on.

i see no reason why an e2180 can not get down to 27 idle @ 3.2 with a good heatsink.

 

[kad]

Non of them
From My experience only Real TEMP 2.5

E 8500 OC'd 3.8 Temps Under Full load Prime 95

 

[Pendergast]

The point of the green box was to show that even though PC Wizard and Speedfan show different core temperatures (43 and 27 degrees Celsius respectively), they do show the same socket temperatures (37 degrees Celsius). Still the question remains - which measurement of the core temps is correct? I'd lean to PC Wizard and HW Monitor because I don't see how an Arctic Freezer 7 Pro with MX-1 could cool down an E2180@3.2GHz@1.35Vcore to 27... There was a little trick with my case that i performed which lowered the core temps by 10 degrees Celsius at both idle and load, but I still think 27 is a little too low. Ambient temperature is around 22 degrees Celsius. Do you want to try and shoot a more accurate TJMax?

 

[nukemaster]

Those 2 temps that line up are because they come from a chip on the board. the cpu does not only because each app is using a different TJmax

How close you are to TJmax is not changing, just the math used to give you a temp.

If you look at the core2 temp guide it looks like your cpu has a TJmax of 85, that would make speedfan right. I have no clue if Intel has changed anything(i would not put it past them to change something in mid production of a product).

Section 6: Scale

Safe and sustainable temperatures vary according to Spec#. The temperature Scales shown below illustrate the Delta between Idle and Load, and the 5c Delta between Tcase and Tjunction. Although the 5c Delta is relatively consistent, Tcase and Tjunction tend to converge at Idle and diverge at Load due to Variables such as Vcore and CPU cooler efficiency. Low Vcore and clock may cause Tcase to Tjunction Delta to indicate as low as 3c at Idle, while high Vcore and overclock may cause the Delta to reach 7c at Load.

If temperatures increase beyond Hot Scale, then ~ 5c below Tjunction Max, Throttling is activated. The Digital Thermal Sensors (DTS) are used to trigger Intel`s TM1 and TM2 technologies for frequency, multiplier and Vcore Throttling within individual Cores. If Core temperatures increase further to Tjunction Max, then Shutdown occurs. Since Tcase indicates CPU Die temperature only, it is not used for Throttle or Shutdown activation, however, as Tcase Max will be exceeded before Tjunction Max is reached, Tcase Max is always the limiting thermal specification.

Use CPU-Z (see Section 8) to read processor information including the Revision field below the Stepping field, then choose a Scale below which matches the CPU being tested. Scales are ordered from highest to lowest Tcase Max.

Scale 1: Duo
E7200: Tcase Max 74c, Stepping M0
E4700: Tcase Max 73c, Stepping G0
E4x00: Tcase Max 73c, Stepping M0
E2xx0: Tcase Max 73c, Stepping M0
E8x00: Tcase Max 72c, Stepping C0
E8x90: Tcase Max 72c, Stepping C0
E6x50: Tcase Max 72c, Stepping G0
E6540: Tcase Max 72c, Stepping G0

-Tcase/Tjunction-
--70--/--75--75-- Hot
--65--/--70--70-- Warm
--60--/--65--65-- Safe
--25--/--30--30-- Cool

So 5c past hot is 5c below TJmax, that says 85 is MAX,
Just keep your self 35-40 away from TJmax that will leave you a safe margin.

Idle

 

[Pendergast]

Sorry, but it just doesn't fit. I've looked at the guide and did some math for comparison:

For example L2 stepping max Tcase 61c, M0 stepping max Tcase 73c
Section 7: Parameters: "(H) Tcase to Tjunction Delta is typically 5c +/- 3c during Prime95 Small FFT`s." L2 stepping's Tjmax is 100c, so if you say M0 is 85c then:
for L2: 100c - 61c = 39c - 5c = 34c -> L2 can go 34c over Tcase before shutting down? No way.
for M0: 85c - 73c = 12c - 5c = 7c -> so M0 can only take 7c more before shutting down? Remember: L2 stepping max Tcase 61c, M0 stepping max Tcase 73c -> that means, M0 can take more heat than the L2, but if you say that M0's Tjmax is 85c, that would be wrong (34c>7c).
Bottom line: I doubt very much M0 stepping (my E2180 too) has a Tjmax of 85c. I'll look into this more in-depth though.

 

[nukemaster]

i give up and i also see no reason why you take the maximum die temp and subtract the maximum temperature measured on the top center(thats all tcase is, and its calibrated[by the mb makers, in some cases they just use another sensor to estimate] and not always accurate).

Intel has NEVER released a TJmax for the core2 cpus since they have a heat spreader. all the software does is try to be as close as they can(thats why some have a TJmax of 100, 85 and even 95 for realtemp).

When there is a heat spreader they pick a max temp at the geometric center on the topside of the processor integrated heat spreader that should not be exceeded with the cpus thermal design power.

do not worry about it, you temps are fine

 

[Pendergast]

I did it to see how much a core can go over Tcase to shut down. 34c is way too high for L2 stepping and 7c is way too low for M0. Now I finally see how it works. If we let Tjmax for L2 be 85c instead of the M0, and the M0 95c, we get these results:

85c - 61c = 24c - 5c = 19c -> L2 cores have to go 19c over tcase to shut down
95c - 73c = 22c - 5c = 17c -> M0 17c, that seems a lot more reasonable than the calculation in my previous post, see what I mean? Now, the steppings have a 12c Tcase difference between them and the cores have a 2c difference between them => M0 can clearly take more heat, which finally makes sense. That means, Tjmax for M0 is 95c, not 85c! (and 85c instead of 100c for L2, but that is irrelevant right now).

MUCH more reasonable, I think you see what I mean 😉.
Btw, I know that my temps are fine, I knew from the beginning, I just want to know what the fu*k's going on so to speak. Thanks for the ideas.