My Tjunction cores just changed order! Defect going on?

[Rocky90]

This is strange.

I'm currently testing what the min voltage is for my Q67@3,2. My PC just crashed as I had too little voltage.

Strangely enough, my Hardware Monitor CPUID would suddenly read them #1#2#0#3, restarting, MISSING 1 CORE and now it says Core #0#3#2#1

I have the usual 2 Hotspot cores on my Q, which are no longer 1 and 2 but now are 0 and 3. This comes to a better spot for the eye lol but still this is strange.

I was wondering wether this could indicate an internal defect going on? Now too little voltage doesn't damage my CPU or anything else...

 

[Lupiron]

Naw, it has been know to do that. I have no clue why it does it, but I have several q6x00s and that happens every once in a while. It doesnt mean anything.

Wanna do me a favor and DL Core Temp, that will give you a nother way to watch them, as well as I would like you to list the VID value of your chip here, if you dont mind? Also, if you bought it as a retail box, or whatever and when!

Thanks!

--Lupi!

 

[CompuTronix]

Real Temp also shows the VID. http://www.techpowerup.com/realtemp/

 

[kad]

RealTemp is reporting core temps 10 degrees less than all other software including CoreTemp
Which one we can rely on??
10 Degrees can make a lot difference

 

[Andrius]

The #core "randomization" is annoying. It happens to me when I put my Q6600 and P35 to 400MHz FSB(I have to raise voltage on (G)MCH and ICH) but goes away if I run them at 367MHz FSB.

I think this is the first sign of instability as random other stuff starts to work strangely at times (linux ethernet failiure, slow USB transfer rates, programs opening slower, ...).

@kad
It's always better to take the higher temperatures as accurate. Just to be safe.

 

[kad]

Andrius:
Thanks,I agree with you

 

[CompuTronix]

Andrius and kad, Real Temp 2.5 will typically indicate 5c cooler on a Q6x00 G0 and 10c cooler on 45nm processors. If you take the time to read all 40 pages of the Real Temp thread - http://www.xtremesystems.org/forums/showthread.php?t=179044 - you may understand and appreciate the validity of the research behind Real Temp, and the sensor linearity problems that it corrects, which the other popular temperature monitoring utilities fail to address.

As the author of the Core 2 Quad and Duo Temperature Guide - http://www.tomshardware.com/forum/221745-29-core-quad-temperature-guide - I endorse and highly recommend Real Temp.

Comp

 

[Andrius]

@CompuTronix
I've read most of your guide in detail and I know how digital temperature sensors work (from own experience - no electronic component is completly linear over a 80°C range but every DTS manufacturer sells them as such).
While RealTemp might be more accurate (and I highly respect your opinion on that) it's safer for us noobs to not go near that accurate (IMO). As you've probably encountered countless times yourself (people do this with almost every limitation) when you say 70°C is the maximum safe temperature they pass their own judgement and call it something like 75° must also still be quite safe. My less (less then me) knowledgable PC user friends tend to do this.
Aside from that, thank you for the heads up on RealTemp (I'll keep it in mind) and read the thread when I have some extra time.

 

[CompuTronix]

@CompuTronix

While RealTemp might be more accurate (and I highly respect your opinion on that) it's safer for us noobs to not go near that accurate (IMO). As you've probably encountered countless times yourself (people do this with almost every limitation) when you say 70°C is the maximum safe temperature they pass their own judgement and call it something like 75° must also still be quite safe. My less (less then me) knowledgable PC user friends tend to do this.

Andrius, I understand completely. Since many users are so brainwashed on Core Temp, they don't realize that there are 2 sets of temperatures (CPU and Cores) at 2 different levels, so they tend to fling temperatures around like so much gorilla poo. Additionally, they don't understand that Intel's specifications for Maximum Case Temperature means CPU temperature (Tcase) and NOT Core temperatures (Tjunction).


From the Guide:

Section 5: Findings

(D) Intel shows Maximum Case Temperature (Tcase Max) in the Processor Spec Finder, which is the only temperature that Intel supports on Core 2 desktop processors.

Section 6: Scale

... as Tcase Max will be exceeded before Tjunction Max is reached, Tcase Max is always the limiting thermal specification.


Comp