• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Q6600 TJunction

Sovereign

2[H]4U
Joined
Mar 21, 2005
Messages
3,098
What is it? I know the author of RealTemp claims it's 95C, but other people say it's 100C. What do you all use (since CoreTemp can be adjusted now)?
 
For my B3 Q6600 I use 85ºC as it gives me about the same temperatures as my friends G0 stepping. For my QX9650 I use 95ºc
 
I have a G0, some guy who claims to work for a "large IT company" (note how I can't verify his claim) is insisting it's 100C. I have mine set to 95C presently in CoreTemp...
 
The temp set at the factory at which the processor asserts a #PROCHOT interupt (AKA Tjunction, AKA Tjmax) is undocumented. It varies from CPU model number to model number from stepping to stepping and perhaps (not documented) even CPU to CPU of same model and stepping.

Only by experimentation can it be determined for a specific CPU. Pick a number that makes you happy, I am not going to type the details up yet again. The "Delta to Tjunction" is the thermal headroom in Deg C until #PROCHOT, the rest are just nice pretty numbers to make you feel good, so pick what you want.
 
My B3 is 85C. I worked this out by using an inline water temp guage and really underclocking my CPU and stopping all the fans.

This made the core temps within 1C of the water temp as the temps slowly raised over time and the system was idle.
 
Can someone explain this all to me? Why is Tjunction important in regards to monitoring temps?
 
The T Junction Max Temperature is the temperature at which Core CPUs are designed to throttle down in order to prevent damage. There is a lot of discussion about it as Intel doesn't disclose exactly what this temperature is for its desktops CPUs.
 
The T Junction Max Temperature is the temperature at which Core CPUs are designed to throttle down in order to prevent damage. There is a lot of discussion about it as Intel doesn't disclose exactly what this temperature is for its desktops CPUs.

I know what Tjunction is, but why does it matter if we're trying to avoid getting temps that high anway?
 
I know what Tjunction is, but why does it matter if we're trying to avoid getting temps that high anway?

There are some like the [H] forum member uncleweb who wrote Real Temp who believes that we are too obsessed with low temps. Running at -20C TjMax should be just fine and that Core 2 processors are very robust. They are designed to protect themselves and that trying to keep it ultra cool isn't really all that necessary.
 
on my q6600G0 the tjunction is set at 100c, this is the valus that came with the processor, does this mean that my processor is safe from damage from heat as long as I keep it below 100c?
 
I think the point you're missing is: no one outside of Intel knows the actual real Tjunction temp on any given Q6600, and Intel ain't telling. So we're all just using 95C or 100C as a basis for comparison - but again, no one outside of Intel knows EXACTLY what the Tjunction temp of any given specific Q6600 is.

That's part of what makes this whole temp reading process so iffy in the first place, we don't know the exact temps, ever. I have a Q6600 G0 stepping and RealTemp comes with 95C as Tjunction - I see no reason to use anything else so that's where I'll let it as default.

As for us being obsessed with low temps, or temps in general, I would have to agree. Build the box, find a stable configuration, and stop fucking around with the damned machines and use them for something.

"How fast is fast?"
 
I think the point you're missing is: no one outside of Intel knows the actual real Tjunction temp on any given Q6600, and Intel ain't telling. So we're all just using 95C or 100C as a basis for comparison - but again, no one outside of Intel knows EXACTLY what the Tjunction temp of any given specific Q6600 is.

That's part of what makes this whole temp reading process so iffy in the first place, we don't know the exact temps, ever. I have a Q6600 G0 stepping and RealTemp comes with 95C as Tjunction - I see no reason to use anything else so that's where I'll let it as default.

According to the Coretemp developer, Tjunction value is stored in an MSR and Tdelta is stored in a second register. Read Tjunction MSR, subtract Tdelta and you have the core's temp. Seems straightforward enough.
 
This is one where published does not equal observed. Have read 100 deg C TJ, but yet I've noticed throttling around 85 deg C on performance indicators of my G0 Quad FWIW. Plus there's no way in holy hell I'm letting a processor get anywhere 100 deg C unless it's sleeping with my girlfriend.
 
Back
Top