Page 1 of 1

Correct Tjmax

Posted: Sun Feb 26, 2012 4:36 pm
by abax2000
What would be the correct Tjmax for a Q6700 G0?

Re: Correct Tjmax

Posted: Sun Feb 26, 2012 5:14 pm
by The Coolest
Core Temp already has the best approximation of the TjMax value for these processors.

Re: Correct Tjmax

Posted: Sun Feb 26, 2012 7:41 pm
by abax2000
That is Tjmax=100C ?
With this I get 10C more than in Speefan (Core Temp reads 33-43 at idle).
Which is closer to truth?

Re: Correct Tjmax

Posted: Sun Feb 26, 2012 8:02 pm
by The Coolest
From numerous observations, what Core Temp shows is closer to the true value in most cases.

Re: Correct Tjmax

Posted: Sun Feb 26, 2012 8:58 pm
by abax2000
Thanx Coolest.
Do these temps make sense for a q6700 at 10*330 (Vcore 1.3625) or is it on the high side?

Re: Correct Tjmax

Posted: Sun Feb 26, 2012 10:46 pm
by The Coolest
It depends on your cooling and the ambient temp, but overall looks reasonable to me.

Re: Correct Tjmax

Posted: Mon Feb 27, 2012 8:09 pm
by GigaByte
Cause SpeedFan is wrong, Intel released the proper TjMax values over a year ago finally after much speculation. The remaining few that cry that even Intel is wrong with their product well then.. can whatever..

Re: Correct Tjmax

Posted: Wed Feb 29, 2012 2:08 pm
by abax2000
ok...
I rechecked my overclock with the default Tjmax (in CorTemp which is 100C).
Max temp is 74C (ambient 17-18C, Zalman aircooling).

1. Does this seem ok (safe) ?
2. [Gigabyte] where I could find the Intel released Tjamx ?

Re: Correct Tjmax

Posted: Wed Feb 29, 2012 3:55 pm
by The Coolest
A bit on the hot side, but still 25c under the tjmax should be fine as long as it's stable.
I'm assuming 74c with full load?

Re: Correct Tjmax

Posted: Wed Feb 29, 2012 6:51 pm
by abax2000
Sorry, I was not clear.
Yes, it is full load under Prime95-"Small FFT" (I think this is the hotter Prime95 torture test).
It worries me because I have read in various places that it is better to max at or below 62C.

Re: Correct Tjmax

Posted: Wed Feb 29, 2012 9:35 pm
by The Coolest
The lower, the better, but 75c is not critical.

Re: Correct Tjmax

Posted: Thu Mar 01, 2012 8:48 am
by abax2000
That's reassuring.

In the meantime, I did some googling on the Tj issue.
I realise that, besides all the debate on Intel published Tjmax, the important thing is the reported distance from Tjmax, as you previously stated.

BTW, Intel product spec pages currently give a Tcase=62.2C, while everywhere else I read that Q6700/Q6600 G0 stepping is 72C. Intel also gives 62.2C at the Q6600 page for G0, but there is no stepping distiction for Q6700. It seems that Intel figures are hard to get.
During testing, my CPU temp (assuming that this is Tcase) is 62C (while Tj is at 74C as previously reported). If the 62.2C Intel figure is valid, this is marginal. If the 72C is valid, then is more or less ok.
Interestngly, if you take into account the widely published difference between Tcase and Tjmax being 5-10C, this would mean for my proc a Tjmax=98-102C (core 0,1) and Tjmax=93-97C (core 1,2).

Reading around, I understand that ambient temp has a direct one-to-one effect on cpu temps.
Taking into account that current ambient is quite low and in the summer can hit 28-30C, it seems that the optimal upgrade for me right now, would be a better ventilated case and/or cpu cooling (since a new rig is not in the plans for the moment).

Re: Correct Tjmax

Posted: Tue Mar 06, 2012 11:11 am
by abax2000
So, finally I moved my system to a CoolerMaster HAF X, along with a Mugen 3.
For the time, it idles at 76C (distance to Tjmax) and goes to 53C (distance again) at full load (ambient 18C).
Much better...

Thanks Coolest for pointing me to the right direction.