CoreTemp vs. RealTemp - Page 10 - Overclock.net

h00chi3 · 03-10-08

Attached is my REAL vs. Core

I wish real temp was right, because I want my load temps to be within 9 degrees of ambient.

23°C in here, load temp 33°C, that would own.

Everest and Coretemp are what I am sticking with.

opt33 · 03-10-08

Quote:

Originally Posted by h00chi3

Attached is my REAL vs. Core

I wish real temp was right, because I want my load temps to be within 9 degrees of ambient.

23°C in here, load temp 33°C, that would own.

Everest and Coretemp are what I am sticking with.

If your cpu temp of 35C (approx tcase) is accurate (either by chance or if you calibrated at ambient using stock or underclock settings), then you are right in sticking with tjmax 100, because core temp can not be less than cpu temp.

h00chi3 · 03-10-08

Though it would be cool if my cores were cooler then my cpu, I would call it reverse cooling, in which you cool pass the IHS directly to the cores without having a naked chip.

GigaByte · 03-10-08

B3 and G0 quads are Tj max 100c.

GigaByte · 03-10-08

Quote:

Originally Posted by opt33

If your cpu temp of 35C (approx tcase) is accurate (either by chance or if you calibrated at ambient using stock or underclock settings), then you are right in sticking with tjmax 100, because core temp can not be less than cpu temp.

Can you please tell that to the CoreTemp devloper? I tried and tried and he still thinks M0 steppings are Tj max 85c, which makes cores cooler than Tcase

I am slowly leaning towards Tj max 95c for M0 steppings, not positive yet.

Schroedinger · 03-10-08

I'm relatively certain that Tjmax for c0 45nm processors is less than 105 C, for two reasons:

1. Theory: 105 C is the number introduced by Intel for Mobile processors, which are generally required to operate at higher heat over long periods of time, and would therefore be expected to function (indeed, need to function adequately) at higher temperatures than desktop processors with giant fans and heatsinks. Intel has said that desktop processors vary (potentially by processor, rather than stepping or SKU) in Tjmax, but this evidence at least implies that Tjmax (C0 45nm Desktop) < 105 C.

2. I get Coretemp readings at 105 C Tjmax of 47-52 C at idle (undervolted as in my sig but at stock speed with Speedstep off), while in BIOS my case temperature is ~27-29 C. I think this is indicative that 105 C is too high as well, although I am unsure how much too high. I'm going to try some underclocking tests to find out. I may be the proud owner of the very first 1.8 Ghz 45nm processor in the history of the world tonight!

opt33 · 03-10-08

Quote:

Originally Posted by GigaByte

Can you please tell that to the CoreTemp devloper? I tried and tried and he still thinks M0 steppings are Tj max 85c, which makes cores cooler than Tcase

I am slowly leaning towards Tj max 95c for M0 steppings, not positive yet.

LOL...not going near that one. After reading some 40+articles on intel thermal documents, many published just in last few years, it is clear intel is making silent revisions all the time in their manufacturing process. And no doubt some of these silent revisions are within the same stepping. Given I have seen the same argument with certain other cpus, both side showing plausible cpu/core correlations at different tjmaxs, makes me think that one possibility is that two different tjmaxes occur in some cases with same stepping. Intel has said, this "may" occur.

I am using speedfan (calibrate, core..) and setting it to tjmax I measure. I am not using any program that I cant calibrate both cpu and coretemp.

But I would be happy to suggest adding a calibration to the coretemp author, as I think that would make both sides happy (those arguing higher and those lower on certain cpus) , and possibly both sides correct.

TheSubtleKnife · 03-10-08

Quote:

1)set your cpu at 1.15 or 1.2 volts (less for 45nm), 6x200 mhz, and boot windows, and keep at idle.

2) check cpu temp, and if reads 20 and ambients are 25, internal ambients 26, then you need to always add 6C to cpu, speedfan can do this, or just remember the difference.

3)Then you really dont need to run temp all the way to DTS=0, just run up to DTS=30 or so (change coretemp settings to read distance from tjmax, ie DTS). This needs to be done at idle to prevent any significant gradient, with heatsink off or at least with fan off (otherwise temps wont reach as high as necessary at idle). Dont use orthos, as at load you may create a gradient from core to cpu. If cpu temp is 65, when DTS=30, tjmax is 95. But basicically just have to add DTS displayed (accurate regardless of what program you use) to cpu (calibrated temp) at idle. This will not work at low temps, as DTS sensors can be inaccurate at low temps, so need to get DTS up to around 35 to 25.

I will do this for the benefit of OCN...lol. I'm not really worried about killing this chip. I got it for $50, and its a meh OCer anyways, so i'll calibrate at lowest idle volts and then run the DTS as high as i can

Edit: what prog should i use to stress?

Xombie · 03-10-08

Coretemp would be correct since you'd very seldom hit 61C with that CPU.

TheSubtleKnife · 03-10-08

okay, i have a termometer that is i have set up measuring ambients, and the temp of the heatsink. be back soon with results

		Overclock.net - Overclocking.net > Intel > Intel CPUs
CoreTemp vs. RealTemp

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)