Real Temp - New temp program for Intel Core processors

helmutcheese · 4 Mar 2008 at 00:44

Quoted :
This is the first public release of Real Temp which is a temperature monitoring program designed for all Intel single Core, dual Core and quad Core processors. Each core on these processors has a digital thermal sensor (DTS) that reports temperature data relative to TjMax which is the safe maximum operating core temperature for the CPU. As your CPU heats up, your Distance to TjMax will decrease. If it reaches zero your processor will start to throttle or slow down so obviously maximizing your distance away from TjMax will help your computer to run at full speed and more reliably too.

Other software like CoreTemp has taken this knowledge and tried to work backwards to convert DTS data to an absolute temperature which most users are more comfortable with and would like to compare.

If the DTS is reading 50 then you know you are 50 degrees away from the throttling point. If the maximum safe operating temperature for your processor (TjMax) is 85C and you are 50 degrees away then your processor must be running at an absolute temperature of 35C.

Absolute Temperature = TjMax - DTS

The formula is dead simple and you don't have to be smarter than a 5th grader to figure it out but there is one small problem.

Intel does not publicly document what TjMax is for desktop processors. TjMax is fully documented for their Mobile processors but not for their desktop Core processors. Without this reference point, your absolute temperature data could be meaningless. Intel has also confirmed that there is no secret bit hidden in these chips that a programmer can read to determine what TjMax really is. The only thing a programmer can do is guess at TjMax and if he gets lucky and the reported temps look more or less believable then everyone is happy. "

You can read the rest at URL bellow.

http://www.xtremesystems.org/forums/showthread.php?t=179044

Cob · 4 Mar 2008 at 02:41

Looks good. He seems to have put a lot of thought into it.

Just need a x64 Vista version now.

Tute · 4 Mar 2008 at 08:32

Cob said:
Just need a x64 Vista version now.

Yeah and this is why i'm avoiding CoreTemp atm. The guy who made it was talking about Vista support months ago, but it's all gone a bit quiet on that front.

Boah · 4 Mar 2008 at 09:09

Tute said:
Yeah and this is why i'm avoiding CoreTemp atm. The guy who made it was talking about Vista support months ago, but it's all gone a bit quiet on that front.

I'm running Vista Ultimate x64 (SP1) and CoreTemp 0.96.1 works without problem for me. I haven't tried it in Vista x86 but I assume it will work there as well. For 0.95.6 I had to download a "special" x64 download which contained a .bat file to run before running the executable (I forget the link, doing a websearch for 'CoreTemp x64' should give you what you want if you're running Vista x64).

Real Temp looks interesting - I will try it once there is Vista x64 support.

Nick1881 · 4 Mar 2008 at 09:46

Just tried it with my Q6600, interestingly it shows 6-7 C cooler than Everest (which for me reads the same as coretemp.)

mOdO · 4 Mar 2008 at 09:52

Quite helpful. Thanks.

DavidB · 4 Mar 2008 at 10:22

Heres a screen shot comparing a few temperature testers on windows xp i just did, to be honest i think this program is showing ~10c too high. Especially when compared to other programs.
This is an E2140 @3.2ghz with a tuniq tower btw.

Tom|Nbk · 4 Mar 2008 at 11:35

So what makes this better than Core Temp, can anyone clue me in.

I've just ran a load test it seems to report 10c lower than Core Temp.

The temperatures do seem a lot more realistic with Real Temp though. Have we finally got a program to read the temp correctly?

How is this guy calculating the TJMax?

Tom|Nbk · 4 Mar 2008 at 11:40

Ahh it seems he has done some testing to find out at exactly what temperature the chips throttle, seems the Core Temp guy was incorrect to assume the TJMax is 105c when it is infact 95c as this guy has proved.

EDIT : Just read up on the link, looks like im binning Core Temp.

BEn_2600+ · 4 Mar 2008 at 11:48

Dosent work with Xp x64 i take it?

mgreaves · 4 Mar 2008 at 12:09

The problem for all these programs is that no one knows what Tjmax really is, and probably never will. Surely the sensible thing to do is ignore the guestimate core temperature that these programs come up with and use the DTS value when overclocking?

MoRT489 · 4 Mar 2008 at 12:12

mr_x_plosion said:
Just tried it with my Q6600, interestingly it shows 6-7 C cooler than Everest (which for me reads the same as coretemp.)

Rapier couldnt help but notice the sig, are you/DA still playing RvS?

Tom|Nbk · 4 Mar 2008 at 12:20

mgreaves said:
The problem for all these programs is that no one knows what Tjmax really is, and probably never will. Surely the sensible thing to do is ignore the guestimate core temperature that these programs come up with and use the DTS value when overclocking?

Erm this guy has ran his E8400 without a heatsink and found the chip throttled at 95c.

He's found out firsthand, I trust him the most at the moment.

pastymuncher · 4 Mar 2008 at 12:24

I just installed this and it shows temps way lower than coretemp with my E4300. In fact i have different temps reported with different programs. I know it's cold up here, but not that cold. Have a look:-

Tom|Nbk · 4 Mar 2008 at 12:30

You obviously didn't read up on the link Pasty

You are supposed to tweak the idle temperature..

pastymuncher · 4 Mar 2008 at 12:35

Tom|Nbk said:
You obviously didn't read up on the link Pasty

You are supposed to tweak the idle temperature..

Nope, did'nt read all of it. Started reading but then got fed up and just downloaded it.

I will go back and read again.

EDIT*** I just hit the ++ thingy and it now says 15/16 degrees but i noticed that none of you in your screenies have it set to ++?

mgreaves · 4 Mar 2008 at 12:48

Tom|Nbk said:
Erm this guy has ran his E8400 without a heatsink and found the chip throttled at 95c.

He's found out firsthand, I trust him the most at the moment.

This same method has also been used to get the Tjmax for other programs, and yet the measured values are all different. The problem is that this method always includes some error, and on top of which there will always be variation between individual CPUs. I am suspicous that the variation between CPUs might be much higher than people expect, which is why Intel don't publish it.

Tom|Nbk · 4 Mar 2008 at 13:16

mgreaves said:
This same method has also been used to get the Tjmax for other programs, and yet the measured values are all different. The problem is that this method always includes some error, and on top of which there will always be variation between individual CPUs. I am suspicous that the variation between CPUs might be much higher than people expect, which is why Intel don't publish it.

No, I spoke to the Maker of Core Temp over MSN, he said that his basing of the TJMax came from some Intel documents yet Intel have never publically released these documents so I'm questioning the credibility of that. Not sure about other programs. But it's stone cold fact the chips throttle at 95c as this is what the guy tested with.

The TJMax does not vary it is a fixed value.

i.e The TJMax will never vary for the E8000 series they will always throttle at that temperature it's just a case of finding out which, which is what Real Temp guy has done.

fornowagain · 4 Mar 2008 at 14:10

Tom|Nbk said:
No, I spoke to the Maker of Core Temp over MSN, he said that his basing of the TJMax came from some Intel documents yet Intel have never publically released these documents so I'm questioning the credibility of that. Not sure about other programs. But it's stone cold fact the chips throttle at 95c as this is what the guy tested with.

The TJMax does not vary it is a fixed value.

i.e The TJMax will never vary for the E8000 series they will always throttle at that temperature it's just a case of finding out which, which is what Real Temp guy has done.

Yes Tjmax is the maximum temperature the various junction points can reach. Unfortunately Tjmax is not the throttling point. The throttling point is a factory calibrated set point called TCC, placed to take account of hotspots and hysteresis (delays in the temp reading to response) and its different for each chip. Each has a different thermal profile as seen from the VID's, so an absolute value is a fudge. More than likely within an acceptable range, although that is unproven. The other thing is that is incorrect is the readings from the IR gun, there has to a temperature gradient from the IHS to the hottest part of the core. If he's measuring 95C as a throttle point its because the TCC is lower that the Tjmax. Just basic thermodynamics, within a homogeneous material there has to a temperature gradient from the hotspot to the point of measurement. Additionally there is also an interface to the solder at the IHS, so a thermal resistance is also overlooked.

Some quotes from the intel specs.

Readings from the DTS are relative to the activation of the TCC.The DTS value where TCC activation occurs is 0 (zero).

The temperature at which the PROCHOT# signal goes active is individually calibrated during manufacturing. Once configured, the processor temperature at which the PROCHOT# signal is asserted is not re-configurable.

In the event of a catastrophic cooling failure, the processor will automatically shut down when the silicon temperature has exceeded the TCC activation temperature by approximately 20 to 25 °C. The temperature where the THERMTRIP# signal goes active is individually calibrated during manufacturing and once configure can not be changed.

Borrowed this from the tech docs, the Tjmax IS the hotspot. Dependent on workload and temperature offsets. There are quite large gradients from the Tjmax to sensor readings. I've asked him on his approach to this, no comment. Personally I'd still pay more attention to the raw DTS values than an assumed absolute value, which by its nature will always be inaccurate.

helmutcheese · 4 Mar 2008 at 14:18

Boah said:
I'm running Vista Ultimate x64 (SP1) and CoreTemp 0.96.1 works without problem for me. I haven't tried it in Vista x86 but I assume it will work there as well. For 0.95.6 I had to download a "special" x64 download which contained a .bat file to run before running the executable (I forget the link, doing a websearch for 'CoreTemp x64' should give you what you want if you're running Vista x64).

Real Temp looks interesting - I will try it once there is Vista x64 support.

Hmm you seem to have that backwards m8, Coretemp does not work in Vista64 without rebooting and choosing no driver signing option or using a Batch file to load Coretemp.