Real Temp - New temp program for Intel Core processors

[jaredpace]

Am I getting this right? I did what Wozzer did and used a different Winring0 to get it to work on Vista 64, but This doesn't seem right.

I'd say it's okay. with your sensors moving 18C, that should be good. Considering that unclewebb wrote that 0C movement is a bad thing and represents a stuck DTS, 18 would be good. i guess?

[TheScavenger]

Any news on when it will work with Vista 32?

Josh

[jason4207]

Q6600 G0 Stepping
1.335v bios voltage, 1.248 ~ 1.256v loaded with prime95 v25.6 SmallFTT test.
Real temp is reading about 5c lower, calibrated the readings using "0"

Is real temp showing me the correct temps ?

Do you see how one program uses 100*C Tjmax and the other uses 95*C Tjmax...there's your answer.

[xgman]

Any news on when it will work with Vista 32?

Josh

It already does, no?. . .

[alex17 GTX]

Great job unclewebb
Seems it's working on Vista X64 SP1 with the compiled drivers posted by WoZZeR999.
I attached two screenshots: first is on XP with RealTemp v.1.0 and the second is on Vista X64 with RealTemp v.2.1 modified.
Note: You must run it as administrator.

[theonlybabyface]

Off topic...but alex17 GTX where did you get the floating Vista bar on the bottom and left side from?

[alex17 GTX]

It's object dock with mac osX leopard skin.

[WiTcHkInG]

Works in vista x64

Moved the 64 bit versions of WinRing0x64 to the real temp folder and it worked, pretty simple really.


Great program unclewebb

edit: after reading back a page or two its allready been pointed out to work on x64. thought i might be first at something for once

[jaredpace]

T7100 Merom Core2duo

realtemp & coretemp.

I finally got it working by right clicking realtemp.exe and selecting run as administrator from within vista 32.

Thanks unclewebb! Rge - wish i had a heatgun!

[unclewebb]

I thought I'd take a day off and do some high temperature testing. No use including features in software if you don't take the time to test them out.

I swapped in my E6400 for a couple of trips up to TjMax and a little beyond.

Other software I have used (CoreTemp, SpeedFan) have a problem when you go beyond TjMax but I'm happy to report that RealTemp was the first program I've used under extreme circumstances that didn't choke. Here's the results:



Intel certainly doesn't support this but in testing I've noticed that the DTS simply wraps around when it reaches TjMax and begins counting down again from 127. Knowing this I built in some code to convert these readings into absolute temps that make sense. The log file is nice and smooth. SpeedFan and TAT will start to display minus 42C when this happens.

The little check marks lit up just as they were designed to do. When temperatures returned to normal, the history bit remained lit up.



Even restarting the program will not clear the history bit. You need to restart your computer. This is an excellent feature. Even if you don't have RealTemp running you can start it up after a throttling incident and it will display the history of it by showing a check mark in the right hand column for each core.

It's also possible that only one of your cores throttles like this.



On this E6400 this happens just before DTS=0. Your results may vary.
At least you've got some Xtreme software now for working in Xtreme situations!

[unclewebb]

I just updated the x64 install folder located here.

http://www.fileden.com/files/2008/3/...ealTemp_64.zip

I copied the x64 drivers directly into this folder this time which seems to be working for other users. You need to run RealTemp as Admin but hopefully someone can try this again to see if this is working properly so anyone with Vista x64 can use this without too much hassle.

[WoZZeR999]

I tested 'as is' from the zip and it didn't work. You need to include the WinRing0.dll file in that package. I believe Winring0.dll detects the x64 OS and then dynamically forwards all of the requests to the x64 conterpart.

[mrcape]

Great Program!

[unclewebb]

OK, I just uploaded the updated zip file. Can you try it one more time to see if it will work as is.

[WoZZeR999]

If you updated the one from post #286, I didn't see any changes and it still didn't run. All it needs is the winring0.dll file as well. You could try having it load winringx64.dll instead of winring0.dll.

[WoZZeR999]

I tried renaming Winring0x64, but It won't identify as WinRing0. I think the easiest way to make it work is to use Winring0.dll as well as the x64 parts.

Edit: The original Real Temp (non x64) works with x64, it just needs to have the x64 dll,sys and vxd included in the package.

[rge]

T7100 Merom Core2duo

realtemp & coretemp.

I finally got it working by right clicking realtemp.exe and selecting run as administrator from within vista 32.

Thanks unclewebb! Rge - wish i had a heatgun!

But you are showing something interesting nonetheless. If your cpu temp is accurate, the gradient between tcase and core at steady state idle, and at steady state orthos load, is so small it gets lost in measurement error, ie 1C or less. The test i linked to earlier with the thermocouple on core versus case temp showed the same, only 0.4C gradient at steady state idle.

Also, the fact your cpu temp scales with core from idle to load, irrespective of absolute cpu calibration, suggests the notion that large gradients of 10-15C occuring on orthos load versus idle, are suspect, and here we have a known tjmax.

Every piece of actual data i have seen has supported unclewebb's conclusion that less than 1C gradient exists between core and tcase at idle, and likely the same occurs at steady state moderate loads like orthos, and steeper gradients occur primarily when load is changing, or may take programs like TAT to load the cores to the point one sees a constant 4-5C gradient. But cant test to know if even that is possible on the new 45nm, since TAT is not compatible with them.

But yeah if you could measure tcase directly and show its same as core, I think that would put any remaining doubt to rest, but for me, not much doubt remains anyways.

[unclewebb]

http://www.fileden.com/files/2008/3/...RealTemp64.zip

The FileDen was putting up a fight. I'd change things and then it would dump my changes so I finally renamed it to trick it. Need one x64 beta tester. All the .dll files are all there, I promise!

[jaredpace]

are there any other test's i could do, like everest full display, or coretemp log's during prime to help support the theory?

I would like to contribute towards finding a working e8400 temp solution in any way possible. thanks again unclewebb & you too, rge!

-jared.

[i43]

thx unclewebb
ver 2.1 works great for me on vista Home Prem. (32bit )
Fujitsu Siemens Laptop mobile c2d T5500 (Merom) /i965 chip.
no problems with dll.. or what-so-ever. just extrct archive and reday to go

no worry but sensor check popup a message abut heat etc. but no prob. for me think it was programmed to run like this on mobile devices ..

[mrcape]

Here's mine -



Sensrors look okay right? I still don't have a great handle on this concept or time to research it.

[HDCHOPPER]

At least you've got some Xtreme software now for working in Xtreme situations!

"UNCLEWEBB"

[karateo]

here is mine

[WoZZeR999]

http://www.fileden.com/files/2008/3/...RealTemp64.zip

The FileDen was putting up a fight. I'd change things and then it would dump my changes so I finally renamed it to trick it. Need one x64 beta tester. All the .dll files are all there, I promise!

Works great, but you can use the regular (non x64 compiled) version. Since all of the calls that would be based on the driver, you don't really need to include the 'Real Tempx64.exe' unless you are planning on including some functionality that would be for the x64 version.

[Demo]

Do you see how one program uses 100*C Tjmax and the other uses 95*C Tjmax...there's your answer.

Yes, So realtemp it is. That is about 8~10c rise above ambient when my processor is idle, is that the normal range for the Q6600 GO's ?