Real Temp - New temp program for Intel Core processors

[k4vz0024]

Here's a quick fix.

I'm not sure if there was one or two problems. So far I think I've fixed one problem. Version 2.58.1 is available for beta testing.
http://www.fileden.com/files/2008/3/...alTempBeta.zip

Try running an XS Bench without Gamer mode on and see if it works and then click on Gamer Mode and see if that causes a crash. Let me know what OS you're using. XP has no complaints but writing information directly on screen might be a no no in Vista. If it isn't allowed in Vista then I'll have to ax this feature.

msgclb: The Not Responding does happen sometimes during the XS Bench due to this process trying to hog the whole computer so the results are repeatable but obviously 2.58 had a big fat bug. I've changed the way RealTemp calculates MHz recently to try and get the CPU MHz option to recognize changes when SetFSB is used. That part is working well on my board and is just as accurate as CPU-Z but I'll have to wait for some feedback to see if it is working for everyone.

I try it Version 2.58.1 and XS Bench without Gamer mode, but it does not respond. I have XP pro with SP2 and Q6600.

[clokker]

clokker: You need to show me a couple of screen shots at idle and full load with RealTemp including the settings window opened up beside CoreTemp.

OK, will do- soon as I get my chip back in (testing a cranky CPU for a friend).

BTW...this new chip (also a q9300) has exhibited a weirdness with the sensors.
When first installed, I ran the "sensor test" and got movement between 3 and 4 across the board (this is clocked up from 2.5 to 3.3GHz).
Later, after a few hours of OCCT stressing, I rechecked them again and saw 2,0,6,7 (oddly, my Core 1 is the least responsive as well).

How does this test work and is it common to get different readings every time it's run?

[unclewebb]

k4vz0024: Does the XS Bench finally complete and give you a score. It's common for it to not respond sometimes during a test but it should complete the test. I'm using XP SP2 and a Q6600 like you and it completes each time. If it sits there and locks your computer up then there is a problem.

clokker: To increase the consistency of the Test Sensor feature you need to keep your testing environment consistent. That means same room temperature and core temperatures when you start the test as well as same core voltage and same core frequency. You also want to keep the background processes to a minimum so nothing is working in the background that might effect your results. All this test does is run the same calculation for a few seconds on each core. In theory, on a Dual or Quad core, each core should increase in temperature a similar amount. If you run the test once and then run it again before the processor has had time to get back to its previous idle temperature then you are likely going to have results that are not consistent.

You might also get more consistent results by turning off your calibration settings. Open up the Settings window and click on the Default button and then click on the Use button so it doesn't overwrite your calibrations. Do your testing and then quit the program and re-start it so your original calibrations are re-loaded.

Richard Dower: He's in the qualification and testing dept. and he swears by Real Temp..

That's good to hear. As long as he's not swearing at RealTemp, I'm happy.

A properly working Gamer Mode that doesn't flicker might happen sometime but it won't be for a while. I have some RivaTuner code that W1zzard told me about but it's going to take some time for me to understand it and maybe get RealTemp working with it. It's almost summer time and I can think of one or two better things to be doing.

Edit: I just ran Test Sensors on my Q6600 at 3000 MHz and 1.368 volts with the Tuniq fan on low. I ran the test 5 times but tossed one of my results out because it was exactly the same. I didn't want anyone to think I was just doing a Copy / Paste kind of test. I followed my suggestions above and got these results:



Seems like a pretty consistent / repeatable test to me.

[TMC2K]

I've read most of this thread and I have to say Great Job to Unclewebb . I was thinking about an appropriate award and noticed there is a precedent:

A Swiss patent office clerk had a general theory about time being relative and won a prestigous award. Almost a hundred years later Unclewebb proves temperature is relative with Intel playing the role of gravity!

I'm sure in a few year's time, we'll get you on the list for that award.

Thanks for a great program and a great thread to learn so much about computing temperatures.

[Warpcore]

Intel playing the role of gravity.

Is that French for being a Black Hole??

[k4vz0024]

[QUOTE=unclewebb;3019446]k4vz0024: Does the XS Bench finally complete and give you a score. It's common for it to not respond sometimes during a test but it should complete the test. I'm using XP SP2 and a Q6600 like you and it completes each time. If it sits there and locks your computer up then there is a problem.

Real temp 2.5 show me the results of XS Bench test without any problems. However, the 2.58 Version just frozen when I run the test. The Real temp 2.58.1 just not show results of XS Bench, but after time not respond.

Why?

Best regards,

[k4vz0024]

I have correct my self: I finally got Xs Score, but different results between Real Temp 2.5 and 2.58.1 for 30 scores less for 2.58.1. And I waiting more time with 2.58 Version of Real Temp.

[unclewebb]

k4vz0024: Version 2.58 is broken and should be deleted. It was quickly replaced by 2.58.1 to correct a bug. These are beta versions with code that I am testing out. Can you post a screen shot of 2.5 with an XS Bench, 2.58.1 with an XS Bench and CPU-Z on the same screen so I can see your problem better. You can upload images to somewhere like www.imageshack.us and then post a link here. Thanks. You can also just e-mail it to me if that is simpler: Real_Temp@yahoo.ca

TMC2K: Thanks. Maybe the new award should be called the FUD Busters Award for trying to get to the bottom of the confusion that Intel has created by being so quiet about this subject. Some smart ass should go create a new entry at Wikipedia.

"unclewebb has here by been nominated for a 2008 FUD Buster for his hard work in trying to make sense out of the black hole of missing documentation by a major corporation."

I'm honored!

[DGill]

Thanks for a great utility. I'm new to overclocking and feel pleased with my achievement (even if no one else is!) of getting an E8400 to 3400 mHz without blowing it up.

I've calibrated RealTemp as you suggest and at a room temp of 21C and a case temp of 25C (at inlet to CPU fan) RealTemp gives 32C for both cores at idle.

Not sure if I should use a -1 idle correction which puts the CPU core temp at 5C above inlet air temp to the cooling fan (sounds correct) or a -2 correction so the core temp is 5C above room temp?

Not a big issue, just curious.

[unclewebb]

DGill: I find the most accurate way to do my test is with the case open so the CPU fan is getting very similar air to your room air. When you have gradients between room temp and the temp inside your case, your calibration results probably won't be as accurate. Try testing again and see what you get. Probably a -1.0 calibration factor would work best.

Here is beta version 2.58.3. I tried to fix the XS Bench back up so it works like it did in version 2.5.
http://www.fileden.com/files/2008/3/...alTempBeta.zip

There have been 101 changes since version 2.5 so slight differences can be expected. If your XS Score has changed by 30 points then post a screen shot of both versions on the same screen so I can have a look. This new version might do the "Not Responding" a little less than before but it should still display a consistent result in the same amount of time even when this does happen.

During testing, I find it is repeatable enough that I can use it to quickly tell a difference when the FSB has been changed with SetFSB by as little as 1 MHz.

333 MHz x 9.0
15.227 s
15.224 s
15.227 s

334 MHz x 9.0
15.182 s
15.186 s
15.189 s

Also fixed in this version is that if you use Start Minimized without a TaskBar, it should go straight to the Tray area without anything briefly popping up on the screen like it's supposed to do.

[msgclb]

DGill:
Here is beta version 2.58.3. I tried to fix the XS Bench back up so it works like it did in version 2.5.

I don't see any problems here using Vista x32 with this Gigabyte system.

[unclewebb]

msgclb: I'm always changing something between releases and sometimes I'll change something while testing and forget to change it back. There was a problem with the original v2.58 but I think that is all taken care of now. Thanks for your testing.

[msgclb]

msgclb: I'm always changing something between releases and sometimes I'll change something while testing and forget to change it back. There was a problem with the original v2.58 but I think that is all taken care of now. Thanks for your testing.

I did see one minor glitch with v2.58.3. Do you see the same?


It is correct where it counts!

[unclewebb]

As I said, I always forget to do something.
Here's the late version of 2.58.3
http://www.fileden.com/files/2008/3/...alTempBeta.zip

[DGill]

Thanks for info Uncle, I'll redo the calibration test as suggested

I found no difference in xs bench between versions (E8400, GA-EX38T-DQ6 & Vista 32bit):

[gearhead]

I dont know about this new program, shows 35-40C at idle. Just installed an E8400. Watercooled. Smart guardian shows low 20's, bios hardware monitor screen confirms this.
My GPU is at 34C idle, (as usual) so I cant see my cpu at 40C at the same time.

[unclewebb]

gearhead: Did you calibrate this program to your new CPU as outlined in the docs? Do you know if the sensors in your CPU are sticking like many of them do on the new 45nm CPUs? You have to do some homework if you want accurate idle temperatures from these CPUs.

Here's a good place to start.
http://www.techpowerup.com/realtemp/docs.php

Try doing the Calibration test and post your results.

Edit: Here's one of my favorite tests of an E8400.
http://www.xtremesystems.org/forums/...&postcount=573

[Talonman]

2.58.1 Movement and XS Score:



2.58.3 Movement and XS Score:



Still looks good on a Q6600.

[jcniest5]

Any ideas why "Run as Administrator" still gives the error that a file is missing on this version anymore?

[loonym]

Any ideas why "Run as Administrator" still gives the error that a file is missing on this version anymore?

If you are using one of the beta packages and you didn't unpack to the original folder you will get that error.

[msgclb]

Any ideas why "Run as Administrator" still gives the error that a file is missing on this version anymore?

http://www.xtremesystems.org/forums/...postcount=1119

[unclewebb]

Hoping to get a new version uploaded to the TechPowerUp site in the next day or two. Testing is 99.9% done. Now I have to fix up the documentation.

[Grandpa]

Nice app any Idea of which is right I am running Vista 64bit. I know it is reading the wrong MHz on my Q6700 but I am not sure about the rest.

[W1zzard]

2.60 released:

http://www.techpowerup.com/downloads...Temp_2.60.html

[Talonman]

I got my copy!!