Real Temp - New temp program for Intel Core processors

[HDCHOPPER]

ok got it ! it dont like me hahaahahaha thanks unclewebb !!

[dark22]

man you are my favorite person here!!!
thank you very good job!!!

[karateo]

great work!
i wish i could run it on vista 64

anyway, thank you for your effort

[ultimeus]

Does he works well with Q6600 G0 ? I have around 31° for cores in realtemp and 36° with coretemp.

Who's right ?

[TEDY]

works

coretemp shows 10°c more...then real temp.

[emoners]

thanks unclewebb!

[ixtapalapaquetl]

Fantastic work amigo!

The logger is working well. The .txt format is not the greatest for Excel, especially given that the values in the file are seperated by spaces. I opened log.txt with word, did a find " " and replace with "," and finally renamed it .csv. Now each measurement gets its own column.

I am going to start collecting data now. Will get back to you with my results.

Thanks again!

[mfcm]

im tryin it out now..hehe worried at 1st to prime 4.3ghz at 1.36v for my e8400 coz saw the temp quite high..prime 4ghz oredi pass 16hours for 1.248v..hehe its time for 4.3ghz..thx for nice software

[Hornet331]

first i want to say, very nice programm and thx for all the effort for getting the real TJmax.

and now i have a littel request: is it possioble that you add (in later versions of the program) the possibility to calibrate each cpu independently(core 0/1 and core 2/3) ?

Im on a quad (QX9650) one core reads "good" idel temps 27/26°C with ~22°C room temp, the other one reads 24/22°C with the same ambient temp.

right now the tool only gives me the possibility to change the calibration factor for the whole cpu.

works

coretemp shows 10°c more...then real temp.

well for everyone who has noticed this... why ddont you guys read unclewebb's first post?

unclewebb usees a Tjmax of 95°Cin his program rather then Tjmax of 105°C which coretemp and everest etc uses.

[eva2000]

Yeah individual core calibration would be nice heh.

System

• Intel Core 2 Duo E8500 Q740A493T 2L7**** / **0909
• CPU Cooling: Dtek Fuzion Rad Tower Box
• Asus Blitz Formula 1402 bios
• 128MB Gainward FX5200 PCI
• 2x1GB Corsair 10000C5D Dominator dual channel kit
• Memory cooling: 120x25mm Thermaltake 81cfm fan
• 80GB Hitachi 7K80 SATAII
• Pioneer DVD-RW
• 1200W Silverstone OP1200
• WinXP Pro SP2 Nlite Fully Updated.

Seems my favourite E8500 Q740A493T with uneven coretemps, is already calibrated correrectly according to Real Temp idle temp calibration method below. Core0 reads correctly within 3-5C of my 27.4C room temps at idle. But Core1 under reports temps by around 11-14C heh.

Setting 6x266FSB and lowest vcore 1.1v bios = 1.064v idle and 1.056v load with 27.4C room ambient temps.



With ++ calibration temps are closer to what Coretemp reports but core1 (2nd core) still under reports temps considering room temp is 27.4C



Prime95 25.6 Load temps with un-calibrated and calibrated idle temps respectively:





Test Sensor results





XS Bench

[Big Lar]

It Works! Thank you Very much unclewebb We needed this..Bad!

Larry

[graysky]

guys, the term 'calibrate' implies that you're going to adjust a reading due to an error. How exactly are you planning to make an individual core adjustment given that you have nothing to calibrate it against. In other words, how to plan to measure the individual core's temp?

[Big Lar]

guys, the term 'calibrate' implies that you're going to adjust a reading due to an error. How exactly are you planning to make an individual core adjustment given that you have nothing to calibrate it against. In other words, how to plan to measure the individual core's temp?

Good point, I am fairly sure that Idle temps will not be "Spot On", However to me anyhow, I could care less as what is important to me, is that Load temps are correct.

Larry

edit; I meant to say also that Real Temp does show my cpu as the E3110 that it is

[SubSurface]

BRAVO! and Thank You!

[rge]

Awesome program! Great idea to show both actual DTS and the temp, and Idle calibration is nice to!

Only issue I have is, I calibrated my cpu temp using lasertemp (+/- 1C accurate and checked against very accurate temp probe) so cpu temp in speedfan always reads within 0-2 C of IR to cpu from 45-95C.

If Tjmax on my E8400 is 95, then there is no gradient from core to tcase on full orthos load, as my calibrated cpu temp is now same temp as core. And I can show for example, that when speedfan cpu shows temp of 53, IR to back of cpu will be same.

Unclewebb, you could easily calibrate your cpu, if have not already, and test same, would be interested in what you get. When I reach 105C core, my tcase measured by IR was 94.8, so I got the same result as you. The only issue is, does any gradient exist between core and tcase at that time. Your control chip, one did not. On my E6850, a gradient of 5-7C existed at all times, and on my E8400 a gradient of 10C existed at all times.

[kadozer]

Works great! my q6600 @3.6 idle:

Anyone with an Abit p35 or quadgt wolfdale wanna try it out and postresults? Would be interesting to see with all the screwy temp reports from every other software.

[Xvys]

I got it installed now, but I'm getting core temps about 10C lower than Core Temp or Everest, 28C/22C? Upon calibration it says I have a bad sensor in Core 0. Hmmmm...

[Kilyin]

nifty program, reports 5 less than Coretemp.

My E6850 is very strange though. TAT reports 13 degrees lower than CoreTemp. Speedfan reports 15 degrees lower than CoreTemp. Everest reports same as CoreTemp. And now this reports 5 degrees lower than CoreTemp.

[Infa]

This is really good! thanks for the work... hope everyone now will trust this and look and these numbers instead of coretemp that has a to high tjmax setting... until he correct this problem, people are goin to use this i hope.. so we all can see correct numbers.

its been proven a couple of times that tjmax really are 95c.. so now be it! 105 are for mobile cpu's and not desktop.

regards.

[kadir_slayer]

here is what ı have



room temp 25-28 ı guess

[PhilDoc]

Very nice. Thanks for the new program and all your hard work. Haven't played with it too much but it does track very well with SG cpu temps under load.

[xploit]

It works fine on VISTA. good! job!!

[unclewebb]

Thanks everyone for the thumbs up! For my first stab in the dark at this issue I think things have gone well.

xploit: Very nice to confirm that it works on Vista.

Can anyone else with a Quad post an idle screen shot. The Test Sensors feature is definitely a work in progress and has only been tested on dual core chips so far.

We can all thank ixtapalapaquetl for his input on this project. He had some great ideas that I was able to implement without too much trouble.

It's quite easy getting the .txt log file into Excel. Using Excel 2003 all I have to do is open up Excel and go into the Open menu and adjust it so it can read All Files. Direct it to the RealTempLog.txt file and when it goes to open it up it will immediately convert it to something Excel is happy with. Click on the Finish button and you're good to go analyze the data.



I'm very interested to see how this new data correlates with CPU temp data.

eva2000: I wouldn't try to jump to too many conclusions when you have two cores that are reporting significantly different temps like your E8500 does. One or both sensors might have issues. When you are running over 4 GHz with some extra core voltage do they ever agree with each other?

If you click on the box in the upper right corner where it says CPUID = ***** it will show you what TjMax I'm using for your processor. I have dropped TjMax from 100C to 95C for some of the previous processors including the newer Quads based on published Intel TCase data. I haven't confirmed this yet with the IR thermometer but I probably will in the near future. I've decided to start with a clean piece of paper and not trust anything other programs say because they don't seem based on any real world testing.

Keep the suggestions and bug reports coming in and I'll do what I can to fix things up for you guys.

rge: I did all my IR testing with the CPU at idle. Even when it was over 100C it was still only at idle. Temperature gradients between TCase and core temp exist but I believe that when at idle, the difference between the true core temperature reading and what you can measure on the top side of the IHS is so small that it is within the accuracy of the test equipment espicially at lower temperatures where I was testing.

: "On my E6850, a gradient of 5-7C existed at all times, and on my E8400 a gradient of 10C existed at all times.

That's why I dropped TjMax for the E6850 from 100C to 95C and dropped the E8400 from 105C to 95C.

Xvys: Can you post a screen of the bad sensor report. It looks like your core0 is stuck at 28C.

eye get a run time error ..... maybe cause eye all clocked up to 4.0

This is the error a user will get if they didn't follow the directions in the first post and install the ia32 driver correctly.

[Kilyin]

Any chance of making the temps display in the system tray in a future release?

[ixtapalapaquetl]

We can all thank ixtapalapaquetl for his input on this project.

You are too kind - this was all you!

I am doing some initial analyses of the log file. There is an issue with Everest, where I am getting my CPU Temp readings from, since it does not read at exact intervals (generally it reads every 5 seconds like Real Temp, but there are frequent wild hairs where it took an extra second). This makes the already statistically risky comparison (due to using two different instruments) into a perhaps completely unreliable correlation. I will see if I can figure something out.

Their should be no problem, however, using average values over time. This still may be interesting. I will let you know if I get anything meaningful.

Once again, fantastic work!

Edit: Unfortunately, I can't use your importing technique because I want to consider the minutes and seconds in separate columns for sorting purposes. Stupid Everest. No biggie - took 10 seconds.