RealTemp General Discussion

[MaxDrDiablo]

Hi everyone, first time posting but have been tracking this forum for a while now...i have read all there is to read from Unlceweb's genorous information..very helpful. i recently setup my i7 920 and have soley been using Realtemp 3.0 to monitor it..i also use core temp, HWMonitor but after reading your insight o Realtemp and how to calibrate it i found the others not as reliable....so i went ahead and performed your cpu cool down test with low voltage settings low heat setting at 12x and speed step activated @1.6ghz....after that i had to calibrate my individual cores to about +6-7 'C to my ambient temp in my room next to my case open...it was about 25'C and setup temps to +6,7 of that at idle of course...then went back reset bios to 3.8ghz, and ran another cool down test @3.8ghx and here are the results and cpuz pics...




My question is, are these temps and settings normal, i did not touch teh TjMax yet cause seems at idle temps are almost identical even though from cool down test only cores 1 and 3 seems identical at lower end but off at higher loads esp core 0...all info i read are on older 45nm chips, not much info on core i7....any insight would be great.

FYI i am running an coolermaster HAF 932 case with zalman 9900 fan, undreloads max temps reach high 70's..i know for this chip that is safe but i would like to know how to more accurately calibrate my temp sensors for better accuracy long term.

Thanks Guys

[HTC]

Originally Posted by HTC

I thought it was possible for those temps on the E6850 since my house doesn't have AC and it was quite cold then, with ambient temp @ about 10º (i have this case).

What surprised me was that, when i placed the E8400, the temps were 15-20º higher, using the same CPU voltage and, @ the time, the same RAM.

The lowest temp ever on the E6850 (that i saw) was 15º on both cores with core 0 reaching that temp and staying there but with core 1, usually, with 4-6º more and spikes down to 15º.

Both CPUs have identical readings while running Linpack (i mean 1 core VS the other: not 1 CPU VS the other).

I'll try to use the same settings as the E6850 CPU for this but i won't be able to do it until this weekend. Hopefully, this might help to calibrate both CPU sensor readings.

A little later then expected but have been ill lately, so i didn't have the opportunity until now.

Here are 2 pics of the E6850:



And



As i said, in previous post, core 0 systematically stays @ a low temp while core 1 stays @ 4-6º higher but with spikes that lower it to the temp of core 0.


Here are 2 pics of the E8400:



And




All of the BIOS settings were the same: the only difference was the CPU itself. The case was open in both tests.

I know the idle temp can be wrong in many CPUs but notice that the full load temp also shows difference between CPUs, which suggests it's indeed a calibration problem.

Unfortunately, i don't have an thermometer to know what the ambient temp currently is but i can tell you that it feels a heck of a lot warmer then it was when i got that 15º screenie with the E6850.

[unclewebb]

MaxDrDiablo: I haven't pointed the laser thermometer at a Core i7 yet, partly because of the extreme amount of heat they put out without a heatsink on but your i7 920 follows a very typical pattern.

From the 101 Core i7 screen shots I've seen, my best guess is that TJMax is only calibrated on the assembly line for core 0. By the time you get down to core 3, most of these are off by about 5C at the TJMax point. I would tend to use TJMax = 100, 1xx, 1xx, 105 for these where the middle two are somewhere between the end two. They're not all like this but lots of them are. I've also frequently seen a high low / high low pattern.

In your case though, your 920 is probably something like TJMax = 100, 103, 103, 106.

For Core i7, run Prime95 Small FFTs for 5 minutes or so until your 4 cores have stabilized at their maximum temperature. Have your calibration factors set to 0.0 and TJMax = 100 for core 0 and then adjust core1, core2 and core3 TJMax until the 4 are balanced. Once that is done, go back to idle at your normal MHz and core voltage that you plan to run at and then adjust the calibration factors to balance your temperatures again. Leave core 0 at 0.0 and adjust the rest of them.

Core i7 seem to have a little bit of slope error but not nearly as much as previous generations.

I've been working lately on updating RealTemp. I switched over to Visual Studio 2008 and added the ability to color the headings which looks better in some situations especially when using a very dark background. I haven't heard any complaints lately so I think the latest beta is working good and is available here if you want to check it out:

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



File_1993: 100 / 93 might be better for you than 107 / 100 and fits my theory that Intel likely only calibrates TJMax for core 0. Intel hinted at different TJMax values in their presentations last year but didn't get into the specifics so all I can do is guess based on my limited testing and from what other users show me. My original interpretation of what they said and what they showed about TJMax is that it is typically 100C but it could go slightly higher. Upon reading some more I think that it could go slightly lower is also a possibility. All Intel did during their presentation was label this difference as X and left everyone, including myself, guessing about how big X really can be and whether X might go up or down.

I wish I had more answers for you but this game has been going on for years now and Intel has only released the bare minimum of information about these sensors to keep users generally quiet. Their official position continues to be that none of these sensors should be used to try and report accurate core temperatures. I think I've been able to get reasonably accurate temperatures out of the processors I have tested but there's no way to prove this, one way or the other.

[MaxDrDiablo]

Hey thank you Unclwebb for your input...and yes i did as you said, left core 0 at 0's and went on from there and once calibrated them all even now, and thanx for the new updated color scheme for realtemp..looks nice....

[HTC]

Originally Posted by HTC

I thought it was possible for those temps on the E6850 since my house doesn't have AC and it was quite cold then, with ambient temp @ about 10º (i have this case).

When i said this, i was guessing because i don't have a thermometer with which to monitor room temp with.

Still, judging by the amount of clothes needed to not be shivering, i would guess it was pretty cold ...

[chooky]

Hope this is the right place but I am having trouble with RealTemp 3.00. I have been using it since the day it was released with no problems but now it has this error

The processor detected is not supported.

GenuineIntel


I am running an E8400 on XP x64 SP2
Version 2.70 works fine.
I have tried reinstalling it with no change.
Thanks for any help.

[unclewebb]

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

Can you try the latest beta to see if it makes any difference?

I know RealTemp 3.00 runs fine on an E8400 when using Vista x64 or Windows 7 x64. I made a switch to simplify processor detection a while ago so if that is causing a problem in XP x64 then I'll have to come up with a fix for you.

[chooky]

Thanks for the quick reply! I tried the beta but it still gives the error. I really like the cool down test so a fix would be great. Do you need any other info/

[unclewebb]

I tried the beta but it still gives the error.

I kind of thought it would.

That's what I get for trying to fix something that wasn't broken. I'll try to use my old code that used to work for you. I'll send you a PM message or post a link to it here when it's ready. Hopefully within 12 hours. Thanks for bringing this to my attention.

[unclewebb]

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

Give RealTemp 3.20 RC1 a try and let me know if it works for you chooky. Post a screen shot of the error message if there are any problems.

Version 3.00 asks the CPU if there are any thermal sensors that it can read temperature data from.
Your E8400 should say "Yes" so I'm not sure why it seems to be saying, "No".

[chooky]

RealTemp 3.20 RC1 works great! Thankyou!

Just a little suggestion. Could we have somewhere in settings to put the path to Prime95 so we don't have to find it every time.

Another thought. CPU load% is not a good indication of the amount of work a CPU is doing as illustrated by the chart of an Everest log, done while running a RealTemp CPU cool down test, below. So is there a better way of measuring load? It makes it really hard to compare apples with apples when people quote a CPU temp at 100% load when their CPU may not be working hard at all.

[burebista]

Originally Posted by chooky

So is there a better way of measuring load?

AFAIK uncle's approach to measure load is the most accurate method. Try RivaTuner plugin to see all cores in load. I love those graphs.

[chooky]

Originally Posted by burebista

AFAIK uncle's approach to measure load is the most accurate method.

Well obviously! So please uncle can you write a letter too Microsoft and let them know how you measure load so they can fix it before Win 7 comes out because the load % found in task manager isn't worth the electons its written with.

Found a new problem uncle. I was going to redo my chart with data collected from the RealTemp.csv file but it does not work while the cool down test is running, any ideas?

Another feature suggestion would be to display your REAL CPU load % in the system tray.

[unclewebb]

chooky: Now that I know RealTemp works on your XP x64 E8400, I can redo the RivaTuner plugin as well to make sure that it works on your system. I ran out of time last night to get it all done. I'm in the process of reinstalling / testing an OS or two and have files, operating systems and C++ compilers everywhere.

So is there a better way of measuring load?

For Intel core based processors, the RealTemp way is the most accurate way in my opinion. The method used to calculate this value does not apply to AMD processors so there's no way that Microsoft will ever use this method in Windows 7 or any other operating system.

I'll look into some of your suggestions and see what I can do.

Edit: RealTemp doesn't seem to have a problem gathering data while running a CPU Cool Down Test. It missed 4 or 5 seconds of data during the initial Sensor Movement Test but the rest of the data looks pretty good. Exit RealTemp after this test to give it a chance to write the data to the hard disk. You might want to try using the HDWrite=1 INI file option. It's explained in the documentation.

  DATE     TIME   C0 C1 C2 C3 NV LOAD%
03/24/09 08:10:36 32 32 32 32 42 0.1
03/24/09 08:10:37 32 32 32 33 42 0.2
03/24/09 08:10:38 32 32 32 33 42 1.0
03/24/09 08:10:39 32 32 32 32 42 0.1
03/24/09 08:10:40 32 32 32 32 42 0.1
03/24/09 08:10:41 32 32 32 32 42 0.1
03/24/09 08:10:46 46 45 45 46 42 85.5
03/24/09 08:10:46 46 45 47 46 42 99.0
03/24/09 08:10:47 46 45 47 46 42 99.8
03/24/09 08:10:48 40 40 40 40 42 89.7
03/24/09 08:10:49 35 35 35 35 42 0.1
03/24/09 08:10:50 34 34 33 34 42 0.1
03/24/09 08:10:51 33 33 33 33 42 0.1
03/24/09 08:10:52 33 33 33 33 42 0.1
03/24/09 08:10:53 33 33 33 33 42 0.1
03/24/09 08:10:54 33 33 33 34 42 2.4
03/24/09 08:10:55 33 33 32 33 42 0.1
03/24/09 08:10:56 33 33 32 33 42 0.2
03/24/09 08:10:57 32 32 32 33 42 0.2
03/24/09 08:10:58 33 33 33 33 42 1.2
03/24/09 08:10:59 32 32 32 33 42 0.1
03/24/09 08:11:00 32 32 32 33 42 0.3
03/24/09 08:11:02 49 49 48 48 42 85.0
03/24/09 08:11:03 51 51 50 49 42 99.3
03/24/09 08:11:04 52 52 51 49 42 99.4
03/24/09 08:11:05 52 52 51 50 42 99.6
03/24/09 08:11:06 52 52 51 50 42 99.7
03/24/09 08:11:07 54 54 52 51 42 99.7
03/24/09 08:11:08 54 54 52 51 42 99.8
03/24/09 08:11:09 54 54 52 51 42 99.7
03/24/09 08:11:10 54 54 52 51 42 99.8
03/24/09 08:11:11 55 55 52 51 42 99.8
03/24/09 08:11:12 55 55 53 52 42 99.7
03/24/09 08:11:13 55 55 53 52 42 99.6
03/24/09 08:11:14 55 55 53 52 42 99.7
03/24/09 08:11:15 55 55 53 52 42 99.6
03/24/09 08:11:16 56 56 53 52 42 99.7
03/24/09 08:11:17 56 56 53 53 42 99.7
03/24/09 08:11:18 56 56 54 53 42 99.7
03/24/09 08:11:19 56 56 54 53 42 99.8
03/24/09 08:11:20 56 56 53 52 42 96.1
03/24/09 08:11:21 56 55 52 53 42 96.6
03/24/09 08:11:22 56 56 54 53 42 98.3
03/24/09 08:11:23 56 56 54 53 42 96.2
03/24/09 08:11:24 56 56 54 53 42 95.8
03/24/09 08:11:25 56 56 54 53 42 95.4
03/24/09 08:11:26 57 57 55 54 42 99.7
03/24/09 08:11:27 57 57 55 54 42 99.5
03/24/09 08:11:28 57 57 55 54 42 99.8
03/24/09 08:11:29 57 57 55 54 42 99.8
03/24/09 08:11:30 57 57 55 54 42 99.8
03/24/09 08:11:31 57 57 55 54 42 99.8
03/24/09 08:11:32 57 57 55 54 42 99.8
03/24/09 08:11:33 58 58 55 54 42 99.8
03/24/09 08:11:34 58 58 55 54 42 99.8
03/24/09 08:11:35 58 58 56 54 42 99.7
03/24/09 08:11:36 58 58 56 57 42 99.7
03/24/09 08:11:37 58 58 56 57 42 99.6
03/24/09 08:11:38 58 58 56 57 42 99.6
03/24/09 08:11:39 58 58 56 57 42 99.7
03/24/09 08:11:40 58 58 56 57 42 99.7
03/24/09 08:11:41 58 58 56 57 42 99.7
03/24/09 08:11:42 58 58 56 57 42 99.7
03/24/09 08:11:43 59 59 56 57 42 99.8
03/24/09 08:11:44 58 58 56 57 42 99.5
03/24/09 08:11:45 59 59 56 57 42 99.6
03/24/09 08:11:46 59 59 56 57 42 99.7
03/24/09 08:11:47 59 59 56 57 42 99.8
03/24/09 08:11:48 59 59 56 57 42 99.8
03/24/09 08:11:49 59 59 57 58 42 99.8
03/24/09 08:11:50 59 59 57 58 42 99.6
03/24/09 08:11:51 59 59 57 58 42 99.6
03/24/09 08:11:52 59 59 57 58 42 99.7
03/24/09 08:11:53 59 59 57 58 42 99.7
03/24/09 08:11:54 59 59 57 58 42 99.7
03/24/09 08:11:55 59 59 57 58 42 99.8
03/24/09 08:11:56 60 60 57 58 42 99.7
03/24/09 08:11:57 60 59 57 58 42 99.6
03/24/09 08:11:58 60 60 57 58 42 99.8
03/24/09 08:11:59 60 60 59 58 42 99.8
03/24/09 08:12:00 60 60 57 58 42 99.6
03/24/09 08:12:01 60 60 59 58 42 99.5
03/24/09 08:12:02 60 60 59 59 42 99.7
03/24/09 08:12:03 60 60 59 59 42 99.7
03/24/09 08:12:04 60 60 59 59 42 99.7
03/24/09 08:12:05 60 60 59 59 42 99.8
03/24/09 08:12:06 60 60 59 59 42 99.8
03/24/09 08:12:07 60 60 59 59 42 99.8
03/24/09 08:12:08 60 60 59 59 42 99.6
03/24/09 08:12:09 60 60 59 59 42 99.7
03/24/09 08:12:10 60 60 59 59 42 99.8
03/24/09 08:12:11 60 60 59 59 42 99.7
03/24/09 08:12:13 60 60 59 59 42 99.7
03/24/09 08:12:14 60 60 59 59 42 99.8
03/24/09 08:12:15 61 60 60 59 42 99.7
03/24/09 08:12:16 61 60 59 59 42 99.6
03/24/09 08:12:17 61 61 60 59 42 99.7
03/24/09 08:12:18 61 61 60 59 42 99.7
03/24/09 08:12:19 61 61 60 60 42 99.7
03/24/09 08:12:20 61 61 60 60 42 99.7
03/24/09 08:12:21 61 61 60 60 42 99.8
03/24/09 08:12:22 61 61 60 60 42 99.7
03/24/09 08:12:23 61 61 60 60 42 99.6
03/24/09 08:12:24 61 61 60 60 42 99.6
03/24/09 08:12:25 61 61 60 60 42 99.7
03/24/09 08:12:26 61 61 60 60 42 99.7
03/24/09 08:12:27 61 61 60 60 42 96.6
03/24/09 08:12:28 61 61 60 60 42 96.0
03/24/09 08:12:29 61 61 60 60 42 95.9
03/24/09 08:12:30 61 61 60 60 42 95.7
03/24/09 08:12:31 61 60 59 60 42 95.6
03/24/09 08:12:32 61 61 60 60 42 99.6
03/24/09 08:12:33 61 61 60 60 42 99.7
03/24/09 08:12:34 61 61 60 60 42 99.8
03/24/09 08:12:35 61 61 60 60 42 99.7
03/24/09 08:12:36 61 61 60 60 42 99.8
03/24/09 08:12:37 61 61 61 60 42 99.6
03/24/09 08:12:38 62 61 61 60 42 99.7
03/24/09 08:12:39 62 61 61 60 42 99.7
03/24/09 08:12:40 62 61 61 60 42 99.7
03/24/09 08:12:41 62 61 61 60 42 99.8
03/24/09 08:12:42 62 62 61 61 42 99.7
03/24/09 08:12:43 62 62 61 61 42 99.8
03/24/09 08:12:44 62 62 61 61 42 99.8
03/24/09 08:12:45 62 62 61 61 42 99.6
03/24/09 08:12:46 62 62 61 61 42 99.7
03/24/09 08:12:47 62 62 61 61 42 99.7
03/24/09 08:12:48 62 62 61 61 42 99.8
03/24/09 08:12:49 62 62 61 61 42 99.8
03/24/09 08:12:50 62 62 61 61 42 99.8
03/24/09 08:12:51 62 62 61 61 42 99.8
03/24/09 08:12:52 62 62 61 61 42 99.6
03/24/09 08:12:53 62 62 61 61 42 99.7
03/24/09 08:12:54 62 62 61 61 42 99.7
03/24/09 08:12:55 62 62 61 61 42 99.7
03/24/09 08:12:56 62 62 61 61 42 99.8
03/24/09 08:12:57 62 62 61 61 42 99.6
03/24/09 08:12:58 62 62 61 61 42 99.7
03/24/09 08:12:59 62 62 61 61 42 99.6
03/24/09 08:13:00 62 62 61 61 42 99.6
03/24/09 08:13:01 62 62 61 61 42 99.5
03/24/09 08:13:02 62 62 61 61 42 99.7
03/24/09 08:13:03 63 62 62 61 42 99.7
03/24/09 08:13:04 63 62 62 61 42 99.7
03/24/09 08:13:05 63 62 62 61 42 99.7
03/24/09 08:13:06 63 62 62 62 42 99.6
03/24/09 08:13:07 63 62 62 62 42 99.7
03/24/09 08:13:08 63 62 62 62 42 99.7
03/24/09 08:13:09 63 63 62 62 42 99.8
03/24/09 08:13:10 63 63 62 62 42 99.7
03/24/09 08:13:11 63 63 62 62 42 99.8
03/24/09 08:13:12 63 63 62 62 42 99.6
03/24/09 08:13:13 63 63 62 62 42 99.7
03/24/09 08:13:14 63 63 62 62 42 99.7
03/24/09 08:13:15 63 63 62 62 42 99.7
03/24/09 08:13:16 63 63 62 62 42 99.8
03/24/09 08:13:17 63 63 62 62 42 99.8
03/24/09 08:13:18 63 63 62 62 42 99.8
03/24/09 08:13:19 63 63 62 62 42 99.8
03/24/09 08:13:20 63 63 62 62 42 99.6
03/24/09 08:13:21 63 63 62 62 42 99.7
03/24/09 08:13:22 63 63 62 62 42 99.7
03/24/09 08:13:24 63 63 62 62 42 99.7
03/24/09 08:13:25 63 63 62 62 42 99.7
03/24/09 08:13:25 63 63 62 62 42 99.8
03/24/09 08:13:27 63 63 62 62 42 99.8
03/24/09 08:13:28 63 63 62 62 42 99.5
03/24/09 08:13:29 63 63 62 62 42 99.7
03/24/09 08:13:30 63 63 62 62 42 99.8
03/24/09 08:13:31 63 63 63 62 42 99.7
03/24/09 08:13:32 63 63 63 62 42 99.7
03/24/09 08:13:33 63 63 62 62 42 96.3
03/24/09 08:13:34 63 63 62 62 42 95.7
03/24/09 08:13:35 63 62 60 62 42 95.8
03/24/09 08:13:36 63 63 62 62 42 99.4
03/24/09 08:13:37 63 63 62 62 42 96.2
03/24/09 08:13:38 63 63 62 62 42 95.8
03/24/09 08:13:39 63 63 63 62 42 99.8
03/24/09 08:13:40 65 63 63 62 42 99.8
03/24/09 08:13:41 65 65 63 62 42 99.8
03/24/09 08:13:42 65 65 63 62 42 99.6
03/24/09 08:13:43 65 65 63 63 42 99.7
03/24/09 08:13:44 65 65 63 63 42 99.7
03/24/09 08:13:45 65 65 63 63 42 99.7
03/24/09 08:13:46 65 65 63 63 42 99.8
03/24/09 08:13:47 63 63 63 63 42 99.7
03/24/09 08:13:48 65 63 62 62 42 97.1
03/24/09 08:13:49 65 65 63 63 42 99.6
03/24/09 08:13:50 65 65 63 63 42 99.7
03/24/09 08:13:51 65 65 63 63 42 99.8
03/24/09 08:13:52 65 65 63 63 42 99.7
03/24/09 08:13:53 65 65 63 63 42 99.8
03/24/09 08:13:54 65 65 63 63 42 99.8
03/24/09 08:13:55 65 65 63 63 42 99.5
03/24/09 08:13:56 65 65 63 63 42 99.7
03/24/09 08:13:57 65 65 63 63 42 99.6
03/24/09 08:13:58 65 65 63 63 42 99.8
03/24/09 08:13:59 65 65 63 63 42 99.8
03/24/09 08:14:00 65 65 63 63 42 99.8
03/24/09 08:14:01 65 65 63 63 42 99.3
03/24/09 08:14:02 65 65 63 63 42 99.6
03/24/09 08:14:03 65 65 63 63 42 99.6
03/24/09 08:14:04 65 65 63 63 42 99.7
03/24/09 08:14:05 65 65 63 63 42 99.7
03/24/09 08:14:06 65 65 63 63 42 99.8
03/24/09 08:14:07 65 65 63 63 42 99.8
03/24/09 08:14:08 65 65 63 63 42 99.8
03/24/09 08:14:09 65 65 63 63 42 99.6
03/24/09 08:14:10 65 65 63 63 42 99.7
03/24/09 08:14:11 65 65 63 63 42 99.7
03/24/09 08:14:12 65 65 63 63 42 99.7
03/24/09 08:14:13 65 65 63 63 42 99.7
03/24/09 08:14:14 65 65 63 63 42 99.6
03/24/09 08:14:15 65 65 63 63 42 99.7
03/24/09 08:14:16 65 65 64 63 42 99.7
03/24/09 08:14:17 65 65 64 63 42 99.8
03/24/09 08:14:18 65 65 64 63 42 99.6
03/24/09 08:14:19 65 65 64 63 42 99.7
03/24/09 08:14:20 66 65 64 63 42 99.7
03/24/09 08:14:21 65 65 63 63 42 98.0
03/24/09 08:14:22 63 63 62 63 42 88.1
03/24/09 08:14:23 63 62 62 62 42 88.1
03/24/09 08:14:24 63 62 62 62 42 88.1
03/24/09 08:14:25 63 62 62 62 42 88.1
03/24/09 08:14:26 63 62 62 62 42 88.1
03/24/09 08:14:27 63 62 62 62 42 88.0
03/24/09 08:14:28 63 62 62 62 42 88.1
03/24/09 08:14:29 63 62 62 62 42 88.2
03/24/09 08:14:30 63 62 62 62 42 88.1
03/24/09 08:14:31 63 62 62 62 42 88.2
03/24/09 08:14:32 63 62 62 62 42 88.0
03/24/09 08:14:33 63 62 62 62 42 88.0
03/24/09 08:14:35 63 62 62 62 42 88.1
03/24/09 08:14:35 63 62 62 62 42 88.1
03/24/09 08:14:37 63 62 62 62 42 88.2
03/24/09 08:14:37 63 62 62 62 42 88.2
03/24/09 08:14:38 63 62 61 62 42 87.1
03/24/09 08:14:40 63 62 62 62 42 86.7
03/24/09 08:14:41 62 62 62 62 42 85.8
03/24/09 08:14:42 62 62 62 62 42 85.5
03/24/09 08:14:43 63 62 62 62 42 85.5
03/24/09 08:14:44 62 62 62 62 42 85.7
03/24/09 08:14:45 62 62 62 62 42 87.9
03/24/09 08:14:46 63 62 62 62 42 88.1
03/24/09 08:14:47 63 62 62 62 42 88.2
03/24/09 08:14:48 63 62 62 62 42 88.1
03/24/09 08:14:49 63 62 62 62 42 88.2
03/24/09 08:14:50 63 62 62 62 42 88.2
03/24/09 08:14:51 63 62 62 62 42 88.2
03/24/09 08:14:52 63 62 62 62 42 88.2
03/24/09 08:14:53 63 62 62 62 42 88.0
03/24/09 08:14:54 63 62 62 62 42 88.0
03/24/09 08:14:55 63 62 62 62 42 88.1
03/24/09 08:14:56 63 62 62 62 42 88.1
03/24/09 08:14:57 63 62 62 62 42 88.0
03/24/09 08:14:58 63 62 62 62 42 88.2
03/24/09 08:14:59 63 62 62 62 42 88.2
03/24/09 08:15:00 63 62 62 62 42 88.0
03/24/09 08:15:01 63 62 62 62 42 88.0
03/24/09 08:15:02 63 63 62 62 42 88.1
03/24/09 08:15:03 63 63 62 62 42 88.1
03/24/09 08:15:04 63 62 62 62 42 88.2
03/24/09 08:15:05 63 62 62 62 42 88.1
03/24/09 08:15:06 63 62 62 62 42 88.2
03/24/09 08:15:07 63 62 62 62 42 88.0
03/24/09 08:15:08 62 62 62 62 42 87.0
03/24/09 08:15:09 61 61 61 61 42 76.9
03/24/09 08:15:10 61 60 61 61 42 76.9
03/24/09 08:15:11 61 60 61 61 42 77.0
03/24/09 08:15:12 61 60 61 61 42 77.0
03/24/09 08:15:13 61 60 61 60 42 77.0
03/24/09 08:15:14 61 60 61 60 42 76.8
03/24/09 08:15:15 61 60 60 60 42 76.7
03/24/09 08:15:16 61 60 60 60 42 77.0
03/24/09 08:15:17 61 60 60 60 42 77.0
03/24/09 08:15:18 61 60 60 60 42 77.0
03/24/09 08:15:19 60 60 60 60 42 77.0
03/24/09 08:15:20 60 60 60 60 42 77.0
03/24/09 08:15:21 60 60 60 60 42 76.7
03/24/09 08:15:22 60 60 60 60 42 76.9
03/24/09 08:15:23 60 60 60 60 42 76.9
03/24/09 08:15:24 60 60 60 60 42 76.9
03/24/09 08:15:25 60 60 60 60 42 76.9
03/24/09 08:15:26 60 60 60 60 42 77.0
03/24/09 08:15:27 60 60 60 60 42 76.7
03/24/09 08:15:28 60 60 60 60 42 76.8
03/24/09 08:15:29 60 60 60 60 42 77.0
03/24/09 08:15:30 60 60 60 60 42 76.9
03/24/09 08:15:31 60 60 60 60 42 76.9
03/24/09 08:15:32 60 60 60 60 42 77.0
03/24/09 08:15:33 60 60 60 60 42 77.0
03/24/09 08:15:34 60 60 60 60 42 76.8
03/24/09 08:15:35 60 60 60 60 42 76.8
03/24/09 08:15:36 60 60 60 60 42 77.0
03/24/09 08:15:37 60 60 60 60 42 76.9
03/24/09 08:15:38 60 60 60 60 42 77.0
03/24/09 08:15:39 60 60 60 60 42 77.0
03/24/09 08:15:40 60 60 60 60 42 77.0
03/24/09 08:15:41 60 60 60 60 42 76.8
03/24/09 08:15:42 60 60 60 60 42 76.9
03/24/09 08:15:43 60 60 60 60 42 76.9
03/24/09 08:15:44 60 60 60 60 42 76.8
03/24/09 08:15:45 60 59 60 60 42 74.2
03/24/09 08:15:47 60 60 60 60 42 73.8
03/24/09 08:15:48 60 59 59 60 42 74.1
03/24/09 08:15:48 60 59 59 59 42 73.7
03/24/09 08:15:49 60 59 59 60 42 76.2
03/24/09 08:15:51 60 59 60 60 42 75.1
03/24/09 08:15:52 60 59 60 60 42 76.8
03/24/09 08:15:53 60 59 60 60 42 76.8
03/24/09 08:15:54 60 60 60 60 42 77.0
03/24/09 08:15:55 59 59 59 59 42 75.1
03/24/09 08:15:56 58 57 57 59 42 65.3
03/24/09 08:15:57 57 57 57 58 42 65.3
03/24/09 08:15:58 57 57 57 58 42 65.2
03/24/09 08:15:59 57 57 57 58 42 65.4
03/24/09 08:16:00 57 57 57 58 42 65.2
03/24/09 08:16:01 57 57 57 58 42 65.1
03/24/09 08:16:02 57 57 57 58 42 65.3
03/24/09 08:16:03 57 57 57 58 42 65.3
03/24/09 08:16:04 57 57 56 58 42 65.3
03/24/09 08:16:05 57 57 56 58 42 65.2
03/24/09 08:16:06 57 57 56 58 42 65.3
03/24/09 08:16:07 57 57 56 58 42 65.2
03/24/09 08:16:08 57 57 56 58 42 65.4
03/24/09 08:16:09 57 57 56 58 42 65.2
03/24/09 08:16:10 57 56 56 58 42 65.3
03/24/09 08:16:11 57 56 56 58 42 65.3
03/24/09 08:16:12 57 56 56 58 42 65.3
03/24/09 08:16:13 57 56 56 58 42 65.4
03/24/09 08:16:14 56 56 56 58 42 65.2
03/24/09 08:16:15 56 56 56 57 42 65.1
03/24/09 08:16:16 56 56 56 57 42 65.3
03/24/09 08:16:17 57 56 56 57 42 65.3
03/24/09 08:16:18 57 57 56 58 42 65.3
03/24/09 08:16:19 56 56 56 57 42 65.3
03/24/09 08:16:20 56 56 56 57 42 65.3
03/24/09 08:16:21 56 56 56 57 42 65.2
03/24/09 08:16:22 56 56 56 57 42 65.2
03/24/09 08:16:23 56 56 56 57 42 65.3
03/24/09 08:16:24 56 56 56 57 42 65.3
03/24/09 08:16:25 56 56 56 57 42 65.3
03/24/09 08:16:26 56 56 56 57 42 65.4
03/24/09 08:16:27 56 56 56 57 42 65.2
03/24/09 08:16:28 56 56 56 57 42 65.4
03/24/09 08:16:29 56 56 56 57 42 65.4
03/24/09 08:16:30 56 56 56 57 42 65.3
03/24/09 08:16:31 56 56 56 57 42 65.2
03/24/09 08:16:32 56 56 56 57 42 65.2
03/24/09 08:16:33 56 56 56 57 42 65.3
03/24/09 08:16:34 56 56 56 57 42 65.3
03/24/09 08:16:35 56 56 56 57 42 65.2
03/24/09 08:16:36 56 56 56 57 42 65.3
03/24/09 08:16:37 56 56 56 57 42 65.3
03/24/09 08:16:38 56 56 56 57 42 65.2
03/24/09 08:16:39 56 56 56 57 42 65.4
03/24/09 08:16:40 56 56 56 57 42 65.3
03/24/09 08:16:41 55 55 55 54 42 64.2
03/24/09 08:16:42 54 54 54 54 42 55.6
03/24/09 08:16:43 54 54 54 53 42 55.7
03/24/09 08:16:44 54 54 54 53 42 55.5
03/24/09 08:16:45 54 54 54 53 42 55.5
03/24/09 08:16:46 54 54 54 53 42 55.7
03/24/09 08:16:47 54 54 54 53 42 55.7
03/24/09 08:16:48 54 54 54 53 42 55.7
03/24/09 08:16:49 54 54 54 53 42 55.7
03/24/09 08:16:50 54 54 54 53 42 55.7
03/24/09 08:16:51 54 52 53 52 42 53.7
03/24/09 08:16:52 54 52 53 53 42 53.9
03/24/09 08:16:53 54 52 54 53 42 55.0
03/24/09 08:16:54 54 52 53 53 42 53.4
03/24/09 08:16:55 54 52 53 53 42 53.2
03/24/09 08:16:56 54 52 53 53 42 53.5
03/24/09 08:16:57 54 52 53 53 42 55.6
03/24/09 08:16:58 54 52 53 53 42 55.6
03/24/09 08:16:59 54 52 53 53 42 55.6
03/24/09 08:17:00 52 52 53 53 42 55.6
03/24/09 08:17:02 54 52 53 53 42 55.5
03/24/09 08:17:03 52 52 53 53 42 55.6
03/24/09 08:17:04 52 52 53 53 42 55.7
03/24/09 08:17:05 52 52 53 52 42 55.6
03/24/09 08:17:06 52 52 53 52 42 55.5
03/24/09 08:17:07 52 52 53 52 42 55.6
03/24/09 08:17:08 52 52 53 52 42 55.7
03/24/09 08:17:09 52 52 53 52 42 55.7
03/24/09 08:17:10 52 52 53 52 42 55.7
03/24/09 08:17:11 52 52 53 52 42 55.7
03/24/09 08:17:12 52 52 53 52 42 55.7
03/24/09 08:17:13 52 52 53 52 42 55.6
03/24/09 08:17:14 52 52 53 52 42 55.6
03/24/09 08:17:15 52 52 53 52 42 55.6
03/24/09 08:17:16 52 52 53 52 42 55.7
03/24/09 08:17:17 52 52 53 52 42 55.7
03/24/09 08:17:18 52 52 53 52 42 55.7
03/24/09 08:17:19 52 52 53 52 42 55.7
03/24/09 08:17:20 52 52 53 52 42 55.7
03/24/09 08:17:21 52 52 53 52 42 55.5
03/24/09 08:17:22 52 52 53 52 42 55.6
03/24/09 08:17:23 52 52 53 52 42 55.7
03/24/09 08:17:24 52 52 53 52 42 55.6
03/24/09 08:17:25 52 52 53 52 42 55.6
03/24/09 08:17:26 52 52 53 52 42 55.7
03/24/09 08:17:27 52 52 53 52 42 55.6
03/24/09 08:17:28 51 51 52 51 42 54.7
03/24/09 08:17:29 50 50 51 51 42 45.4
03/24/09 08:17:30 50 50 51 50 42 45.5
03/24/09 08:17:31 50 50 51 50 42 45.6
03/24/09 08:17:32 50 50 51 50 42 45.5
03/24/09 08:17:33 50 50 50 50 42 45.4
03/24/09 08:17:34 50 50 50 50 42 45.4
03/24/09 08:17:35 50 50 50 50 42 45.5
03/24/09 08:17:36 50 50 50 50 42 45.6
03/24/09 08:17:37 50 50 50 50 42 45.6
03/24/09 08:17:38 50 50 50 50 42 45.6
03/24/09 08:17:39 50 50 50 49 42 45.2
03/24/09 08:17:40 49 49 50 50 42 45.4
03/24/09 08:17:41 49 49 50 50 42 45.4
03/24/09 08:17:42 49 49 50 50 42 45.5
03/24/09 08:17:43 49 49 50 49 42 45.4
03/24/09 08:17:44 49 49 50 49 42 45.6
03/24/09 08:17:45 49 49 50 49 42 44.6
03/24/09 08:17:46 49 49 50 49 42 45.6
03/24/09 08:17:47 49 49 50 49 42 45.2
03/24/09 08:17:48 49 49 50 49 42 44.3
03/24/09 08:17:49 49 49 50 49 42 44.8
03/24/09 08:17:50 49 49 50 49 42 44.7
03/24/09 08:17:51 49 49 50 49 42 44.5
03/24/09 08:17:52 49 49 50 49 42 44.6
03/24/09 08:17:53 49 49 50 49 42 44.0
03/24/09 08:17:54 49 48 49 49 42 44.8
03/24/09 08:17:55 49 49 50 49 42 44.6
03/24/09 08:17:56 49 49 50 49 42 44.5
03/24/09 08:17:57 49 49 50 49 42 44.4
03/24/09 08:17:58 49 49 49 49 42 43.0
03/24/09 08:17:59 49 49 49 49 42 43.0
03/24/09 08:18:00 49 49 49 49 42 43.0
03/24/09 08:18:01 49 49 49 49 42 42.0
03/24/09 08:18:02 49 49 49 49 42 42.8
03/24/09 08:18:03 49 49 49 49 42 44.4
03/24/09 08:18:04 48 48 49 49 42 44.5
03/24/09 08:18:05 49 49 49 49 42 44.5
03/24/09 08:18:06 49 49 49 49 42 44.9
03/24/09 08:18:07 49 49 49 49 42 44.3
03/24/09 08:18:08 49 49 49 49 42 44.5
03/24/09 08:18:09 48 48 49 49 42 45.4
03/24/09 08:18:10 48 48 49 49 42 45.6
03/24/09 08:18:11 48 48 49 49 42 45.4
03/24/09 08:18:13 48 48 49 49 42 45.4
03/24/09 08:18:13 48 48 49 49 42 45.6
03/24/09 08:18:14 48 48 49 48 42 44.8
03/24/09 08:18:16 47 47 48 48 42 38.6
03/24/09 08:18:17 47 47 48 48 42 38.6
03/24/09 08:18:18 47 47 48 48 42 38.6
03/24/09 08:18:19 47 47 48 48 42 38.4
03/24/09 08:18:20 47 47 48 48 42 38.4
03/24/09 08:18:21 47 47 48 47 42 38.6
03/24/09 08:18:22 47 47 48 47 42 38.4
03/24/09 08:18:23 46 46 48 47 42 38.6
03/24/09 08:18:24 46 46 47 47 42 38.5
03/24/09 08:18:25 46 46 47 47 42 38.4
03/24/09 08:18:26 46 46 47 47 42 38.4
03/24/09 08:18:27 46 46 47 47 42 38.2
03/24/09 08:18:28 46 46 47 47 42 38.5
03/24/09 08:18:29 46 46 47 47 42 38.6
03/24/09 08:18:30 46 46 47 47 42 38.5
03/24/09 08:18:31 46 46 47 47 42 38.6
03/24/09 08:18:32 46 46 47 47 42 38.5
03/24/09 08:18:33 46 46 47 47 42 38.4
03/24/09 08:18:34 46 46 47 47 42 38.4
03/24/09 08:18:35 46 46 47 47 42 38.4
03/24/09 08:18:36 46 46 47 47 42 38.5
03/24/09 08:18:37 46 46 47 47 42 38.6
03/24/09 08:18:38 46 46 47 47 42 38.6
03/24/09 08:18:39 46 46 47 47 42 38.6
03/24/09 08:18:40 46 46 47 47 42 38.4
03/24/09 08:18:41 46 46 47 47 42 38.4
03/24/09 08:18:42 46 46 47 47 42 38.5
03/24/09 08:18:43 46 46 47 47 42 38.5
03/24/09 08:18:44 46 46 47 47 42 38.5
03/24/09 08:18:45 46 46 47 47 42 38.5
03/24/09 08:18:46 46 46 47 47 42 38.6
03/24/09 08:18:47 46 46 47 47 42 38.4
03/24/09 08:18:48 46 46 45 46 42 38.1
03/24/09 08:18:49 46 45 45 46 42 37.2
03/24/09 08:18:50 45 45 45 46 42 38.5
03/24/09 08:18:51 45 45 45 46 42 38.5
03/24/09 08:18:52 45 45 45 46 42 38.5
03/24/09 08:18:53 45 45 45 46 42 38.6
03/24/09 08:18:54 45 45 45 46 42 38.3
03/24/09 08:18:55 45 45 45 46 42 38.2
03/24/09 08:18:56 45 45 45 46 42 38.3
03/24/09 08:18:57 45 45 45 46 42 38.0
03/24/09 08:18:58 45 45 45 46 42 38.1
03/24/09 08:18:59 45 45 45 46 42 38.6
03/24/09 08:19:00 45 45 45 46 42 38.5
03/24/09 08:19:01 45 45 44 46 42 37.6
03/24/09 08:19:02 44 44 44 45 42 29.0
03/24/09 08:19:03 44 44 43 45 42 29.2
03/24/09 08:19:04 44 44 43 44 42 28.2
03/24/09 08:19:05 44 43 43 44 42 28.2
03/24/09 08:19:06 43 43 43 44 42 28.1
03/24/09 08:19:07 43 43 43 44 42 28.2
03/24/09 08:19:08 43 43 43 44 42 28.2
03/24/09 08:19:09 43 43 43 44 42 29.2
03/24/09 08:19:10 43 43 43 44 42 29.1
03/24/09 08:19:11 43 43 43 44 42 29.4
03/24/09 08:19:12 43 43 43 44 42 29.2
03/24/09 08:19:13 43 43 43 44 42 29.1
03/24/09 08:19:14 43 43 43 44 42 29.4
03/24/09 08:19:15 43 43 43 44 42 29.2
03/24/09 08:19:16 43 43 43 44 42 28.7
03/24/09 08:19:17 43 43 43 44 42 29.2
03/24/09 08:19:18 43 43 43 44 42 29.2
03/24/09 08:19:19 43 43 43 44 42 29.3
03/24/09 08:19:20 43 43 42 44 42 29.3
03/24/09 08:19:22 43 43 42 44 42 29.4
03/24/09 08:19:22 43 43 42 42 42 29.2
03/24/09 08:19:24 43 43 42 42 42 29.1
03/24/09 08:19:24 43 43 42 42 42 29.2
03/24/09 08:19:26 43 43 42 42 42 29.2
03/24/09 08:19:27 43 43 42 42 42 29.3
03/24/09 08:19:28 42 42 42 42 42 29.1
03/24/09 08:19:29 42 42 42 42 42 29.4
03/24/09 08:19:29 42 42 42 42 42 29.4
03/24/09 08:19:31 42 42 42 42 42 29.4
03/24/09 08:19:32 42 42 42 42 42 29.4
03/24/09 08:19:33 42 42 42 42 42 29.3
03/24/09 08:19:34 42 42 42 42 42 29.4
03/24/09 08:19:35 42 42 42 42 42 29.4
03/24/09 08:19:36 42 42 42 42 42 29.3
03/24/09 08:19:37 42 42 42 42 42 29.0
03/24/09 08:19:38 42 42 42 42 42 29.2
03/24/09 08:19:39 42 42 42 42 42 29.4
03/24/09 08:19:40 42 42 42 42 42 29.3
03/24/09 08:19:41 42 42 42 42 42 29.4
03/24/09 08:19:42 42 42 42 42 42 29.3
03/24/09 08:19:43 42 42 42 42 42 29.4
03/24/09 08:19:44 42 42 42 42 42 29.4
03/24/09 08:19:45 42 42 42 42 42 29.4
03/24/09 08:19:46 42 42 42 42 42 29.2
03/24/09 08:19:47 42 42 42 42 42 29.4
03/24/09 08:19:48 40 40 41 41 42 27.1
03/24/09 08:19:49 38 38 39 39 42 0.3
03/24/09 08:19:50 38 38 39 39 42 0.2
03/24/09 08:19:51 38 38 38 39 42 0.2
03/24/09 08:19:52 38 38 38 39 42 0.2
03/24/09 08:19:53 37 37 38 39 42 0.1
03/24/09 08:19:54 37 37 38 37 42 0.2
03/24/09 08:19:55 37 37 38 37 42 0.2
03/24/09 08:19:56 37 37 38 37 42 0.4
03/24/09 08:19:57 37 37 38 39 42 0.5
03/24/09 08:19:58 37 37 38 37 42 0.2
03/24/09 08:19:59 37 37 38 37 42 0.2
03/24/09 08:20:00 37 37 38 37 42 0.2
03/24/09 08:20:01 37 37 38 37 42 1.4
03/24/09 08:20:02 37 37 38 37 42 0.2
03/24/09 08:20:03 37 37 38 37 42 0.1
03/24/09 08:20:04 37 37 38 37 42 0.1
03/24/09 08:20:05 37 37 38 37 42 0.5
03/24/09 08:20:06 37 37 37 37 42 0.2
03/24/09 08:20:07 37 37 37 37 42 0.1
03/24/09 08:20:08 36 36 38 37 42 0.2
03/24/09 08:20:09 36 36 37 36 42 0.3
03/24/09 08:20:10 37 37 38 37 42 1.4
03/24/09 08:20:11 36 36 37 37 42 1.3
03/24/09 08:20:12 36 36 37 37 42 1.4
03/24/09 08:20:13 36 36 37 36 42 1.3
03/24/09 08:20:14 36 36 37 37 42 1.4
03/24/09 08:20:15 36 36 37 36 42 0.5
03/24/09 08:20:16 36 36 37 36 42 0.2
03/24/09 08:20:17 36 36 37 36 42 0.2
03/24/09 08:20:18 36 36 37 36 42 0.1
03/24/09 08:20:19 36 36 37 36 42 0.1
03/24/09 08:20:20 36 36 37 36 42 0.2
03/24/09 08:20:21 36 36 37 36 42 0.2
03/24/09 08:20:22 36 36 37 36 42 0.1
03/24/09 08:20:23 36 36 37 36 42 0.4
03/24/09 08:20:24 36 36 37 36 42 0.2
03/24/09 08:20:25 36 36 37 36 42 0.3
03/24/09 08:20:26 36 36 37 36 42 0.1
03/24/09 08:20:27 36 36 37 36 42 0.4
03/24/09 08:20:28 36 36 37 36 42 0.1
03/24/09 08:20:29 36 36 36 36 42 0.1
03/24/09 08:20:30 36 36 36 36 42 0.2
03/24/09 08:20:31 36 36 36 36 42 0.1
03/24/09 08:20:32 36 36 36 36 42 0.1
03/24/09 08:20:33 36 36 36 36 42 0.1
03/24/09 08:20:34 35 35 36 35 42 0.1
03/24/09 08:20:35 35 35 35 35 42 0.2
03/24/09 08:20:36 34 34 35 34 42 0.1
03/24/09 08:20:37 34 34 35 34 42 0.1
03/24/09 08:20:38 34 34 35 34 42 0.1
03/24/09 08:20:39 34 34 35 34 42 0.1
03/24/09 08:20:40 34 34 35 34 42 0.1
03/24/09 08:20:41 34 34 35 34 42 0.1
03/24/09 08:20:42 34 34 35 34 42 0.1
03/24/09 08:20:43 34 34 35 34 42 0.1
03/24/09 08:20:44 34 34 33 34 42 0.1
03/24/09 08:20:45 34 34 35 34 42 0.2
03/24/09 08:20:46 34 34 35 34 42 0.1
03/24/09 08:20:47 34 34 33 34 42 0.1
03/24/09 08:20:48 34 34 35 34 42 0.1
03/24/09 08:20:50 33 33 33 34 42 0.1
03/24/09 08:20:51 33 33 33 34 42 0.1
03/24/09 08:20:52 33 33 33 34 42 0.1
03/24/09 08:20:53 33 33 33 33 42 0.1
03/24/09 08:20:54 33 33 33 33 42 0.1
03/24/09 08:20:55 33 33 33 33 42 0.1
03/24/09 08:20:56 33 33 33 33 42 0.2
03/24/09 08:20:57 33 33 33 33 42 0.2
03/24/09 08:20:58 33 33 33 33 42 0.1
03/24/09 08:20:59 33 33 33 33 42 0.1
03/24/09 08:21:00 33 33 33 33 42 0.1
03/24/09 08:21:01 33 33 33 33 42 0.7
03/24/09 08:21:02 33 33 33 33 42 0.1
03/24/09 08:21:03 33 33 33 33 42 0.1
03/24/09 08:21:04 33 33 33 33 42 0.1
03/24/09 08:21:05 33 33 33 33 42 0.1
03/24/09 08:21:06 33 33 33 33 42 0.2
03/24/09 08:21:07 33 33 33 33 42 0.1
03/24/09 08:21:08 33 33 33 33 42 0.1
03/24/09 08:21:09 33 33 33 33 42 0.1
03/24/09 08:21:10 33 33 33 33 42 0.1
03/24/09 08:21:11 33 33 32 33 42 0.1
03/24/09 08:21:12 33 33 32 33 42 0.1
03/24/09 08:21:13 33 33 32 33 42 0.1
03/24/09 08:21:14 32 32 32 33 42 0.1
03/24/09 08:21:15 32 32 32 33 42 0.1
03/24/09 08:21:16 33 33 32 33 42 3.9
03/24/09 08:21:17 33 33 33 33 42 4.1
03/24/09 08:21:18 34 34 33 34 42 4.1
03/24/09 08:21:19 35 35 35 35 42 4.1
03/24/09 08:21:20 34 34 36 35 42 1.2
03/24/09 08:21:21 33 33 32 33 42 3.0
03/24/09 08:21:22 33 33 32 33 42 0.1
03/24/09 08:21:23 32 33 32 33 42 0.2
03/24/09 08:21:24 32 32 32 33 42 0.1
03/24/09 08:21:25 32 32 32 33 42 0.1
03/24/09 08:21:26 32 32 32 33 42 0.1
03/24/09 08:21:27 32 32 32 33 42 0.2
03/24/09 08:21:28 32 32 33 32 42 0.1
03/24/09 08:21:29 32 32 32 32 42 0.1
03/24/09 08:21:30 32 32 32 32 42 0.1
03/24/09 08:21:31 32 32 32 32 42 0.2
03/24/09 08:21:32 32 32 32 33 42 0.2
03/24/09 08:21:33 32 32 32 32 42 0.1
03/24/09 08:21:34 32 32 32 32 42 0.1
03/24/09 08:21:35 32 32 32 32 42 0.2

[chooky]

Well couldn't the OS identify the CPU and calculate it your way for Intel and some other way for other brands?

OK posted this half way through your post. Will try that and get back to you.

[unclewebb]

Originally Posted by chooky

Well couldn't the OS identify the CPU and calculate it your way for Intel and some other way for other brands?

That's easy enough to program but the majority of users could care a less about whether the load meter in Windows is 100% accurate or not. As long as you are not using Clock Modulation, the Windows Task Manager load graph is good enough for most users.

The RealTemp method is based on me reading between the lines of some Intel documentation. When burebista and I were testing, it seemed to be a lot more responsive during rapid load transitions than the code used in the traditional load meters. If you like it too then I guess you'll have to use RealTemp. The method I came up with is a little too complicated for Microsoft.

[unclewebb]

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

This download includes the same version of RealTemp as above but I've included an updated RTCore.dll file which should work on your computer chooky. You will need to open up the Settings window in RealTemp and click on the RivaTuner button to install this version over top of any older versions of RTCore.dll in RivaTuner. As long as you're not graphing in RivaTuner, you can install this new version even if RivaTuner is already running. This will allow you to draw some nice load graphs when you are running a Cool Down Test.

The RTCore.dll will have a version number of 3.2.0.0 which you can check by doing a Properties on it.

Here's how a Cool Down Test looks using RivaTuner and the RTCore.dll plugin:



Here's how LinX looks during a test. Pretty obvious that only core 0 gets fully loaded while the other 3 oscillate from full load to zero with each test:



LinX is great for creating heat but it's not the best tool for comparing core temperatures from one core to the next.

[F7P]

Hello,
yesterday armed with a LCD Digital thermometer with probe i decided to calibrate my system as my temps were really high in RealTemp but not in the BIOS. RealTemp was giving me 44 , 39 , 43 , 43 . ! And thats odd coz i have an Arctic Freezer 7 Pro and its just been done with MX-2 and the BIOS reports idle at 23 degrees. Mobo 30 degrees. Anyway.... the clock speed wasnt overclocked and set to 333mhz. Voltage at 1.10 and multiplier to 6.0.
I checked the ambient temp with the probe and it was reading between 22 and 23 degrees. Same as BIOS.
So i rebooted to windows. RealTemp was still showing temps of 42 , 37 , 41 , 41. So i worked out .... ambient temp between 22/23. Plus the 7 degress for my high end air (fan) and that would of been 29/30 degrees. So RealTemp was wayyyy out. !!
Then i opened RealTemp and Settings and Idle Calibration and put the values as -13 - 7 - 11 - 11 . ( Working out the difference) and RealTemp now told me .. Temps . 29 , 29 ,29 ,29 . I reset the BIOS and they idled at around 29 to 30 respectively . Under load with Prime95 they only went up to a max 42 degrees during the heat up/cool down.

Did i do this all correctly and does it sound more like it should for my system ?

Thank you so much.


Running ....
Windows XP Pro SP3
Antec 900 Gamer Case Ultimate
Intel Core 2 Quad Q9300 1333FSB 6MB Cache Quad Core Core 64 Bit Processor
ASUS P5N-T Deluxe Motherboard
4GB DDR2 800MHz Dual Channel Ram
640GB (2 x 320GB) Serial ATA II Hard drive with 16MB Buffer
Dual Nvidia 8800GTX 1024MB PCI-E Graphics Running In SLI Mode

[unclewebb]

F7P: The first thing you need to do is a CPU Cool Down Test. A lot of the Core 2 Dual and Quad 45nm sensors start getting stuck at about 35C to 40C so they might not be giving you an accurate idle temperature. If that's the problem then you shouldn't be trying to calibrate them with this method. If none of them are sticking then you've gotten lucky with your CPU.

If the BIOS reports a CPU temperature of 23C when your room temperature is 23C then obviously it's reporting too low. The CPU BIOS temperature can vary a lot depending on how each manufacturer calibrates that sensor. This sensor is separate from the on chip core sensors. The CPU BIOS temperature is supposed to represent the TCase temperature but is not always correct. BIOS writers apply a correction factor to this data and what you end up with is anyone's guess and can change from one bios version to the next. On my board, at full load, this sensor reports a higher temperature than my core temperature which is impossible and not accurate.

TCase temperature at full load when running a stress program like Prime95 Small FFTs can be 20C less than your actual core temperature. The difference is even greater when stressing your CPU with Linpack / LinX / IBT. That's how quickly CPUs cool down under load when you move less than 1 cm from the hottest point on the core to the geometric center of the IHS where Intel specifies that TCase should be measured.

What load percent was RealTemp reporting when testing at idle?

So RealTemp was wayyyy out. !!

RealTemp is never out.
It's Intel's crappy 45nm sensors that are way out.

It's typical for 45nm sensors to be out by 5C or 10C at idle. Some of the error is because TJMax is not exactly equal to 100C on all 4 cores and the rest of the error is slope error. Trying to figure out what's causing most of the error is a bit of a guessing game. Doing a calibration check definitely helps out but 100% accurate temperatures from these sensors from idle to TJMax isn't possible using RealTemp or any program because these sensors weren't designed for that purpose and also because Intel does not fully document these sensors.

Do a Cool Down Test and upload a screen shot to www.imageshack.us and post a link to it here.

[F7P]

Hi unclewebb
sorry about saying RealTemp was wayyy out ! I meant it just seemed to report my temps a little high . ( esp compared to RealTemp 2.70) If my ambient ( room) temp is 23, then my chip at idle shouldnt be 44 degrees should it ! ? Also what should i set RT at to balance the cores if its out ?

Anyway.. i did the cooldown test for you and i also reset RealTemp to defaults for it.

All the cores at idle are now 45 , 39 , 43 , 43 ,

Oh and the load at idle was 0.2

Thank you and many thanks for an excellent program to.


Heres the test.

http://img166.imageshack.us/img166/3807/cooldowntest.jpg

Or

[chooky]

Back again! Thought I would see what your thoughts are on this. I tried OCCT ("OverClock Checking Tool") 3.0.2 Beta 2 from here and their Linpack test is able to get my E8400 to 75 degrees at 97.3% CPU Load as measured by RealTemp while the cool down test only gets the CPU to 63 at 99.3% CPU Load as measured by RealTemp. So what are those sneaky russians doing to my CPU. I thought Prime95 was the ultimate test to really load up your CPU but OCCT seems to be able to load it up by another 12 degrees, which is quite a lot.





I am water cooling my PC in big way. See it here http://www.techpowerup.com/gallery/2067.html

[unclewebb]

sorry about saying RealTemp was wayyy out !

No problem. I was just joking with you.

RealTemp 2.70 used TJMax=95C and the later versions use TJMax=100C because that's what Intel said is correct. Changing TJMax by 5C, bumps your reported temperatures up by 5C.

Unfortunately, Intel didn't mention how big the error is in that number so their news release turned out to be somewhat meaningless. Now when anyone asks them about TJMax or complains they now say with a straight face, "We've already released that information."

Based on my observations, I think there can be 10C of error in TJMax from one CPU to the next and even from one core to the next on the same CPU. TJMax=100C might actually be plus or minus 5C or +10C / -0C or maybe +0C / -10C. We'll never know and I can't test enough processors to come to any meaningful conclusion.

My E8400 also reads significantly too high like yours which is why I went with TJMax=95C. I thought it was a good compromise at the time. It probably is somewhere close to 100C on both cores but that means there is an extra 5C of error at idle that needs to be corrected for.

One interesting thing is that you don't need to have an exact number for TJMax. You can be out be a few degrees and using the calibration factors, you can end up with pretty much the same thing in the normal temperature range if you calibrate both of them. You can also run two instances of RealTemp with different calibration settings to see how they compare at idle and at full load.

Do you plan to overclock your Quad? If so then run another cool down test at higher MHz and higher core voltage to create some more heat. If you get your CPU about 10C or 15C hotter then the 4 cores should balance out. When they don't balance out, that's typically a sign that TJMax is not consistent for all 4 of your cores. I use TJMax = 100/100/105/105 on my Q6600.

When you compare the 0.5% level to the 87.9% level for cores 1, 2 and 3; you get a temperature change of 6.4, 6.5 and 6.5. This is a sign that the temperature slopes of these 3 are almost exactly the same. They may all have slope error but the calibration factors you use for these 3 should all be exactly the same or very close to exactly the same. The difference is TJMax error.

You're right that an idle temperature of 44 is probably too high. Make sure you have your case open for best results. Now you just need to try and find out what the main cause of the error is; slope error or error at TJMax. Use Idle Calibration to correct for slope error and Set TJMax to balance out the TJMax error.

Core 0 moved 6.9 from 0.5% to 87.9% so its calibration factor will need to be slightly different than the other three. You'll never get perfect temperatures out of these things but you can get a lot closer to the truth by following the calibration guide lines. Post some more data and I'll have a look.

[unclewebb]

For creating maximum heat, Linpack blows Prime95 Small FFTs away. Linpack creates far more heat.

These sensors report the hottest point of the core so Linpack is likely working the floating point unit much harder than other testing software does. Prime95 is designed to find Prime numbers. It's not designed to create maximum heat. Linpack is designed to create maximum heat. I think it's impossible for any software to heat a CPU up more. That's why Intel designed Linpack like that. Perfect for testing the most extreme situation but a little over kill compared to what the majority of users will ever run on their computer.

[F7P]

Originally Posted by unclewebb

No problem. I was just joking with you.

RealTemp 2.70 used TJMax=95C and the later versions use TJMax=100C because that's what Intel said is correct. Changing TJMax by 5C, bumps your reported temperatures up by 5C.

Unfortunately, Intel didn't mention how big the error is in that number so their news release turned out to be somewhat meaningless. Now when anyone asks them about TJMax or complains they now say with a straight face, "We've already released that information."

Based on my observations, I think there can be 10C of error in TJMax from one CPU to the next and even from one core to the next on the same CPU. TJMax=100C might actually be plus or minus 5C or +10C / -0C or maybe +0C / -10C. We'll never know and I can't test enough processors to come to any meaningful conclusion.

My E8400 also reads significantly too high like yours which is why I went with TJMax=95C. I thought it was a good compromise at the time. It probably is somewhere close to 100C on both cores but that means there is an extra 5C of error at idle that needs to be corrected for.

One interesting thing is that you don't need to have an exact number for TJMax. You can be out be a few degrees and using the calibration factors, you can end up with pretty much the same thing in the normal temperature range if you calibrate both of them. You can also run two instances of RealTemp with different calibration settings to see how they compare at idle and at full load.

Do you plan to overclock your Quad? If so then run another cool down test at higher MHz and higher core voltage to create some more heat. If you get your CPU about 10C or 15C hotter then the 4 cores should balance out. When they don't balance out, that's typically a sign that TJMax is not consistent for all 4 of your cores. I use TJMax = 100/100/105/105 on my Q6600.

When you compare the 0.5% level to the 87.9% level for cores 1, 2 and 3; you get a temperature change of 6.4, 6.5 and 6.5. This is a sign that the temperature slopes of these 3 are almost exactly the same. They may all have slope error but the calibration factors you use for these 3 should all be exactly the same or very close to exactly the same. The difference is TJMax error.

You're right that an idle temperature of 44 is probably too high. Make sure you have your case open for best results. Now you just need to try and find out what the main cause of the error is; slope error or error at TJMax. Use Idle Calibration to correct for slope error and Set TJMax to balance out the TJMax error.

Core 0 moved 6.9 from 0.5% to 87.9% so its calibration factor will need to be slightly different than the other three. You'll never get perfect temperatures out of these things but you can get a lot closer to the truth by following the calibration guide lines. Post some more data and I'll have a look.

Hi unclewebb,
i'm afraid i still havent a clue what to do to get my temps about right, i do know that even using RealTemp at defaults and idling at 44 i never go above 58 degrees when pushed hard with prime95. Havent tried Linpack... is it safe ?
I dont want to overclock, i really didnt see much difference when i did and for a frame here or there, no point really.
Anyway heres my temps. What should i do ?
Thank you.

http://img240.imageshack.us/img240/1677/realtemp3.jpg

Or

[unclewebb]

Linpack was written by Intel so it must be safe.

I learned one thing during this project. When not overclocked, Core 2 Dual and Quad core processors can reliably run at pretty much any temperature you throw at them. Intel might have cheaped out on the temperature sensors that they're using but the CPUs themselves are close to bullet proof.

I once ran Prime95 Small FFTs on my E8400 for 3 hours. I deliberately disconnected the CPU fan to try and create some excitement. The core temperature went up to 98C but my computer continued to run flawlessly at this temperature until I got bored and finally stopped this test after 3 hours. I was more afraid I was going to warp my board than hurt my CPU.

The sensors are far from accurate and as long as your computer is 100% stable then there's no reason to give your core temperature a second look. I hate to tell people not to run my program but in your case, there's no need to run it. The software you use is never going to get the core temperature up to a dangerous level even if the CPU fan fails.

For calibration I'd use TJMax = 97, 104, 100, 100 and use calibration factors of about -8.0 for cores 1, 2 and 3. For core 0, use whatever calibration factor is necessary so that the 4 cores are pretty much equal at idle.

If your thermal paste is fresh and your case was closed during testing then I think the above numbers might be pretty close. Give it a try and see how your temps look at idle and when running Prime95 Small FFTs. All 4 core temperatures should be fairly close at idle or at full load.