Real Temp - New temp program for Intel Core processors

[radaja]

i had it working in win 7 6956 build too but then while overclocked i started getting the same message need to run as administrator.it worked a few times running as administrator then it just stopped starting altogether.i have since went back to vista

[unclewebb]

it says i need to be logged in as administer to access drivers. so i'm guessing it's not yet compatible with beta windows 7?

In XP and Vista you have to run RealTemp as an Administrator so I'm not sure if there is any difference. You'll definitely have to do the same in Windows 7. There's no other OS legal way for the open source WinRing0 driver that RealTemp uses to gain access to the temperature sensor data.

In Vista the best way to start RealTemp when you boot up is using the Task Scheduler.
http://technet.microsoft.com/en-us/a.../aa906020.aspx

This way you can assign Administrator rights and not have to worry about this. The Task Scheduler works great and I'd be very surprised if that option is not available in Windows 7. The RealTemp folder needs to be located in a directory that you have privileges to run programs from.

[stasio]

Windows 7 build 6956

[unclewebb]

stasio: Thanks for clearing that up.

[Venturer]

Here we go guys!!!

First, MERRY CHRISTMAS TO EVERYONE!!!!

So.....my christmas gift was.....a brand new Thermalright Ultima 90i heatsink plus 92mm Papst PWM top fan.

I just installed this beast in my case and just repeated the realtemp cooldown test, but this time with the cpu at full 2.93ghz speed.

The results are here



What do you think about temps now?

I think i have a moderate improvement (previous tests where done with the cpu at maximum 2.13ghz speed).

[unclewebb]

When Intel originally released TJMax information for 65nm, it was so full of inconsistencies that I decided it would be best to ignore it. They came out with an updated version on November 18th. I still don't agree 100% with their numbers but I think it's better than what it was. Users also have to keep in mind that Intel calls these numbers TJ Targets and actual TJMax might be higher.

Here's Intel's November 18th update for the 65nm Quads:



The B2 / G0 heading comes from the Dual Core section. For Quads it's actually B3 / G0. They originally had the B3 at 80C but with this update they increased that to 90C. I agree with that.

They originally said,
"TJ increased on G0 stepping to enable lower cost heatsinks or quieter systems (slower fan speed)"

I agree with that statement but their updated chart doesn't show that. I think the G0 Quads like the Q6600 are TJMax = 100C.

On the second row they list a G0 QX processor but the QX6700 only came as a B3.

The third line covers the QX6800 which came as both a B3 and a G0.

I don't know how to interpret this information. Maybe this line should read 80 for the B3 and 90 for the G0 based on their quote above or maybe both versions really are 80C for the QX6800. This line did not change between their original TJMax news release and this updated version. Either this line was already correct, no one complained like they did for some of their other 65nm processors or else they simply forgot.

Do you see what I mean by inconsistencies in the Intel documentation? This is actually an improvement compared to having no TJMax documentation at all but for many CPUs, there is still far too much guessing needed.

What I notice about your results Venturer and what I've noticed about a lot of Quads is that it seems that TJMax is not consistent for all 4 cores. The TJ Target might be 80C but the actual TJMax looks like it is 5C higher for core2/core3 compared to core0/core1. The data coming from my Q6600 has a similar difference so I tried setting TJMax to 100C for core0/core1 and to 105C for core2/core3 and the 4 cores seem to track each other much better when set like this.

For your QX6800, why not try TJMax = 80C for core0/core1 and TJMax = 85C for core2/core3. When going between idle and full load, I think you will see that your 4 cores line up much better when set this way.

Intel admits that TJMax is not consistent from one CPU to the next but hasn't said anything about TJMax being different between the two Dual Cores within a Quad. It's just a theory I have at the moment.

I've seen too many screen shots that are difficult to explain. Sensor slope error only explains a small part of it. It's possible that something in the design of the Quad causes core2/core3 to always run 5C cooler from idle to TJMax but after learning about these sensors, it's far more likely that TJMax is being set differently at the factory between the two Dual Cores.

[Venturer]

First, thanks for you great and exaustive answer, Uncle!

So i have tried to lower tjmax of core 0 and 1 to 80c (while core 2 &3 remain at 85c). As you said, temps now are perfectly balanced through all 4 cores (only 1c of difference between all cores).

What i want to ask you is about Realtemp "LOG" feature. Unfortunatly, even with the new heatsibk, when i play games for about half an hour Realtemp tells me thermal throttling has been engaged on the first 2 cores (max temp logged was 85c for core 0 and 82x for core 1). So my question is: Does the thermal throttling log feature of Realtemp depend by Tjmax value that is set in the setting window or does Realtemp read thermal throttling directly from CPU sensors? If the first case is true lowering tjmax for the first two cores should not appear Thermal throttling LOG. I have to do tests with lower tjmax yet.

Thank you and merry christmas again!

[unclewebb]

The bit that signals that thermal throttling has taken place is stored within the CPU itself. It doesn't matter how RealTemp is set up or what TJMax really is. The CPU keeps a record of any throttling episodes until you re-boot your computer. You don't even need to have RealTemp running. If you start up RealTemp after thermal throttling has occurred then it will immediately read this information from the CPU and display it as LOG in the RT Thermal Status area.

This flag gets set generally when Distance to TJMax is 2 or 3. It varies slightly from one CPU to the next.

When the B3 Quads first came out, some users had problems with thermal throttling, especially with the OEM cooler. This might be one of the main reasons why Intel increased TJMax by 10C for the newer G0 series.

If you can, go into your bios and set your core voltage to about 1.10 volts if your board allows you to and boot up with it locked at 266 x 6.0. Open your case as well. Using the 80/85 TJMax, see what temperature your computer idles at with your CPU fan on high and compare that to your room temperature. That's usually a good way to see how reasonable your TJMax is. I think some software still assumes that TJMax is 100C for your CPU but it's definitely not that.

[jerreece]

Unclewebb, are you saying that even though Intel supposedly published the G0 Q6600 as having a TjMax of 90c, you don't think it is accurate? CoreTemp and RealTemp both set my Q6600 as having a TjMax of 100c. Assuming SpeedFan does as well based on the temps I show on all three. I just got done doing the whole SpeedFan calibration deal that CompuTronix has over on the Tomshardware Forums.

http://www.tomshardware.com/forum/25...-help#t1786165

Anyhow, ultimately I just want to know what temp my CPU really is. LOL! There's a pretty big difference between 90c and 100c as far as TjMax settings go. I guess I'm wondering where I should really be setting things.

[unclewebb]

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

I've decided to re-work the RealTemp code that displays the cpu multiplier. It's easy to read the multiplier from a Core CPU but waking a CPU up to ask it what multiplier it's at, or was at, can change the results. It might be asleep at 6.0 for 99.99% of a sample period but waking it up to read it could cause it to jump up and report 9.0 or some other number.

Intel provides counters in Core 2 and Core i7 processors and shows an easy method that can be used to accurately and quickly compute the average multiplier during a sample period.

When I first saw CPU-Z reporting a multiplier of 22.5 for rge's Core i7-940, I thought it was a bug since I can't find any Intel documentation that shows half (0.5) multipliers exist for Core i7. I think the half multi only exists in the 45nm Core 2 series.

After thinking about this for a while I've decided to join CPU-Z and also start reporting half multipliers for both Core i7 and Core 2. Technically, they may not be 100% accurate but I think they do provide users with some additional and useful information.

RealTemp used to round off multipliers to the nearest whole number for Core i7. When the Turbo mode would first start to become disengaged on some Core i7 motherboards, CPU-Z would start to show a half multiplier like 22.5 while RealTemp would continue to show 23.0. It may be impossible for a Core i7 CPU to be at a multiplier of 22.5 but reporting that does inform a user that something is going on and Turbo mode is no longer fully engaged. I have to agree with CPU-Z that a 0.5 multi in this situation is useful information. On some Core i7 motherboards as you reach the thermal wall, the multiplier will oscillate between 22 and 23 as Turbo mode kicks in and out. RealTemp can measure the exact multiplier so rounding off to the nearest 0.5 multiplier seems like a good idea.

When playing around with my Q6600 I also noticed that when software reports a single number for the multiplier, it may not be telling you what's really going on. Depending on Windows power settings, I found that one Dual Core in my Quad could be fairly steady with a 9.0 multiplier while the other Dual Core was mostly at 6.0. Software that reports either 9.0 or 6.0 may not be telling the whole story. Here's an interesting example.

My Q6600 can be tricked so one of the dual cores runs with a 6.0 multiplier while the other one is set to a 9.0 multiplier even at full load running Prime. I used CPU-Z to read one core from each dual core. My picture shows the confusion but RealTemp comes to the rescue. It averages the multiplier for all cores and reports 7.5. If one dual core is running at 2000 MHz and the other one at 3000 MHz then the average is reported as 2500 MHz with an average multiplier of 7.5.



That sounds a little weird but I think this accurately tells a user exactly what's going on inside their CPU. This code that was originally designed for the Core i7 guys also provides useful information for Core 2. If you see your multi jumping all over the place at idle in RealTemp, that's a sign that your bios and Windows power settings are fighting against each other.

I've also made a few changes to the MHz for much better support of the mobile processors. Metric gave me a for getting his P8400 right now.
------------------------------------------------------------------------
jerreece: Intel told the world that the Q6600 has a TJ Target of 90C and that TJMax might actually be higher than that number. With this new play on words, I guess TJMax could be just about anything.

The Q6600 G0 is a popular processor so I bought one, retail box, and tested it with an IR thermometer. For my CPU, there isn't a snow ball's chance in hell that it is TJMax = 90C. rge's extensive testing showed that the core temperature has to be about 5C hotter than the IHS temperature. Things cool down very quickly as you move even a small distance away from the hottest point on the core so that sounds reasonable to me.

When the IHS of my Q6600 measures 80C, to get RealTemp to display 85C, I have to use TJMax = 100C. If I use TJMax = 90C then my core temperature is going to be displayed as being 5C cooler than my IHS surface temperature which is impossible. If TJMax = 95C then the reported core temperature and the measured IHS temperature would be exactly the same which isn't very likely either. It makes the most sense if the core creating the heat is hotter than the IHS.

Intel has also stated that TJMax is not a fixed value. It varies from one CPU to another even with the same model number but they haven't stated how big the range of TJMax might be for an individual processor. This range seems larger for 45nm than 65nm but I'm just guessing at that.

[thaw]

I just set up a i7 920 system with a gigabyte UD5 motherboard, running a stock air cooler. It's not in the case yet, just on the desk with a big fan blowing towards it. The only bios change is that is was updated to version F4. Should the temperatures reach this hot? I do see the multiplier is jumping to 21 and reaching 2833mhz when at 100%. Is that the reason why the temps are so high?

[IcemaN22]

Hey guys, what is the TJ max I have to set in Real temp? Default was 95, but I read it was 100 on a other place :confused

Please, where can I find this information that is correct?

[unclewebb]

The OEM cooler is barely adequate. Your temps look like it might not be seated correctly. Have a look at that and run CPU-Z to make sure the core voltage isn't too high. You've already got two cores thermal throttling which isn't good.

IcemaN22: What CPU? Download the latest RealTemp version from a couple of posts ago and click on the Defaults button in the Settings window to see if that changes it. TJMax documentation does not seem to be 100% accurate.

Anyone that doesn't like to see their multiplier jumping up and down should head to their bios and turn off SpeedStep and C1E or head into their Power Options and set it properly so these features work as intended.

[IcemaN22]

The OEM cooler is barely adequate. Your temps look like it might not be seated correctly. Have a look at that and run CPU-Z to make sure the core voltage isn't too high. You've already got two cores thermal throttling which isn't good.

IcemaN22: What CPU? Download the latest RealTemp version from a couple of posts ago and click on the Defaults button in the Settings window to see if that changes it. TJMax documentation does not seem to be 100% accurate.

Anyone that doesn't like to see their multiplier jumping up and down should head to their bios and turn off SpeedStep and C1E or head into their Power Options and set it properly so these features work as intended.

On my E6600 default TJ max accoring to real temp is 85 and on E8400 95.
Both of them are around 35 degre idle, not sure what is wrong or correct...
I used to follow this: http://www.tomshardware.co.uk/intel-...ews-29460.html

[unclewebb]

Download the latest version of RealTemp and it should be using different TJMax values than that as Defaults. That information by Intel has been somewhat corrected. Unfortunately there are 101 errors in that data.

[thaw]

Well Cpu-z is showing voltage is jumping around 1.152-1.232 around full load. As well I did cover the entire contact of the cpu with artic silver 5 with a flat edge, trying to get it as thin as possible. I just took a look at the instructions on artic's site and they suggest running a thin line across the cores and letting the heatsink spread it out properly and thin. I'll give that a try and hope it shows somewhat of a improvement.

[IcemaN22]

Download the latest version of RealTemp and it should be using different TJMax values than that as Defaults. That information by Intel has been somewhat corrected. Unfortunately there are 101 errors in that data.

The new beta looks good, It gave 100 to the E8400 that probably is correct... but it gave my E6600 to much, 90 ! That can't be right...

[carepolice]

The new beta looks good, It gave 100 to the E8400 that probably is correct... but it gave my E6600 to much, 90 ! That can't be right...

You can always change TJmax value from settings

[randomizer]

Download the latest version of RealTemp and it should be using different TJMax values than that as Defaults. That information by Intel has been somewhat corrected. Unfortunately there are 101 errors in that data.

People need to ignore the data in that article, it's way off. Even if it was "fixed" it wouldn't really be fixed. *grumbles*

[IcemaN22]

You can always change TJmax value from settings

yeah, only problem is that I dont know what value is right for e6600. My article sucks and the beta is giving wrong too...

[randomizer]

Regarding the power management settings in Windows, I don't actually have "Processor Power Management" in there. It doesn't require EIST to be enabled for it to show up does it? I only use C1E.

RC4 looks great unclewebb (once I got it to stop downloading the old one like last time ). It seems my average multi doesn't go below 6.5x and occassionally spikes up to 7-8.5x with no active programs running.

[jerreece]

jerreece: Intel told the world that the Q6600 has a TJ Target of 90C and that TJMax might actually be higher than that number. With this new play on words, I guess TJMax could be just about anything.

The Q6600 G0 is a popular processor so I bought one, retail box, and tested it with an IR thermometer. For my CPU, there isn't a snow ball's chance in hell that it is TJMax = 90C. rge's extensive testing showed that the core temperature has to be about 5C hotter than the IHS temperature. Things cool down very quickly as you move even a small distance away from the hottest point on the core so that sounds reasonable to me.

When the IHS of my Q6600 measures 80C, to get RealTemp to display 85C, I have to use TJMax = 100C. If I use TJMax = 90C then my core temperature is going to be displayed as being 5C cooler than my IHS surface temperature which is impossible. If TJMax = 95C then the reported core temperature and the measured IHS temperature would be exactly the same which isn't very likely either. It makes the most sense if the core creating the heat is hotter than the IHS.

Intel has also stated that TJMax is not a fixed value. It varies from one CPU to another even with the same model number but they haven't stated how big the range of TJMax might be for an individual processor. This range seems larger for 45nm than 65nm but I'm just guessing at that.

Awesome, thank you for the timely response. I'll be leaving my TJMax setting as 100c then. What you've said makes complete sense. It's just to bad Intel can't be more accurate for forthcoming with their data. CompuTronix over at Tom's Hardware has confirmed my SpeedFan Calibration is correct, so I'm just going to go with what it's showing me now, and continue on with life.

Thanks again.

[unclewebb]

yeah, only problem is that I dont know what value is right for e6600. My article sucks and the beta is giving wrong too...

The updated TJMax news release from Intel shows that TJ Target for an E6600 is 80C. I believe that the actual TJMax is higher than that number and that's why RealTemp is set to use 90C by default.

That number is based on IR thermometer testing of my E6400 B2 stepping which is very similar to your E6600 B2 stepping except the E6400 has half of the L2 cache disabled. The cores are identical beyond that.

I tested my E6400 the same way I tested the Q6600 I mentioned above. For both the E6400 and the Q6600, Intel's TJ Target number is 10C lower than the actual TJMax.

Flip a coin and pick whatever TJMax you wish. You can even average those two numbers and use 85C like I believe Core Temp and most other programs are still using. It's really not that important. As long as your CPU is stable and not thermal throttling, then there's no need to be concerned about the core temperature of the CPU. It's just a semi-random number coming from poorly documented sensors that were never designed to give out accurate temperature information.

Regarding the power management settings in Windows, I don't actually have "Processor Power Management" in there. It doesn't require EIST to be enabled for it to show up does it? I only use C1E.

I think that option disappears if you don't have EIST enabled. If you want your multiplier steady at 6.0 when idle, you need to enable C1E and EIST and set up Windows appropriately. If I have the Minimum processor state set to 100%, my 4 cores on my Q6600 will continuously dance between 6.0 and 9.0. Depending on your setup, your multiplier might not be as stable as you've always thought it was. I've decided to let RealTemp tell it like it is so users can make adjustments to their settings to get the multi they like.

[unclewebb]

My friend rge sent me some info about a new program that is designed to create some serious heat in those cores.

With a name like Core Damage, I should have known better.
Here's what happens to my Q6600 after running Core Damage for two minutes.



I'll have to remember to plug in the CPU fan next time. I guess that means the thermal throttling flag is definitely working in RealTemp.

During earlier testing, as it sat at the maximum temperature, the reported multiplier slowly decreased from 9.0 to 6.0 in a gradual, more believable fashion.

I'll think twice next time before accepting software from a guy that considers drilling holes in the top of his IHS a hobby.

[rge]

coredamage worked pretty well for sensor test, raised temps 10C higher than prime previous test in link...but I left my fans on
http://www.xtremesystems.org/forums/...postcount=2877