Intel TAT / CoreTemp / IDCC all different temperatures.... [Archive]

fgw

01-21-2007, 05:36 PM

thats what i have posted over in the abit aw9d-max forum to clear things up:

Case Temperature
intel thermal specs always references a case temperature (TC). this TC is defined as the temperature measured at the geometric center of the package on the surface of the IHS.

X6800 TC = 60,4C
C2D with 4MB L2 cache TC = 60,1C
C2D with 2MB L2 cache TC = 61,4C

PROCHOT#
the system bus signal PROCHOT# will go active when the processor temperature of either core exceeds its maximum operating temperature. this indicates the Thermal Control Circuit (TCC) has been activated. the temperature at which the PROCHOT# signal goes active is individually calibrated during manufacturing. once configured, the processor temperature at which the PROCHOT# signal is asserted is not re-configurable.

THERMTRIP#
the processor will automatically shut down when the silicon temperature has reached its operating limit. at this point the system bus signal THERMTRIP# goes active and power must be removed from the processor. the temperature where the THERMTRIP# signal goes active is individually calibrated during manufacturing. once configured, the temperature at which the THERMTRIP# signal is asserted is neither re-configurable nor accessible to the system.

Digital Thermal Sensor
the Digital Thermal Sensor (DTS) is the on-die sensor to be used for fan speed control (FSC). each core has its own DTS. the DTS is monitoring the same sensor that activates the TCC. readings from the DTS are relative to the activation of the TCC. the DTS value where TCC activation occurs is 0 (zero).

Core Temp
as far as i know, thats the point where core temp kicks in. core temp just reads two registers. the first one is the one which holds the temperature where TCC gets activated, the second one is the register holding the DTS value itself. all what core temp has to do, is subtract the DTS value from TCC activation temperatur to get current core temperatures. as i understand, core temp displays the TCC activation temperature as "Tjunction" and the actual temps of both cores under "Core #0" and "Core #1".

thats how i understand this after reading intel C2D processor specs, intel C2D thermal guidelines and core temp description ...

so, as long as TCC is not activated (the cpu is not throttling) temps are below the 85C core temps shows as "Tjunction". i guess throttling could also be seen within cpu-z as reduced clockrate. have not yet tested this, anybody seen this already?
in this case we safely can assume TAT is wrong. somebody mentioned already, TAT was not written for this processors. it is a tool designed to test mobile cpu's. as this temps are read from registers, it's likely TAT reads the wrong register ...

alpha0ne

01-22-2007, 12:37 AM

After playing with my E4300 I too am reasonably sure that CoreTemp is NOT reading these chips correctly as I have stated in my E4300 quick test shown HERE (http://www.xtremesystems.org/forums/showthread.php?t=131054)

The Coolest

01-22-2007, 06:46 PM

impar

01-24-2007, 02:25 AM

Greetings!

Nice thread! :clap:

I was also having trouble figuring out which reading software was correct in my E4300.

Speedfan, CoreTemp and TAT were all showing different temperatures. :confused:

impar

01-25-2007, 08:19 AM

andylihaha

02-02-2007, 12:31 PM

Greetings!

Enabled CPU TM Function and PECI in BIOS.

The E4300 is at 2,7Ghz, 1,35v.
Ambient temperature is 17-18ºC

In idle:
CoreTemp reports 12ºC+16ºC
Speedfan 9ºC
TAT 28ºC+30ºC

Its impossible that, using a Zalman 9700 as heatsink, the CPU is bellow ambient temperature. :confused:

so from your results, TAT is more accurate than the others.
this makes things even more confusing now. So the accuracy is depended from board to board.

impar

02-02-2007, 01:23 PM

Greetings!
so from your results, TAT is more accurate than the others.
this makes things even more confusing now.Yep. I cant get myself to believe CoreTemps readings.
Currently, the ambient temperature is 15ºC and CoreTemp reads 13-16ºC.

So the accuracy is depended from board to board.
Motherboard is P5B-Deluxe with 0910 BIOS.

A similar thread turned up in OCForums:
http://www.ocforums.com/showthread.php?t=497920

Also, SpeedFan is definitely off. With PECI enabled it read 9ºC, with it disabled it read 22ºC, and after one minute of Orthos it reads 79ºC now, after I turned Orthos off 10 minutes ago.

Could Miwo had the throttling feature disabled when he performed the test?

PS
Another thread in another forum discussing the same issue:
http://www.hardforum.com/showthread.php?t=1151311

Battle_Rattle

02-26-2007, 09:52 AM

fgw

02-26-2007, 10:29 AM

Many experienced people are contradicting each other here and in other threads...

Is there any consensus clarity on what is the most accurate core temperature monitor? Say for a Conroe and a P5B-D? CoreTemp the most accurate?

As it stands ... my PC probe II, TAT, and CoreTemp beta 94 are markedly different... While I understand the rationale, I dont think the answer "trust monitor X because it shows the highest temperature" is a good answer.
unfortunately you are out of luck!

i have read a lot of documents and forums on this issue. to sum it up:
no matter what tool you are using, all this programs have to rely on sensors integrated within the cpu to measure temps. intel supplies two ways to measure temps:

the old fashioned thermal diode, readable via PECI interface
digital thermal sensor (dts, introduced with conroes), readable through registers

the problem is as follows:
the thermal diode let you read an absolute temperature BUT is not very accurate AND needs to be calibrated in order to show correct temps! programs using this method will show slightly different readings depending how accurate the program developer calibrates his software.

on the other hand, dts is very accurate. as intel is intended to use this sensor for fanspeed control, this sensor DOES NOT DELIVER ABSOLUTE TEMPERATURES! i gives a reading how far the core is away from the point the cpu starts to throttling. all programs using this accurate sensor have to calculate absolute temps by subtracting dts readings from the mentioned throttling activation temperature. to make things even worse, intel did not supply this temperature value in any register as far as we know yet. thus, software has to "assume" this temperature value and do the calculations.
you can figure out yourself the probability, the programmer hits the correct value just for your very own processor.
this temperature value is burnt into the processor at manufacturing and may vary from die to die.
as said before: as this temperature value is not readable via any register, you are out of luck!

in your situation, you can of course use any program (i would recommend coretemp) and compare readings from one situation to the other. e.g. when changing the fan you can see the change in temps. this change is very accurate. but it is impossible to compare temps between different processors even if they are of the same type ...

tzitzos

02-26-2007, 10:48 AM

very interesting :)

take a look at the screen below

temp on asus p5b dlx with waterchiller ( liquid -3C)

i think that coretemp it's the most accurate also.

http://i126.photobucket.com/albums/p91/tzitzos/General/Test%20E4300%20Acqua%200%20gradi/th_002.jpg (http://i126.photobucket.com/albums/p91/tzitzos/General/Test%20E4300%20Acqua%200%20gradi/002.jpg)

Millyons

02-26-2007, 10:53 AM

cadaveca

02-26-2007, 11:21 AM

In RD600 cpu throtlles @ 85c via smartguardian. Shutdown temp set in bios also reads same temp. this is not the same temp that Coretemp displays(61c).

So, unless it's the bios @ fault on RD600, Coretemp is not reading correctly. I have same issue too, reading 19c idle in 25c ambient, using AIRCOOLING.

this being said, coretemp reports same temp across many different boards for same cpu, when voltages/cooling are the same. Although the temp may not be perfectly accurate, it's the best we got ATM.

Millyons

02-26-2007, 11:23 AM

core temp showed normal temperatures for me....it was asus probe that showed 1/2 temp

fgw

02-26-2007, 03:37 PM

I dont even care about temps, i just try to stay away from the throttling point by about 10c for myself and other than that i dont care if my temps are xx or xx ....... but core temp is the one that is comparable to other mobos and setups, you just cant compare asus probe to gigabytes whatever ... other software like speedfan, everest, mbm actualy only shouw what probe (in this case) would show.... o another note when i put in a quad cpu probe showed 1/2 temps

i agree completely! the point is to keep some sort of safety margin to the throttling point. as i said already, this point is different from cpu to cpu AND can't be read by software. so software like coretemp is using either 85c or 100c as throttling point, which might be correct on some cpu's but is definitely wrong on others.

don't get me wrong, im not blaming the writer of coretemp! this is completely intels fault for not providing documentation of this features to the public. i'm still a great supporter of coretemp!

would be a nice feature if coretemp would also display the dts reading directly just showing the room left to the throttling point. i already requested this at the coretemp forum. lets see if "The Coolest" will implement this in a future release.

In RD600 cpu throtlles @ 85c via smartguardian. Shutdown temp set in bios also reads same temp. this is not the same temp that Coretemp displays(61c).

So, unless it's the bios @ fault on RD600, Coretemp is not reading correctly. I have same issue too, reading 19c idle in 25c ambient, using AIRCOOLING.

this being said, coretemp reports same temp across many different boards for same cpu, when voltages/cooling are the same. Although the temp may not be perfectly accurate, it's the best we got ATM.
this is a good example. i'm pretty sure, coretemp is using the wrong throttling temp on your specific cpu and therefore fails calculating the absolute temperature correctly.

Yes, i too think it's the best ATM, as long as you are using it on a cpu where coretemp gets the throttling point correctly!

TheMafioso

03-07-2007, 02:27 PM

Just Got a e4300+p5b, and i'm facing the same issue, TAT, Coretemp, PC Probe all showing different temperatures w/ temp reported by TAT higher than about 15 degrees compared to coretemp...and PC Probe shows an inbetween value...

So any updates on the issue....what to trust..:confused:

impar

03-07-2007, 02:58 PM

Greetings!

I decided to trust TAT.
CoreTemp was showing sub-ambient temepratures on idle, using a Zalman 9700.

sparkY1000

03-07-2007, 03:37 PM

TheMafioso

03-08-2007, 02:55 AM

^^Ahh DMm w/ temp probe...u can end this mystery for once and for all..
Would be looking forward for u'r results :toast: ...Do take u'r time

Regards,

SLi_dog

03-09-2007, 04:06 AM

Use TAT for E4300s :)

I will be interested to see your results though sparky1000 :)

fgw

03-09-2007, 05:27 AM

I am running the latest speedfan, coretemp and TAT, speedfan and coretemp agree and at stock under full load with 20C ambient and stock HSF I am reading 35C which I think is very reasonable TAT would have me believe it is nearer 60C which I dont think is right.
TAT was not meant for the E4300 and seems to have problems reading the temps whereas the latest coretemp is supposedly updated for the E4300.
If you have an E6XXX series CPU TAT probably is correct but I think for E4XXX it is not trustworthy.
I have a DMM with temp probe and I will try to get it onto the IHS if I can and see what that gives us, may take a few days though as im a bit busy at the moment :(
i doubt this temps are correct. 35c under load seems pretty low to me. well, depends what you understand under full load, but the 4300 i tested runs way hotter. don't remember vcore i used, have to check, but it was somewhere around 1.40v. coretemp readings were in the mid 50ies and i have to add 15c in order to get something near to real temps, as coretemp is using a wrong temperature offset in temperature calculations on certain processors. thus, in your situation i would assume temps near 50c which at least to me is far more realistic than 35c and still are good temps for stock hsf.

waiting for the next release of coretemp as "The Coolest" probably will add the feature of showing real dts values in addition to absolute temps. this will allow us to read relative but accurate temps which are correct for ALL processors supporting dts!

reading relative temps we are loosing the ability to compare temps from one processor to another but analyzing load temps or cooling efficiency on very one processor will do just fine!
don't care about the absolute temp values coretemp shows. focus on temp changes reported by coretemp as this changes are accurate: e.g. if temps go up by 10c than temps changed by exactly 10c. point! if this change was from 50c absolute up to 60c absolute or from 53c absolute up to 63c absolute - you simply can't tell!

Truckchase!

03-09-2007, 11:54 AM

New speedfan out... direct C2D support.

TheMafioso

03-11-2007, 10:15 AM

^^New Speedfan's reading is same as CoreTemp's....
So it means, coretemp was right all the time..?

sparkY1000

03-11-2007, 10:22 AM

I think until we have some definite answers believe the highest temp as you will be alot less likely to kill your chip that way :(

unclewebb

03-14-2007, 12:47 PM

sparkY1000

03-15-2007, 03:54 AM

I hope it gets sorted soon as my E4300 really wants to fly! :D

mezcal

03-15-2007, 04:13 PM

I'm one with CT reporting idle temps below ambient with 4300. From reading here and elsewhere I'm convinced that CT is performing its calculations based on an incorrect t-junction for this proc. As mentioned previously, this introduces a ~15 degree error. Using TAT 'til fixed-better safe than sorry.

sparkY1000

03-15-2007, 05:09 PM

mezcal

03-16-2007, 01:02 PM

Use the new speedfan and offset your core temps by 15C it is alot nicer to use than TAT and speedfan can sit in your taskbar giving you warnings if your CPU gets hot.
Also is it normal for E4300's to be v hot!? I am running @3.0GHz 1.3375V (in BIOS about 1.29V real) with F@H running and the stock heatsink seeing 60C loaded temps!
I am getting water sometime next month but would love to know what other people are getting with the stock cooler under load.

I read in the big BX2 thread that speedfan had been updated and now works with the badaxe2. I'll give it a try when the box is back up. Currently I'm suffering from either a global shortage of female AUX molex pins or some sort of conspiracy to prevent me from acquiring any!
Mine was running pretty warm as well-~55-58C (TAT) Orhos mix priority 9. This as in my sig. Orthos failed in a few minutes @ 333 but I didn't play with voltages. Will mess with it more when I'm back up of course.
BTW-using stock cooler.

mezcal

03-17-2007, 10:00 PM

It appears that "The Coolest" has released coretemp 0.95 today which purports to fix the temp reporting problem with the 4300s.

SLi_dog

03-18-2007, 05:28 AM

It appears that "The Coolest" has released coretemp 0.95 today which purports to fix the temp reporting problem with the 4300s.

Technically it's not a "fix" just an adjustment of the Tjunction value from 85C to 100c for E4300s which will mean CoreTemp will read 15C hotter now.

Does anyone know what the Tjunction value should be for E4300s?

sparkY1000

03-20-2007, 07:14 PM

New Coretemp causes my system to restart! No matter what as soon as I open the new coretemp up it restarts my machine! LOL!

SLi_dog

03-20-2007, 07:32 PM

would be a nice feature if coretemp would also display the dts reading directly just showing the room left to the throttling point. i already requested this at the coretemp forum. lets see if "The Coolest" will implement this in a future release.It gives the Tjunction value and the dts in only a calculation based on that value. You can quiet easily work out what the DTS is actually reading.

New Coretemp causes my system to restart! No matter what as soon as I open the new coretemp up it restarts my machine! LOL!

yeah, that seems to be common.

http://www.alcpu.com/forums/viewtopic.php?t=237

ted3

03-21-2007, 01:21 AM

http://img63.imageshack.us/img63/9260/tortureuc4.png

On my P5B Deluxe with an early Conroe E6400 CoreTemp seems to be reporting 100% correctly. At 85C thermal throttling was active on both cores and it couldn't go any higher just as Intel intended. I think TAT was reporting 2C too low.

CoreTemp seems to have a problem with the E4300. If it reports your idle temps lower than ambient then I would suggest adding a correction of 15C (100C-85C) to the reading and it should be right. Not all E4300 chips are being misread but there is definitely a problem on some boards and not others.

Sorry man, thats not throttling. When throttling occurs TAT will show you the throttled speed of the CPU.

My experiment with throttling on E6600 on BadAxe revealed that Throttlewatch didnt work. Anyone seen it working with C2D's?

Heres my previous CPU (E6600) without fan on the cooler at 1.6volt, its supposed to be running at 3.6GHz, as you can see in the top right corner. Now thats what i call throttling =)

http://i8.photobucket.com/albums/a5/ted3/TAT_Throttle.jpg

fgw

03-21-2007, 02:26 AM

It gives the Tjunction value and the dts in only a calculation based on that value. You can quiet easily work out what the DTS is actually reading.
i agree! just would be handy to show plain dts value without recalculation! if i understand the new feature list corrcectly, the coolest already implemented this feature.
unfortunately coretemp 0.95 reboots my system as soon as i run it, so i'm not able to verify this.

anyway, i think we all should get rid of this "absolute temperature measuring" thing:

there is no way to read tjunction, in fact its thermal control circuit (TCC) activation temperature, although everybody calls it tjunction
this TCC activation temp is calibrated during manufacturing, IS NOT software readable after calibration AND might differ from die to die! the 85c or 100c value is just an approximation of TCC which results in all temps calculated from this value are more or less accurate. monitoring software could not read this 85c or 100c directly, as there is no register to read it from. instead a single bit in a register is read and depending on this bit if set or not 85c or 100c is used during temperature calculations. as of intel, this works for mobile cpu's but not for desktop and server processors. on desktop and server processors, this bit might hold any information and thus on some steppings lead to the assumption of 85c on others to 100c!
using plain dts value instead maintains accuracy provided by dts readings and shows how far the processor actually is away from TCC. intel implemented temperature readings on c2d to use it for fan speed control and here they just need to know how far is the temp away from TCC activation.

i agree, it would be nice to read absolute temperatures directly, but as the way this is implemented currently, this is simply not possible!

fgw

03-21-2007, 02:31 AM

Anyone seen it working with C2D's?
yes, rmclock shows the multi and vid dropping as soon as throttling kicks in.
tat is not working on my abit board ...

psy4fun

03-21-2007, 07:38 AM

SpeedStep is supposed to be enabled in BIOS to allow TAT throttling, isn't?

Elsewhere, 85C is not an absolute value. Recall what fgw said to best understand:

the temperature at which the PROCHOT# signal goes active is individually calibrated during manufacturing. once configured, the processor temperature at which the PROCHOT# signal is asserted is not re-configurable

So, throttling can occur at 85C or something else.

TheMafioso

03-21-2007, 08:32 AM

Well so how many of you are OCing the e4300....I've been waiting for this bug to get sorted out, before starting to OC it, but now from the looks of it...the mystery only deepens with new coretemp readings..

But Somehow i believe the temps reported by earlier coretemp(or new speedfan) are the correct one, b'coz other guys are OCing this chip to around 3.6-4GHz, surely it would run very hot at this speed, if this chip runs this hot(as suggested by new coretemp) and chips would start throttling but SuperPI scores dosen't seems to suggest so...

fgw

03-21-2007, 09:23 AM

Elsewhere, 85C is not an absolute value. Recall what fgw said to best understand:

So, throttling can occur at 85C or something else.

Alex
you got it! the way coretemp works, throttling occurs at the point where one of the cores shows exactly the same value as is displayed as tjunction in coretemp. this might be any other value in programs reading their temperature off the thermal diode. i assume rmclock does so, as rmclock shows slightly different values than coretemp.
but nobody can tell if this temperature displayed in coretemp as 85c is actually 85c, 83c, 88c, or even near 100c (assuming coretemp is using the wrong tjunction) ...
you see, playing around with absolute temps leads nowhere. thats, why i requested The Coolest to implement displaying plain dts values in coretemp. was not able to verify this yet.

GGuyZ

03-22-2007, 02:49 AM

Technically it's not a "fix" just an adjustment of the Tjunction value from 85C to 100c for E4300s which will mean CoreTemp will read 15C hotter now.

Does anyone know what the Tjunction value should be for E4300s?

This does not make sense at all. Since Tjunction temperature does not exist in desktop processors, why would a fix be required? If you are already assuming that a Tjunction temperature exists(which is fault assumption too, as we all agreed), why not use the same Tjunction for E4xxx as well as E6xxx. I can understand why reporting C2Qs as 100c Tjunction makes sense, but I don't understand this logic with the E4300.

This would confuse many users that will now believe their E4300 operates at a much higher temperature. If any fix was to be applied, then it should be on TAT itself. Or instead, reporting only the DTS value(as fgw suggested) might be a better solution.

In RD600 cpu throtlles @ 85c via smartguardian. Shutdown temp set in bios also reads same temp. this is not the same temp that Coretemp displays(61c).

So, unless it's the bios @ fault on RD600, Coretemp is not reading correctly. I have same issue too, reading 19c idle in 25c ambient, using AIRCOOLING.

this being said, coretemp reports same temp across many different boards for same cpu, when voltages/cooling are the same. Although the temp may not be perfectly accurate, it's the best we got ATM.

This is very fishy and may damage the integrity of the conclusions which we all now believe in. While the absolute temp reading CT gives may be incorrect in many cases, it should always start throttling at 85C with C2D cpus, and 100C on C2Q cpus and some E4xxx. Obviously this is not real temperature, but it does indicate the throttling point. If this is indeed so, how is it possible that this guy encountered throttling at 61c in Coretemp?

fgw

03-22-2007, 04:37 PM

Does anyone know what the Tjunction value should be for E4300s?
no, nobody does! the only way to find out is to push this cpu to its limits and see what temps are reported by coretemp when throttling starts...

Originally Posted by SLi_dog
Technically it's not a "fix" just an adjustment of the Tjunction value from 85C to 100c for E4300s which will mean CoreTemp will read 15C hotter now.

This does not make sense at all. Since Tjunction temperature does not exist in desktop processors, why would a fix be required? If you are already assuming that a Tjunction temperature exists(which is fault assumption too, as we all agreed), why not use the same Tjunction for E4xxx as well as E6xxx. I can understand why reporting C2Qs as 100c Tjunction makes sense, but I don't understand this logic with the E4300.
most of the time PROCHOT# (TCC activation) temperature erroneous gets referenced as tjunction! coretemp uses this TCC activation temperature as a reference point to calculate temperatures from. TCC activation temperature can't be read by software, BUT there is a bit in a register (we discussed this a few posts earlier) which was implemented by intel. this bit is valid on mobile processors and defines if TCC activation temperature is near 85c or near 100c! i use the term "near" because in reality, this temperature is calibrated on a processor basis and differs from die to die. on one processor die it might be 87c while on an other processor die it might be 83c! coretemp, as i assume all other programs using DTS, uses this bit to decide if for a certain processor 85c or 100c has to be used in the calculations. although this is valid only for mobile processors, this seems to work also for most current BUT not all desktop processors too. we have seen more and more new processors coming out where this assumption is not correct anymore. thats the situations, where coretemp reads temps about 15c off!
so, to get coretemp back to more valid readings, the only thing to to is to change the reference temp from 85c to 100c instead on relying on the obviously wrong set bit in the register. that was done in coretemp 0.95 for e4300 and may be some other processors too.

This would confuse many users that will now believe their E4300 operates at a much higher temperature. If any fix was to be applied, then it should be on TAT itself. Or instead, reporting only the DTS value(as fgw suggested) might be a better solution.
well, in fact the new 15c higher readings might be the correct readings. i have a second rig with one of this e4300. unfortunately i have not yet pushed it into throttling. this is the only way to figure out which temp is right. for now i would assume the "new" 15c higher reading is correct. i can hardly believe, that my e4300 running on air keeps temps at 55c under full load! its far more realistic, that it should read 70c instead. will verify this as soon as i get my hands on this rig ...

Originally Posted by cadaveca
In RD600 cpu throtlles @ 85c via smartguardian. Shutdown temp set in bios also reads same temp. this is not the same temp that Coretemp displays(61c).

So, unless it's the bios @ fault on RD600, Coretemp is not reading correctly. I have same issue too, reading 19c idle in 25c ambient, using AIRCOOLING.

this being said, coretemp reports same temp across many different boards for same cpu, when voltages/cooling are the same. Although the temp may not be perfectly accurate, it's the best we got ATM.

This is very fishy and may damage the integrity of the conclusions which we all now believe in. While the absolute temp reading CT gives may be incorrect in many cases, it should always start throttling at 85C with C2D cpus, and 100C on C2Q cpus and some E4xxx. Obviously this is not real temperature, but it does indicate the throttling point. If this is indeed so, how is it possible that this guy encountered throttling at 61c in Coretemp?
unfortunately i don't have any information on smartguardian and so cant say anything here. you are right, this does not fit very well into this picture. different polling intervals might play a role here.

cadaveca, did the cpu throttle at 85c or did the bios shutdown the system as the bios is set to 85c too? may be smartguardian did not signal throttling correctly? there is a tool called rmclock, which i use to check throttling. may be you can give it a try http://cpu.rightmark.org/products/rmclock.shtml

GGuyZ

03-23-2007, 05:35 AM

most of the time PROCHOT# (TCC activation) temperature erroneous gets referenced as tjunction! coretemp uses this TCC activation temperature as a reference point to calculate temperatures from. TCC activation temperature can't be read by software, BUT there is a bit in a register (we discussed this a few posts earlier) which was implemented by intel. this bit is valid on mobile processors and defines if TCC activation temperature is near 85c or near 100c! i use the term "near" because in reality, this temperature is calibrated on a processor basis and differs from die to die. on one processor die it might be 87c while on an other processor die it might be 83c! coretemp, as i assume all other programs using DTS, uses this bit to decide if for a certain processor 85c or 100c has to be used in the calculations. although this is valid only for mobile processors, this seems to work also for most current BUT not all desktop processors too. we have seen more and more new processors coming out where this assumption is not correct anymore. thats the situations, where coretemp reads temps about 15c off!
so, to get coretemp back to more valid readings, the only thing to to is to change the reference temp from 85c to 100c instead on relying on the obviously wrong set bit in the register. that was done in coretemp 0.95 for e4300 and may be some other processors too.

You probably did not understand me correctly. First, where did you read that Tjunction is calibrated specifically for each processor. This does make sense, and while I'm aware that PROCHOT# and THERMTRIP# temps are calibrated in the making process, I'm uncertain the same applies to Tjunction. The reason I'm uncertain of this is because the MSR bit you were talking about(bit 30 of MSR 0xEE to be exact), have held constant information(either 85C or 100C as you mentioned) in the mobile processors.

Second, this bit may be valid in many desktop processors, but it has no official reference in Intel docs! It is possible that this bit information has been left by mistake, but it has no real use. Why am I saying this? Well, because in the intel dev forum it has been made clear that trying to derive the real temp from the pseudo Tjunction Temp is wrong. What I don't understand is why the CoreTemp program developer keeps following this mistake, while he knows this unfortunate fact.

well, in fact the new 15c higher readings might be the correct readings. i have a second rig with one of this e4300. unfortunately i have not yet pushed it into throttling. this is the only way to figure out which temp is right. for now i would assume the "new" 15c higher reading is correct. i can hardly believe, that my e4300 running on air keeps temps at 55c under full load! its far more realistic, that it should read 70c instead. will verify this as soon as i get my hands on this rig ...

throttling tests cannot help much in determining whether the CoreTemp reading is closer to reality when calculating 100c or 85c. In the previous case, throttling would have started at around 85C core temp. Now it will require 100C to begin throttling, but this has no indication on real temps.

unfortunately i don't have any information on smartguardian and so cant say anything here. you are right, this does not fit very well into this picture. different polling intervals might play a role here.

cadaveca, did the cpu throttle at 85c or did the bios shutdown the system as the bios is set to 85c too? may be smartguardian did not signal throttling correctly? there is a tool called rmclock, which i use to check throttling. may be you can give it a try http://cpu.rightmark.org/products/rmclock.shtml

I'm very interested in this as well.

joebuffalo

03-30-2007, 12:02 PM

fgw

03-30-2007, 05:11 PM

I am totally onboard. Who cares what the absolute temperature is. It requires us to guess what the 'tjunction' is set at. But we do know, without any guessing, what the 'delta to tjunction' is thanks to Core Temp 0.95. So that's what we should be looking at, yeah.
i see we are talking the same language ... ;)

Riddle me this, though. How small can we let that delta get before we are pushing it? Does anybody have any recommendations for how close is too close for a load (100% TAT load) temperature?
well, i will handle it this way:

i don't want my processor to start throttling -> keeping dts readings greater 0 would be enough in this case! you might add some safety margin, lets say 10c or 15c, just in case. this would be fine as long as the processor is not overclocked or better said as long as vcore is at default setting.

as soon as vcore is raised, this might be a little different: some time ago, i have read a story on thg about extreme overclocking and there was this link to the physics behind electromigration. don't want to comment on the thg story as i have the feeling they did not interpret it right, and also don't want to warm up old dishes again, but to make a long story short and if i understood the readings correctly: if vcore is increased, current flow increases and thus does current density. double the currentdensity requires 20c lower temps to NOT run into increased electromigration! currently i'm running a vcore of 1.50v and coretemp 0.94 (unfortunately 0.95 fails on my system) reads 50c to 55c under full load. this would translate to a dts reading in the range of 30c to 35c and seems sufficient to me.

joebuffalo

03-30-2007, 10:06 PM

fgw

03-31-2007, 02:37 AM

Wow, you must have some great cooling going on there.

What are you using to load your system?

I am running an e4300 at 2.8-GHz with a Ninja/Yate Loon cooler and I'm getting DTS readings of about 30 when loaded with TAT.
dont forget i'm on water!

i use orthos to load my system. small ffts to produce maximum heat and blend test to verify stability. currently running some blend tests at various vcore settings.

regarding your e4300: in a second rig i have an aw9d-max/e4300 running on air with a tt big type 120vx. the e4300 definitely tends to run hotter. have not tried coretemp 095 here as this version crashes my ab9quadgt/e6400 system. seems you are using coretemp 095 successfully. can you post screenshots or just the different readings (temp, dts, tjunction, cpuid, revision, processor) of coretemp 094 and 095? in coretemp 094 temps on the e4300 are going up to 70c. this would translate to dts readings of 15c! if i'm right coretemp 95 should read about 85c in the same situation.

also what voltage do use on the e4300?

joebuffalo

03-31-2007, 08:03 AM

dont forget i'm on water!

i use orthos to load my system. small ffts to produce maximum heat and blend test to verify stability. currently running some blend tests at various vcore settings.

regarding your e4300: in a second rig i have an aw9d-max/e4300 running on air with a tt big type 120vx. the e4300 definitely tends to run hotter. have not tried coretemp 095 here as this version crashes my ab9quadgt/e6400 system. seems you are using coretemp 095 successfully. can you post screenshots or just the different readings (temp, dts, tjunction, cpuid, revision, processor) of coretemp 094 and 095? in coretemp 094 temps on the e4300 are going up to 70c. this would translate to dts readings of 15c! if i'm right coretemp 95 should read about 85c in the same situation.

also what voltage do use on the e4300?

haha...your cooling is listed right there in your sig...I was just too tired to nitice I guess.

These screenshots are at idle, 60% TAT load, and 100% TAT load.

I'm at stock voltage, 1.325-V

http://img393.imageshack.us/img393/2549/15515518gp3.jpg

http://img393.imageshack.us/img393/3771/14286848fv4.jpg

http://img136.imageshack.us/img136/8918/20453105lt5.jpg

GGuyZ

04-02-2007, 12:43 AM

Obviously, all of those who know the logic behind this mechanism will use the Delta temp and not the absolute temp.

But still, getting closer to 'the truth' behind this is intriguing to say the least. I'm especially curious, why did CoreTemp v0.95 decided to change L2 rev chips Tjunction temp to 100c? Where was this logic derived from? It doesn't make much sense because as we can see it seems that L2 chips are much much hotter. If it's a newer revision, going through the same manufactoring process, why is it significally hotter? I can understand ~5c. but not 15c.

joebuffalo

04-02-2007, 03:19 PM

To match the results displayed via TAT. Obviously TAT is assuming a Tjunction_max of 100.

fgw

04-02-2007, 03:59 PM

... your screenshots ...
thanks for your screenshots. can't use tat on my board and coretemp 0.95 doesn't work either ...
from looking at your screenshots: thats what i have expected.

Obviously, all of those who know the logic behind this mechanism will use the Delta temp and not the absolute temp.

But still, getting closer to 'the truth' behind this is intriguing to say the least. I'm especially curious, why did CoreTemp v0.95 decided to change L2 rev chips Tjunction temp to 100c? Where was this logic derived from? It doesn't make much sense because as we can see it seems that L2 chips are much much hotter. If it's a newer revision, going through the same manufactoring process, why is it significally hotter? I can understand ~5c. but not 15c.
guess the assumption here is, tjunction could be either 85c or 100c and as 85c was obviously to low it must have been 100c then!
btw. from reading your post at the coretemp forum i have seen you also followed the thread at the intel software network forum...

To match the results displayed via TAT. Obviously TAT is assuming a Tjunction_max of 100.

the point, as we all know in the meantime, is tat was designed for mobile processors only! the use on desktop processors might produce results that look right but nobody knows if they are. i really doubt it.
from all what i have read on this so far:

the decision if 85c or 100c are to be used, based on a single bit in a msr, is not valid for desktop processors and thus wrong.
even on desktop processors the temp is not exactly 85c or 100c. its just some where near this point.
this temp is calibrated on a per chip basis and thus different from die to die.
there is no way to read this temperature by software
the only reliable way is to use the dts value directly

although not that precise you can use a monitoring program that's reading the thermal diode. comparing this readings to coretemp readings while the processor is forced into throttling. this could give an idea about the value of tjunction. in fact its not tjunction, its the temp where thermal control circuit gets activated, but thats a different story. at least you can figure out if its near 85c or near 100c ...
i tried this on my e6400. throttling kicked in at coretemp readings near 85c. don't remember the correct value. might have been 86c or 87c, but it definitely was not 100c!

joebuffalo

04-02-2007, 04:55 PM

Dude ran his e4300 up to 97-C (according to TAT) before he encountered throttling.

http://forumz.tomshardware.com/hardware/TAT-reading-high-hit-97C-throttling-toasty-ftopict230942.html

Obviously TAT is GREAT for putting max load on CPUs. But it is terrible for reading temperatures because it was not made for desktops and we have no idea what Tj_max it is assuming when it calculates an absolute temperature. Long live DTS!

unclewebb

04-05-2007, 09:33 PM

ted3: Sorry man, thats not throttling.

When I brought my E6400 up to 85C it was definitely throttling. On the Intel C2D there are two types of throttling.

http://www.intel.com/design/processor/manuals/253668.pdf
Here's a quote from page 13-5 of the Intel® 64 and IA-32 Architectures Software Developer's Manual, Vol. 3A:

"One mode modulates the clock duty cycle; the second mode changes the processor’s frequency. Both modes are used to control the core temperature of the processor."

When a C2D processor first starts to approach Tmax the duty cycle will be reduced from 100% to 87.5%, 75% and continuing down to 12.5% in 12.5% increments. If the temperature of the processor still can't be controlled then the second mode kicks in which reduces the processor frequency (MHz) like you experienced. The first "duty cycle" throttling happens approximately 3C or 4C before Tmax.

My testing has shown that CoreTemp 0.95 which uses a Tjunction of 100C for the E4300 is reporting the temperature correctly.

Rather than take a processor up to the boiling point I decided to do the opposite. I set out to try and run the processor as cool as possible.

To do this I dropped the FSB to 266 MHz ( 200 MHz for the E4300 ), I dropped the multi to 6X and then I also dropped the core voltage down to 1.175 volts. I booted up and set the Intel CPU fan at full speed to cool the processor as much as possible.

At idle I was able to get my E6400 down to 2C - 3C above ambient temperature which makes sense. A C2D puts out minimal heat at idle and even the OEM cooler is sufficient to dissipate that heat.

When I did a similar test on an E4300, CoreTemp 0.94 was reporting the core temperature significantly below ambient temperature which is impossible. An air cooled processor can't be running cooler than the surrounding air so this is a sign that CoreTemp 0.94 was reading too low.

By this test, the 15C correction introduced with CoreTemp 0.95 is correct for the E4300.

It would be interesting to see if anyone else can duplicate these numbers. This is the best and safest way I've found so far to prove that a software monitoring program is wrong.

CoreTemp 0.95 seems to be the most accurate program to date. If you don't believe the absolute value reported then just switch it to "Show Delta to Tjunction temp". I like to leave at least 10C to 15C of headroom to Tmax for maximum performance and stability.

joebuffalo

04-06-2007, 12:03 AM

The Tjmax for nay core 2 chip is not necessarily 85 or 100 exactly. It is calibrated for each chip at the factory. That's why the DTS reading is so damn awesome.

And I *think* you weren't throttling. In the Intel docs, it states that the DTS reading is the number of degrees between the current core temp and TCC (throttling). So unless you see a DTS of 0 in CoreTemp 0.95 you shouldn't be throttling. I think. (And since you said you were at 85 degrees, that's somthing like 10 or 20 degrees DTS) Maybe I misunderstood something.

SLi_dog

04-06-2007, 12:20 AM

This does not make sense at all. Since Tjunction temperature does not exist in desktop processors, why would a fix be required? If you are already assuming that a Tjunction temperature exists(which is fault assumption too, as we all agreed), why not use the same Tjunction for E4xxx as well as E6xxx. I can understand why reporting C2Qs as 100c Tjunction makes sense, but I don't understand this logic with the E4300.

This would confuse many users that will now believe their E4300 operates at a much higher temperature. If any fix was to be applied, then it should be on TAT itself. Or instead, reporting only the DTS value(as fgw suggested) might be a better solution.

I supposed you have to ask yourself what's more believable, allendale core CPUs (E4300 or L2 E63/6400) that idle at or below ambient temperature or that the "reference" Tjunction value for them needs a +15C adjustment?

Anyway, isn't it just a reference value which gives a far better idea of CPU temperatures than we've ever had before to better maintain optimal performance when overclocking? Who really cares if it's 100% accurate?

unclewebb

04-06-2007, 09:51 AM

The Tjmax for nay core 2 chip is not necessarily 85 or 100 exactly.
Is there any documentation from Intel or any evidence from other users to support that? I'm not saying it's not true. I'm just interested like everyone else in what's really going on.

For my E6400 Revision B2, CoreTemp uses this formula:
reported temp = 85 - DTS

CoreTemp reports 85C when DTS equals zero.

During testing, TAT first started to report, "Thermal Monitor Active" in red for Core0 when DTS was at 2 so CoreTemp was reporting 83C. At this point the reported MHz is not yet reduced but the performance of the processor starts getting reduced internally to try and control the temps. CPUz will still be reporting full MHz but performance is reduced.

If the temperature of the processor can still not be controlled using this first thermal throttling technique, then the processor MHz will also be reduced. If the temperature still can't be controlled then finally the "catastrophic shutdown detector" is enabled and the processor shuts down.

Both methods of throttling will result in reduced performance for your processor. At stage 1, the reported MHz might still show your CPU at full speed but performance is already being reduced.

unclewebb

04-06-2007, 10:01 AM

....why did CoreTemp v0.95 decided to change L2 rev chips Tjunction temp to 100c? .......as we can see it seems that L2 chips are much much hotter. If it's a newer revision, going through the same manufactoring process, why is it significally hotter?
The reason Tjunction or Tmax was raised from 85C to 100C for the E4300 isn't because the E4300 puts out any more heat or runs hotter than previous C2D chips. A chip built on the same manufacturing process, being supplied the same voltage and running at the same frequency is going to put out pretty much the same heat as previous C2D processors.

Intel raised the maximum so this chip can run 15C hotter before thermal throttling will take place.

The E4300 is designed as a budget chip which will be used in small budget computers that typically have poor air flow and system cooling. This chip gives computer manufacturers more head room to design a small computer without having to worry about the processor hitting the thermal throttle and slowing down performance.

joebuffalo

04-06-2007, 12:22 PM

For my E6400 Revision B2, CoreTemp uses this formula:
reported temp = 85 - DTS

Mea culpa. Sometimes I forget that not everybody has an e4300.

Is there any documentation from Intel or any evidence from other users to support that? I'm not saying it's not true. I'm just interested like everyone else in what's really going on.

I *thought* I read that in the spec sheets somewhere, but I can't seem to find it in my notes. Maybe I just read it on a forum? I did find this:

http://documents.irevues.inist.fr/bitstream/2042/6579/1/TMI23.pdf

"The DTS is calibrated at manufacturing conditions and the reference point is set to this test temperature. Functionality, electrical specifications and reliability commitments are guaranteed at maximum Tj as measured by the DTS. Any test inaccuracy or parameters variance are already accounted for in the DTS set point."

Which sort implies what I claimed, but certainly doesn't say it clearly. Either way, the paper is a good read.

unclewebb

04-06-2007, 01:58 PM

"The DTS is calibrated at manufacturing conditions and the reference point is set to this test temperature. Functionality, electrical specifications and reliability commitments are guaranteed at maximum Tj as measured by the DTS. Any test inaccuracy or parameters variance are already accounted for in the DTS set point."
I read the above paragraph and of course came to the opposite conclusion.

I think if Intel is making a batch of processors with a Tjmax of 85C then they calibrate the DTS during manufacturing so that it reads 0 when the hottest point in the core is at exactly 85C. Same thing for the E4300 at 100C. This happens at the Filtering and Post Processing stage in the diagram where an offset can be added to the DTS reading to correct for any variation in the digital sensors.

To me it makes no sense that Intel would manufacture a batch of processors and then bin them based on what temp they should be able to run at.
"This one looks like it can run at 85C so we'll set Tjmax=85 and then the next one gets set to 87 or 90 or ....."

It makes a lot more sense that Intel determines for long term reliability and for stability that the first generation of C2D processors can run fine at up to 85C so they set that value to every processor to avoid excessive warranty claims. With further testing and with improved manufacturing they feel that their second generation of processors built on the Allendale core can run reliably up to 100C so they bump Tjmax up 15C.

Whether my interpretation or yours is right isn't too important. CoreTemp 0.95 tells you the DTS reading directly and references that to either 85C or 100C depending on your processor. It then calculates a core temperature based on one of those two reference points.

My processing power first gets reduced when DTS=2. If you're not overclocked then you should be able to continue and get to DTS=0 without any loss in stability. When fully overclocked, I found that the length of time that Orthos could run without any errors decreased when DTS dropped below about 13. That's the important number for my computer. I have great long term stability at DTS=15 and not so great when DTS=10.

Thanks for that article. Every little bit of information helps.

fgw

04-06-2007, 02:02 PM

Is there any documentation from Intel or any evidence from other users to support that? I'm not saying it's not true. I'm just interested like everyone else in what's really going on.
joebuffalo, might have referenced the following paragraph in the Thermal and Mechanical Design Guidelines to be found here:
http://www.intel.com/design/processor/designex/313685.htm

4.2.8 THERMTRIP# Signal

The temperature where the THERMTRIP# signal goes active is individually calibrated during manufacturing. The temperature where THERMTRIP# goes active is roughly parallel to the thermal profile and greater than the PROCHOT# activation temperature. Once configured, the temperature at
which the THERMTRIP# signal is asserted is neither re-configurable nor accessible to the system.

i would interpret this in the same way. individually calibrated during manufacturing let me assume this temperature is different from one processor to the other.

also i have read, but unfortunately don't remember where: "this thermtrip value is calibrated during manufacturing on a per chip basis by burning some shunts on the processor and may vary from die to die."

another interesting read on this issue is the thread on the intel software network forum:
http://softwarecommunity.intel.com/isn/Community/en-US/forums/thread/30222546.aspx

look at the post before the last one:
...Note Tj is not a fixed value and the IA32_TEMPERATURE_TARGET[15:8] value can vary from part to part. Tj is also not software readable....

...Some steppings of the mobile Intel® Core™2 processor do indicate Tj to be approximately 85 or 100 via a single bit in the EXT_CONFIG register (msr 0EEh) but desktop, workstation and server processors do not....

fgw

04-06-2007, 02:10 PM

unclewebb

04-06-2007, 09:07 PM

Thanks fgw & joebuffalo for the additional info. Looks like I got some more reading to do!

joebuffalo

04-06-2007, 09:20 PM

fgw

04-07-2007, 12:33 PM

It certainly looks like Tj_max is higher on the L2's...do we agree on this?

So my new question: Is it *wise* to run yourn e4300 at the same DTS as an e6600? Just because the Tj_max was raised, does that mean we *should* be running our L2 processors at a higher absolute temperature? Obviously we can...but should we?
i would say it's always good to keep temps as low as possible. besides that i would not hesitate to increase vcore, and temps obviously too, until dts approaches 0c. intel would not have changed this tjmax if reliability cant be maintained at this temps.
anyway, to keep some safety margin would be a good idea for sure. i personally would keep this margin in a range of 10c to 15c.
as vcore is increased, temps need to be decreased in order to keep electromigration and thus reliability under control. so i would add another 20c to this safety margin.
for coretemp, this leads to a dts reading of about 35c or

a temp reading of about 50c if your tjunction in coretemp is 85c and to
a temp reading of about 65c if your tjunction in coretemp is 100c.
i know there is some talks about this issue (electromigration) going on in various threads so i wont comment on that right now. found some documents explaining this and will start a thread on this later on.

joebuffalo

04-07-2007, 01:12 PM

imho, that's a HUGE DTS.

A very wise (and conservative) man wrote this Core 2 Temperature Guide:

http://forumz.tomshardware.com/hardware/Core-Duo-Temperature-Guide-ftopict221745.html

and he seems to think that 65-C is AOK for the Tj_max = 85 systems. That's a DTS of 20. Right now my DTS is ~30 with my e4300 at 2.7-Ghz (stock 1.325-V) topped with a Scythe Ninja / 1300 RPM Yate Loon fan and AS5.

So I'd say a DTS of 30 leaves a LOT of headroom. A DTS of ~20 seems reasonable based on the opinions in the thread above.

I'm still just not sure that we should be running L2's at the same DTS as the older Core 2 CPUs. Should we run the L2's x-degrees hotter just because Tj_max went up x-degrees. (That's why I am keeping my DTS at ~30 for now)

TheMafioso

04-15-2007, 12:03 PM

BUMP
r we reachin on any conclusion ???

fgw

04-16-2007, 04:36 PM

A very wise (and conservative) man wrote this Core 2 Temperature Guide:

http://forumz.tomshardware.com/hardware/Core-Duo-Temperature-Guide-ftopict221745.html

and he seems to think that 65-C is AOK for the Tj_max = 85 systems. That's a DTS of 20. Right now my DTS is ~30 with my e4300 at 2.7-Ghz (stock 1.325-V) topped with a Scythe Ninja / 1300 RPM Yate Loon fan and AS5.
i would not rely too much on this post over at the thg forums. there are a lot of misinterpretations in it!
lets start with the things you can confirm by looking at intel datasheets:

Datasheet
http://www.intel.com/design/processor/datashts/313278.htm

Thermal and Mechanical Design Guidelines
http://www.intel.com/design/processor/designex/313685.htm

yes, there are more than one sensors measuring temperatures
one thermal diode located on the processor die
two digital temperature sensors within each core on c2d
four digital temperature sensors within each core on quadintel defines tcase at the temperature measured on top of the ihs and at the center of the ihs.

afik, bios or software is not displaying tcase! the temperature displayed there is simply the processor temperature measured via the thermal diode. the thermal diode as implemented by intel is delivering a voltage proportional to processor temperature. the software using this signal has to calibrate this readings and convert it to a temperature. as different software, or bios, is calibrating this readings differently, you get different temperatures but by no way tcase as described above.

the digital temperature sensors on the other hand, are located directly within the cores near the hottest places. this sensor has a smaller footprint and thus can be located easier within the core. the reading, as opposed to the thermal diode, is not a voltage! digital temperature sensors convert the temperature directly into a binary value (dts reading) and store it in a register within the processor thus eliminating the need of calibrating this sensor readings.

the problem here is, this dts reading is not absolute. its a relative temperature to the thermal control circuit activation temperature (prochot#). as this thermal control circuit activation temperature is calibrated during manufacturing on a per part basis and is not read by software it is impossible to convert dts readings into an absolute temperature.

on mobile processors tjunction as a reference temperature instead of thermal control circuit activation temperature. also, there were just two possible tjunction values used on mobile processors in the past (don't know intel plans on future mobile processors). on this mobile processors, there is a bit in a register which can be read to check if tjunction is at 85c or if it is at 100c! absolute temperature can now simply calculated by substracting dts reading from tjunction. thats how coretemp, and i assume all other programs using digital temperature sensors readings, calculate absolute temperatures.

using this method on desktop or server processors might result in a temperature reading that look right, but this does not mean it is correct.
there is only one correct temperature reading on intel core technology processors: relative temperature (dts reading) as introduced incoretemp 0.95

everything else, besides mobile processors as discussed above, is speculation!

and so is most of the following:
Tcase + 15c = Tjunction its more like: dts reading + 15c = thermal control cicuit activation temperature
Thermal Case Temperatures of 60c is hot, 55c is warm, and 50c is safe. well, this might be true, but there is no way to measure tcase besides a temp probe on the ihs! do you use one? i don't.
The difference between Tcase (BIOS, Motherboard Utilities and SpeedFan: CPU or Temp X) and Tjunction (TAT and SpeedFan: Core 0 / Core 1) is Tcase + 15c = Tjunction
Vcore should not exceed ~ 1.5v.
Tjunction is always ~ 15c higher than Tcase.
Tcase is always higher than Ambient. as any temp as long as you are on air or water ...
Tcase Idle should be ~ 1 to 15c higher than Ambient.
Tjunction Idle should be ~ 15 to 30c higher than Ambient.
Tcase Load should not exceed ~ 55c with TAT @ 100% Load. again, how is it measured?
Tjunction Load should not exceed ~ 70c with TAT @ 100% Load.
Idle to Load Delta should not exceed ~ 25c.
Tjunction Results are Hottest Core Idle and Load.
Vcore will typically sag ~ .025 volts under Load. depends on motherboard
Any hardware and / or software may misreport Tcase and / or Tjunction temps.
965 chipsets may misreport Tcase and Tjunction temperatures with +/-15c offsets. the chipset is not involved at all - all sensors are within the cpu. well, calibration of thermal diode is done in bios
If TAT will not run, then Orthos Priority 9 Small FFT’s simulates 88% of TAT ~ 5c lower.
C2D’s manufactured with concave Integrated Heat Spreaders may report high Deltas and temps. true
An improperly seated CPU cooler is the leading cause of abnormally high temperatures. also true, but again - no surprise hereadded comments to some of them although almost all are either questionable or obvious ...

joebuffalo

04-16-2007, 05:10 PM

fgw

04-16-2007, 05:33 PM

I agree with a lot of what you said there.

Although I think the mobo does come into play when talking about Tcase (keep in mind he is not defining Tcase the same way Intel is...is is talking about the thermal diode output) because the mobo is where the thermal diode reading is converted into a temperature reading.

I didn't post his link because I think that thread is correct. In fact I disagree with a lot of it...that's how I wound up here on the DTS bandwagon. My point was just that he recommends a DTS of 20 (doesn't actually SAY that, but you and I know that is the jist of it) and I think he is pretty conservative (maybe it's because I've read his opinions on many overclocks) when it comes to temperatures.

I think your 30-degree DTS is VERY safe. Too safe for a bunch of overclockers...imho.
you are right! but i mentioned this 30c by two reasons:

electromigration: to maintain the same reliability, temperature has to be decreased by 20c when current density gets doubled. i know, increasing vcore from 1.28v up to 1.50v does not result in doubling the current density, but the scale is logarithmic, so in my case temps might be decreased by 17c or may be 15c!
simply the fact, that my watercooling system is able to maintain this sort of safety margin ...;)

unclewebb

04-21-2007, 11:21 PM

there is only one correct temperature reading on intel core technology processors: relative temperature (dts reading) as introduced incoretemp 0.95 everything else, besides mobile processors as discussed above, is speculation!
I agree with the above statement 100% but.......

I still think that CoreTemp 0.95 which uses either a fixed 85C or 100C Tjunction value in its calculation results in a pretty accurate absolute temperature for the desktop processors.

I decided to come up with a simple test so users can determine if the absolute temperatures reported by CoreTemp are valid or not.

On my E6400, revision B2, CoreTemp 0.94, CoreTemp 0.95 and SpeedFan 4.32 all report the same temps. For this test I used SpeedFan for its graphing capabilities.

The goal is to run your processor as cool as possible. Start by opening your case. Next lower the FSB to 200 MHz, lock the cpu multiplier in the bios at 6 and reduce the core voltage. Setting core voltage to 1.200 volts or less is adequate. With my P5B Deluxe I was able to get the core down to 1.104 volts as reported by CPUz.

You can do this with any Core2Duo so this test is consistent no matter what processor you have.

I believe that when you run an air cooled processor at reduced speed and voltage in an open case, the core temperature will start to approach the ambient temperature. I used SpeedFan to keep the cpu fan at its maximum.

With a room temperature of 18C as measured with a consumer grade digital house thermometer, I ended up with a reported core temperature of 17C during this test.

http://img207.imageshack.us/img207/6513/testvd4.png

In theory with air cooling you shouldn't be able to get lower than ambient temperature but if you keep in mind that both values are integer values and are at best only accurate to +/- 0.5 degrees and that my $20 Radio Shack thermometer is far from perfect then I think my results confirm that:

An air cooled C2D at 1200 MHz and 1.1 volts will put out so little heat at idle that the core temperature and the ambient temperature will be very close to equal.

For my E6400, the CoreTemp / SpeedFan assumed Tjunction value of 85C, seems to produce a very accurate absolute temperature value.

This test can be used by anyone with any Core2Duo processor to at least give them an idea if their temperature monitoring software is in the ball park or if it's totally out to lunch.

joebuffalo

04-22-2007, 12:05 AM

Fair enough. But riddle me this, enclewebb:

What good is the absolute temperature? How do you know what is a safe absolute temperature? That's the beauty of the DTS, it tells you the 'Degrees To oh Shoot'

biatch0

04-22-2007, 12:08 AM

Hey, just want to get some input from you guys about my CPU temps. I was previously using the stock Intel cooler that came with my QX6700... but after a bunch of temp readings that were confusing, I took the safe route and assumed the higher temp readings were the accurate ones. Below are screenshots from TAT/CoreTemp/Everest from the time I was using the stock HSF:

http://www.waddekao.com/files/biatch0/everest.jpg
http://www.waddekao.com/files/biatch0/pcprobe-coretemp.jpg

And here's a screenshot of those after I installed the new ASUS Silent Square Pro:

http://www.waddekao.com/files/biatch0/allin1.jpg

The temps I'm getting are still a little on the high side which worries me... anyone with an aircooled QX6600/6700 with temp screenshots?

unclewebb

04-22-2007, 09:15 AM

What good is the absolute temperature? How do you know what is a safe absolute temperature? That's the beauty of the DTS, it tells you the 'Degrees To oh Shoot'
Users seem to like knowing the absolute temperature of their cpu, even if that number isn't 100% accurate.

Using my testing method you can at least determine if your Tjunction is about 85C or about 100C and which version of CoreTemp is reporting your absolute temperature closest to reality.

Going by DTS is great in theory but already there are discussions of what is a good DTS number to shoot for and a 101 qualifiers like whether you are overclocked or over volted and by how much, etc., etc. I've already discovered that if you are at default MHz and voltage that you can run reliably at a higher core temperature than if you're fully overclocked which makes sense.

People have never been able to agree on a safe absolute core temperature any more than they will ever be able to agree on a safe DTS value.

My E6400 with an approximate Tjunction of 85C showed Orthos stability started to decline when DTS went below 15C and declined further when it got to 10C. I have absolutely no idea whether those results will apply to an E4300 which has a Tjunction of 100C. It's possible that a well overclocked C2D looses stability when its absolute core temperature exceeds 70C to 75C regardless of what Tjunction Intel has rated it at.

Hopefully some E4300 owners can shed some more light on this topic by doing their own testing.

unclewebb

04-22-2007, 09:25 AM

biatch0: You didn't mention whether those are idle temps or if you are running something like Orthos at full load.

If those are idle temps then I think you should look at re-mounting your cooler.

If those temps are while running Orthos then temps in the low 60C range with all 4 cores loaded at 3000 MHz looks pretty good to me. The Quads put out more heat and are going to run hotter at the same load and same frequency compared to a Core2Duo.

fgw

04-22-2007, 10:12 AM

In theory with air cooling you shouldn't be able to get lower than ambient temperature.
thats true! its also true for watercooled systems, except its not ambient temp but water temp that can't be beaten!

would not mind if temps are off by a few degrees, but if the processor runs at stock, or even below that and you get core temps 10c or more below ambient than tjunction in coretemp is wrong and thus your absolute core temps!

yes, your simple test can quickly verify if coretemp or whatever software used, gets the reference temp (tjunction in coretemp) correctly!

Fr3ak

04-22-2007, 04:35 PM

I somehow have the feeling that CoreTemp 0.95 does not work "correct" for all E4300. I built 2 E4300 rigs last week using the Infinity and the idle temp at stock volts/stock clock was 42-46C on both rigs. So I am having the feeling thats not the real temp for one of the best air coolers and a C2D at stock. With 1.34V real I had 80C according to coretemp at 3150Mhz after 19hours of Orthos. Those temps dont seem to be right to me. 15C less than that sounds more like it. Unfortunately I have only remote access to those rigs, so I cant test much from here :/

joebuffalo

04-22-2007, 05:00 PM

Fr3ak

04-22-2007, 05:07 PM

Most of it.
From what I understood, 0.94 reads 15C to low for the E4300, so it was changed in 0.95 which reads a 15C higher temp.

Now what if the 15C issue is not working for all E4300 batches?

unclewebb

04-22-2007, 05:15 PM

I somehow have the feeling that CoreTemp 0.95 does not work "correct" for all E4300.
That's what I've found as well. They seem to come with either a TjMax of 85C or 100C and no software so far can tell the difference.

If it guesses wrong your reported temps will be wrong by 15C. Lucky a smart guy figured out a possible way to determine which one is right. :D

The only thing I've found is it might be related to the manufacturing date code. I found an early E4300 with date code L640 which seemed to have a 100C TjMax and one with a date code of L644 which had a TjMax of 85C. Between week 40 and week 44 of 2006 is anyone's guess.

joebuffalo

04-22-2007, 06:06 PM

Nobody knows what the Tjmax for Core2 Duo desktop processors is. As such, it is impossible to determine the absolute core temperature of the processor.

I think your method, unclewebb, although logical, is flawed. You are still *assuming* many things to make you calculation.

You are also assuming that there is no thermal gradient within the processor. You assume that Tjunction and Tcase are equal.

You assume that the air inside your case is the same temperature as the air outside the case.

You are assuming that Tjmax is either 85 or 100.

I see what you are shooting for. But you method is based on a LOT of assumptions that I don't see the point. Why do you need to know the absolute temp, based on assumptions, when you already know the relative temp with certainty?

CoreTemp 0.95 works PERFECTLY with any Core2 processor because it measures, and can be configured to report, the DTS...which is all you need be concerned with.

fgw

04-23-2007, 03:56 AM

joebuffalo,

second that!

SLi_dog

04-23-2007, 04:21 AM

CoreTemp 0.95 works PERFECTLY with any Core2 processor because it measures, and can be configured to report, the DTS...which is all you need be concerned with.

You mean like this?

http://slidog.bigblog.com.au/data/2/11934/image/CoreTempDTSreadout4561520070423202957.png

It would be seriously great if it is. :)

It just seems like it's based on the same Tjunction value set by the software, which is the cause of this argument, but it just reads in reverse :confused: :)

fgw

04-23-2007, 05:52 AM

You mean like this?

http://slidog.bigblog.com.au/data/2/11934/image/CoreTempDTSreadout4561520070423202957.png

It would be seriously great if it is. :)

It just seems like it's based on the same Tjunction value set by the software, which is the cause of this argument, but it just reads in reverse :confused: :)
jumping in here as i requested this feature (show delta to tjunction temp) over at the coretemp development forum.

anyway, YES this is the feature!

all temp monitoring software which is using dts (digital thermal sensor) is reading the temperature from a register within the processor. this register holds the plain dts value showing the temperature offset to

tjunction temperature for mobile processors
thermal control circuit (tcc) activation temperature for desktop and server processors

absolute core temperatures are simply calculated by the following formula:
core temperature = tjunction - dts or core temperature = tcc - dts

it doesn't really matter if you call it tjunction or tcc as in both cases this temperature is just an assumption made by the writer of the monitoring software. the value used in the formula counts, not the name ...

accuracy depends on the correct chosen value for tjunction or tcc which is obvious by looking at the above formula. as intel does NOT make this values available to public nor is this value readable by software, software developers have to guess what value should be used for tjunction.
thats the difference in coretemp 94 and coretemp95! both versions of coretemp simply use different values for tjunction, either 85c or 100c. thats why readings in coretemp might be off by 15c on certain processors.

using the delta to tjunction temp instead, displays plain dts readings without conversion to an absolute temperature value thus being completely independent from probably wrong selection of tjunction value in monitoring software. using this feature in coretemp makes you completely nonindependent from tjunction and gives correct values on ALL processors supporting dts.

this delta to tjunction temp just says how many degrees is the processor away from tcc activation (throttling) and thus looks reverse to regular coretemp readings: large dts means cool processor, small dts means hot processor!

the last question here: what is a save dts value?
well as with regular coretemp readings this depends on the environment your processor is running in. my very personal rule is as follows:

running on high vcore, which to me is everything above 1.5v, i would try to keep dts above 30c. (this translates to about 55c in coretemp94)
on lower vcore, a dts value above 15c should be fine and easy achievable. (70c in coretemp94)
this assumptions take into account that higher vcore results in higher current and thus in higher current density which besides temperature is the most responsible parameter for reliability or failures due to electromigration. reducing operating temperatures maintains reliability even if current density (vcore) is raised.
the general rule is roughly: decrease temps by at least 20c for every current density increase of 100%! yes, there is some safety margin in my personal rules.

again, this is my very personal opinion gathered from reading through a lot of documents on this issues ...

unclewebb

04-23-2007, 08:29 PM

joebuffalo: You've brought up some valid points. I'm going to do some more testing and hopefully in a day or two I might have some more information for you.

I was just testing RightMark CPU Clock Utility. Version 2.25 is now capable of monitoring core temperatures of the Core2Duo.

http://www.softpedia.com/progDownload/RightMark-CPU-Clock-Utility-Download-18249.html

This utility seems to track very closely the absolute temperatures reported by Tat and CoreTemp from ambient to over 80C. It also shows you when TM1 and TM2 throttling become activated and by what percent.

The interesting thing is that it is the first utility I've seen for the C2D that displays core temperatures with a resolution of 0.2 degrees.
ie. 40.0, 40.2, 40.4, 40.6, 40.8, 41.0

Edit: Upon further testing I've discovered that this program averages the previous 5 temperature values which gives the look of better temperature resolution but it is still using the same Intel documented DTS integer value and guessing at a Tjunction value like every other program.

SLi_dog

04-23-2007, 08:54 PM

this delta to tjunction temp just says how many degrees is the processor away from tcc activation (throttling)I guess that's my question though, is it reporting the actual DTS sensor and its approach to throttling or is it just calculating the readings in an alternate method based on the application determined Tjunction?

I really have a limited idea of how these sensors work and/or I may not be configuring it right however I'm wondering if it would simply read a different "delta to Tjunction" value if I changed to a CPU that CoreTemp believes has an 85C Tjunction value? :confused:

Cheers :toast: :)

joebuffalo

04-23-2007, 11:49 PM

DTS is read *directly* from a CPU register. No calculations are involved. No assumptions are made. It is the ONLY temperature reading that is reported directly from the CPU.

EVERY software tool that says it can report an absolute core temperature reads the DTS (they all read the exact same DTS value), assumes a tjmax, and calculates an absolute temperature.

SLi_dog

04-24-2007, 12:39 AM

DTS is read *directly* from a CPU register. No calculations are involved. No assumptions are made. It is the ONLY temperature reading that is reported directly from the CPU.
So.......if it was somehow possible to change the CoreTemp Tjunction value from 100C to 85C, the Delta to Tjunction wouldn't end up being 15C lower as it's purely reporting the DTS value? :confused:

If not, how is it different to CoreTemp reporting based on an assumed Tjunction value?

Just trying to understand this, thanks for the patience :)

fgw

04-24-2007, 03:56 AM

So.......if it was somehow possible to change the CoreTemp Tjunction value from 100C to 85C, the Delta to Tjunction wouldn't end up being 15C lower as it's purely reporting the DTS value? :confused:
correct!

If not, how is it different to CoreTemp reporting based on an assumed Tjunction value?
the point here is, you simply don't know if the assumed tjunction value which is used in temperature calculation is correct or not.

e.g. if you have a dts (Delta to Tjunction) of 30c you get a core temperature displayed in coretemp of either 55c or 70c depending on tjunction value assumed by coretemp for your processor. is it now 55c or 70c your cores are running at?
on the other hand, if you are looking at plain dts (Delta to Tjunction) you can see you are still 30c away from throttling and this holds true regardless what tjunction value was selected by coretemp as no calculations and assumptions are involved.

btw. this is the same for all other programs using dts to read core temperatures. afik, coretemp is the only program giving you the ability to display dts (Delta to Tjunction) directly and thus is the preferred choice!

SLi_dog

04-24-2007, 05:01 AM

correct!

the point here is, you simply don't know if the assumed tjunction value which is used in temperature calculation is correct or not.

e.g. if you have a dts (Delta to Tjunction) of 30c you get a core temperature displayed in coretemp of either 55c or 70c depending on tjunction value assumed by coretemp for your processor. is it now 55c or 70c your cores are running at?
on the other hand, if you are looking at plain dts (Delta to Tjunction) you can see you are still 30c away from throttling and this holds true regardless what tjunction value was selected by coretemp as no calculations and assumptions are involved.

btw. this is the same for all other programs using dts to read core temperatures. afik, coretemp is the only program giving you the ability to display dts (Delta to Tjunction) directly and thus is the preferred choice!But don't you find it strange that the "Delta to Tjunction" throttle point with the quad core co-incides exactly with the software written Tjunction value of 100C? :confused:

I guess I'm still not convinced but I'll bow to your knowledge guys. Thanks for the help and patience :toast: :toast:

coldpower27

04-24-2007, 09:21 AM

My Delta to Tjunction at load is about 25 Degrees, with it being 50 Degrees at idle.

unclewebb

04-24-2007, 10:48 AM

Here's a guy on the [H] forum that decided to take my test using his E6400 revision L2.

http://www.hardforum.com/showpost.php?p=1030957763&postcount=87

Though his heatsink and fan are far superior at cooling then my OEM heatsink and fan, he ended up with the exact same results.
His absolute core temperature with an assumed Tjunction of 85C was equal to his ambient temperature.

The assumed Tjunction of 100C that CoreTemp 0.95 uses for his L2 E6400 processor doesn't seem right. That's all I'm really trying to prove.

joebuffalo: "You assume that the air inside your case is the same temperature as the air outside the case.

No I didn't. The case was open and the temperature probe was beside the cpu fan.

"You are assuming that Tjmax is either 85 or 100.""

The Core2Duo mobile processors with the exact same CPUID of 6F6 have a documented TjMax of a fixed 100C or 85C. Why would Intel change their manufacturing process for the Core2 Duo desktop processors that have the same 6F6 CPUID? They're the same processor built on the same manufacturing process. The only difference is the packaging they're shoved into with a different pin count for a mobile or desktop application.

Assuming TjMax isn't a fixed value of 85C or 100C is no different than assuming it is a fixed value. Until documentation is released by Intel to clarify this, neither side can be proven to be right or wrong.

Try taking my Tjunction test and let's see what kind of results you get. I'm very interested in results from E4300 users.

joebuffalo

04-24-2007, 11:50 AM

No offense, unlcewebb, but not only will I not take your test, I will advise others against taking your test. It confuses the matter (for those that don't understand Core2 temperature readings) and serves no purpose (for those that do understand Core2 temperatures).

The only thing anybody with a Core2 Duo desktop processor should do is read their DTS directly from CoreTemp 0.95

joebuffalo

04-24-2007, 11:58 AM

But don't you find it strange that the "Delta to Tjunction" throttle point with the quad core co-incides exactly with the software written Tjunction value of 100C? :confused:

Actually, it makes *perfect* sense:

The software reads the DTS directly from the CPU as '55'
The software *assumes* a tjmax of '100'
The software *calculates* a core temperature of '45'

cmos.gr

04-25-2007, 12:29 PM

kwalker

04-25-2007, 06:36 PM

Why not just use the ''clasic'' diode from PECI ?...

Mine results with an 6400 at 3600@ 1.417v watercooled 24C room 26-27C Radiators(3x dual nexXxos) are...

idle core temp 39-41c (ok i will tell the max of one core only)
idle PECI from AI SUITE 32-33

Full load orthos after 30s
Full core temp : 65c
Full PECI AI SUITE 55C

After some tests i realize that core temp was WRONG.It says 10C More under load and 7-8C on idle.

In my opinion if all previous posts are right them...coretemp is just for some cores/cpus.
One reason you won’t be able to rely on the PECI reading of the on die thermal diode is that Intel has discontinued or disabled the Diode on newer CD2 processors and only the DTS is used for thermal management and fan control...
Perhaps better support is inevitable for developers of software monitoring core temperature.
Every processor package varies slightly and T-junction being set per package during manufacturing and not software readable makes this a chore to determine the actual thermal design limit.
If you run Ortho’s for any length of time at max load you will notice the margin of error is less from core temp to the diode reading and this varies from build to build.
For now that’s what we have and it is not perfect
In some cases just wrong.
I rely on the motherboard software more than Core temp because of the discrepancies.
If it’s stable why worry.:cool:

cmos.gr

04-26-2007, 08:15 AM

uhm so...

If someone want to see if his core temp or any prog reports right temperature the only thing he has to do is to stop his cooler's fan or pump for watercool and see if at 85 (reported) temp his CPU will start throttling..
If yes then temp progs like core temp are fine.

Right ?

Edit:

I just test mine and...
At reported temp of 83C° (core temp) cpuz gone from 2128mhz to 1600mhz (default 6400 to make sure TM is working fine)

Now what ? The right temp is not only higher (than PECI) but +2C° more than core temp's reported temperature ?

Sh..t thats TOO bad uh ? :S :S :S

kwalker

04-26-2007, 08:20 AM

In theory Yes.
If your a brave soul:)

cmos.gr

04-26-2007, 08:56 AM

Lol thats not a prob..ill not cry for them :P

Ok so test # 2 (from a guy at hard forums)

266fsb*6multi@ 1.07vcore.

Radiator Temperature 25C (+-1 or 2 cause its from a gatewatch panel...(i need to find fast a good tool for temp)
IDLE
Motherboard temp(or peci I really don't know wtf is that temp) : 24C (impossible ?! but NEAR radiator temperature)

Core 1 :31C
Core 2 :30C

Its +- 5C° more than radiator temperature.
I cant believe this can be right with these settings.Only 1600mhz and 1.07vcore IDLE...

Argh I'm confused:confused:
Or finaly I must say to my self that I have a really hot chip.what can I say :S

kwalker

04-26-2007, 09:39 AM

Those temps seem to be in line considering the applied voltage.
As for core temp TAT or speed fans functionality I’ll do a quick comparison between platforms.
My core duo lap top with TAT or Orthos running full load and no overclock the temperature reported by TAT and core temp never exceed 47C .
This is in a little flat black box with virtually no air flow and a heat pipe with a fan on the end exhausting out the back..
Using TAT and core temp in my desktop with excellent air flow and high a high end liquid cooling solution the temperatures will soar upward of 60C(TAT) 51C orthos.
This could be the fact that TAT was designed for the mobile platform and similarly core temp seems to be more accurate at low applied voltages or higher clocks at max load.
The discrepancy I have noted is in the area where Fan speed control is commanded or TCC activation.
Is core Temp accurate?
In many cases yes.
Then again Intel states that the Bios or Intel interrogator software is accurate within 3C.
Some how I believe in the very near future you will see Core temp to be very reliable but at this time there is no 100% accurate temperature reporting utility.
Like I said before
If its stable why worry.

cmos.gr

04-26-2007, 10:08 AM

That was a nice example...

Only a word on this

If its stable why worry.

If it was stable at your temp ( 51C orthos )I wont say anything...:(
But here I see a 8-10C difference between 3 Sensors (numbers lets say better).

Please if anyone can to this test...post his results.:toast:
Post the CPU , Core 0 and Core 1 idle from SpeedFan 4.32
Then
Enable PECI from bios and post the same temps Cpu, core 0 and core 1.
I test that in 2 systems with same mobo different cpus and i got different results.:confused:
On my system CPU temp (peci anyway) had +-10 temp less than core 0-1
On other system CPU temp was exactly the same with core 0-1.:stick:

And one more thing.
Tjunction is at 85C ?Thats a standard ? 100% ?
Lets say...it cant be at 80 ? so coretemp or any program uses DTS report wrong temp ?Cause mine for example was not as 85 but in 83 (IF temp was right then why Tjunction enabled at 83 and not at 85 for my cpu?...:rolleyes: )

I'm really stuck on that now :dammit:

joebuffalo

04-26-2007, 05:24 PM

Here are the things we know 100% for certain regarding Core2 Temperatures:

1. Nobody knows for sure what the Tjmax is for a Core2 Desktop Processor

2. DTS readings are always 99% percent accurate and should not be questioned

3. CoreTemp 0.95 can be configured to display DTS readings directly

4. Absolute temperatures are unnecessary and confusing. Learn to embrace DTS.

cmos.gr

04-27-2007, 01:38 AM

2. DTS readings are always 99% percent accurate and should not be questioned

So why I have still 2-3 C to Tjunction but CPU started already?
Who says that the real Tjuncton isn't at 75C ?(at this temp +- (i don't remember for sure)reported from Tcase diode.)
CoreTemp still calculates with report point 85C which isn't the real Tjunction.

And if..

3. CoreTemp 0.95 can be configured to display DTS readings directly

DTS readings (from CoreTemp)in my case are wrong.
I tested.:nono:
Its just the reverse:nono:

fgw

04-27-2007, 04:53 AM

So why I have still 2-3 C to Tjunction but CPU started already?
first of all, on desktop and server processors the reference point used is not tjunction but tcc (thermal control circuit) activation temperature instead! tjunction is used on mobile processors only.
i know coretemp is calling this reference temperature tjunction on all processors. tat does the same, but as tat is designed for mobile processors only it is naming this reference temperature correct. guess thats the reason many people call this reference point tjcunction.
anyway, this does not change anything in the resulting temperatures ...

so why have you seen throttling starting at a coretemp reading of 83C? coretemp uses a sample rate of 1 second by default. the last time coretemp read the dts value coretemp might have read 2C and thus calculated 83C absolute temperature, but as dts and tcc reacts very fast it started throttling even before coretemp was able to read the next dts value a second later.
you can try to set the temperature read interval in coretemp to a lower value and you might see different temperatures where throttling starts.

besides that, the temperature where tcc gets activated is calibrated during manufacturing on a per part basis and thus might differ from processor to processor. regardless of coretemp, which will read anything near 85C or near 100C (depending on the assumed reference temperature), your processor might start throttling at 83C, 84C or even 87C. you simply cant tell, cause its impossible to convert dts readings into an absolute temperature.

thats the reason i requested the display of plain dts values through coretemp. which was implemented in coretemp095. instead of trying to keep your processor below a certain temperature you should try to keep dts readings ABOVE a certain value which means staying away from tcc activation by this value.

Who says that the real Tjuncton isn't at 75C ?(at this temp +- (i don't remember for sure)reported from Tcase diode.)
there is no tcase diode in c2d processors! the td (thermal diode) you are referring is like the dts (digital thermal sensor) located on the die itself and thus reports processor temperature directly from the die. the dts has a smaller foot print than the td and thus can be located closer to hotspots on the die. also dts reacts very fast and dts temperatures can be read directly as a 7bit value from a register. in contrast td needs two pins on the processor socket where an electric signal can be measured which needs to be calibrated and converted into a temperature! thus accuracy of temperature readings via td rely on calibration ond conversion. as this is done in bios, its up to the bios what temperatures are read via td. the bios might calibrate this temp to simulate a temperature similar to tcase definitions by intel or it might be calibrated to simulate a temperature which is close to core temperatures. you simply don't know what the bios writer had in mind.

regarding tcase:
intel defines tcase as the temperature measured at the geometric center of the ihs on top of the ihs surface! there is no temperature sensor nor a diode reporting this temperatur. it's just a place outside the processor where temperatures may be measured ...

CoreTemp still calculates with report point 85C which isn't the real Tjunction.

And if..

DTS readings (from CoreTemp)in my case are wrong.
I tested.:nono:
Its just the reverse:nono:
as dts temperatures are an offset to tcc activation, coretemp as all other software using dts has to calculate absolute temperatures by subtracting dts readings from tcc activation temperature. the problem here is this tcc activation temperature, as mentioned above, is different from part to part and once configured during manufacturing can't be changed nor read out by any software! this tcc activation temperature simply has to be assumed somehow!
on mobile processors a bit in a register exists which tells if tjunction (which is used on mobile processors only) is at 85C or at 100C.
this bit is NOT valid on desktop and server processors. depending on the content of this single bit, temp monitoring software assumes if 85C or 100C has to be used in subsequent calculations. this assumption might be close to reality on many processors, but definitely is about 15C off on some other processors.

DTS readings are always 99% percent accurate and should not be questioned
yes, DTS readings are correct! the problem is in the conversion into absolute temperature values as explained above. here comes inaccuracy into play ...

SLi_dog

04-27-2007, 04:57 AM

DTS readings (from CoreTemp)in my case are wrong.
I tested.:nono:
Its just the reverse:nono:I originally thought that this was the case as well but according to the creator of CoreTemp, the configured "Delta to Tjunction" temperature are direct DTS readings only, no calculation is involved.

3. CoreTemp 0.95 can be configured to display DTS readings directlySeems you were right Joe :toast:

unclewebb

05-03-2007, 08:54 AM

so why have you seen throttling starting at a coretemp reading of 83C? coretemp uses a sample rate of 1 second by default. the last time coretemp read the dts value coretemp might have read 2C and thus calculated 83C absolute temperature, but as dts and tcc reacts very fast it started throttling even before coretemp was able to read the next dts value a second later.
I think what cmos.gr has noticed is that processor throttling begins at a CoreTemp 0.95 DTS value of 2.

By controlling the cpu fan speed I can bring the temperature up very gradually and TAT will first report "Thermal Monitor Active" in red when DTS=2. RightMark also reports at this point "CPU core overheat detected" and shows that some form of throttling is taking place.

By adjusting the fan speed I can hold DTS=2 without letting it ever get to zero. If I turn up the fan speed a little to cool things down, as soon as CoreTemp reports DTS=3, the TAT Thermal Monitor Active light turns back to green and changes to Thermal Monitor Idle.

This testing seems to show that processing power first gets reduced at DTS=2. I don't think the TAT reported MHz gets reduced until DTS=0 but that doesn't mean that the processor isn't already throttling.

http://img209.imageshack.us/img209/2059/tattestingep4.png

Note: Just to clear things up. I booted up at 425 X 8 = 3400 MHz and then used ClockGen to reduce it to 400 X 8 = 3200 MHz for better stability at full load and full temp before starting this test. CoreTemp reports the wrong MHz but the 3200 MHz in TAT is correct. This first level of throttling does not reduce the reported 3200 MHz the processor was originally running at.

On a side note, does anyone know why TAT grays out CPU1 sometimes as in the above picture?

cmos.gr

05-03-2007, 09:44 AM

Actually my cpu reduced mhz at DTS = 2

I think your setup/configure is not able to slow down your processor...

This happen to me..Only with default setting it could slow down cpu speed.

unclewebb

05-03-2007, 01:28 PM

I did another test and this time I let the temperature go higher until DTS=0.

TAT held steady at 3200 MHz but CPUz showed it being reduced to 3167 MHz and once it dropped down to the low 2000 MHz range but I missed a screenshot of that.

CoreTemp showed that the VID had been reduced from 1.325 to 1.1625 volts. SpeedFan and CPUz showed the core voltage steady at 1.392 volts so the Asus P5B might lock the CPU voltage and ignore any requests to lower it when it is not set to AUTO in the bios.

TAT and SpeedFan occasionally reported absolute temps of -40C to -42C when the temps seemed to go higher.

http://img235.imageshack.us/img235/6770/tattesting2jq3.png

joebuffalo

05-04-2007, 12:09 AM

Wow, now that's a hot processor!

This is interesting work you are doing and I it will surely help us to better understand the Core2 Desktop processors. I would point out though that these new observations should not be misconstrued as evidence that the DTS reading is innaccurate. I made notes before of when other people who ran their processors to crazy high temps like you. I'll see if I can find those so we can do some comparisons.

Do you have both TM1 and TM2 enabled in your BIOS (or maybe they are disabled in some software - I do know I've seen check-boxes for these in RightMark)? That could possibly affect what happens as you approach DTS=0.

joebuffalo

05-04-2007, 12:14 AM

Here is one link where sombody else saw throttling at DTS=2:

http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=1588903#1588903

joebuffalo

05-04-2007, 12:39 AM

I had not seen this doc before (though you posted the link a long time ago) so I thank you for that unclewebb.

http://www.intel.com/design/processor/manuals/253668.pdf

Reading this document makes it clear that if we have a clever programmer and an intrepid tester, we can learn a lot about how this stuff is working. That MSR contains a lot of info: has thermal monitoring taken place recently? the ability to enable thermal throttling via software, etc. Very interesting.

joebuffalo

05-04-2007, 12:50 AM

I had mentioned a few days (weeks?) back that I thought that the Tj_max was *individually* set for *each* processor. If that were the case, we can all agree that absolute temperature are ABSOLUTELY POINTLESS, right? If you have no clue what Tj_max is, you can never accurately convert a DTS reading into an absolute temperature. NEVER. Do we agree?

Good. Now read this. I can't believe I read this like a month ago and totally forgot about it. I should have shared this with more people. It's a MUST READ for those who want to understand Core2 temperatures.

http://documents.irevues.inist.fr/bitstream/2042/6579/1/TMI23.pdf

Allow me to quote my two favorite lines:

"To achieve measurement accuracy, each sensor is calibrated at test time. Calibration is done for the Maximum Tj and the linearity of the readout slope. The temperature reading is post processed for filtering out random noise and generating the H/W activated thermal protection functions."

and

"The DTS is calibrated at manufacturing conditions and the reference point is set to this test temperature. Functionality, electrical specifications and reliability commitments are guaranteed at maximum Tj as measured by the DTS. Any test inaccuracy or parameters variance are already accounted for in the DTS set point"

Allow me to summarize: Absolute temperture *calculations* on Core2 processors are *pointless*. DTS *readings* are accurate and factory calibrated.

Embrace DTS. ;)

joebuffalo

05-04-2007, 12:58 AM

While I'm at it I might as well link to everything else useful I've collected on this topic:

http://www.intel.com/design/core2duo/documentation.htm

http://www.intel.com/technology/itj/2006/volume10issue02/art01_intro_to_core_duo/p06_thermal_design_point.htm

http://www.intel.com/technology/itj/2006/volume10issue02/art03_power_and_thermal_management/p03_power_management.htm

ftp://download.intel.com/design/intarch/designgd/31527902.pdf
Intel® CoreTM 2 Duo E6400 and E4300 Processors for Embedded Applications

4.2.10 Digital Thermal Sensor
The Intel® Core™2 Duo desktop processor E6000 sequence introduces the Digital
Thermal Sensor (DTS) as the on-die sensor to use for fan speed control (FSC). The DTS
will eventually replace the on-die thermal diode used in pervious products. The
processor will have both the DTS and thermal diode enabled. The DTS is monitoring the
same sensor that activates the TCC (Refer to Section 4.2.2). Readings from the DTS
are relative to the activation of the TCC. The DTS value where TCC activation occurs is
0 (zero).
The DTS can be accessed by two methods. The first is via a MSR. The value read via the
MSR is an unsigned number of degrees C away from TCC activation. The second
method which is expected to be the primary method for FSC is via the PECI interface.
The value of the DTS when read via the PECI interface is always negative and again is
degrees C away from TCC activation.

4.2.2 Thermal Control Circuit
The Thermal Control Circuit portion of the Thermal Monitor must be enabled for the
processor to operate within specifications. The Thermal Monitor's TCC, when active, will
attempt to lower the processor temperature by reducing the processor power
consumption. In the original implementation of thermal monitor this is done by
changing the duty cycle of the internal processor clocks, resulting in a lower effective
frequency. When active, the TCC turns the processor clocks off and then back on with a
predetermined duty cycle. The duty cycle is processor specific, and is fixed for a
particular processor. The maximum time period the clocks are disabled is ~3 µs. This
time period is frequency dependent and higher frequency processors will disable the
internal clocks for a shorter time period. Figure 7 illustrates the relationship between
the internal processor clocks and PROCHOT#.
Performance counter registers, status bits in model specific registers (MSRs), and the
PROCHOT# output pin are available to monitor the Thermal Monitor behavior.

4.2.3 Thermal Monitor 2
The processor supports an enhanced Thermal Control Circuit. In conjunction with the
existing Thermal Monitor logic, this capability is known as Thermal Monitor 2. This
enhanced TCC provides an efficient means of reducing the power consumption within
the processor and limiting the processor temperature.
When Thermal Monitor 2 is enabled, and a high temperature situation is detected, the
enhanced TCC will be activated. The enhanced TCC causes the processor to adjust its
operating frequency (by dropping the bus-to-core multiplier to its minimum available
value) and input voltage identification (VID) value. This combination of reduced
frequency and VID results in a reduction in processor power consumption.

ftp://download.intel.com/design/processor/datashts/31327803.pdf

Intel® Core2 Extreme Processor X6800 and Intel® Core2 Duo Desktop Processor E6000 and E4000 Sequences

5.1.1 Thermal Specifications

5.2.4 PROCHOT# Signal

An external signal, PROCHOT# (processor hot), is asserted when the processor core
temperature has reached its maximum operating temperature. If the Thermal Monitor
is enabled (note that the Thermal Monitor must be enabled for the processor to be
operating within specification), the TCC will be active when PROCHOT# is asserted. The
processor can be configured to generate an interrupt upon the assertion or deassertion
of PROCHOT#.
As an output, PROCHOT# (Processor Hot) will go active when the processor
temperature monitoring sensor detects that one or both cores has reached its
maximum safe operating temperature. This indicates that the processor Thermal
Control Circuit (TCC) has been activated, if enabled. As an input, assertion of
PROCHOT# by the system will activate the TCC, if enabled, for both cores. The TCC will
remain active until the system de-asserts PROCHOT#.

5.2.5 THERMTRIP# Signal

Regardless of whether or not Thermal Monitor or Thermal Monitor 2 is enabled, in the
event of a catastrophic cooling failure, the processor will automatically shut down when
the silicon has reached an elevated temperature (refer to the THERMTRIP# definition in
Table 26). At this point, the FSB signal THERMTRIP# will go active and stay active as
described in Table 26. THERMTRIP# activation is independent of processor activity and
does not generate any bus cycles.

Tabel 26

THERMTRIP#

In the event of a catastrophic cooling failure, the processor will
automatically shut down when the silicon has reached a
temperature approximately 20 °C above the maximum TC.
Assertion of THERMTRIP# (Thermal Trip) indicates the processor
junction temperature has reached a level beyond where permanent
silicon damage may occur. Upon assertion of THERMTRIP#, the
processor will shut off its internal clocks (thus, halting program
execution) in an attempt to reduce the processor junction
temperature. To protect the processor, its core voltage (VCC) must
be removed following the assertion of THERMTRIP#. Driving of the
THERMTRIP# signal is enabled within 10 µs of the assertion of
PWRGOOD (provided VTT and VCC are valid) and is disabled on deassertion
of PWRGOOD (if VTT or VCC are not valid, THERMTRIP#
may also be disabled). Once activated, THERMTRIP# remains
latched until PWRGOOD, VTT, or VCC is de-asserted. While the deassertion
of the PWRGOOD, VTT, or VCC will de-assert THERMTRIP#,
if the processor’s junction temperature remains at or above the trip
level, THERMTRIP# will again be asserted within 10 µs of the
assertion of PWRGOOD (provided VTT and VCC are valid).

5.3 Thermal Diode

The processor incorporates an on-die PNP transistor where the base emitter junction is
used as a thermal "diode", with its collector shorted to ground. A thermal sensor
located on the system board may monitor the die temperature of the processor for
thermal management and fan speed control. Table 31,Table 32, and Table 33 provide
the "diode" parameter and interface specifications. Two different sets of "diode"
parameters are listed in Table 31 and Table 32. The Diode Model parameters (Table 31)
apply to traditional thermal sensors that use the Diode Equation to determine the
processor temperature. Transistor Model parameters (Table 32) have been added to
support thermal sensors that use the transistor equation method. The Transistor Model
may provide more accurate temperature measurements when the diode ideality factor
is closer to the maximum or minimum limits. This thermal "diode" is separate from the
Thermal Monitor's thermal sensor and cannot be used to predict the behavior of the
Thermal Monitor.

http://download.intel.com/design/processor/designex/31368502.pdf

Intel® Core™2 Duo Desktop Processor E6000? and E4000? Sequence Thermal and Mechanical Design Guidelines

4.2.10 Digital Thermal Sensor
The Intel® Core™2 Duo desktop processor E6000/E4000 sequence introduces the
Digital Thermal Sensor (DTS) as the on-die sensor to use for fan speed control (FSC).
The DTS will eventually replace the on-die thermal diode used in pervious products.
The processor will have both the DTS and thermal diode enabled. The DTS is
monitoring the same sensor that activates the TCC (see Section 4.2.2). Readings from
the DTS are relative to the activation of the TCC. The DTS value where TCC activation
occurs is 0 (zero).
The DTS can be accessed by two methods. The first is via a MSR. The value read via
the MSR is an unsigned number of degrees C away from TCC activation. The second
method which is expected to be the primary method for FSC is via the PECI interface.
The value of the DTS when read via the PECI interface is always negative and again is
degrees C away from TCC activation.

joebuffalo

05-04-2007, 01:05 AM

OK, I was re-reading the info in the above post. Important line to read: "The DTS value where TCC activation occurs is 0 (zero)."

cmos.gr

05-04-2007, 07:07 AM

So...When someone see DTS=0 that doesn't mean his cpu is at 85C.
He may have 60C...or 70C or 80C or 90C.

Right ?

joebuffalo

05-05-2007, 05:03 PM

So...When someone see DTS=0 that doesn't mean his cpu is at 85C.
He may have 60C...or 70C or 80C or 90C.

Right ?

That is the gist of it.

unclewebb

05-08-2007, 07:38 AM

joebuffalo: Thermal Monitoring is enabled in my bios and Rightmark shows that both TM1 and TM2 are enabled.

Here's some more high temp experimenting:

http://img238.imageshack.us/img238/9083/tattesting4ht3.png

What I found interesting today is that once again TAT went red and showed Thermal Monitor Active when CoreTemp got up to DTS=2 but TAT never reported more than 79C to 80C even as I held it at the boiling point for several minutes.

Here's a partial look at the TAT log:

http://img67.imageshack.us/img67/1011/temploguu3.gif

The other day a CoreTemp reading of DTS=2 was getting me TAT reported temps of 83C and today only 80C for the exact same DTS value. There doesn't seem to be too much logic when a computer program varies its output depending on the day of the week or who knows what.

joebuffalo

05-08-2007, 09:31 AM

From my experience, TAT tends to read a degree or two below CoreTemp

In the last screen shot you posted (previous page) you had TAT=83 when DTS=0

unclewebb

05-08-2007, 12:25 PM

I find it odd that my first screen shot on the previous page, TAT + DTS = 85 and then on the second screen shot TAT + DTS = 83 and now this last screen shot TAT + DTS = 82.

If TAT is using the same DTS value that every other program is using then on the same computer, it should be consistent from day to day but it's not.

I can understand when there are timing differences when a sensor is read but in each case I tried to stabilize the temps before going for a screen shot.

Some people trust TAT because it has the nice Intel logo on it but I find that CoreTemp is a lot more consistent.

joebuffalo

05-08-2007, 02:29 PM

Agreed. Reason I don't like TAT (for reading temp) is because it will not display DTS directly and we do not know how it using DTS to calculate the absolute temperature. Two unknowns :(

RealTelstar

05-26-2007, 05:13 PM

I'd like to bump this very informative thread :)

Coffee29

05-26-2007, 10:51 PM

This is probably the best thread on desktop core 2 duo temperatures on the web. I spent a week trying to find out why my temps were reading 15 degrees one way or the other depending on the program. My search led me to the intel forums, product documentation, and to the "temperature guide" from that other forum, Im sure you know the one.

I added up all of the facts and compared them to what was being said and I realized that most of the info on the web about reading core2duo temps is misinformation. I finally concluded that there is NO way to get an absolute temp on these desktop chips, but I also realized that this is ok because one can accurately utilize DTS and temp readings on former chips probably were not that accurate anyway. DTS is act ually a great thing.

IT was at this point that I googled DTS and C2D chips and found this thread. It was refreshing to see you guys coming down to the very same conclucion...you are right on....everyone needs to embrace DTS!

Coffee29

05-26-2007, 11:07 PM

a few more things that I think need to be restated due to all the misinformation, please correct me if I am wrong:

1. TAT is NOT intended to be used with desktop C2D processors and WILL NOT give you accurate temps because temps for desktop c2d cannot be known. Mobile c2d's have a set tjunction and TAT works fine for them.
2. The word "Tjunction" itself only applies to mobile c2d's.
3. The whole reason programs are giving the 15 degree variance is because they are assuming tjunctions of 85 or 100...refer to number 2--->tjunction DOES NOT apply to desktop c2d's.
4. I am not sure about this but I think tjunction DOES apply to quads---> correct me if Im wrong. If it does and if they are documented at 100 TAT would work just fine I would assume.
5. It seems that the consensus is that DTS readings, which can be obtained from the newest coretemp are safe if above 20 or so, generally speaking.
6. Coretemp confuses the issue with its even mentioning "tjunction", this word and its use needs to be deleted from the entire program and the program should only read DTS. Even when you set it to read dts it states x dts to tjunction....this should instead read so x dts to TCC. TCC is the trip noted in the intel documents.
7. Changing the tjunction in coretemp should not affect the DTS, it looks like it would and for this reason the program is confusing to mnay...so remember DTS is not a calculated number, it is simply a readout of what the diods are reporting.

Please correct me where I went wrong, but from the forums around town it seems like there still is A LOT of confusion over this.

redrol

05-29-2007, 06:01 PM

This thread is awesome. I have embraced DTS. :D

unclewebb

05-31-2007, 09:45 PM

I finally bought an IR Thermometer to do some testing with. I've posted some interesting information over on the [H]ard forum but unfortunately the site is down at the moment.

I have an early revision B2 E6400 which every program including TAT assumes a TjMax of 85C. Using my method of lowering the core voltage and lowering the MHz as low as they go I was able to remove the fan off of the OEM heatsink and get a very accurate temperature reading directly off the copper cone in the center of the heatsink. With CoreTemp reporting 21C, the measured temperature of the heatsink was 27C.

This of course is impossible. A heat source at 21C can't possibly heat up the heatsink to 27C. After this test and others I seriously doubt if there are any C2D processors with a TjMax of 85C. All mobile Core processors, Solo, Duo, Core 2 Duo, all have an Intel documented TjMax of 100C. I continue to believe that the Desktop Core processors also use the exact same TjMax of 100C.

DTS is still the best way to go but for those that are interested in an absolute temperature, I would only use software that assumes a TjMax = 100C.

Next step is to try and run the processor without a heatsink with only a highspeed 80mm hand held fan keeping it cool. If it's possible to run a C2D like this without the core temps going through the roof then I will be able to get a direct temp reading of the cores without a heatsink getting in the way, reducing the true core temperature.

Should have some interesting results by next week if nothing catches on fire!

Prefixxx

06-01-2007, 01:15 AM

Curious about your findings unclewebb :)

Me also having problems with the understanding of my own E6600 temps.

Hardware :

- E6600
- Asus P5B-e
- Asus 8800 GTS 320 mb
- Antec NSK6000 (front 2x 92 mm fans, bottom 120 mm intake fan @ 800 rpm and back outtake 120 mm AC fan).

Temps are :

Ambient 21.3 celcius

- IDLE
CPU : 42
Motherboard : 40
Core 0 : 23
Core 1 : 23

(on the screenshots after a cold boot)

- STRESSED
CPU : 56
Motherboard : 41
Core 0 : 41
Core 1 : 41

As you can see the CPU temp is higher then Core temp.
Are those Core temps realistic with this Ambient temperature?.

I started a thread on Tom's Hardware asking about this and some of their members say that CPU temp is way to high (55 is max for Orthos) as I have +/-56).

Quote from another member @ Tom's Hardware

Dude, the CPU temp of 56-60 is what is reading accurately, your two core temps are the ones which are inaccurate and need to be offset. Read the C2D temp guide. By the way, your accurate cpu temp of 56 is too high, the guide explains that you should never exceed 55 with orthos, 60 with TAT.

..

What that means is your actually loading @56/71-71 <---> 60/75-75 which is too high. Stay bellow 55/70-70 using Orthos.

For @56/44-44 to be accurate or what ever, that would mean you are water cooled. By deductive logic, the 44s are then the ones reading inaccurately. 44 Tjunction doesnt make sense on Orthos unless you are watercooling, meaning it definately needs to be offset by +~27. I had to offset mine by +~25. Same boat as you.

As you can see and already sayed in this thread, it's get a bit (alot) confusing :)

-- Screenshots --

http://img108.imagevenue.com/loc1146/th_58409_29mei_idle_no120sidefan_ambient_21_122_11 46lo.jpg (http://img108.imagevenue.com/img.php?image=58409_29mei_idle_no120sidefan_ambien t_21_122_1146lo.jpg) - http://img41.imagevenue.com/loc785/th_58417_29mei_idle_no120sidefan_ambient_21_2_122_ 785lo.jpg (http://img41.imagevenue.com/img.php?image=58417_29mei_idle_no120sidefan_ambien t_21_2_122_785lo.jpg) - http://img41.imagevenue.com/loc943/th_58428_29mei_stressed_no120sidefan_ambient_21_12 2_943lo.jpg (http://img41.imagevenue.com/img.php?image=58428_29mei_stressed_no120sidefan_am bient_21_122_943lo.jpg) - http://img153.imagevenue.com/loc850/th_58439_29mei_stressed_no120sidefan_ambient_21_2_ 122_850lo.jpg (http://img153.imagevenue.com/img.php?image=58439_29mei_stressed_no120sidefan_am bient_21_2_122_850lo.jpg)

kwalker

06-01-2007, 07:32 AM

Note the inconsistencies between the Tcase temp and core temp in the last two replies
This is on the rare side and related to motherboard bios revision and or software corruption.
Referring to Tcase as T-junction is just plain silly.
T-juntion (or TCC activation) is fixed and between 85 and 100 C it is not software readable but closes enough to reference the actual digital temperature reading which the software calculates for you.
A Tcase reading in the 50s under load is well within the thermal guidelines and a temperature of 60C max is fine. (The advice was incorrect)
I have no opinion of TAT that I can share being that its design was intended for the use in the mobile platform
The problem is the discrepancy where Tcase is higher than core temp.
This makes both readings unreliable.
Update your bios and use the motherboard reporting software to monitor your temps.
Perhaps this will resolve the issue but there are no guarantees.
Uncleweb……
You’re bold.:D
Just an opinion,
Using an IR thermometer is acceptable for cooking a Turkey or even approximating ram temperature but highly inaccurate for thermal analysis that refers to internal temperatures or at least Tcase which requires an appropriate heat sink to stay within the thermal design guidelines.

blacksun1942

06-01-2007, 07:38 AM

56C is NOT TOO HIGH, whoever told you that you shouldn't exceed 55C is feeding you BS.

56C is WELL WITHIN NORMAL RANGE when running Orthros.

What that means is your actually loading @56/71-71 <---> 60/75-75 which is too high. Stay bellow 55/70-70 using Orthos.

This is meaningless.

...he says "your *accurate* cpu temp of 55 is too high"?? If 55C is accurate then you're absolutely within normal and safe limits. He has NO CLUE what he's talking about.

Trust me, you're fine.

unclewebb

06-01-2007, 07:45 AM

This is how software tries to determine your TjMax:

http://softwarecommunity.intel.com/isn/Community/en-US/forums/permalink/30222546/30228327/ShowThread.aspx#30228327

"Next you have to assume that the Tjunction is 100 degrees Celsius and if bit 30 of MSR 0xEE is set then it is 85 degrees Celsius."

The Intel rep says at the end of this thread that this is NOT valid for Desktop processors but I think temperature monitoring software continues to do this.

I used CrystalCPUID to do a RDMSR 0xEE and my E6400 shows bit 30 is set.

http://img477.imageshack.us/img477/8653/rdmsr0xeegv8.png

I can understand TAT reading this bit since TAT was designed specifically for mobile processors and has never been updated for the desktop core processors but if CoreTemp and SpeedFan are using this method to determine that my processor has a TjMax = 85C then they are wrong. All of these programs are in complete agreement but that doesn't make them all right. Instead it makes them all wrong.

At idle if my heatsink is 6C above my reported core temperature then my reported core temperature is wrong. The DTS represents the hottest point on the core. The heat transfer from the core through the IHS, through the AS5 paste and through the copper heatsink should result in a lower temperature for the heatsink and definitely not higher.

Using a TjMax = 100C for my processor gives me a lot more realistic number for my absolute core temperature.

I believe that TjMax is a fixed value but the other guys on this forum could be correct and my TjMax might actually be 95C or some other number around 100C. If I can get the heatsink off I think the IR thermometer should get me some temperatures very close to the real core temperature.

All I know so far is that assuming a TjMax=85C for my revision B2 Conroe core processor is wrong.

Also keep in mind that I am only using the Intel OEM heatsink which certainly isn't the most efficient these days. I have a Tuniq tower arriving later today which typically results in even lower reported core temperatures.

unclewebb

06-01-2007, 08:03 AM

Using an IR thermometer is acceptable for cooking a Turkey or even approximating ram temperature but highly inaccurate for thermal analysis...
When I first got this toy and started using it I was getting some very inaccurate results. The shiny copper cup at the center of the heatsink could not be read properly so I covered it in some thin masking tape to take the shine off of it.

After doing that, the readings I'm getting now are very consistent. I unplugged my computer and let it sit for over an hour. When I came back the temperature of this cone was exactly equal to the ambient temperature so my readings of this cone are accurate.

The only way I'll be able to run a C2D with the heatsink off is to have a hand held fan blowing cool air on it. At low MHz and voltage, my C2D is only putting out about 20 watts of heat so this is certainly possible. If I can do this and can get a stable CoreTemp reading then I might be able to go in with the IR gun and get a reading for comparison. If nothing else it will certainly be interesting! Worst case, I see a new E6420 in my future. :D

My old Pentium III lived through worse torture than this.

RealTelstar

06-01-2007, 08:21 AM

Considering simple matters as the ambient temp, frequency and cooling that I have, I'm pretty confident of my assumptions.

coretemp 0.95 reports the following:
core0: 31°C (delta 54°C)
core1: 28°C (delta 57°C)
read interval 1000ms

My ambient is 21°C measured by analogic quality thermometer. The assumed TCC is 85° (by coretemp) and it seems accurate to me.

Prefixxx

06-01-2007, 08:32 AM

56C is NOT TOO HIGH, whoever told you that you shouldn't exceed 55C is feeding you BS.

56C is WELL WITHIN NORMAL RANGE when running Orthros.

This is meaningless.

...he says "your *accurate* cpu temp of 55 is too high"?? If 55C is accurate then you're absolutely within normal and safe limits. He has NO CLUE what he's talking about.

Trust me, you're fine.

Thanks for the confirmation for those temps :up:
Still find it strange that CPU is higher then both Cores wich I understood should not be. Searched Google for an answer and didn't got any, but 1 thing it did notice me is that it occured on (most) Asus P5B boards.

Hehe when enabling PECI, I get lower temps for CPU (below the core temps). 16 degrees Celcius looks nice for CPU ;) :P

unclewebb

06-01-2007, 01:37 PM

The assumed TCC is 85° (by coretemp) and it seems accurate to me.
It used to seem pretty accurate to me too until my IR thermometer proved that it can't be 85C.

The new Tuniq just arrived. It transfers heat better than the OEM cooler so the core temp readings at idle, based on TjMax=85C, should decrease and look even more ridiculous.

At 1200 MHz and 1.100 volts at idle, my OEM heatsink goes from 22C to a steady 62C with the fan disabled. Those little cores in there are putting out a huge amount of heat even at those low settings. The DTS is always reporting from the hottest part of the core so I think people are under estimating the true core temperatures.

Even with the computer sitting in Stand By mode the cores put out enough heat to raise the OEM heatsink by 6C above room temperature.

joebuffalo

06-01-2007, 03:57 PM

Let's try to keep this thread on topic folks.

I think anybody new to this thread should read the ENTIRE thing a couple of times before posting.

Let's not turn this into a 'help me figure out my temperatures' thread. Start a new thread if that's why you are here.

joebuffalo

06-01-2007, 04:09 PM

It used to seem pretty accurate to me too until my IR thermometer proved that it can't be 85C.

I told you that low voltage / reduce multiplier test which proved your processor had a tj_max of 85 was not kosher. :p:

I ran across that hardforum thread the other day but have not had a chance to read it yet. I see it is not accessible right now.

I continue to be impressed with all the abuse you put your processor through so that we may all better understand temperatures.

RealTelstar

06-01-2007, 04:13 PM

Hmm... perhaps I should invest in a thermometer :)

unclewebb

06-01-2007, 04:45 PM

I told you that low voltage / reduce multiplier test which proved your processor had a tj_max of 85 was not kosher. :p:
You were 100% right joebuffalo. My previous test seemed like a good idea at the time!

It has come in useful though. That is where the biggest temperature difference is found which proves that the accepted TjMax=85C simply isn't right.

SuperKeijo on the [H]ard forum used my test and was able to lower his core voltage down to about 0.69 volts and get his C2D to run at 600 MHz on his DS3 board. It became very obvious that TjMax=85C was not right for him either. He got things started by using an IR thermometer to confirm his numbers so I picked up a Fluke 62. I calculated at the reduced MHz and voltage that his C2D is likely only consuming 9 watts so he might be the first to pull his heatsink off for some true core temperature measurement. :D

Hopefully by early next week we'll all learn a little more. Either that or I'm going to have the first C2D keychain on my block.

joebuffalo

06-01-2007, 05:00 PM

You are a gentleman and a scholar. And a CPU abuser!

unclewebb

06-01-2007, 05:19 PM

One thing you can also use the CrystalCPUID MSR Editor program for is to read your DTS directly.

The raw DTS data is located in IA32_THERM_STATUS which is located at 0x19C.

http://img234.imageshack.us/img234/9114/readingdtsnq5.png

Bits [22..16] contains the DTS info. To extract that in the above example you chop off the 4 least significant zeros to the far right in EAX. The number 28 is the DTS. 28 hex = 40 C so on this core I am 40 degrees away from TjMax whatever that turns out to be. You can use this program to read this data for any core.

It confirms that CoreTemp is reading this value and displaying it correctly in real time. The TAT reported numbers are not always consistent with the DTS.

Location 0x19C is used in ALL Core processors, Solo, Duo, Core 2 Duo, Quad and I believe it was first used in the Pentium 4 Prescott series but I don't have one of those to confirm. The 85C TjMax number might have actually originated from the P4 era.

Miwo

06-02-2007, 11:50 AM

wow, had no idea my thread was still alive and kicking, I hardly visit this section of the Intel Forums :p

Very nice to see that many of you are testing/investigating and finding out a lot of new information on this matter. Keep up the good work!!! :)

unclewebb

06-02-2007, 01:06 PM

wow, had no idea my thread was still alive and kicking,
The mystery of the absolute temperature of a Core 2 Duo has still not been solved.

I had a look at your results on the first page of this thread and would have to disagree with you. I believe your processor has a TjMax of 100C and you really were running it at over 90C.

These are just numbers and nothing to get too worried about. I've found that as long as your Orthos stable your C2D will be fine. I ran Orthos for half an hour one day and adjusted the CPU fan to keep it at 3 degrees from TjMax. In my case that was likely 97C and my computer was rock solid stable.

Yours is one of the few processors that I've seen that TAT is reading properly as TjMax=100C. Would it be possible for you or anyone that has a processor that TAT shows numbers above 85C to do a RDMSR on location 0xEE and post your results.

Refer to this post to understand what I'm talking about:
http://www.xtremesystems.org/forums/showpost.php?p=2224761&postcount=139

My C2D does not start to throttle until DTS=2. With a TjMax=100C that would be an absolute temperature of 98C. Your test showing your processor running at 94C and not throttling makes perfect sense. It wasn't hot enough yet.

Bail_w

06-02-2007, 05:42 PM

http://img53.imageshack.us/img53/9646/tempwn3.th.jpg (http://img53.imageshack.us/my.php?image=tempwn3.jpg)

Look at this.

Coffee29

06-02-2007, 10:46 PM

Thank you for the MSR infor unclewebb. I knew the address of the MSR but was not sure how to really use it. For those people that have problems getting coretemp 95 to run without crashing your system, this is a solution to getting your DTS.

RealTelstar

06-03-2007, 04:36 AM

http://img53.imageshack.us/img53/9646/tempwn3.th.jpg (http://img53.imageshack.us/my.php?image=tempwn3.jpg)

Look at this.

I tend to believe to coretemp 0.95.

Also, I think newer C2D have 100° tcc (tjunction). I'd like to hear from people with supposedly ~100°C tjunction which week is their c2d.

unclewebb

06-03-2007, 01:18 PM

Also, I think newer C2D have 100° tcc (tjunction).
I don't think TjMax=85C applies to any of the Core family of processors, mobile or desktop. I think it came about because a programmer or two misinterpreted bit 30 at location 0xEE and the misinformation spread from there.

My E6400 is one of the early ones which CPUz shows as revision B2 and is based on the original Conroe core.

E6400 - SL9S9 (http://processorfinder.intel.com/details.aspx?sSpec=SL9S9)
Batch#: L633A988
Pack Date: 12/05/06 ( May 12, 2006 )

kwalker

06-03-2007, 01:44 PM

I don't think TjMax=85C applies to any of the Core family of processors, mobile or desktop. I think it came about because a programmer or two misinterpreted bit 30 at location 0xEE and the misinformation spread from there.

My E6400 is one of the early ones which CPUz shows as revision B2 and is based on the original Conroe core.

E6400 - SL9S9 (http://processorfinder.intel.com/details.aspx?sSpec=SL9S9)
Batch#: L633A988
Pack Date: 12/05/06 ( May 12, 2006 )

You may be right
no one really knows at what point TCC activation occurs for actual core temp calculations read from the DTS.
with the introduction of the Q6600 and a few other newer processors the elimation of the Thermal diode takes away another factor (Tcase)
the temperatures are read directly from the DTS and discrepencys are surfacing in the software readings we thought were reliable.
I will keep following this thread
perhaps some light can be shed on this subject in time.

Prefixxx

06-07-2007, 01:04 AM

RealTelstar

06-07-2007, 04:02 AM

doh!
we've been ignoring the "cpu" temp and looking at the cores so far...

Elisha

06-07-2007, 04:51 AM

Just FYI

Send an email to Intel asking about temps and this what I get back.

Please pay attention to the CPU temperature.
You do not need to pay attention to the core temperature.

:shrug:

I agree there....because when i test Orthos and the CPU temp goes above 70.....Orthos fails nomatter what the Core Temp is....Core temp has gone upwards to 79 and has been stable.
but when i open my case side panel.....cpu temps do not exceed 63 degrees. with it closed it get to 70 and above and i have placed a fan everywhere i can think a sensor was located since i thought speedfan was actually reading the NB or PWN temp as CPU temps. but nomatter where i place the fan....the temps do not decrease.

fgw

06-07-2007, 11:45 AM

Just FYI

Send an email to Intel asking about temps and this what I get back.

Hello xxxxx,

Thank you for contacting Intel(R) Technical Support.

Please pay attention to the CPU temperature.
You do not need to pay attention to the core temperature.

If the CPU temperature is lower than 60.1 everything should be OK.

Sincerely,

Axxxx S.
Intel(R) Technical Support

:shrug:
the question here is: where is this cpu temperature measured?
looking at the figure of 60.1c i assume this intel guy is talking about tcase as documented in the processor specs. besides tcase is different for different processor families, unfortunately no software is showing this temp anywhere. the recommended method of measuring tcase is by putting a temp probe on the ihs. from looking at intel docs, this is a very tricky task and i doubt many people out here will do this ...
the cpu temperature displayed in bios or read by any other software is simply the temperature measured via the thermal diode and calibrated by some unknown factor. this reading would vary from motherboard manufacturer to motherboard manufacturer and even from one bios rev to the other. don't assume to reading tcase here!

unclewebb

06-07-2007, 12:19 PM

I finally got around to doing some more testing today. I'm not quite sure what I've proven besides human stupidity has no bounds but I did end up with lots of interesting data.

I purchased a Fluke IR thermometer recently and a crazy guy at the [H]ard forum jokingly suggested that to get an accurate core temperature reading I should pull off the heatsink and the IHS so I can get a direct shot at the processor. That sounded like an interesting idea :rolleyes: so I decided to be the first on my block with a naked Core 2 Duo. I left the IHS on but after it booted up I pulled off the OEM heatsink and fan while it was running. Two strips of masking tape over top of the processor takes away the shine and helps ensure repeatable temperatures from the IR thermometer.

http://img465.imageshack.us/img465/2126/testbenchsg0.jpg

To keep the processor from getting too hot I used an 80mm high speed hand held fan and had no trouble keeping the processor running in the 60C to 70C range.

http://img217.imageshack.us/img217/2541/66cqc1.jpg

CoreTemp 0.95 which uses a TjMax=85C for my E6400 was in agreement with the IR readings I was getting at 60C, 70C, 80C and beyond.

To explore when the throttle point kicks in I decided to put the hand held fan down and take pictures as the processor warmed up. Here was the last picture I took before things started to get serious.

http://img233.imageshack.us/img233/7682/84uh0.jpg

When the IR reported my cores getting near 85C, CoreTemp started spitting out these errors every couple of seconds.

http://img233.imageshack.us/img233/6979/warningfk7.png

I thought for sure a crash was imminent but it kept running as Intel Thermal Monitoring kicked in trying to regain control of this processor by throttling it.

http://img78.imageshack.us/img78/9858/badnewsqg6.jpg

We need a time out to explain the temperature numbers being displayed by these programs.

The DTS is a 7 bit number. What seems to be happening is DTS gets to zero and then when the processor gets 1 degree hotter the value in DTS goes from 000 0000 to 111 1111 binary. That is equivalent to 127. CoreTemp shows with a question mark that there is something wrong here and then reports the value in DTS directly.

CoreTemp -> DTS -> Real_Temp
84 -> 1 -> 84
85 -> 0 -> 85
127 -> 127 -> 86
126 -> 126 -> 87
125 -> 125 -> 88

SpeedFan and TAT work differently. They continue to use the standard formula of Real_Temp = TjMax - DTS so in this case they display:
Temp = 85 - 127 = -42

They report negative numbers starting at -42 and then as the processor continues to heat up the DTS continues to decrease just like usual.

The CoreTemp log file looks interesting.

http://img233.imageshack.us/img233/7986/logfilezy5.png

Core0 maxes out at a DTS value of 116 which, if my calculations above are correct, is actually 97C. Core1 seems to be displayed at a constant 85C reading when temperatures are out of control.

In the following graph, SpeedFan shows after it gets to 85C there is a sharp drop to -42 and then it continues to increase from there. Before my processor caught on fire I decided to put the hand held fan back on it and was able to regain control of this overheating processor.

http://img232.imageshack.us/img232/1388/finalgraphdo0.png

After taking this processor to hell and back I learned something that I already knew.
The Core 2 Duo is a hell of a processor. :D

Elisha

06-07-2007, 12:44 PM

very nice find.....so what is the CPU temp measurement in Speedfan?
i find that Orthos fails when that hits 70 nomatter what the Core0 and Core1 temps are and nomatter what the cpu voltage is at 3.6Ghz.

unclewebb

06-07-2007, 12:57 PM

I've gone back and forth but after seeing the results I would have to say that my TjMax really is 85C. I had no trouble stabilizing my processor at various temperatures from 60C to 80C and each time the IR thermometer reported a temperature within 1C of what CoreTemp with an 85C TjMax was displaying.

In my case, if TjMax was 100C like I had thought then that would have to mean that the individual cores are running 15C hotter than the IHS which is touching them. Without a heatsink attached, I can't believe that there is 15C of heat dissipation happening in only the thickness of the IHS.

The masking tape used on the processor for testing has previously proven to provide no difference in the temperature of the object underneath it. With my test computer shut down and unplugged for the last couple of hours, the IR thermometer reports the CPU to be exactly equal to the ambient temperature which validates my testing procedure.

If the above is all true and DTS = 85C for my processor then I would also have to conclude that the DTS sensor is not accurate at lower temperatures. Intel designed the DTS and calibrated it to accurately control the Thermal Monitoring functions when a processor gets near the TjMax point. It shouldn't be too surprising if it's not very accurate at reporting very low idle temperatures.

That also explains why Intel refuses to document an absolute core temperature for a Core 2 Duo. Any calculation based on the DTS wouldn't be accurate for the entire operating temperature range of the processor.

Most importantly, the XS guys seem to be 100% right, again! :D Trust the DTS and unless you want to invest in an IR gun, don't put too much faith in any absolute temperatures for the Core 2 Duo.

I'll try to post some temps of the copper plate on my new Tuniq when I have that up and running but I think I've learned all that can be learned when it comes to C2D temperatures.

unclewebb

06-07-2007, 01:17 PM

Elisha: On my E6400 I have run Orthos stable for half an hour at 3200 MHz with my processor within a degree of starting to throttle. At 3400 MHz the absolute core temperature can't go beyond 75C while running Orthos before it crashes within seconds but it can run Orthos stable for hours at 3400 MHz as long as the core temperature stays down in the low 70C range.

If you are at or near your maximum MHz overclock then you will need to run your processor at a lower core temperature to maintain stability. The CPU temperature reported by SpeedFan is based on an inaccurate diode reading. Intel designed the DTS because the diode based temperature readings weren't accurate enough so ignore the SpeedFan reported CPU temperature.

joebuffalo

06-07-2007, 01:33 PM

How did you come to the conclusion that DTS was not as accurate at lower temperatures?

unclewebb

06-07-2007, 01:43 PM

How did you come to the conclusion that DTS was not as accurate at lower temperatures?
Here is the [H]ard forum where I've been discussing this issue:
http://www.hardforum.com/showpost.php?p=1031127676&postcount=165

I did a previous test with just the OEM heatsink attached and then I used my 80 mm hand held fan to get my core temp down. This allowed me to get in there and measure the temperature of the copper core inside the heatsink without a fan being in the way.

Based on a TjMax=85C, CoreTemp was reporting a very steady 21C while the IR thermometer was reporting that the center copper cone of the heatsink was at 27C. That's impossible. The heatsink can't be warmer than the processor it's cooling.

I thought that must mean that TjMax for my processor is 100C but that doesn't seem possible after my test above.

From the data gathered my only conclusion is that the DTS can't possibly be 100% accurate from idle temps to DTS=0 temps. I'm really open to any other possible explanation for what I've found.

unclewebb

06-08-2007, 11:30 AM

Just for fun I took my final graph yesterday and decided to use Photoshop to cut out the part of the graph when SpeedFan started reporting negative numbers. I moved that piece of the graph straight up and it fit into the upper part of the graph perfectly.

http://img53.imageshack.us/img53/3517/finalgraph2ki4.png

This seems to show that both cores continued to run even when the core temperature was up over 90C.

RealTelstar

06-08-2007, 04:44 PM

I think i want to invest in a IR thermometer for fun :)
There are other components besides the cpu that I want to check (ram and my hot scsi controller on top). what do u suggest under $50 and available in EU?

Coffee29

06-08-2007, 10:34 PM

Unclewebb...just a thought.

If I understand your logic you seem to have argued the following: your previous testing suggested that a tjmax of 85 could not be right and you thus assumed it to be 100. However, your more recent testing is in favor of the 85 over the 100, thus you have now concluded that your tjmax may in fact be 85. In order to remedy the inconsistency of the former testing you conclude that DTS may not be accurate at lower temps.

Correct me if I am wrong about this summation, however, if I have stated the basic argument correctly, isn't a more plausable explanation simply that the tjmax you are looking for does not exist? Possibly DTS is accurate all of the time and the reason that you are getting conflicting results is not due to invalid DTS readings at lower temps, but instead because you are maintaining that your processor uses a tjmax variable of 85 or 100?

unclewebb

06-09-2007, 12:29 AM

Every C2D has to have some fixed TjMax value associated with it where the processor begins throttling. It might be 85C or 100C or some number in between but it has to be some number. The problem I'm having is coming up with one number that explains what I'm seeing at both idle and at high temperatures.

Assuming a TjMax of 85C for my E6400 results in absolute temperatures reported by CoreTemp that are within 1 degree of what my IR thermometer is reading pointed directly at and hovering just above the cores.

Yesterday I had no idea what I was in for and the core temp went sky high as I was busy taking pictures and saving screen shots. I put the camera aside today and just concentrated on comparing the readings from CoreTemp to the IR thermometer. By adjusting the distance and angle of the fan to the cpu, I was able to run at a variety of temperatures between 48C and 70C without a lot of variation. I had plenty of time to slowly position the IR thermometer directly over the cores and obtain readings while the core temperatures remained steady.

The IR readings were generally the same as CoreTemp or within 1C at any point from 48C to 70C.

I installed a Tuniq Tower this evening and so far I've just been running it at low MHz and low volts. The ambient temperature is 23C and CoreTemp using an 85C TjMax is reporting 19C. That's not possible. Worse than this is that the copper plate on the bottom of the Tuniq which presses against the IHS is at 29C. For this to be possible, CoreTemp needs to be reporting at least 10C higher so that implies TjMax needs to be 95C for the idle temps to make sense.

TjMax = 95C is impossible at higher temperatures over 50C. How could two cores that are both at 60C only be heating the IHS up to 50C? The IHS is designed so that it is physically contacting both cores and the IR thermometer is located within 1 mm or 2 mm of the IHS.

That's the dilemma. There's no single TjMax number that covers both situations at idle and at high temps. The only thing I can conclude from this is that the DTS is calibrated and very accurate at the trip point but isn't as accurate at very low idle temps.

SuperKeijo on the [H]ard forums has also removed his heatsink and is experiencing this same dilemma during testing of his E4300 with an IR thermometer. Hopefully he'll have some more results this weekend.

RealTelstar: Extech makes IR thermometers and are available new on EBay for about $50 to $60. I think an IR-205 (http://www.extech.com/instrument/products/alpha/IR205.html) would be sufficient for this kind of testing.

unclewebb

06-09-2007, 10:31 AM

....so what is the CPU temp measurement in Speedfan?

Here is some proof that the CPU Temperature number based on the on board diode is a completely meaningless number.

http://img371.imageshack.us/img371/9625/speedfanfz8.png

With an ambient temperature of 22C the bios displays 13C when it reads this diode and SpeedFan reports 12C which are beyond meaningless. Turn it off and ignore it because inaccurate sensors can't provide you with any reliable information that can be used to determine anything.

joebuffalo

06-09-2007, 11:22 AM

Before I jump in on this discussion again, I want to state the point of this thread one more time for any new readers who have not taken the time to read all the posts:

Absolute temperature are worthless. All you need to know is DTS. Stop worrying about anything else!

unlewebb, your assumption that you are reading the core temperature as measured by DTS when you use your IR thermometer on the heatsink or even the IHS is just wrong. There is DEFINITELY a thermal gradient that you are simply ignoring. The reason Intel implemented multiple digital temperature sensors is because they knew that there are multiple hot-spots which could not be captured by a singe sensor.

From http://documents.irevues.inist.fr/bitstream/2042/6579/1/TMI23.pdf
(emphasis is mine):

"It can be seen that large temperature gradients exist on the
die. It also can be noted that some workloads display high
temperature gradients while other have no offset. Thermal
control algorithms need to prevent the hot spot from
exceeding the max temperature specification. It is possible
to mitigate the temperature difference by applying a fixed
offset to the diode reading. This obviously is a non
optimal solution as the workloads with low offset will be
panelized by the unnecessary temperature offset. The use
of digital thermometer provides improved temperature
reading, enables higher CPU performance within thermal
limitations and improves reliability."

If there are larger temperature gradients ON THE DIE surely they exist across the TIM, IHS, more TIM, heatsink, etc.

TheMafioso

06-09-2007, 03:57 PM

SuperKeijo on the [H]ard forums has also removed his heatsink and is experiencing this same dilemma during testing of his E4300 with an IR thermometer. Hopefully he'll have some more results this weekend.

What does his tests about the e4300 conclude, has he been able to figure out if it has TjMax of 85C like your e6400 or is TjMax 100C as taken by tat/coretemp 0.95

(I looked for it in the [H] thread, but couldn't find/missed his results)

kwalker

06-09-2007, 04:18 PM

unclewebb

06-09-2007, 04:53 PM

Absolute temperature are worthless. All you need to know is DTS. Stop worrying about anything else!

With TjMax and absolute core temperatures still undocumented by Intel, that statement remains true.

unlewebb, your assumption that you are reading the core temperature as measured by DTS when you use your IR thermometer on the heatsink or even the IHS is just wrong. There is DEFINITELY a thermal gradient that you are simply ignoring.

Here I disagree. I definitely haven't ignored the potential gradients and have even done some testing of this. Gradients exist when an application works one part of the core while other parts of the same core are mostly idle. For example this will happen if a program is working on a large calculation that involves mostly floating point operations. The floating point area of the core is going to be running at a hotter temperature than other surrounding parts of that same core. Here's how I tried to determine how big this gradient might be.

The process that is known to create the most amount of heat in a Core 2 Duo is MeromMaxPowerVer0p3.exe which is the name of the process that TAT runs when you tell it to run a load. I decided to run this process on one core while the second core was idle and then I switched the Affinity of this process while it was running to get it to immediately switch over and run on the second core so the first core now became idle. The difference in temperature between the two cores averaged 6C with the OEM heatsink and fan. With a better cooling solution the difference would likely be slightly less.

http://img112.imageshack.us/img112/6098/tempdifferenceog6.png

To me that represents a worst case scenario. Having one core running the most demanding program at full load heats up the other totally separate core to within 6C even though the second core is idle doing nothing. That surprised me because I was expecting a larger gradient to exist. This shows that the heat of each individual core is being shared with the other core and more importantly the graph also shows how quickly a core will decrease in temperature to its surroundings when it becomes idle. CoreTemp logging showed that this took less than a second for each core to stabilize at its new temperature.

With virtually zero load on both cores, with the heatsink and TIM removed and being cooled with a hand held high speed fan, the gradient between the two cores drops to zero. Both cores report that they are operating at the exact same temperature. Gradients are caused when part of a core is under load while other parts of the same core are mostly idle but if the entire core is idle than the gradients do not exist or are so small that they can be ignored.

The hot spots shown in the Intel diagrams are for a processor that is working but those hot spots don't apply to a processor that is idle and allowed to sit and stabilize at a fixed temperature.

A consumer hand held IR thermometer is not accurate enough to acquire rapidly changing temperature data from a processor that is under load but at idle, it's a different story.

The temperature of the cores at idle becomes the same and they also heat up the center of the IHS to a similar temperature. The round opening in the IR thermometer is approximately 1 square inch so it is taking data from both cores and the space between them and averaging it out to display one temperature number. My opinion is that at idle, the temperature gradient between the two individual cores and the IHS where I'm taking readings from becomes insignificant and certainly less significant than the 1% sampling error of the IR thermometer.

unclewebb

06-09-2007, 05:07 PM

Sorry to disappoint. Not only were there no flames but this heat baked E6400 with the new Tuniq is running better than ever. More proof that you don't have to worry about overheating one of these things even if the heatsink accidentally falls off! :D

http://img207.imageshack.us/img207/720/3640mhzps6.png

Anyone who goes to Google to learn about C2D temperatures will sooner or later end up here so I think it's good to have some more facts which they can interpret as they wish to come to their own conclusions.

So far the E4300 being tested with a different IR thermometer is showing temperatures similar to what CoreTemp 0.94 is reporting which uses an assumed TjMax=85C.

joebuffalo

06-09-2007, 05:14 PM

Get a copper rod. Place one end in ice water. Heat the other end with a blow torch. There will be a thermal gradient along the bar. Yes? One end will be very hot, one end will be cold. In the middle of the bar, the temperature will depend on the location along the length of the bar.

We have the exact same thing going on here.

Let's neglect all the thermal conductivity killing interfaces that are present between the core and the heatsink for the moment. Let's assume we have a perfect heat conduction path from the core to the air.

You have a heat source (the core)
You have a heat 'sucker upper' (the ambient air)

There WILL be a thermal gradient along the path from the core to the ambient air. The underside of the IHS will be hotter than the top side of the IHS. This is a fact of life.

I *think* you may be getting caught up by assuming a steady state (where everything is the same temperature) is reached. But it is not. Heat is always flowing from the source to the sink. As a result, there will always be a thermal gradient. As such, you are not measuring the core temperature when you measure the temperature of the IHS.

unclewebb

06-09-2007, 05:35 PM

There WILL be a thermal gradient along the path from the core to the ambient air. The underside of the IHS will be hotter than the top side of the IHS. This is a fact of life.
I was faced with the same argument when I was taking temperatures of the copper cone at the center of the OEM heatsink. I agree that there is a gradient but the temperature being reported by the IR thermometer has not really changed.

I removed the heatsink so I could get a reading off of the IHS directly which takes the copper heatsink and the TIM out of the equation but the temperature results didn't significantly change. There was less than one degree difference when measured with the IR thermometer compared to when I measured the core temperature by measuring the other side of the copper heatsink.

unclewebb

06-09-2007, 06:47 PM

joebuffalo: The C2D is individually calibrated on the assembly line. What likely happens is the processor is put into a socket and fed a minimal amount of voltage and run at a slow MHz similar to what I'm doing. SuperKeijo had no trouble running Orthos with his heatsink off because his motherboard allows him to run at a lower voltage than I can. The temperature can then be checked on the assembly line, likely with an IR thermometer or something similar and compared to real time readings from the DTS. Any necessary offset to correct the DTS to the measured temperature of the processor could be programmed into an on chip PROM type of memory which can't be changed later by the user.

My method of trying to read the core temperature with an IR thermometer might not provide me with the true core temperature due to the temperature gradients that you've mentioned but this method might be quite similar to how Intel calibrates their processors during manufacture. These are just my wild guesses based on nothing.

It really doesn't matter though. DTS is still king that no one can argue against.

What I do know now is that for my processor, when TjMax is assumed to be 85C, CoreTemp 0.95 reports absolute temperatures that are exactly equal to the temperature of the outside of my IHS as measured with an IR thermometer from about 45C to 85C. That may not be the ultimate true, blue core temperature but the CoreTemp reported absolute temperature number has plenty of meaning to me.

Testing has been fun. Thanks for all of your input. :)

abe786

06-16-2007, 02:22 PM

joebuffalo

06-16-2007, 06:56 PM

Excellent thread guys..very informative :up:
I have gone through all the testing which you did unclewebb..I've concluded that speedfan 4.32 and coretemp 0.94 both report the correct temperature for c2d's since they assume TjMax of 85 degrees which you're testing have resulted....so is that correct and I can breath easy by just noting there values...i own an c2d E4300 proccessor myself btw..

NO!!!!

You missed the entire point of the thread. Just use CoreTemp 0.95 and check out the DTS readings directly.

Even if you agree with all of unclewebb's testing, HE DOESN'T HAVE AN e4300!

Coffee29

06-16-2007, 11:55 PM

i have an e4300 and you CANNOT go by the temp readings. Go by the DTS reading on core temp 95. The temp programs assume variables that are not documented and may not even exist, DTS is documented and is accurate.

abe786

06-17-2007, 03:22 AM

NO!!!!

You missed the entire point of the thread. Just use CoreTemp 0.95 and check out the DTS readings directly.

Even if you agree with all of unclewebb's testing, HE DOESN'T HAVE AN e4300!

Yes, i've gone through the most of the thread and I do agree that DTS values is the way to go, but i wanted to find the safe temperature operating range for my CPU, since i was wanting to find out max overclock i can get with tempratures staying under control to prevent any long term damage...how can I do that with just DTS values, like for c2d its written that it shouldn't temprature shouldn't exceed 61 degrees for safety of the CPU...so to calculate it, I believe I need the TjMax value...

Unclewebb does mention that e4300 does have TjMax of 85 in post#176 (http://www.xtremesystems.org/forums/showpost.php?p=2242798&postcount=176) (maybe he's borrowed that result from the hardforum guy whose also testing the CPU with IR thermometer)

Coffee29

06-17-2007, 04:56 AM

the documented c2d temp of 61 degrees is an absolute temperature. This information is 110% useless to you in determining your safe zone for OCing because it is 100% impossible to ever determine an absolute temperature. Moreover, discussing tjmax makes little to no sense. The DTS number is a value that tells you very accurately how far your chip is from TCC activation. Temps guessed off of this are at best relative.

A good way to stay safe is by keeping your DTS over 20-30 in my opinion. Also use your fingers, check your ram, northbridge and HSF. Also make sure your system is stable, crashing is a good sign it is too hot.

What OC do you want and what cooler are you using? I didn't take my e4300 over 2.4 for a 24/7 OC on the stock hsf. With my new scythe mine I'll probably do a 3.0 24/7 OC.

Above all else, instead of attempting to use relative temps arrived at by guessing variables, skip the guesswork and learn to use what they are relative to--->the DTS

abe786

06-17-2007, 05:49 AM

^^I'm running with stock cooler for now and OC is same as urs ie 2.4g, i get 22 as max DTS reading w/ TAT full load running on both cores...I will be getting a better cooler in coming weeks, with that I plan to run around 3.0ghz 24/7, maybe even more, if things are favourable...its a week 41 chip

By the way, the 20-30 DTS reading would corrospond to 65-55 coretemp 0.94 reading, so it may/may not be correct reading, but target's the same :D

unclewebb

06-18-2007, 10:32 AM

TjMax exists for all processors but without this number being documented by Intel, all software is left guessing when it comes to reporting an absolute core temperature. If you don't want to invest in an IR thermometer and risk damaging your processor by running it naked and if you really want to see an absolute temperature number then I'd use software that assumes TjMax = 85C like SpeedFan 4.32 and CoreTemp 0.94 do. This is only for the Core 2 Duo and not the Quads and don't use this number to compare your results with anyone because they might be meaningless.

If you want a number that is 100% accurate, documented and guaranteed to have some meaning when it comes to C2D temperatures then stick with CoreTemp 0.95 which can be set to report the DTS directly, in real time, without any further bastardization of the number.

http://img234.imageshack.us/img234/3154/coretempdtszt0.png

The Core 2 Duo has a huge amount of head room and there are only two things you need to be concerned about.

#1) Throttling. When a C2D gets too hot it will throttle and slow down which no one wants to happen. My E6400 doesn't start to throttle until DTS=2 which is far hotter than what 99.999% of people run their cpu at so it's not really an issue, especially if you're overclocking.

#2) Stability. If you can run Orthos for 4 hours and if your processor is not throttling then you've got a pretty good system that will be able to run most real world applications without any problems. If you are Orthos stable and not game stable then you should concentrate on your GPU and making sure it is not over heating.

You're CPU is going to need more temperature headroom the more you overclock to maintain stability. Here's what my E6400 needs.

3200 MHz ~ 1.30 volts ~ DTS=5
3400 MHz ~ 1.40 volts ~ DTS=15
3600 MHz ~ 1.50 volts ~ DTS=25

Those voltages are approximate Orthos load voltages and the DTS values are approximately how much headroom I need to maintain stability. Everyone knows that as the MHz goes up you have to use more core voltage to maintain stability but I was a little surprised to find out that there is also a direct relationship when it comes to temperature. You need more temperature headroom the faster you go and reading the DTS directly is the most accurate way to monitor that.

tal0n

06-18-2007, 11:55 AM

Oh! unclewebb thanks for sharing...

I really suspect this, thats the reason I can not hit 3600 Mhz in my quadcore because the diference at 1,55 Vcore is no more than 15 º to Tjunction.

My Q6600 temps at load with 4 primes is 75º 75º 71º 70º at 8 x 400 = 3200 Mhz. That´s with thermalright Ultra 120 eXtreme.

That temps are at 1.375 VCORE set in bios with case closed, only one fan in Ultra 120 at 1400 rpm. I will try 3400 Mhz I think I can hit with only 1.425 VCORE in bios but my temps are near 80 º.

My temps are correct? Or is someone obtaining less.

Thanks

joebuffalo

06-18-2007, 08:27 PM

tal0n, how are you obtaining those temperature measurements?

If you just reported DTS we wouldn't have to ask that question.

Grinch

06-18-2007, 08:32 PM

so is this ok?

http://s2.supload.com/thumbs/default/010-20070618223216.jpg (http://s2.supload.com/free/010-20070618223216.jpg/view/)

joebuffalo

06-18-2007, 09:43 PM

A DTS of 44 is completely safe.

tal0n

06-19-2007, 01:51 AM

tal0n, how are you obtaining those temperature measurements?

If you just reported DTS we wouldn't have to ask that question.

That's my temperature obtained with core temp but not DTS. DTS must be setting in core temp to show how much temp is far from TJUNCTION.

If I have a 75º temp in my hottest core I have a DTS of 25º because my TJUNCTION in core quad is 100º.

cya

vera

06-19-2007, 05:08 AM

Hi, I bought a E6400 "L628B103"... This cpu is nice O/c'er , I use it 3500mhz def. vcore , and my cooler is TT bigtyphoon , but When I run the orthos small fft , it's up to 80 C !, before I have a E6420 and I used it 3.4 ghz and it's never to pass 60C ? I can't understand ?? please help !

joebuffalo

06-19-2007, 08:39 AM

vera, READ THE THREAD. Your absolute temperature are meaningless.

tal0n, if you are running at stock voltage, DTS=25 is a good number. I personally am at 30 but would not have a problem running at 20 one of these days.

I think unclewebb has shown us that you are OK as long as your processor is stable and you have just enough DTS that you don't start throttling.

BUT, just because the systems is 'OK' doesn't mean that you aren't putting extra wear and tear on the CPU. So leave some a margin with a bit of DTS if you want your processor to not have any problem down the road.

EVERYBODY, PLEASE STOP WITH THE "My processor is at 80-C. Is that OK?" POSTS? THEY ARE POINTLESS IF YOU WOULD JUST READ THE THREAD. Please don't turn this informative thread into junk.

tvdang7

07-02-2007, 09:25 PM

so whats the verdict? im in a dilema on which temps to trust cuz core temp is like 15 degrees hotter than everything else. maybe i might want a new heatsink.

Coffee29

07-06-2007, 10:31 PM

tvdang7 the answer is right above your question posted by Joebuffalo on 6/19. Please read the entire thread.

Timbosan

07-08-2007, 11:56 PM

Interesting thread. I cant believe intel has stated an absolute max temperature of 61.4*C for my cpu (e6300), however they have not provided anyone (developers ets) with the appropriate tools / information with which they could design a temperature monitoring program. So they have said "keep your processor under XX*C for reliable operation, but we have no way of measuring the absolute temperature. Once processors start being affected by degradation caused by overheating, what will they do? Ask the end user if they ran it out of spec! How do we know, we dont know its temperature! Intel appear to have designed a loop hole, sidestepping their responsibility to the customer. If they stated a DTS value and said keep over that for reliable operation it would be different. However even using the DTS value is dubious - it is simply a fixed number (85 or 100) at which the program tells you the cpu shuts off at - the REAL temp may be above or below that number. It is calibrated so when any given processor is at its individual and unique max temp, DTS = 0.

unclewebb

07-09-2007, 10:04 AM

Intel: The TDP specification should be used to design the processor thermal solution. The TDP is not the maximum theoretical power the processor can generate.

Timbosan: That 61.4C number is the maximum recommended case temperature (Tcase) at TDP which is measured on top of the IHS at the geometric center. It's not the maximum safe temperature the processor can operate at. That's controlled solely by data from the on chip digital thermal sensors (DTS).

Intel: As an output, PROCHOT# (Processor Hot) will go active when the processor temperature monitoring sensor detects that one or both cores has reached its maximum safe operating temperature.

PROCHOT# is asserted whenl DTS=0 which Intel documents to be "the maximum safe operating temperature."

Processor shutdown or THERMTRIP# does not occur until approximately 20C after TjMax for the Desktop core processors or at a documented absolute temperature of approximately 125C for the mobile Core 2 Duo processors.

Intel: Regardless of enabling the automatic or on-demand modes, in the event of a catastrophic cooling failure, the processor will automatically shut down when the silicon has reached a temperature of approximately 125 °C. At this point the THERMTRIP# signal will go active.

The Core 2 Duo has a huge amount of temperature head room and as long as the DTS reports a positive number you are operating it within the Intel specifications and you aren't voiding your warranty by doing so.

The safe operating temperature, processor throttling and shutdown are all controlled by the DTS. It's the only temperature data you need to know.

Timbosan

07-11-2007, 05:59 AM

Ok i asked intel what the maximum operating temperature of the e6300 is, as i was worried about keeping it within its specifications. This was the response:

"I understand that you are concerned about the operating temperature of the Intel(R) Core(TM) 2 Duo processor E6300.

The maximum operating temperature of the Intel(R) Core(TM) 2 Duo processor E6300 is 61.4 degrees Celsius. As long as the processor is operating under this temperature it is operating within specifications."

That implies the cpu should not exceed 61.4*C or it is "out of spec".

I then asked what software i should use to monitor the temperature with, as i would need to be able to ensure the cpu was operating within spec. This is the response:

"Regarding your first inquiry about monitoring software, thermal management and sensors are built into the motherboard; therefore, the motherboard manufacturer should be the ones providing a monitoring solution that can read the information from the processor.

Intel would only count with monitoring software that is specific to Intel(R) desktop boards and not to the Intel(R) processors; I would recommend contacting your board manufacturer for the application you need to run to be able to verify on the processor temperature and voltages.

As well, the PECI option is a feature that is particularly handled by the system chipset and BIOS meaning that Asus* (i told them i have an asus mobo) would be the ones recommending whether this option should be enabled or disabled in the BIOS"

This is a problem, as how could asus have been able to design accurate temperature monitoring software? AFAIK there is no accurate extrapolation possible of any values the thermal monitors output into absolute temperature, without a TCC (TjMax) being documented for the core 2 duo.

unclewebb

07-11-2007, 07:04 PM

Timbosan: Sounds like Intel hired a summer student and put him in charge of the C2D Temperature FUD program.

The green guy you talked to has never read through the Intel data sheets. To properly measure the case temperature of a processor you have to cut a groove in the heat spreader (IHS) and run a thin thermocouple wire to the geometric center of the processor. Sounds like a lot of work for the average person to see how their C2D is running. I think I'll stick with the much simpler and fully documented DTS readings.

Nistomax

07-12-2007, 05:10 PM

I'm getting multiple readings on my phase cooled e6600 as well.
Motherboard sensor said -36C so i figured okay to boot.
CoreTemp says around 4C for cores 1 and 2, as does speedfan. It is interesting, however, that speedfan reports -27C for CPU temp.
Does this mean the cpu is running at 4 or -27 or both depending on where it is measured?

joebuffalo

07-13-2007, 12:45 AM

He wants to know if his CPU is running at -27 or 4 degrees C...

muahahahahahahahaha

Read the thread, nistomasx, use CoreTemp, and learn to love DTS.

cowlickkid

07-13-2007, 06:05 PM

Core Temp .95 gives me a Tjunction of 100C for my e2160. Does this sound correct?

unclewebb

07-13-2007, 06:28 PM

Nistomax: Here's a good example why I don't put too much faith in the CPU temperature reported by most motherboards.

http://img57.imageshack.us/img57/1025/biostempho0.jpg

13C is a great temperature but in a 22C environment with an air cooled processor it's impossible. SpeedFan reads this sensor and reports it as CPU but it's obviously meaningless.

For my E6400, when software guesses that TjMax=85C, I think the DTS is providing a pretty good approximation of what my absolute core temperature is from say 45C to 85C but I'm not convinced that the DTS is accurate at lower temperatures. It was designed and calibrated to be accurate at the throttling point but I have no data to prove how accurate it is at low temperatures like you're running.

Even if it's not accurate for reporting absolute temperatures, it is always accurate at reporting relative temperatures. If CoreTemp 0.95 reports that your Delta to Tjunction is 80C one day and then you do a few mods and it reports that your Delta to Tjunction is 85C the next day, your processor is definitely running 5C cooler than what it was running at before.

The DTS was designed to report how far you are away from TjMax and for that purpose, it's fully documented, calibrated and it's the best temperature information available for users.

Khapz

07-13-2007, 06:40 PM

e4300 revision l2 have a 85 or 100 tjunction?

unclewebb

07-13-2007, 06:41 PM

Core Temp .95 gives me a Tjunction of 100C for my e2160. Does this sound correct?

Probably not. The new E2160 is identical to an E4300 but with half the cache disabled ( 1MB vs 2MB ). CoreTemp 0.95 assumes all revision L2 desktop processors like yours have a TjMax = 100C but they are all very likely TjMax = 85C. SpeedFan 4.32 assumes TjMax = 85C so use it if you want to see an absolute core temperature.

That's just my guess and whether I'm right or wrong isn't important. Read this thread and set CoreTemp 0.95 to "Show Delta to Tjunction temp." That reads the sensor directly, tells you how much head room you have and then there's nothing to argue about.

cowlickkid

07-13-2007, 07:17 PM

I have read the thread and understand it, especially your well done explanation on page 8 Uncle. The reason why I was curious about the reported 100C tjunction was if there was some physical characteristic of the e2160 that would give 15C more headroom compared to other Core2 processors?

unclewebb

07-13-2007, 08:13 PM

In theory a maximum Tjunction of 100C would give you 15C more headroom than a processor with TjMax of 85C.

There's no documentation from Intel to confirm that the TjMax for desktop processors is 100C or 85C or anything in between. When CoreTemp was updated from version 0.94 to 0.95 it decided to start guessing that all of the revision L2 Core 2 processors are TjMax=100C and the revision B2 are TjMax=85C. There's a pretty good chance that CoreTemp is wrong.

Other users here will argue that it might even be a completely different value. No one knows for sure so the only thing we can go by is "Delta to Tjunction" which is correct for all Core based processors.

impar

07-14-2007, 01:43 AM

Greetings!
e4300 revision l2 have a 85 or 100 tjunction?
In CoreTemp 0.95, 100ºC.
In CoreTemp 0.94, 85ºC.

When the E4300 arrived, CoreTemp 0.94 showed a too low temperature so the author bumped tjunction to 100ºC.
Funny thing is that TAT automatically adjusted to the new L2 core. And is probably the most reliable temperature reading program for C2Ds.

If THG did this right, the L2 stepping, with all throttling options enabled in BIOS, also throttles at 85ºC:
http://img292.imageshack.us/img292/3197/temperaturverlaufcore2dam8.th.png (http://img292.imageshack.us/my.php?image=temperaturverlaufcore2dam8.png)
After 3 minutes 15 seconds, the B2 stepping began to reduce its speed due to overheating (87°C), protecting itself by throttling the clock speed.
By comparison, the L2 stepping held out a full 5 minutes before its internal heat sensor triggered the throttling feature. (http://www.tomshardware.com/2007/07/11/energy-efficiency-intel-left-out-in-the-cold/page8.html)

unclewebb

07-14-2007, 02:22 PM

impar: THG is usually pretty good but they didn't mention what software they're using to measure temperatures. Without knowing that and what assumptions that software made about TjMax, their results are meaningless.

You are right about what caused CoreTemp to change its assumption about TjMax from 85C to 100C. The graphs show the L2 having a lower idle temperature and the slope of the graph confirms that the L2 uses less watts and therefore produces less heat than the B2. The L2 E4300 also had a lower idle MHz when SpeedStep and C1E enabled, ( 1200 MHz vs 1600 MHz ) and because of their reduced idle wattage and heat output, absolute core temperatures with a good cooler were being displayed that were below the ambient temperature which is impossible.

The programmer of CoreTemp's solution was to add 15C to the maximum Tjunction which corrected this problem at low temperatures. I don't agree with that decision though. I think the real problem is a DTS that is not accurate at low temperatures and modifying the TjMax may correct the low end but results in absolute temperatures being misreported too high at full load.

With the voltage and MHz dropped down, my B2 E6400 can also report absolute temperatures below ambient and a good 10C below the temperature of the copper plate on the bottom of my heatsink, both of which are impossible. Using a 100C TjMax for my processor to try and cure this problem results in completely unbelievable reported core temperatures at or near TjMax.

Without any documentation from Intel saying what TjMax really is for the Desktop processors, all software calculating an absolute temperature based on an undocumented TjMax is only guessing.

TAT may have the Intel logo on it but it does not track the on chip DTS exactly and depending on the temperature or the day of the week, it does vary. It was never designed for the desktop processors and has never been updated since the Core 2 Duo desktop processors were introduced.

unclewebb

07-14-2007, 02:54 PM

I originally posted this at the [H]ard forum but I think it's relevant and this procedure can be used to test how accurately programs that people take for granted actually track the DTS.

I use CrystalCPUID (http://crystalmark.info/software/CrystalCPUID/index-e.html) to read the on chip model specific register ( MSR ) that contains the raw DTS information. The 7 bits of temperature data are located in bits [22..16] of MSR 0x19C in all Core based processors.

Here's my E6400 ( 3600 MHz 1.504 volts ) at idle in a 22C room with the Tuniq Tower fan at its lowest speed of 1150 rpm.

http://img237.imageshack.us/img237/2691/rmclocktest2mz1.png

When you enter 0x19C in the MSR Number box and click on RDMSR ( Read MSR ) you will get the value of the DTS. In this example bits [22..16] corresponds to 36 hexadecimal which is ( 3 X 16 + 6 ) = 54 in decimal. This means at idle I am 54C degrees away from TjMax which is the Intel documented highest safe operating temperature for a C2D cpu.

If RMClock is using the DTS and calculates the absolute core temperature to be 33C then it must be assuming that my TjMax = 33C + 54C = 87C. That's certainly possible but now let's check it again at full Orthos load.

http://img156.imageshack.us/img156/3694/rmclocktest1yp7.png

At 3600 MHz and 1.480 volts it is happily running Orthos within 18C of TjMax. RMClock reports a core temperature of 63C so now it must be assuming that TjMax = 63C + 18C = 81C.

The maximum junction temperature of your processor ( TjMax ) is a fixed value, set at the factory and doesn't change yet RMClock is using two different values depending on whether the processor is hot or cold. That's not right. I also performed this test with TAT and the calculated TjMax could change from day to day which is also impossible.

Now you can run CoreTemp 0.95 or SpeedFan 4.32 and you will see that both of these programs track the on chip digital thermal sensors exactly. CoreTemp may be guessing wrong at TjMax but it can be used to report the DTS directly. When it is set like that, it reports the temperature data in processor register 0x19C exactly.

This is the only Intel documented core temperature information for a Core 2 Duo desktop processor.

Edit: On the main page of CrystalCPUID you can set it to read any of your cores. TAT is using different TjMax values at idle, at full load and even CPU 0 and CPU 1 are not consistent.

joebuffalo

07-15-2007, 12:21 AM

This is the only Intel documented core temperature information for a Core 2 Duo desktop processor.

:up:

...just though it deserved to be repeated...

SLi_dog

08-14-2007, 10:39 PM

heh heh heh, apologies for digging out this old thread but it's just an invaluable resource for Core temperature/DTS monitoring info :D

Before I start I should make it clear I understand that all Core2 desktop CPUs have no Tjunction value and as a result they should probably all be configured to read the DTS values however there is already a massive amount of info out there in regards to the Conroe Core 1066Mhz FSB Core2Duo and Kentsfield Core 1066Mhz FSB Core2Quads that have a "safe" CoreTemp load threshold of 65C and 75C respectively. Changing that perception, imho, would take more work than any benefit it may achieve.

As I'm sure you're all aware, the new 1333FSB processors are another CPU that seems to be having incorrect readout issues with the normal configuration of CoreTemp. I also have a sneaking suspicion that this applies to the E2140/E2160 CPUs as well.

This is all well and good as all who have followed this thread know you can just configure version CoreTemp v0.95 to report the DTS readings.

The problem comes with new users and what appears to be a lack of an easy to find thread which details the processors that are affected by this Tjunction issue and also a detailed method on how to configure CoreTemp to show the DTS.

I admit my understanding of actual CPUs affected by this is limited compared to many who post here, but I am assuming that:

E4300/E4400/E4500
E2140/E2160
E6x50 (1333FSB based CPUs)

Should all have their core temperature monitored by reading the DTS.

Another general assumption I'm working with is that the Core2 CPUs should be run no hotter than within 20C of the DTS 0C mark.
- This assumption is based on the first Conroe CPUs where initial testing seemed to show that staying under 65C (20C lower than the 85C nominated Tjuction) was the best threshold for a "safe" overclock.

I have also put together this image on how to configure CoreTemp v0.95 to show the DTS readouts.

http://www.geilunleashed.com/~sli_dog/Help_Files/CoreTemp0.95_DTS_readout.png

This kind of information would make a handy sticky in the "Intel Info And Support" section for easy reference to new users of Core2Duo/Core2Quad.

I can start a thread if people here think it's worthwhile but I'd prefer to get some feedback on whether the CPUs listed above are the main ones affected by this issue.

There's also need to be a Mod to sticky it otherwise it's pointless.

bing

08-15-2007, 05:37 AM

Doing it the correct and Intel's way DOES REALLY matter !

Ever wonder how are those Intel CPU engineers measure Intel's own CPU in their uber leet lab ?

Warning, cpu pron pics inside, not advisable for minor ! :D

Intel® Core™2 Extreme Quad-Core Processor and Intel® Core™2 Quad Processor Thermal and Mechanical Design Guidelines (http://www.intel.com/design/processor/designex/315594.htm) Download the PDF file.

For those who likes quick shortcut, just go to page 81 and beyond !

After you've read that and if you believe it, next time you will start chuckling & gigling back there when you hear these kinda comments : :rofl: :ROTF:

"TAT cpu temp reading rocks ! I ignore others"
"Coretemp is the best and I believe it !"
"Temp diode reading is the most accurate one"
"The Tjunction should be xxx, not yyy .... "
..bla..bla..

For those non believers even after you've read it, c'mon, if this Intel's document can not be trusted, then what else ? :fact: :stick:

Or that program that shows the temp that you "want" to believe ? :slap:

Conclusion ? Software based for absolute CPU temp reading is NEVER ACCURATE, period !

Edit:Repost.

unclewebb

08-15-2007, 10:48 AM

SLi_dog: Another general assumption I'm working with is that the Core2 CPUs should be run no hotter than within 20C of the DTS 0C mark.

The general assumption that DTS=20 or 65C is a good number to shoot for when overclocking is too vague. 65C is an undocumented absolute temperature and people's definition of overclocking varies. My E6400 runs fine at 80C (DTS=5) when overclocked to 3200 MHz but 60C (DTS=25) is a good upper limit when I'm pushing it to 3600 MHz and a little beyond, God willing. ;)

Software based for absolute CPU temp reading is NEVER ACCURATE, period !

That's my conclusion and it applies to all Core processors. There's a reason why Intel engineers outline such an elaborate process to come up with an accurate Tcase temperature.

The new revision G0 processors, especially the dual cores, use significantly less watts at idle than the previous B2 processors ( 8W vs 22W ) when C1E is enabled. That's even less than the 12 watts that the L2 processors ( E4x00 ) are rated at. This is resulting in some impossible looking idle temps that are well below ambient when air cooled. For me, it's just more proof at how inaccurate the DTS is at reporting low idle temps which it was not designed for. I believe the DTS may start losing accuracy below 50C so even load temps for well cooled G0 dual core processors at default MHz and voltage may not be accurate.

Using the DTS to report absolute core temperatures the way SpeedFan and CoreTemp do is not documented by Intel and is just plain wrong. No matter what TjMax is chosen, I don't believe you can get 100% accurate core temperatures this way across the full range of temperatures that a Core processor can operate at.

The DTS was designed and calibrated to signal and control the thermal throttling and shut down points of an Intel Core processor and for that purpose, it works excellent. Using CoreTemp 0.95 as shown above to keep track of temperatures relative to the throttling point is the only thing users should be keeping track of and using the DTS for.

SLi_dog

08-15-2007, 09:26 PM

The general assumption that DTS=20 or 65C is a good number to shoot for when overclocking is too vague. 65C is an undocumented absolute temperature and people's definition of overclocking varies. My E6400 runs fine at 80C (DTS=5) when overclocked to 3200 MHz but 60C (DTS=25) is a good upper limit when I'm pushing it to 3600 MHz and a little beyond, God willing. ;)It may appear to be more vague that it really is. By definition, running your CPU no closer than 20C from the DTS 0C point means exactly that. You'd have to factor in maximum CPU load and maximum ambient operating temperatures into that threshold. eg, don't go setting the 20C DTS rule as an idle temp at 6am in the middle of winter ;)

I kinda threw it out there as a rough value for beginner users but it is well and truely open for discussion and I actually wouldn't apply it myself in the case of a Core2Quad. I guess the question then is, if not 20C, then what? Maybe 25C as you use? :)

Also, I personally wouldn't run a CPU closer than 20C to the 0C DTS point at 24/7 settings regardless of whether it was at 2.4Ghz or 3.4Ghz so I may have misunderstood what you're saying there.

Conclusion ? Software based for absolute CPU temp reading is NEVER ACCURATE, period !

That's my conclusion and it applies to all Core processors. There's a reason why Intel engineers outline such an elaborate process to come up with an accurate Tcase temperature.Absolutely and that's fine for you, Bing, myself and everyone else who has followed this thread but I guess you're missing the point of my post :). My point was that this information needs to be easily accessible to the average overclocker and it doesn't appear to be.

The new revision G0 processors, especially the dual cores, use significantly less watts at idle than the previous B2 processors ( 8W vs 22W ) when C1E is enabled. That's even less than the 12 watts that the L2 processors ( E4x00 ) are rated at. This is resulting in some impossible looking idle temps that are well below ambient when air cooled. For me, it's just more proof at how inaccurate the DTS is at reporting low idle temps which it was not designed for. I believe the DTS may start losing accuracy below 50C so even load temps for well cooled G0 dual core processors at default MHz and voltage may not be accurate.This seems to contradict your previous statement agreeing that software based absolute temperature are inaccurate. If they're inaccurate for absolute temperatures, I can't understand how they can be used to indicate that the CPU temps are lower than ambient?
Admittedly my understanding of thermodynamics is reasonably limited but I find it hard to believe that any amount of current could exist at a lower temperature than the air that's being used to cool it (Note: I understand about surface area and heat transfer)

At the risk of repeating myself, I guess you're missing the point of my post. My point was that this information should be easily accessible to the average overclocker. From the large amount of threads on this subject here and at other online OC forums, it currently doesn't seem to be.
From following this thread I assumed that there was enough know-how and experience here to put something together that could be of great use to the wider OC community.

bing

08-15-2007, 11:06 PM

Absolutely and that's fine for you, Bing, myself and everyone else who has followed this thread but I guess you're missing the point of my post :).

My point was that this information needs to be easily accessible to the average overclocker and it doesn't appear to be.

Agree, if its accurate, it will be every OC-er dream isn't it ?

But unfortunately the answer is "NO", you won't and can't get that info since Intel doesn't have it for them self ! ;)

1st thing pop out from the mind after reading my above statement is "Huh ? Why & How the heck I know that ?" :D

Ok, here are my explanations, its just basic electronic that every EE learned on their 1st year and "common sense". ;)

1st, you have to understand that popular "tdiode" in that silicon die is basically a plain jane diode, and it is the very basic structure of silicon electronic.

A diode consist of "PN junction" (try wiki yourself, since it will be too long here) that has an interesting physical characteristic, it has a predictable and very linear behaviour response on it's temperature coefficient which is approx. 1.22 mV/deg F or 2.2 mV/deg C.

So whenever there is a changes in the temperature, we could track the changes (delta) temp quite accurate, hence it is used intensively to control the cooling parameter in Intel CPU mechanism, for example adjusting fan speed accodingly, or throttling up/down the PWM controller and etc. The word PECI come in mind ! :D

Now you might ask how about the absolute temp reading ?

This is the weakest point, cause no diode in this world have a same temperature offset even they were produced in the same batch. Unless you could make that diode by assembling atom by atom at the precise location for all diodes. :D

So to make an accurate absolute reading, we must calibrate each diode in a controlled temperature room at high & low temp to get the offset temp for "EACH" diode.

If Intel want to give us an accurate temp reading on "EVERY TDIODE" in the CPU they sell, they must calibrate "EVERY TDIODES" that is spreaded in that CPU at silicon die level. Fyi, each CPU has more than one Tdiodes. :D

Layman term, they have to toast every cpu die and measure every tdiode in it at the temperature controlled room at cold and hot in order to get the offset of each tdiode and write/burn down that offset value into the CPU die.

Sounds complicated isn' it ? :D Ok, even it is not complicated, you must agree that this "extra" process needs an extra production cost right ?

Now, imagine you were Intel management, that always struggling to optimize your production cost at each CPU. Do you want to do that ?

Remember, Intel "officially" never endorse or encourage overclocking on their CPU.

Hope this helps.

SLi_dog

08-16-2007, 12:44 AM

Hope this helps.It does, thanks. This is a great thread and is full of great and useful information and your post is no exception :)

In regards to the variations in the individual diodes used, I understand the nature of that but I had the impression that each DTS was calibrated on a per CPU basis already, possibly for that reason? :confused:

If the DTS read out indicates a CPU's approach to it's throttle point (which is defined by Intel themselves), what's would be the potential danger of giving a value of say 20C from 0C DTS as a rough guideline for Core2 CPUs? I think a statement like that would do much more good than harm. It would also be much better than what is currently being applied where the "actual temperature" value of 65C is used across the board regardless of whether people are "informed" about the flaws of certain applications with certain Core2 CPUs :)

bing

08-16-2007, 03:18 AM

In regards to the variations in the individual diodes used, I understand the nature of that but I had the impression that each DTS was calibrated on a per CPU basis already, possibly for that reason? :confused:

If the DTS read out indicates a CPU's approach to it's throttle point (which is defined by Intel themselves), what's would be the potential danger of giving a value of say 20C from 0C DTS as a rough guideline for Core2 CPUs? I think a statement like that would do much more good than harm. It would also be much better than what is currently being applied where the "actual temperature" value of 65C is used across the board regardless of whether people are "informed" about the flaws of certain applications with certain Core2 CPUs :)

Intel die is using MOS process, its just a type of silicon processing.

If you watch closely "almost" all other electronic chip's datasheet that is using MOS process, has the "absolute" maximum operating temperature on the junction at least 100 C or even more.

Now, Intel's specification at their datasheet stated that the maximum TC or IHS temp at TDP 130 Watt is at 54.8 C, this figures quoted from QX6800 datasheet, imo, currently this is their hottest CPU that they released for "enthusiast" market since they're supposed the owners will OC them. :D

What strange is, they stated the max temp at the TC (IHS) temp instead of Tjunction temp as common industry standard. Dunno why, don't ask me.

But after some "quick & dirty" thermo dynamic calculation on the IHS and of course some safe "assumptions" included, it's like they're blessing the max tjunction at approx. 85C.

If you ever did DIY in electronic or experiment by using common diode as temp probe, and more than one of course, you will know that the offset deviations for diodes as I mentioned at my previous post rougly hovering about + or - 15 to 20 C. Voila ! :D

So simple conclusion, the accuracy for any uncalibrated diode as a temp probe is about +/- 15 to 20 C.

In the worst case scenario, if they're off by even - 20C, meaning the cpu is actually hotter 20C more than the reported temp by the tdiode, still the cpu is in the safety margin/limit before the thermal/throttling mechanism kicks in.

Another quote from Intel CPU datasheet :

The temperature at which Thermal Monitor activates the thermal control circuit is not user configurable and is not software visible

I can only speculate that Intel just use "cheap & quick" testing and verification method to determine that "throttling temp" since they're still at the safety margin.

Now other view angle, that crappy Intel stock HSF thermal resistance at full speed fan is 0.2 C/W. (this is the official number from their datasheet too)

OT, I hate that these days, popular HSF makers don't like to mention this very important specification (C/W) in their product anymore like in the old day. :down:

Now the calculation as usual and using 130 Watt power dissipation :

IHS Temp = (Ambient Temp) + ( (Heatsink C/W) X (Power Dissipation) )

hence

IHS Temp = (Ambient Temp) + ( 0.2 C/W X 130 Watts) = Ambient Temp + 26 C

So "on paper" the max. ambient temperature for this QX6800 beast using stock HSF at 130 Watt full load is at 28.8 C

Of course we all know this is strictly "on paper", real life those limits are much higher and more relax since they just want to play it safe at the "official" paper. :D

Now, if the CPU is not Extreme version which has lower TDP, I'm quite confident that they (Intel) is also making a safe and calculated assumption like this :

Ignore or don't care at the tdiode values, with a cpu running "at stock speed/not OCed", with stock Intel HSF, and at full load in the room temp as high as 30 C, it is very safe to assume that the CPU is still on safe temp limit.

Don't you agree ? ;)

This is how I see this overall issue. :D

The0men

08-16-2007, 03:41 AM

Just to through another spanner in the works, i use asus PC Probe (good for voltages, picks up droop well) NVMonitor and coretemp. Both asus probe and NvMonitor read exactly the same, coretemp reads 2 degrees lower, but shows both cores. I always assume the worst, so i just take into account these temperatures, but assume the highest is correct, then i know i should be safe.

SLi_dog

08-16-2007, 06:06 AM

If you ever did DIY in electronic or experiment by using common diode as temp probe, and more than one of course, you will know that the offset deviations for diodes as I mentioned at my previous post rougly hovering about + or - 15 to 20 C. Voila ! :D

So simple conclusion, the accuracy for any uncalibrated diode as a temp probe is about +/- 15 to 20 C.

In the worst case scenario, if they're off by even - 20C, meaning the cpu is actually hotter 20C more than the reported temp by the tdiode, still the cpu is in the safety margin/limit before the thermal/throttling mechanism kicks in.A +/- 20C error margin is damn scary. That would be a worse case senario though yeah? I'd hope that Intel used used better diodes than that :eek:
That said, the accuracy of the diode is not something we as end users can control and if it was misreading, it would affect the CPU as it approached the throttle point regardless of whether it was overclocked or not so it kinda a mute point to me :)

Now other view angle, that crappy Intel stock HSF thermal resistance at full speed fan is 0.2 C/W. (this is the official number from their datasheet too)

OT, I hate that these days, popular HSF makers don't like to mention this very important specification (C/W) in their product anymore like in the old day. :down:

Now the calculation as usual and using 130 Watt power dissipation :

IHS Temp = (Ambient Temp) + ( (Heatsink C/W) X (Power Dissipation) )

hence

IHS Temp = (Ambient Temp) + ( 0.2 C/W X 130 Watts) = Ambient Temp + 26 C

So "on paper" the max. ambient temperature for this QX6800 beast using stock HSF at 130 Watt full load is at 28.8 C

Of course we all know this is strictly "on paper", real life those limits are much higher and more relax since they just want to play it safe at the "official" paper. :Dhmm interesting.

I'm quite confident that they (Intel) is also making a safe and calculated assumption like this :

Ignore or don't care at the tdiode values, with a cpu running "at stock speed/not OCed", with stock Intel HSF, and at full load in the room temp as high as 30 C, it is very safe to assume that the CPU is still on safe temp limit.

Don't you agree ? ;)

This is how I see this overall issue. :DI do agree that Intel likely aren't that concerned about whether the diode is that accurate because they've built in a fair amount of headroom.
However, if you replace the stock intel HSF with a good quality one and reduce the vcore to only what the CPU needs, you may be able to keep within or below this headroom even when overclocked.

Example 1: A Q6600 at 2.4Ghz, on the stock Intel HSF, in a poorly ventilated case, with auto set vcore (usually around 1.35v) might have max load temps of 20C from the DTS = 0C point.

Example 2: The same Q6600 at 3Ghz, on a Tuniq Tower, in a well ventilated case, with 1.21vcore might also have max load temps of 20C from the DTS = 0C point.

I don't see how one is worse than the other? I'm actually more inclined to believe that the second example is better for the CPU as it uses less voltage which reduces electron migration.

bing

08-16-2007, 07:00 AM

A +/- 20C error margin is damn scary. That would be a worse case senario though yeah? I'd hope that Intel used used better diodes than that :eek:
That said, the accuracy of the diode is not something we as end users can control and if it was misreading, it would affect the CPU as it approached the throttle point regardless of whether it was overclocked or not so it kinda a mute point to me :)

Actually, with the precision level at 65nm technology that Intel has, I believe they can narrow down that margin, by how much, no body knows except Intel. Or maybe other readers that is Silicon Wafer Processing Engineer want to jump in to shed some light here ! :up:

Well, I've done in the past on Pentium D, the throttling point if I'm not mistaken (its too long ago) was about 90 - 95 C ! :eek: According to software based temp reading.

Did it by yanked out the stock HSF fan, and holding it right above the HSF with one hand and other hand ready on the power switch for emergency shutdown. And then run the proggy to load the CPU and ThrottleWatch for observing if throttling started, and marked down the temp when it started to throttle.

While running at max load, I sort of waived the fan farther and closer to the HSF while watching the CPU temp raised until it started to throttle. That was how I figured that throttling temp for my cpu. :D

If you feel gutty, you should try it yourself someday ! :cool:

I do agree that Intel likely aren't that concerned about whether the diode is that accurate because they've built in a fair amount of headroom.

I believe that headroom is much-much higher than we thought, especially after with what I've experienced on that waiving fan game. he..he.. :D

However, if you replace the stock intel HSF with a good quality one and reduce the vcore to only what the CPU needs, you may be able to keep within or below this headroom even when overclocked.

Example 1: A Q6600 at 2.4Ghz, on the stock Intel HSF, in a poorly ventilated case, with auto set vcore (usually around 1.35v) might have max load temps of 20C from the DTS = 0C point.

Example 2: The same Q6600 at 3Ghz, on a Tuniq Tower, in a well ventilated case, with 1.21vcore might also have max load temps of 20C from the DTS = 0C point.

I don't see how one is worse than the other? I'm actually more inclined to believe that the second example is better for the CPU as it uses less voltage which reduces electron migration.

Why not do some simple math to see which one is better ! :D

Fyi, nowdays most mobos do cheat when on Auto vcore, ok, not cheating, but they deliberately mark up the vcore for some headroom like compensating weak PSU factor, cheats on benchmark like Asus which is known always does OC on the Clock Gen chip with abit higher on stock speed and etc.

Lets take latest Q6600 G0 for example, average VID is about 1.2850 Volt, (you might check this as well HERE (http://www.ocforums.com/showthread.php?t=520284)) :D, for the sake of nice rounded number lets put it on 1.3000 Volt for VID at stock speed.

Now, before we start calculating, assuming all factors are ruled out like it has ideal PSU, ideal PWM with nice flowing VCore juide, ideal heatsink mount.

Fortunately, you've pick good HSFs both with thermal spec according to the manufacturer which are :

Data :
Q6600 Q0 Intel's TDP = 95 Watt
VID = 1.3 Volt
Stock Intel HSF = 0.2 C/R at highest fan speed
Tuniq Tower 120 = 0.12 C/W at 2000 RPM stock fan

Lets the math begin with this formula :

Power Dissipation (PD) = TDP x (OCSpeed / StockSpeed) X ( Square(VCore) / Square(VID) )

Example 1 :

PD = 65 Watt X (2.4 Ghz / 2.4 Ghz) X (Square(1.35 Volt) / Square(1.3 Volt) ) =
PD = 65 Watt X 1 X (1.8225 Volt / 1.69 Volt) = 70.096 Watt

Using Stock Intel HSF at 0.2 C/W
IHS Temp = Ambient + (0.2 C/W X 70.096 Watt) = Ambient + 14 C

Example 2:

PD = 65 Watt X (3 Ghz / 2.4 Ghz) X (Square(1.21 Volt) / Square(1.3 Volt))
PD = 65 Watt X 1.25 X (1.4641 / 1.69) = 70.39 Watt

Using Tuniq Tower 120 at 2000 RPM = 0.12 C/W
IHS Temp = Ambient + (0.12 C/W X 70.39 Watt) = Ambient + 8.4 C

Normally I would make the TDP a bit higher by adding more 10% or 20% to be more realistic.

So congratz, you're right, the example 2 with higher clock and also cooler temp ! :)

rge

08-19-2007, 07:33 AM

I guess i will just add to this thread, instead of making another.

My question (at end) stems from ? contradictory statements in two intel spec sheets vs simply variance of measurement error.

first - has some interesting info including temp gradients
http://documents.irevues.inist.fr/bitstream/2042/6579/1/TMI23.pdf

here intel says
"The DTS also reduces the other temperature readings
errors, which are not shown in this paper. The DTS is
calibrated at manufacturing conditions and the reference
point is set to this test temperature. Functionality,
electrical specifications and reliability commitments are
guaranteed at maximum Tj as measured by the DTS. Any
test inaccuracy or parameters variance are already
accounted for in the DTS set point."

So sounds like I should be able to let the DTS go as low as I wish and be within specs, irrespective of actual temps, assuming voltage is within specs, and I can be stable.

In another spec sheet (previously quoted on here), it says Tcase max is 72C. When I gun the back of my cpu with my laserpro or use my wifes accurate touch probe and get 72C, the DTS by coretemp reads 18C (interestingly my cpu Tcase sensor says 71C so not bad, intel under one of its graphs mentions Tcase diode for 50-80C range, though with sample of one mine may just be "accurate" by chance).

Currently I am at 3.83 (E6850 on h20) with 1.4375 in bios, 1.41 measured on water with tcase load 58C (orhthos load). If I go to 3.94 at ~1.47 measured I get Tcase temps of ~72C on TAT (not orthos) full load. DTS reads 16-18.

I am probably going to stay at 3.83 anyway. But question is if I run at 3.94 and if my Tcase temps are running near max, am I slowly cooking my processor, or since my DTS is around 16-18 I am within specs?