Nice work The Coolest!
CPU sensor reads ~10C higher then my DFI here :)
Nice work The Coolest!
CPU sensor reads ~10C higher then my DFI here :)
Yes I see that. :p: Funny to think about isn't it? A common proceedure even here at XS(which made sense at the time) is to adjust the heatsink/water block while watching the temperature and getting the lowest temp. If the thermal diode read from motherboard is an average of both diodes it would be possible to still get a lowest "average temperature", but averaged way out of balance per core.(i.e 14 and 27 would average 20c) Seems a worthless technique now. If the accuracy of "Core Temp" is in question, it's ability to check for even cooling is not. Another thing is a bit interesting is Sorrow13's Opteron 165 above mine, also has a large difference between the cores. I'll have to go back over the 8 pages and look for other 165's to see any pattern. Kind of mind boggling too that this mounting (5th time) was the only mounting I could prime for 4 hours at 3150Mhz at 1.6v, where I could bearly work in windows with other mountings, and had coolest average temp in BIOS. But yet it looks the worst mounting with this application.Quote:
With that big of a Gap between your cores I would try to reseat your water bloack as you probably have bad contact.
Not sure how to respond to this........but.... that's the point. There are 3 or 4 people that are getting below ambient readings where it is impossible to do in the real world. I think we all know that and are questioning the completeness of this application in it's current build.Quote:
There is no way u could cool below ambient with water cooling without cooling it actively like with a chiller or sticking the radiator in a frigde/bucket of ice. things like that.
Anyone have any idea why it's not reading my newcastle correctly? still reads 40c at idle with watercooling when the ambient is a lot lower AND I had ice water running through it.
Ok lets see.
Regarding calculation of the temp. The TCaseMax - Value, that's for Intel CPUs. On AMD the app reads a value from the CPU, substructs 49 from it and that is supposed to be the temperature.
I am not sure why there are so much reports of two cores with drastically different readings, and why some go below ambient. The only logical thing I can think of is bad callibration on AMDs part, as this feature was still undocumented and not used officially for these chips.
Regarding the Newcastle. As stated above not all DTS are perfect, so I guess not all of them will work properly. The Newcaste is 3 revisions behind the official announcement of this feature.
Regarding even temp readings on dual cores, the only time I've seen someone really equalize his temps was in the first post on the 5th page. The are many IHS-less running with very different temperature though.
Quote:
Originally Posted by turbotoaster
just incase anybody missed it, for those with different temps on each core
my HSF doesnt have screws.. it has clips. I'll try to press it in different ways to see how the temp changes.
edit:
The cores are reverse.. core 1 in taskmanager = core 1 in CoreTemp, core 0 in task manager = core 2 in CoreTemp. I tried to press the hsf from different sides but no use.. my 2nd core gets very hot for some reason when i run prime (60C+) When i run prime on only first core then temps are about same on both cores.. but when im priming on both or 2nd core.. then that one is ~10C more. Im afraid to run prime now :(
guys again can anyone email me this prg since the download link doesn't work for me
is there something wrong for IPs from Egypt ??
email is ahmad.okda (AT)gmail.com
It's a great app, but the thread should read AM2 and Intel Core temp monitoring tool. My poor 'old' Opteron won't work ;(
That's what I'm beginning to believe. I don't understand why one of the cores almost always seem to give results that are probably close to being accurate though.Quote:
Originally Posted by The Coolest
That's not always a solution. But it would be nice if by adjusting the HS I could bring both cores below ambient temps (lol).Quote:
Originally Posted by turbotoaster
Email sentQuote:
Originally Posted by SaFrOuT
It works for most K8 chips and should work on your Opteron. Maybe post up your text dump for the coolest to look at.Quote:
Originally Posted by Rickster_64
I was hoping it'd work on Dothans too but I guess not. :(
I'm still totally lost as what these temp readings should do or prevent from happening... the Tcore temp seems to have much more sense then the TCaseMax thing..
All AMD chips have a TControlMax temp which ensures proper operation until that temp is reached. TControlMax is documented in AMD's tech files with specifics per CPU... mine is set for 70°C
TCaseMax is generated by factors unknown... core contact, TIM, IHS.. who knows...
For us it's much more important to know the TControlMax and the TDP so we know how much heat we need to dissipate from the CPU to stay well below safe margins... as we all know it's best to keep your CPU under 50°C....
Also if my TCaseMax represents my max oporational temperature (57°C) before my CPU starts to degrade then why doesn't DFI let me set a thermal shutdown in my BIOS for 50°C or 55°C?
Don't get me wrong I like the program... even though I already had the coretemp feature in Everest but I think people are making the wrong assumptions when it comes to link the temperatures to the CPU.
I tried on my system, and am getting 35c and 40c respectively on core1 and core2... Not a huge difference...
I like this, I also tryed NXSENSOR, and it reported the same thing, the Fintek chip on the mobo, and the CPU diode, which matched the coretemp app., and what speed fan reported.
Load:
http://f10.putfile.com/thumb/7/19714423730.jpg
Idel:
http://f10.putfile.com/thumb/7/19714431237.jpg
Hmm my temp reading with coretemp so high....77c... IHS removed + watercooled with BIX3
Idle
http://img148.imageshack.us/my.php?image=idleam7.jpg
Load
http://img148.imageshack.us/my.php?image=loadgg9.jpg
any advise?
i like this program,
alot of people reporting HIGHER temperatures than what speedfan/mbm is reporting,
im actually getting a few degrees lower, die temperatures are about -28 in coretemp and -26'ish in mbm when idling,
strange thing is though, my core2 reads the above, but my core1 is always reading -65, which couldnt possibly be right!
Ok..d/led new vewer version
Idle
core 1 22
core 2 24
Large fft P95
start P95 core 1
core1 only temp goes to 38c
core2 36C
P95 both cores
start core 2
core 1 47/48C
core 2 39/40
stop...
start P95 core 2 only
core1 45C
core2 36C
something seems very weird....why would core 1 temp rise so much with P95 only running on core 2....
the temp sensor/calibration can be correct.....
-65? The minimum this program should be able to report on an AMD is -49C. can you upload a few screenshots and a txt dump of GCPUID (after Core Temp ran)
I'm also getting a 10 degree difference on my x2 4200, but I know its not bad IHS contact because I took it off, lapped it (not the core side, the HSF side), and re-attached it with AS5.
ya.. just wondering with 3200+ venice @ 2.2 + watercooling on load i get 77c with coretemp... imagine if using standard HSF :confused:
Works great on my system! I'm about 3-4°C under the reading from my dfi board and it shows much higher temps, then 99% of the other ones. Thx! :)
WOOHOO!!! I finally tried this and i actually get REAL readings!! 45C load, 32 idle. 48c load with fan @ 5v (completely silent)
Here is my experiences with Core Temp:
Opty 170 (see sig for details)
at idle:
(core1 =36degC, core2=41degC)
http://img124.imageshack.us/img124/1...idleyl6.th.jpg
w/ toast.exe running on core 0:
(core1 =43degC, core2=52degC)
http://img123.imageshack.us/img123/1...ore0rc3.th.jpg
w/ toast.exe running on core 1:
(core1 =61degC, core2=55degC)
http://img518.imageshack.us/img518/5...ore1yq1.th.jpg
w/ 2x toast.exe running (1 per core):
(core1 =66degC, core2=64degC)
http://img205.imageshack.us/img205/1...oreswa5.th.jpg
some questions:
1) how can my opty 170 only be 35 watts?
2) if TcaseMAX is correct (49degC), why doesn't my cpu crash?
3) why does core 2 (in this program) read almost loaded temp, when core 1 is the only core w/ a load?
I must fix that.
49C is the result of '0' reported by the register, previously there were specs for 49C TCM, in a later revision of the document AMD said that 49C was not used anymore. If I had to guess it'd be that chips that have over 71C TCM (max that can be reported), are given '0', so they put out more wattage probably than what the max AMD rating is.
But I am not AMD and I can't be 100% sure, this is just my opinion.
Next rev will have 49C\35W removed.
Thanks for clarifying this coolest ...I'm looking forward to the revision.Quote:
Originally Posted by The Coolest
Hey man:
For some reason the forums where this is posted to download are down.
Can you mirror somewhere?
If the file is small, I can provide a mirror as well on my school web-space.
-Ghent
The server went down tonight because of a bad update, it is now back up :)
Sorry for the inconvenience
At stock, CoreTemp shows 10C higher on idle and temp than Speedfan (which seems right to me). Speedfan: 26/36 CoreTemp: 36/46
But when overclocked, CoreTemp shows 10C higher on idle and 20C higher on load :eek: than Speedfan. That really doesn't look right to me. Speedfan: 30/40 CoreTemp: 40/60
Help?
Speedfan doesn't show temp of each core.
That's a wrong way to compare.
Point taken, but I'm not really interested in comparing the values as I am comparing the temp increase.Quote:
Originally Posted by The Coolest
If both cores are really increasing 20C, shouldn't speedfan which reads the temp sensor on the board show an increase of greater than 10C? My temp probe also only shows an increase of about 10C.
I've already popped off the IHS and replaced it after changing the thermal compound with arctic silver, so my temps should be better than what CoreTemp gives me. At least, I'd like to hope so :D
EDIT: The hotter core has an increase of 20C on load while the cooler core is more like 25C according to CoreTemp.
Also, I'm not knocking the program one bit. I'm just trying to figure out which one is right before my rig has a meltdown :)
this program shows idle temp that is 11 degress more then the temp that the MB software shows
and i understend that the core temp shuald be higher then but the is the 11 C diferent is OK?
i have TT BT
http://img458.imageshack.us/img458/6885/untitledgx6.png
In my Experience Speedfan reads 10 degree's low on Idle and almost 20 degree's low at full load.
And if you think about it, it makes perfect sense that this reads higher than any other program as it is reading the Temperature from the Actual cores and not some sensor on the outside.
That's the kinda info I'm looking for. Thanks!Quote:
Originally Posted by CanadianTSi
The sensor is on the outside but it is reading the output from the thermal diode inside the processor. AMD says in their white papers that the thermal diode can read as much as 14C low and that's assuming that the offset is properly applied and there is no error in the onboard sensor.Quote:
Originally Posted by CanadianTSi
am i missing something... my 165 is naked and if i apply more pressure to diff areas on the block the two cores tempos dont change so its not a contact issue...
core 1 reading = 28C (on water this is seems good)
core 2 reading = 38c (seems a lil high)
Yes. Read through this thread. Temps such as yours are not because of faulty IHS or HS contact on DC processors. It seems to be a problem with the calibration of one of the thermal sensors.Quote:
Originally Posted by bcbooter
http://img106.imageshack.us/img106/7598/temphr9.jpg
strange .. isnt it ?
How is that strange?
everybody says speedfan reads lower temps than "core temp" except me ...
Hmm.. who knows. You do have watercooling though, right?
yes, external radiator o.O
Quote:
Originally Posted by $ilver
Actually I also had it read lower than speedfan when i was running it on an amd 64 3000+ Idle was lower than speedfan. It was about 32 in the core temp checker while it was about 35 in speedfan.
It was the other way around on my opty 165 tho
Maybe the reason why I can't OC any more is because of the temps. Here I thought I was at 45*C when actually using this app I see I'm reaching 50*C. According to the app my TCasemax is 55*C so this all makes sense now. Guess I'm gonna adjust the RPM on my fan on my CNPS9500, I have it set really low right now (I don't like noise).
Hmm, so this means when my fan was speeding up to full @ 47*C when the room was really warm I was actually near that 55*C TCaseMax temp? Think I should get a new cooler.... or remove the IHS.... Hmm...
I love this programme. However, is there any chance of making the task bar icon when you minimize it show the CPU temperature, as opposed to having to hover over the icon? I reckon something like ATI Tool's temperature display would be awesome.
As you can see from the attached screenshot, Core Temp reports the exact same temperature as the "CPU Diode" value in Everest Ultimate Edition. Are both programs reading the same sensor or is this just a coincidence?
cool program. do we have a sticky of all these good programs? if not, we should
Both programs are reading the same sensor. It's mislabeled in Everest. The label "CPU" is actually the diode.Quote:
Originally Posted by Raillex
I kind of thought so, as that value jumps around (and skyrockets under load) alot like the CPU Diode readout did on my old Asus NF2 board. The good news, for me at least, is that I have a lot more overclocking headroom than I though I had since the "CPU" readout on Everest is always 4-5 degrees hotter than the "Diode" readout under heavy load.Quote:
Both programs are reading the same sensor. It's mislabeled in Everest. The label "CPU" is actually the diode.
BTW, MBM agrees with Everest as far as what it reports as a CPU socket temp. Is is possible that MBM is confusing socket and diode temps as well?
The bios and programs like MBM use the sensor on the motherboard to read the value of the thermal diode. CoreTemp uses the thermal sensor inside the processor for temperature readings. This sensor is used to shut down the processor in case the high temp threshold is exceeded. Until CoreTemp, and now Everest and NextSensor, no program used this sensor for temperature reporting.
Question for DFI LanParty NF4 users (Ultra-D, SLI-D, SLI-DR, whatever):
How far off is your idle temp from what CoreTemp is telling you?
I'm using the MBM from DFI-Street, which most people probably use too because how else do we get it to wrok with our DFI boards.... and i'm reading 33 idle for my Opteron 170, but CoreTemp is telling me I'm at 37/39 for the 2 cores. Once I shoot up stress prime, my load reads 56/52 for the 2 cores. MBM is telling me I'm only at 41.
?????????????????
Which is correct?
Are you guys using any offset in bios?
Edit: With my Panaflo running full blast and an XP-120, should my temps really go that high?????
Edit #2: It says my casemax is 49 degrees, does this mean I'm killing my CPu because its running in the 50s according to CoreTemp?
Ahh, so yea just curious what excalty does this program read to produce the microarchitecture code name.... I just ran it on my secondary (3700+) and it Reads "Toledo" now my stepping for that chip is CCBIE which is a dual core opteron stepping... so whats it reading the chipitself or stepping?
Do i have a hidden core somewhere?
http://i7.photobucket.com/albums/y282/yaddam205/piv.jpg
Yep, you have a failed dual core chip. Nice TDP though.
@The Coolest
can you tell me which register you are using? or atleast point me to the data sheet where you got the info from?Quote:
AMD chips report the temperature by a special register in the CPU's NB. The way the temperature is calculated like this: 'Core Temp = Value - 49'.
They can report temperatures between -49C and 206C
thx in advance
In the RevF K8 BIOS Design GuideQuote:
Originally Posted by Dragoran
Its the Thermtrip Status in the NB registers, in my PDF its on page 163
As was said above, you have a failed dual core.Quote:
Originally Posted by yaddam205
I chose to make the program report the code name of the actual core that is used and not the code name for a given chip (in your case it'd be San Diego).
1) I think that the higher temps are probably the correct ones, but we can't be 100% sure.Quote:
Originally Posted by dmo580
2) The TCaseMax == 49C now appears as "Unused" in official AMD docs, it will not be shown in the next release, I would just ignore that.
And one more thing.
Thanks Praz for having answers and sharing them with everyone. I don't think there's anything more I could add to them :)
I'm getting a 10 degree difference between cores on my x2 4200, but I know its not bad IHS contact because I took it off, lapped it (not the core side, the HSF side), and re-attached it with AS5.
Is this a technical issue with dual core chips?
@The Coolest
thx I have found it ;)
Have you come across anything that either states or can be inferred what the accuracy of the thermal sensor is? I have read in several AMD papers the accuracy (or lack there of) of the thermal diode but nothing on the sensor. Since the thermal sensor is being used to set Thermtrip I would assume the accuracy is pretty good. But there still has to some type of margin of error.Quote:
Originally Posted by The Coolest
Nope, I haven't found any indicators on accuracy either.
Ambient: 20c
Core #1: Idle 22-26c
Core #2: Idle 26-30c
Something's not right, my coolant temperature is 25.5c right now O.o
*double post*
As I've said like a million times now:Quote:
Originally Posted by creidiki
The program is only as accurate as the sensor. It could be poorly callibrated, and underreport.
It´s the sensor that reads the temps the program only report them.
My processor seems a little too hot!!! I've never seen this 10°c extra before. Where does it come from? Is it legit? 62°c seems way exagerated. But maybe it's the real value? I'm puzzled. It doesn't make sense to have a Tcasemax of 55 and gettign to 62 (and eve 65...) no prob for a whole day... It's literally 10°c added to the temp I'm used to read.
So if I understand it properly, this is the right temp, and the 10°c less is the wrong temp. So, in my bios, I shoudl adjust the fan speedup according to this temp not the temp I'm used to read which is 10°c less(?). Say if it's above 40 (meaning 50 from this prog), then i should speedup the fan to cool.
Also A64Info 0.61 seems to report same temp reading from it's core 0 section then Core Temp Beta 0.9.0.91. This coudl explain why I would get hangs at cpu temps of 48-50°c.
it says i idle about 7c higher than mbm says. thats not cool. first time i used core temp, the sensors flipped out, going from -20 to 200c, and my computer shut off. if my temps are really that high, i guess i have to underclock :( my load temp according to coretemp is 60c. i dont believe that. the heatsink gets warm to the touch, 60c wouldnt be just warm.
theres a 4-6c difference between core 1 and core 2, core 2 idles higher and loads 1-2c lower. core temp says my processor has a 65w TDP, but its a 110w :(
People are reading way too much into this. All these programs are accurate to the point of reporting the temperatures that they are receiving. The only thing that is certain at this time is there is no way of knowing how accurate any temperature data is from any AMD processor. All these reading should be used as guidelines only. I think all the authors of these programs will agree.
you have a 110w chip, the TDP isn't the same thing as the 110w rating...Quote:
Originally Posted by breakfromyou
Exactly.Quote:
Originally Posted by Praz
one thing to consider is that no matter what your thermal contact is like, the heatsink will still be removing about the same amount of heat: the amount of heat the chip is producing.Quote:
Originally Posted by breakfromyou
if you have bad thermal contact, the chip will still be putting out roughly the same amount of heat. it'll just get a lot hotter due to the thermal resistance from ihs -> heatsink ;)
so now, tell me the point of all these programs then? most people already have speedfan or something else which is just as inacurate as this little app.Quote:
Originally Posted by Praz
1. I prefer the temps it gives me.Quote:
Originally Posted by Dave_Sz
2. It uses minimal mem. (Good for Screen Shot while benching)
3. Only displays what I need.
I dont really know what program to trust anymore.
http://i3.photobucket.com/albums/y55...eTempvsMBM.jpg
I also prefer the temperatures reported by the thermal sensor.Quote:
Originally Posted by yaddam205
Although there is really no way of knowing the accuracy of the sensor or the diode I think a couple of assumptions can be made. The thermal sensor is what AMD decided to use to supply temperature data to the ThermTrip register. Evidently they feel this sensor is accurate enough to protect the processor in case of overheating. By their own admission they state the thermal diode could read as much as 14C low in a worst case scenario . They also state that manufacturers should not use the diode for thermal management design.Quote:
Originally Posted by Dave_Sz
Since the sensor is used to protect the processor a person wouldn't expect an error of accuracy of more then ± a few degrees. AMD states the diode reads ~4C cooler then the center of the CPU junction (which is the hottest area) and ~1.8°C additional thermal margin should be used when using the thermal diode instead of the die surface temperature. Instead of a worst case scenario for the diode if a person uses a 6 or 7 degree of inaccuracy then both the sensor and the diode come close to reporting the same temperature.
I just tested another proc with this program. I trust the program but I'm wondering if there's any reason why 1 core would have about 13C difference than the other. I've remounted and repasted about 3 times but no difference.
Just wondering if there are any theories as to why 1 core would be so much different to the other besides the mount. I"m just using the stock amd cooler on a X2 3800.
Which core is reading off so much? Usually it's core 0. It appears that AMD only calibrated 1 sensor on dual core processors. Core 0 for the processor in my sig always read a couple of degrees below ambient at idle.
the new version of speedfan, 4.29, has also the core temp reading:
http://img452.imageshack.us/img452/5855/coretempzr8.jpg
Yes it does.
Nice work.The temperature is about 1C higher what's shown in Speedfan. I consider it very accurate.
http://img438.imageshack.us/img438/2...emp1kg6.th.jpg
One problem though, why it doesn't show the TDP and TCaseMax for my Orleans 3000+?
Edit: Oops, sorry about the double post
If I remember correctly 49C is not a valid value. If the processor reports this value I believe The Coolest decided to ignore it in this release.Quote:
Originally Posted by dickeywang
I must say my TCaseMax and TDP value seems somewhat low for this chip... An energy efficient Opty 165... yea right. :D
EDIT: nvm I see it's an already known bug and meant to be fixed in next version.
http://img284.imageshack.us/img284/7583/coretempyg3.jpg
I just join to post this information regarding tcasemax and core temperatures...
tcasemax physically represents the maximum temperature allowed either at the top of the die or at the top of the IHS depending on how Intel or AMD did the thermal modeling.
core temperature since from the definition in the op is measured in the die, can only represent the junction temperature.
so, actually comparing core temp. to tcasemax would be incorrect.
you should compare tcase to tcasemax
again, tcase is physically the temperature either on the top of the die or on top of the IHS.
To convert, junction temperature (tjunction or core temp) to case temperature (tcase) you would need to know the thermal resistance of the die (silicon) and the thickness of the die if tcase is defined as the temp on top of the die. Or additionally, you would need to know the thickness of TIM, thermal resistance of TIM, thickness of IHS, and thermal resistance of IHS if tcase is defined as top of the IHS.
Max junction temperature is defined by foundry process only. Wafer subcons like TSMC defined this to be either 90C, 110C, 125C or 150C depending on different process. This is unknown for Intel or AMD since they fab their own wafers....
here's mine
http://i69.photobucket.com/albums/i4...ncoretemp2.jpg
i have a opteron 165 and it seems to show core #2 temp in speedfan 4.29.
Lovely little app. Thanks for your contribution, keep up the great work! :)
Hello,
i'm just wondering why in most cases Core temps (SpeedFan) are higher than Cpu tems???
Your conventional motherboard monitoring programs should be reporting the CASE TEMPERATURE (Tcase) as the CPU temperature. Again, depending how it is defined, it can be the maximum temperature at the top of the die (dead center temperature of the die) or the maximum temperature at the top of the IHS (dead center temperature of the IHS). The key point about Tcase is that it is can be physically validated by a temperature measuring tool (ie thermal couple).Quote:
Originally Posted by martin_ezz
The core temperature as defined in the original post here should be the JUNCTION TEMPERATURE of the die. Compared the CASE TEMPERATURE, JUNCTION TEMPERATURE is less useful to the average system builder or consumer because it cannot be measured physically without destroying the part.
Since Intel and AMD are not system builders (end products are ICs), they defined their products' design limit with Tcasemax (because Tcase can be measured by the customer). This is an industry standard way of reporting thermal limitation of ICs. Of course, they CAN also define Tjunctionmax since Tcase is derived from Tjunction, but customers who buy these ICs would not be able to verify the actual Tjunction. The ability to verify Tcase is very important to system builders (I don't mean average system builders but Dell, HP...) because they need verify the thermal design limit of the system.
Tjunctionmax is usually defined by wafer foundries based on the process technology of the wafer. For example, 0.35um process would have a Tjunctionmax of 150C, 0.13um process would have Tjunctionmax of 125C (don't quote me on this; just for example)...etc And you can see this in Nvidia graphic driver where they set the max core temperature to 125C.
Thus, if you know the Tjunctionmax of the CPUs, then knowing the core temps would be useful.
Or if you know (for example if Tcase is defined as top center of the die) the die thickness and the thermal resistance of the die per thickness and power draw of the die, Tcase can be calculated from the core temperature.
For the newer Flip Chip type IC modules (basically all CPUs and GPUs nowadays), the core temperature would be close the case temperature but they would not be identical since the thermal resistance of bare silicon has to be taken into account.
Under dry ice
http://lites.no.sapo.pt/coretemp.jpg
Nice program, thanks! :)
Would it be possible to add an icon to the program please?
so how does this work with the new Intel Core duo CPUs ?
are the readings we are getting accurate at all ? there seem to be too many discussions at the moment about which program is accurate ...
If you're getting the results of the core temperature, you cannot verify it. You just have to take Intel or AMD's word for it.Quote:
Originally Posted by Omarko
I'm just pointing out that in this program, the core temperature is compared to Tcasemax which is comparing two different things. I'm not trying to discourage the author of this program, I think the program it is great, but I want to point the inaccurate way of how this program report results so everyone is aware and may be the author will make an update.
Holy ****!! i'm only doing 23-25c on core 1 and 31 on core 2. Smart guardian is showing 33-34c. Are you sure this is accurate stuff? MY room temp is 25c, so theres no way water cooling can dip below room temp.
Seems everyone's speedfan reads almost exactly the same as this program, but mine is completely off.
I have written a similar programm for linux...Quote:
Originally Posted by Lolento
you only read a value from a register and calculte the temp ( regvalue - 49), so the programm only reports what your cpu provides (in case of amd, I don't have a core2 duo ; and also havn't read the core 2 dup docs)
Since we are talking about inaccuracies we need to discuss the thermal diode. AMD states in their white paper that because of the location of the diode it may read up to 4C cooler then the junction temperature. Then they go on to state that the accuracy of the diode is +/- 10C. They also say that these values do not include any inaccuracies of the sensor on the motherboard. AMD also cautions manufacturers not to use the diode temperature for thermal design calculations.
None of us know how accurate the thermal sensor is built into the processor that Core Temp is reading. What we do know is that the temperature reported by the bios and programs that we have used in the past is at best a general indication of what the actual temperature may be. This information is nor speculation. It's presented by AMD in their design papers.
This is exactly where I'm getting at. System builders will not rely on reports of junction temperature because it cannot be validated.Quote:
Originally Posted by Praz
Junction temperature can be verified. AMD outlines the procedure in one of their papers. They state the temperature reading will be within 2C of the junction temp. Manufacturers are told not to validate their systems using the thermal diode. The same diode that the bios, MBM and similar programs rely on that we have been using.
what's a thermal couple???Quote:
Originally Posted by Lolento
Junction temperature can be verified. But you have to destroy the chip by drilling a hole from the pin side of the cpu to place the thermal couple.Quote:
Originally Posted by Praz
Case temperature can be verified without any destructive technique.
AMD tells how to measure the junction temperature indirectly. Following their guidelines they claim the readings will be within 2C of actual junction temperature.
Two things we all know for sure. Core Temp is only as accurate as the thermal sensor. No where has any info been reported from AMD as to the accuracy of the sensor. But the info released from AMD states the diode can read as much as 6C high to 14C low. And as AMD says this does not take into account any inaccuracy of the motherboard sensor or the software being used.
yes, the indirect method IS measuring the CASE TEMPERATURE, and then reverse derive the JUNCTION TEMPERATURE of the chip. But knowing the JUNCTION TEMPERATURE would not help you since you don't know what is the TjunctionMax.Quote:
Originally Posted by Praz
Any time you measure the surface temperature of the chip, you are measuring the CASE TEMPERATURE.
Again, the CASE TEMPERATURE can be defined differently (however, always on the surface of the IC package) but the easiest definitions are either top of die or top of IHS. This is because these locations make it easy to calculate the JUNCTION TEMPERATURE.
On the other hand, if you define the CASE TEMPERATURE on the bottom of the IC package, it will be more difficult to calculate the JUNCTION TEMPERATURE because new IC packages use organic substrate instead of ceramic (like in the old days). The thermal resistance of organic substrate is hard to characterize due to asymetry in substrate design (copper thickness versus dielectric thickness makes a big difference). On the other hand, older ceramic substrate is much easier to characterize due to tungsten thickness and ceramic thickness have little effect on thermal resistance.
Anyhow, since Tcasemax is given for the AMD chips. You should just follow AMD's direction to measure Tcase. Knowing Tjunction does not tell you anything.
If you read AMD's paper, they assume you understand what is CASE TEMPERATURE and what is JUNCTION TEMPERATURE. So before you read them, read a book on IC thermal management.
I highly suggest you understand these meanings before you read any thermal management papers. Aside from Intel and AMD, other design houses use ThetaJA, ThetaJC, PsiJT...to define thermal design limits. You will not know how to build a reliable system if you don't know what these mean.
Definitions:
Case Temperature (not the pc case temperature) is the surface temperature of the IC package. This is usually defined as the top surface temperature of the IC package. The limitation of case temperature is defined from the maximum junction temperature allowed and the configuration of the IC package.
Junction temperature is the temperature on the active circuitry of the die (...how do you physically measure this without destructive technique once the die is packaged in flip chip form???). The limitation of junction temperature is defined by the wafer techonlogy.