Well... VID is not VCore. VID is the Voltage ID setting on which the mobo bases the VCore setting on.
What I am interested in is why the temperature is different between TAT and Core Temp?
Printable View
Well... VID is not VCore. VID is the Voltage ID setting on which the mobo bases the VCore setting on.
What I am interested in is why the temperature is different between TAT and Core Temp?
coretemp and tat give identical results for me.. ~88c core 1, 91c core 2, 98c zone 1.... :D
i seeQuote:
Originally Posted by The Coolest
well both my E6600 B1 and E6700 B1 show TAT under report temps under load and over report temps at idle see http://i4memory.com/showthread.php?t=3082
What we need to do is figure out at what temp 'CoreTemp' reads to be the safe temp of 60c. Right now no one knows what the hell temp there cpu is really at since there is no one way to measure it. I say we all use coretemp, one it's free, works on most rigs (speedfan and tat do not work on my rig so they are not a good choice, lol) and is easy to use.
So what everyone needs to do is find out what temp in coretemp is = to intels 60c and we can all rest easy:toast:
the link to the prg doesn't work ever for me, i am located in Egypt and don't know what si wrong
can you please host it on filefront or here on the forum for example so i can get it
sorry for bothering but i have been facing this problem since the start of your thread with all versions
Are you reading the dts output from the msr or through the peci interface?Quote:
Originally Posted by The Coolest
Could be this a reason for different readings.
Also i would like to know where you read TJunction, it is from a msr? I can't find nothing related in all the datasheets of core 2 processors. Well it's obvious that the DTS is referenced to a temperature higher than TCasemax, but in the datasheets intel did not mention nothing about this temperature...so maybe you have a friend at intel :)
The core 2 duo processors can take 85 C of TJunction without problems, because intel has planned this and even inserted a circuit that throttle the processor at this temp. When the processor reach a critical temperature (higher than 85) another control circuit kicks in and shut down the processor.Quote:
Originally Posted by rodman
please correct this error when we enable EIST :
http://www.persianpcshop.com/EIST.JPG
another Mirror for CoreTemp v0.93 :Quote:
Originally Posted by SaFrOuT
I upload it here for you : ;)
http://www.savefiles.net/d/mpdynjqh3kmq.html
That error is due to the fact that core temp is not able to read the FSB from the mobo pll (well The Coolest correct me if i'm wrong :p: ), is the same bug of sandra that made some people think the northbridge clock was linked to the cpu multiplier :DQuote:
Originally Posted by HamidFULL
Quote:
Originally Posted by HamidFULL
Thank you, Thank you, Thank you
seems that i am not the only one facing the same problem since it has been already downloaded for 6 times
The temps are read from the MSR. Regarding the Tjunction... Well I wish I had a friend at Intel, heh. But I do have a friend who does, so he sends me some interesting bits of information, also I spoke to someone who decompiled TAT and found out that it was looking at the same MSR as I am and checking it for 85C or 100C.Quote:
Originally Posted by astaris
That is unfortunatly still a problem.Quote:
Originally Posted by astaris
Franck (CPUz), had to write a completely new way of detecting CPU clock speed using a special hardware based timer, which means some major recoding of that part.
I don't have much free time anymore, and I still have some other parts of the program that I want to improve or add, so we'll see what gets the priority.
I know this is a confusing and sometimes looks weird/funny but there's not much I can do right now. All I can say is that I will get it done eventually.
Well, thank you for your answer. Still, it is really strange that tat and you prog give different temps. Also some people reported that after updating the bios, coretemp would give totally different temps...but sincerly i can't see how this can happen...Quote:
Originally Posted by The Coolest
Anyway thank you for your great prog, i used it for my beloved opty 165 and i'll use it for my 6600. And about the fsb thing, i think this is really a minor issue, after all i use coretemp for the temps, not to monitor my system frequency. Well i mean, i know my fsb, i don't need a prog for that :D
another Mirror for CoreTemp :
Download v0.93 Here 100% Uptime
astaris:
Thank you :)
HamidFULL:
Thanks, both mirrors are now posted on Core Temp's frontpage now.
my temps are a couple degrees lower than TAT, and TAT is a couple degrees lower than asus probe
my coretemp software's been crashing lately 10secs following startup of app everytime......:shrug:
old and new versions
asus probe runs perfect but reads 3-4C higher consistently no matter what temp range
Go into the options, and try to disable the logging feature.
yeah that fixed it :toast: lolQuote:
Originally Posted by The Coolest
So I (we) understand it as to keep temps (using coretemp) below 80c to be on the safe side? LOL, no problems there;) Are we all in agreement?
FWIW TAT and coretemp give me different temps too... about a 3-5C difference with coretemp being reported hotter.
Yes under 80 and you are safe, but as always in this case lower is better :DQuote:
Originally Posted by rodman
See post 102 just above yours by 5 posts: http://www.xtremesystems.org/forums/...&postcount=102Quote:
Originally Posted by astaris
I run ~85c when gaming, up to 91c when running both CPU cores under load and no GPU load.......
Are they ever going to get this proggy to read negative temps from the bios?
And you are sure that your cpu does not throttle? Also do you have enabled Thermal monitor control in bios?Quote:
Originally Posted by STEvil
After 6h Prime95.
Isn't nice temperatures :-(
http://sweb.cz/EXTREME.OVERCLOCKER/c...%20pime%20.JPG
I will searching better cooling.
your temps are so fine under 60C is very goodQuote:
After 6h Prime95.
Isn't nice temperatures :-(
is there a reason why core temp is very inaccurate for me? it shows both cores ~45c when speedfan shows 31/32c, bios shows 32/34c, eztune 5 shows 32/33c.
core temp is within 1-2c of everest and TAT for me
The only reason is that CoreTemp/TAT (or Everest *Ultimate*) is able to show the readout (translated to temp) from the digital diode(s) in the core(s).Quote:
Originally Posted by ryaan
Bios, Speedfan, EZTune does not show coretemp.. well.. it didn't last time I checked. :)
I prefer looking at my speedfan temps. They're a good 13 degrees lower. =P
That's the case for me also with CoreTemp being lower.Quote:
Originally Posted by Tallman
I'm using Asus P5WDH with bios 1503 & E6700 .
While orthos priming with stock cooler and 1.45Vcore , i read a temperature of 124-128C . I don't know why , but straight after 85-86C it jumps to 125C .
Ouch, that is very odd.Quote:
Originally Posted by Stelios
What does TAT read?
What does your motherboard report?
How accurate is Core Temp?
TAT doesn't work as i use Win2k3 .Quote:
Originally Posted by The Coolest
Asus Aibooster and all other similar programs shows 67-70C . That's approx 18C below core temp (when core temp shows right temp) .
See the attached pictures .
I think your computer is on fire or has turned into some kind of a fission reactor, have you seen smoke coming out? (Kidding)
But seriously, 67-70c in Aibooster and you are not scared for 24/7. What is temp when running prime95 or orthos?
OK . Small update . :)
I managed to run Intel TAT . Just needed to change the compatibility mode to win2000 . After 85C , TAT reads >85C . No 86C or anything else . Just >85C . At the same time Core temp gives 124-126C . So , mystery solved . :p:
Previously i haven't encountered this , as all previous versions of core temp had the overheating pop up window which made orthos prime to loose priority.
:D :D :D :DQuote:
Originally Posted by Yoxxy
These temps are while running orthos prime at 3.5Ghz & 1.45Vcore (Real vcore as i have made vdroop mod) .
Of course i'm not running this setup at 24/7 . My default cooling is a single stage (subzero cooling) . I'm just testing some CPU's . :rolleyes:
Uhm i remember from intel datasheet that when temperature is higher than the maximum operating temperature (i.e. TJunction) an out of spec status bit is set in the DTS MSR. Maybe this is the cause of the odd behaviour of coretemp.Quote:
Originally Posted by The Coolest
I'm pretty sure that chip is OVERHEATINGQuote:
Originally Posted by Stelios
when a motherboard sensor shows you close to 70C that is very very bad.
Yes. That is probably it. the (?) at the end of the temp suggests it couldn't get a proper reading, so this value can be ignored.Quote:
Originally Posted by astaris
I'll see that this gets fixed in a future release.
Thanks.
thanks dude for this URL(The Way It Works)..
I'll send a feedback after trying it :)
Good, the new version isn't "crashing" after 1min. The system is in the sig. Previously it would close by itself after a while. Quite annoying.
0.93 that is.
Grats', mate!
One request (sorry if it has already been discussed) - could you please make the sys tray show temp instead of, errr .... nothing at all. 10x, keep up the good work!
Isn't there some way to get this to read the minus temps that clearly show in the asus bioses?
Yes, it has been requested dozens of times :p:, and I'm working on it.Quote:
Originally Posted by Mord
Well... Core Temp shows what is read from the DTS, its a known issue with Intel CPUs for whatever reason, they just don't go lower than a certain temp, but never goes near 0C or under.Quote:
Originally Posted by xgman
I've got no solution, I'm guessing that this is the DTS limitation.
My article which discusses motherboard temperature sensors and DTS has been posted just now here:
http://www.overclockers.com/articles1378
Quote:
Originally Posted by The Coolest
The latest Everest release reads minus cpu tems and so does MBM5 if you set the MB to the ASUS P4800. I don't know where they are reading the temps from, but they do see minus temps in Windows somehow. How can we see where MBM5 is reading this from? If you could duplicate that somehow maybe it would work?
If Core Temp doesn't read negative temps, no other program I know of will read negatives using the DTS either.
MBM5 and Everest are probably using the standard sensors.
hmmm . . so could core temp have an option to read those as well for the minus crowd at least? Where is the asus bios reading the minus temps from? Funny how even the asus utiliies can't see the minus temps that their own bioses shows clearly.Quote:
Originally Posted by The Coolest
If you read my article, you'll notice that's Core Temp does not read, nor has access to any Super I/O chip. So, sorry, I'm afraid I can't
is there anyway that this will tell the true volts of the cpu not the ones that are meant for it?
Same as above :-\
I updated the first post with the article link and a "Digg it" link so to increase Core Temp popularity. so please click on it and digg it:
http://digg.com/hardware/Dual_Core_C...itor_EACH_core
I would also. Those are crazy temps for a Yonah. Those are about the same temps I'm seeing on my 6600 running at 3.6 on Air.Quote:
Originally Posted by ex.treme
Yes, it freaked me out the first day I got my P5B Deluxe, but since then I have learnt it is just calibrated closer to Core Temp than other boards..Quote:
Originally Posted by The Coolest
Notes on P5B vs P5W-DH temps - why only Core Temp matters
Thanks for your program and hard work...
Has anyone figured out why TAT's temps are identical to coretemps at idle (on Asus P5wdh) but at load, Core temp is up to 4c higher (In dual prime, at least; in most 'real world' programs, its usually just 1-2c higher than TAT)?
If both coretemp and TAT are measuring the exact same diode, why does coretemp report up to 4C higher, but only at max load ?
This was asked a few pages back and then seemed to fall to the wayside...
I still have no clue why that happens.
Duggin' :p:Quote:
Originally Posted by The Coolest
I have a question having just moved to w/c... my temps as per coretemp have decreased significantly. I am much more comfortable with my temperatures now, stable where I have not been before, alot more headroom etc.
However - although coretemp's core 1 + 2 max out at 50c (heaviest load, opteron 165 ccbwe 0551 upmw.... I know :stick: ) my motherboard (epox 9NPA+SLI)'s "CPU" temp reading has started reaching 67c, whereas previously it has stuck fairly close to the temps given by coretemp, often slightly below.
I was wondering whether this would be because the lesser airflow around the mobo sensor with the move to watercooling from air? I mean mobo sensors are like a best guess I suppose, but still...
And most importantly, do you guys think I should be concerned about this?
Thanks :D
1) spot on!Quote:
Originally Posted by WigSgiW
2) if there is no airflow around the PWM area , you might be getting instabilities due to overheating of that area and fluctuation voltage levels
Good point... dammit I was trying to move away from fans! :rolleyes:
I'll try and fashion a method of getting a fan to send some air in that direction... This board already has 0.1v vdroop from idle to two cores fully loaded (measured with DMM from measuring points on motherboard)... so I don't really want to increase that sort of thing to be honest ;) Gotta get that vdroop mod done dammit... just such tiny legs to solder to!
Also, really nice to see 49c 35W removed from core temp as tcasemax... makes me feel much more comfortable about things... still want to keep it sub 50c... so pointless question really but... do you think the absolute, ball-breaking (stress testing) max should be 65c as per AMDs docs? Or am I talkin out my ass? (yes, I have a small horse out the back :D)
Build is:
M2N32-SLI Deluxe
AMD 4200+ X2 Windsor
EVGA 7900 GT
2 GB Corsair XMS 6400
OCZ 600W PSU
For a while I was using v0.91 and didnt really put the 2 together, but occasionally my computer would just shut off, usually when alt tabbing between apps. I could run prime95 all night, folding@home all day, rthdribl etc. with heat not being the problem.
So a few days ago I tried v0.93 and within 20 seconds computer just turns off. So I tried using it again. Only this time the program wont run for more then 5 seconds and gives some weird error message(sorry I didnt write it down). Now since then it seems to run ok except fot the fact that when I do anyting requiring cpu load and having core temp running computer just turns off. This is the only program that causes this.
Oh and one night I left core temp on. In the morning my temps were 50C on both cores and core temp was using alot of cpu time in task manager.
hmmm.
Well first of all this is in the wrong thread :p
Try to disable temperature logging in settings, see if that does anything.
What is the verdict here? What is the Tj(max) for a E6600? I can find it for other processors, but not the 4 MB cache Core 2 Duo line.
This...
http://www.benchmarkhq.ru/english.html
... indicates you can read Tj(max) out of the processor. Anybody knows how that works?
But I cannot find anything about that in manual 3 for the Core2Duo's. They say look it up in the processor spec, but the processor spec only gives a table (table 28, figure 18), ending at 65 watts and 60.1 C. (31327802.pdf)
It's labeled "Tc". Is that supposed to be Tj(max) for the E6600?
If so, did they really give up on even reporting temps above 60 C?
P.S. I have Linux code to read the digital temperature sensor. It works, I am just not sure what the reference point is supposed to be.
in "chip" jargon, Tc is T case...
TCaseMax is different from Tjunction.Quote:
Originally Posted by uOpt
TCaseMax refers to the maximum temperature between the CPU (core or IHS, depending on config) and the heatsink.
TjunctionMax refers to the maximum temperature between the CPU core and the PCB of the CPU.
That is why its a much higher number than the TCaseMax.
Hi.
I have a Problem with coretemp:
My CPU: Allendale e6300 ES Step4 B0.
(Overclocked @ 2,8GHz @ default Vcore)
My Board: Asus P5W DH Deluxe SN66...
I am using a very good watercooling setup and coretemp reports 75 degrees @ idle and 87 degrees @ load.
Also the "Tjunction" is 100 degrees and not 85 degrees.
Because I thought my cooler fitted not perfect I reinstalled it 3 times but the temps don't change at all.
So is this a bug or is the temperature really that high?? :confused:
Thx 4 help!
What does TAT report?
i have a slightly different problem , TAT and CT give me the same numbers BUT the cores have a Temp difference in idle and load modes , have a look
http://img147.imageshack.us/img147/7...titled4ba4.jpg
any ideas are welcomed :D
Same temps as CoreTemp, 75@idle and 86@loadQuote:
Originally Posted by The Coolest
CoreTemp doesn't work in Vista =/
Some would say it could be a problem with the mount or a thermal compound destribution issue.Quote:
Originally Posted by ea6gka
Well, its a B0, their readings are usually messed up and there's not much I can do about it. just try to substract the 15C off the temps that you get.Quote:
Originally Posted by HugeXL
[ur=http://www.thecoolest.zerobrains.com/forums/viewtopic.php?t=171]Read this[/url] about how to get it to work in Vista.Quote:
Originally Posted by saboya
The link didn't work for me, but I tried to run with admin rights and it worked (doh) :)Quote:
Originally Posted by The Coolest
under or over the HeatSpreader?? :D . cause i have remounted my heatsink and i still get the same teps. damn it:mad:Quote:
Originally Posted by The Coolest
Its the same with mine:confused: . But when both cores are loaded the gap closes. At first I thought I did a bad lapp job...this wasn't the case.Quote:
Originally Posted by ea6gka
Version 0.94 - 16th November, 2006
- Fix: Temperature log file not created in EXE folder.
- Fix: Temperature log file left "Processor" field blank.
- Add: An actual icon for the EXE/System tray.
- Add: An option to display temperature in Fahrenheit.
- Add: An option to start Core Temp minimized.
- Change: Only list physical processors in CPU selection.
- Change: Improve Intel CPU detection.
- Remove: GCPUID button.
Older versions
I know that many of you have expected to see a system tray icon that would display CPU temps.
I'm just way too busy to get any work done lately, I started some course so its work then come home and do homework.
Next release should have an OSD display if I can get it working quickly enough.
Hope you like the update, Intel users will get better CPU recognition and Revision detection.
Your work is very much appreciated. Cheers! :toast:
-phil
P.S. I dig the new icon.
Great!! !thank you!!
[QUOTE=The Coolest]Some would say it could be a problem with the mount or a thermal compound destribution issue.
QUOTE]
i) And they would probably be wrong.......!!!
When idling the cpu is not under 0% load...at least core 0 isn't.
ii) In the same way that 2 cpus tested in the same rig will vary in core temp. so 2 cores on the same chip may also.
iii) it could be the thermal compound distibution, but this is very very unlikely to say the least
Where can you download the new version?
Same place, same link. In the first post
Retarded question, but I have to ask. Is "Tjunction" the max temperature that the CPU the program is running on is meant to handle?
I'm not sure what the limits are on the core 2 duo so far. It's even more confusing when software temp sensors using the motherboard give me 33c for the CPU and this program gives me 46c at the same level of load.
Tjunction is the max temp before the CPU starts throttling I think... Or before the CPU shuts itself down.
Thanks for the update on this. ;)
So this means I have all the way until 85c on the e6400 before I should back down? I'm up to 71c right now at full load - is this pushing it? I'm using aircoling before I receive the rest of my watercooling stuff.
Well, The Coolest just answered these questions a few posts back (#164 actually) :
TCaseMax is different from Tjunction.
TCaseMax refers to the maximum temperature between the CPU (core or IHS, depending on config) and the heatsink.
TjunctionMax refers to the maximum temperature between the CPU core and the PCB of the CPU.
That is why its a much higher number than the TCaseMax.
Can you upload the program on rapidshare please?
I can't access your site .
I havent read the entire thread, but has anyone tested the latest Core Temp with a Kentsfield? All temperature readings seems really low (I am overclocked at 3.2ghz and all 4 cores are under full load with temsp under 38???)
Omar, go to overclockers.ru
ea6gka , i had the same problem with my setup (opteron 165) but...Quote:
Originally Posted by ea6gka
i had this problem with my hyper6 hs only, when i check for a while its stock hs i saw -of course- higher tems but the diversion between the cores' temps was minimal 1-2 C at max) , so.. i assume it has to do with the increased weight of my hs , which "presses" the bottom part of the cpu (suppose there is the core0 ?:P)
Exellent Program. I used it to figure out why my x2 would not OC. core 1 was 7c hotter than core 0 that was pushing core 1 to around 52-55c under load at even a mild OC.
I took the lid off, and the temp plumeted but core1 is still aound 5c hotter at all times than core0. I'm using a DD TDX and AS5 (very thin layer). Anyone else have an amd dual core with a temp spread of 5 to 7c at all times??
Just scan through this thread you'll notice that you're not alone.
Guys i have question , is that possible?
I am using QX6700 & Tuniq Tower?
http://img84.imageshack.us/img84/8876/tempdf9.jpg
There's a bug in CoreTemp 0.94. For me, with E6600ES at stock speeds, it shows frequency as 2400MHz, which is correct, but it's displayed as 6x400MHz, which is wrong, as I have my CPU at default FSB and multi.
.93 does it too but i think all of them do itQuote:
Originally Posted by Alexandrus
wow... used it for the first time today... and i must be doing something right... cause the cpu core temp was way lower than i expected!
another mirror for CoreTemp v0.94 :
http://www.persianpcshop.com/CoreTemp94.zip
That's not a bug. Disable EIST and then run Core Temp again and that should correct the core clock display problem.Quote:
Originally Posted by Alexandrus
They fixed the tray icon. Only if it allowed to show the temp in tray...
that would be lovely :DQuote:
Originally Posted by JoeBar
Thanks, finally a version that not give me a "application must be closed" error when CPU is loaded :D
is it accurate for QX6700?? i mean i am on water and it gives me 60C idle.....