TheRapture
Supreme [H]ardness
- Joined
- May 31, 2000
- Messages
- 6,885
Damn straight bro, let us know.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
C2D temps revisited....
- Thread starter TheRapture
- Start date
Damn straight bro, let us know.
CpuMan: Is there any way you can get a more accurate room temperature reading, preferably closer to the fan on your cpu? I know in my room, the temperature displayed on the digital thermostat can be a degree or two different from the air entering my computer at the floor level.
The processors with an actual Tjunction of 100C that CoreTemp 0.94 was using a Tjunction of 85C to calculate the core temperature were generally about 10C below room temperature during this test making it very obvious that CoreTemp 0.94 was very wrong.
With your data I tend to believe CoreTemp 0.94 and SpeedFan 4.32 which both use a Tjunction of 85C as the one you should be believing.
By the way, I suggest you close your computer case back up.
If one of your fans that is running at 1,350,000 rpm ever gets loose it's going to take someone's head off!
I'll have to see if I can find a thermometer. I have a outside thermometer but not sure I trust it?
The pc is at the same level as my monitor in the bedroom and the thermometer I used was the thermostat in the livingroom, and it read 68 degrees Fahrenheit, which would be 20c. I said 18c to 20c because the windows in the bedroom were open, but in the living room they were closed.
Oh yea, the cpu fan is one of those super ultra duty rpm fans, LOL.
cpuman....nice and cold in your room eh? Man if my room was that chilly.....yikes. 25c to 27c here, but then again, it's in south texas, and already the days are getting to the high 80f range....summer here sees 98f-100f alot of the time and keeping a house any cooler than ~75f gets to be expensive and the damn unit runs all the time.
Looks like what unclewebb says is right.
It was fairly cool this morning but in a few weeks I am sure we will probably have to shut the house up and crank up the AC, it gets pretty hot in the summer here to in central Arkansas.
I will let you know but unfortunately it won't be tonight anymore. I had some probs with the original flash of the BIOS and I re-flashed just recently to see if it would cure the probs. I got more testing to do to insure stability before testing the temps. I will hopefully get back to ya with results tomorrow. BTW, I grabbed my indoor/outdoor digital thermometer and stuck the outdoor probe into my case and got the side shut. I got the probe hanging right in front of the cpu fan. I'll prolly post up a pic along with my findings.
CpuMan: Your reported CoreTemp 0.94 value of 18C might be below your ambient temperature so I decided to do some more testing to see if I could come up with a reasonable explanation for what you're seeing. I think I have one.
First of all, the heat produced by a processor is proportional to the MHz you run it at and it is also proportional to voltage squared. That's why adding voltage to a processor jacks up the heat in a hurry but reducing voltage does the opposite and removes a lot of heat.
You ran your test with less voltage than us so I decided to do the same and ran my second test with CPUz reporting the core at 1.104 volts. I also had no problem running my E6400 at a FSB of 200 MHz X 6 for a total of ~1200 MHz.
My nothing special, $20 Radio Shack digital thermometer reported an ambient temperature of 18C when placed beside the open case and this was my result:
CoreTemp & SpeedFan reported 17C or 1C below my ambient temperature. In theory it's not possible for an air cooled cpu to end up cooler than the ambient temperature but people have to keep in mind that both my digital thermometer and the DTS value from the Core2Duo are integer values with an error of +/- 0.5 degrees. If you combine two readings both with an error of +/- 0.5 degrees then the combined error is +/- 1.0 degrees so 17C reported by one device and 18C by the other is within the margin of error and should be considered equal.
In both cases we don't know how these temperatures are rounded off. If the real ambient temperature was say 17.8C my digital thermometer might round that up to 18C whereas a C2D running at that same temperature might truncate off the decimal part and report that as 17C. It's also impossible to say how accurate my thermometer really is.
This new testing has convinced me that any Core2Duo processor running at a minimum speed of 200 MHz X 6 = 1200 MHz and 1.1 volts should see idle temperatures approach if not equal the room temperature give or take a degree or two. This method is a perfect way to calibrate one's processor and will prove what version of CoreTemp is telling the truth.
It's possible that some E4300 cpus have a Tjunction of 85C and some have a Tjunction of 100C. Intel has issued no documented way to tell the difference between the two for desktop processors. Without a known Tjunction you can't calculate an absolute core temperature value and any presently available monitoring software can only guess by picking one Tjunction or the other.
Use the above test and the guessing will finally be over.
First of all, the heat produced by a processor is proportional to the MHz you run it at and it is also proportional to voltage squared. That's why adding voltage to a processor jacks up the heat in a hurry but reducing voltage does the opposite and removes a lot of heat.
You ran your test with less voltage than us so I decided to do the same and ran my second test with CPUz reporting the core at 1.104 volts. I also had no problem running my E6400 at a FSB of 200 MHz X 6 for a total of ~1200 MHz.
My nothing special, $20 Radio Shack digital thermometer reported an ambient temperature of 18C when placed beside the open case and this was my result:
CoreTemp & SpeedFan reported 17C or 1C below my ambient temperature. In theory it's not possible for an air cooled cpu to end up cooler than the ambient temperature but people have to keep in mind that both my digital thermometer and the DTS value from the Core2Duo are integer values with an error of +/- 0.5 degrees. If you combine two readings both with an error of +/- 0.5 degrees then the combined error is +/- 1.0 degrees so 17C reported by one device and 18C by the other is within the margin of error and should be considered equal.
In both cases we don't know how these temperatures are rounded off. If the real ambient temperature was say 17.8C my digital thermometer might round that up to 18C whereas a C2D running at that same temperature might truncate off the decimal part and report that as 17C. It's also impossible to say how accurate my thermometer really is.
This new testing has convinced me that any Core2Duo processor running at a minimum speed of 200 MHz X 6 = 1200 MHz and 1.1 volts should see idle temperatures approach if not equal the room temperature give or take a degree or two. This method is a perfect way to calibrate one's processor and will prove what version of CoreTemp is telling the truth.
It's possible that some E4300 cpus have a Tjunction of 85C and some have a Tjunction of 100C. Intel has issued no documented way to tell the difference between the two for desktop processors. Without a known Tjunction you can't calculate an absolute core temperature value and any presently available monitoring software can only guess by picking one Tjunction or the other.
Use the above test and the guessing will finally be over.
CpuMan: Your reported CoreTemp 0.94 value of 18C might be below your ambient temperature so I decided to do some more testing to see if I could come up with a reasonable explanation for what you're seeing. I think I have one.
First of all, the heat produced by a processor is proportional to the MHz you run it at and it is also proportional to voltage squared. That's why adding voltage to a processor jacks up the heat in a hurry but reducing voltage does the opposite and removes a lot of heat.
You ran your test with less voltage than us so I decided to do the same and ran my second test with CPUz reporting the core at 1.104 volts. I also had no problem running my E6400 at a FSB of 200 MHz X 6 for a total of ~1200 MHz.
My nothing special, $20 Radio Shack digital thermometer reported an ambient temperature of 18C when placed beside the open case and this was my result:
CoreTemp & SpeedFan reported 17C or 1C below my ambient temperature. In theory it's not possible for an air cooled cpu to end up cooler than the ambient temperature but people have to keep in mind that both my digital thermometer and the DTS value from the Core2Duo are integer values with an error of +/- 0.5 degrees. If you combine two readings both with an error of +/- 0.5 degrees then the combined error is +/- 1.0 degrees so 17C reported by one device and 18C by the other is within the margin of error and should be considered equal.
In both cases we don't know how these temperatures are rounded off. If the real ambient temperature was say 17.8C my digital thermometer might round that up to 18C whereas a C2D running at that same temperature might truncate off the decimal part and report that as 17C. It's also impossible to say how accurate my thermometer really is.
This new testing has convinced me that any Core2Duo processor running at a minimum speed of 200 MHz X 6 = 1200 MHz and 1.1 volts should see idle temperatures approach if not equal the room temperature give or take a degree or two. This method is a perfect way to calibrate one's processor and will prove what version of CoreTemp is telling the truth.
It's possible that some E4300 cpus have a Tjunction of 85C and some have a Tjunction of 100C. Intel has issued no documented way to tell the difference between the two for desktop processors. Without a known Tjunction you can't calculate an absolute core temperature value and any presently available monitoring software can only guess by picking one Tjunction or the other.
Use the above test and the guessing will finally be over.
I would say your assessment sounds reasonable. Thanks for all the work you have put into this.
Hopefully Intel will decide to document the Tjunction so we could determine for certain the temps of these cpu's. But in the mean time, doing what you suggest should give us a fairly accurate idea.
whatisthetjunction
n00b
- Joined
- Apr 24, 2007
- Messages
- 2
I've got an E6400 L2 and I was having the same problem which temperature to believe.With 300 FSB and Vcore at 1.175 = 2.4 Ghz i got the following temp:
Idle
Core Temp 0.94 25°C
Core Temp 0.95 40°C
TAT 42°C
Speedfan 25°C
PC Wizard 2007 25°C
Everest 26°C
Load
Core Temp 0.94 39°C
Core Temp 0.95 52°C
TAT 52°C
Speedfan 37°C
PC Wizard 2007 37°C
Everest 37°C
I searched Internet until I found this Thread and decided to run your test.
With a FSB of 200, a multiplier of 6 and a vcore of 1.104 i got the temps you can see on the screenshot. Important: Room Temperature was 22.4°C
So I think Coretemp 0.95 and TAT are wrong!
EDIT: Forgot to say that the CPU Cooler is a Noctua U12F (Excellent Tower) which is supported by four 120mm Fans.
Idle
Core Temp 0.94 25°C
Core Temp 0.95 40°C
TAT 42°C
Speedfan 25°C
PC Wizard 2007 25°C
Everest 26°C
Load
Core Temp 0.94 39°C
Core Temp 0.95 52°C
TAT 52°C
Speedfan 37°C
PC Wizard 2007 37°C
Everest 37°C
I searched Internet until I found this Thread and decided to run your test.
With a FSB of 200, a multiplier of 6 and a vcore of 1.104 i got the temps you can see on the screenshot. Important: Room Temperature was 22.4°C
So I think Coretemp 0.95 and TAT are wrong!
EDIT: Forgot to say that the CPU Cooler is a Noctua U12F (Excellent Tower) which is supported by four 120mm Fans.
Nice name.
You have a much more powerful cpu heatsink & fan(s) than the OEM one I'm using yet your results at idle when running low volts and 1200 MHz is exactly the same.
Core temperature and ambient temperature are equal proving that your Tjunction is 85C and not 100C.
Now we just need some people with an E4300 to run this test. Does it really have a Tjunction of 100C or is it 85C too?
There seems to be some of each out there so run the test and see if you can determine what you've got.
If you have an E4300, try to post the date code from your processor box as well if you have it. ie. Q640...Q644
whatisthetjunction
n00b
- Joined
- Apr 24, 2007
- Messages
- 2
Thanks. I registered just for this Thread so I took a suitable nickNice name.
But whats exactly the problem? Is is just Coretemp 0.95 which failes to compare the Tjunction for E6300 / E6400 L2 revision with its database or is it the fact, that nobody really knows how to determine the right Tjunction?
EDIT: I another Board i found this one: E4300 Allenale L2 with 85°C Tjunction
http://img177.imageshack.us/img177/2545/e4300fp3.jpg
The problem is that Intel has no documented way to determine Tjunction from a Core2 Duo desktop processor so software developers can only guess.
CoreTemp 0.94 used to guess at 85C and seemed to be right until the E4300 was released. Some of those seem to have a Tjunction of 100C so CoreTemp 0.94 which uses Tjuction = 85C, will report the absolute temperature wrong.
The programmer then decided to set Tjunction to 100C for all Revision L2 E4300 /E6300 / E6400 processors but users soon discovered that not all of the new E6300/E6400 processors have a Tjunction of 100C. I'm not even sure if any of them actually do so CoreTemp 0.95 won't report the correct absolute temperature for many of the L2 processors. That includes the E4300 because some of them seem to have a Tjunction of 85C while some are 100C.
I hope this sums up this confusing mess!
CoreTemp 0.94 used to guess at 85C and seemed to be right until the E4300 was released. Some of those seem to have a Tjunction of 100C so CoreTemp 0.94 which uses Tjuction = 85C, will report the absolute temperature wrong.
The programmer then decided to set Tjunction to 100C for all Revision L2 E4300 /E6300 / E6400 processors but users soon discovered that not all of the new E6300/E6400 processors have a Tjunction of 100C. I'm not even sure if any of them actually do so CoreTemp 0.95 won't report the correct absolute temperature for many of the L2 processors. That includes the E4300 because some of them seem to have a Tjunction of 85C while some are 100C.
I hope this sums up this confusing mess!
Hi All. I could not replicate your test as my E6400 L2 Rev. would not operate at 1200Mhz. I can tell you this though. Currently I am overclocked to 2950 Mhz @ 1.325 Vcore, Ambient temp is 24.5C and Coretemp Beta 0.94 has me at 20C. I think that's supposed to be impossible. Coretemp 95 - 36C and TAT is 38C. All temps are hottest core. I have an aftermarket cooler. It was much hotter with the stock HSF. Also, my case has great airflow. Nvidia Ntune showed my temp at 28C but I don't trust it and removed it for messing up other applications.
You guys have done a lot of research and it is greatly appreciated, but I'm not sure anyone will ever know why these L2s are so hot, with less vcore than B2s take.
You guys have done a lot of research and it is greatly appreciated, but I'm not sure anyone will ever know why these L2s are so hot, with less vcore than B2s take.
My test allows a user to prove if a software program is out to lunch and your testing proved that CoreTemp 0.94 which uses a Tjunction of 85C for your processor is not right. CoreTemp 0.95 uses a Tjunction of 100C and that seems to work correctly with your L2 E6400.
All of the B2 E6400 cpus that I've seen have a Tjunction of 85C. The L2 E6400 seems to come in both 85C and 100C Tjunction models. There is no software available to determine what a processor really is so temperature monitoring programs can only guess.
The difference in Tjunction does not mean that one processor runs hotter than the other. It's only a rating of how hot a processor can run before automatically throttling itself. A 100C Tjunction processor is able to run 15C hotter before it gets throttled back. Both processors in the same case, being fed the same voltage at the same MHz with the same heatsink and fan are going to run at pretty much the exact same temperature.
Tangdalun: I set my computer up to similar specs as yours. 1.328 volts at idle and 2950 MHz.
With a room temperature of 20C, my B2 E6400 idles at 31C.
Add on 5C to my temps to correct for the difference in our room temps and my B2 idles at the same temp as your L2.
That's interesting. I've been disappointed because I read all these articles about E6400s and evga 680i sli mobo and tuniq tower 120 and I expected my setup to more or less mirror those, but then I received the L2 instead of the B2, which I've gotten over already, but then was wondering why I wasn't seeing idle temps of 27C like all the tests indicated. I was sure I didn't seat the cooler properly or used too much thermal grease. I guess I'm just wrong all around
On another note...aren't you afraid to run 1.45 vcore? Intel's max for your processor is 1.35, right?
On another note...aren't you afraid to run 1.45 vcore? Intel's max for your processor is 1.35, right?
The Intel specs for the C2D states a maximum core voltage of 1.55 volts.
You won't gain too many extra MHz by going this high but you also won't hurt anything either.
This number comes from:
Intel® Core™2 Extreme Processor
X6800Δ and Intel® Core™2 Duo
Desktop Processor E6000 and
E4000Δ Sequences
Section 2.6.1 Absolute Maximum and Minimum Ratings
Table 4
ftp://download.intel.com/design/processor/datashts/31327804.pdf
Not many people with a room temp of 25C are seeing idle temps of 27C so don't be too sad.
You won't gain too many extra MHz by going this high but you also won't hurt anything either.
This number comes from:
Intel® Core™2 Extreme Processor
X6800Δ and Intel® Core™2 Duo
Desktop Processor E6000 and
E4000Δ Sequences
Section 2.6.1 Absolute Maximum and Minimum Ratings
Table 4
ftp://download.intel.com/design/processor/datashts/31327804.pdf
Not many people with a room temp of 25C are seeing idle temps of 27C so don't be too sad.
unclewebb: I noticed your temps were measured with speedfan so I downloaded it and here is what I got.
well i can't figure how to get my image to show up so I'll just have to tell you. Speedfan shows my hottest core at 25c, that's same as ambient running the same specs as above: 2950mhz, 1.325 vcore. core 0 is at 24c. So I guess for me, TAT and coretemp 95 are the way to go?
well i can't figure how to get my image to show up so I'll just have to tell you. Speedfan shows my hottest core at 25c, that's same as ambient running the same specs as above: 2950mhz, 1.325 vcore. core 0 is at 24c. So I guess for me, TAT and coretemp 95 are the way to go?
SuperKeijo
Gawd
- Joined
- Jan 18, 2007
- Messages
- 705
I guess coretemp 0.95 was right after all. From the middle of the ninja heatsink with an IR thermometer this is what I got with an ambient temperature of 18.6 just outside the case.
One thing I forgot to ask unclewebb is where does this 3-5c difference between idle temps and ambient temp come from? Just what tests seemed to show?
Now there seems to be a ~10 degree difference between the heatsink temp and core temps.
Ambient: 23,4:
Taking the ambient reading much closer to the case, seems like coretemp 0.94 does show temps below ambient. And much below the actual heatsink temp.
One thing I forgot to ask unclewebb is where does this 3-5c difference between idle temps and ambient temp come from? Just what tests seemed to show?
Now there seems to be a ~10 degree difference between the heatsink temp and core temps.
Ambient: 23,4:
Taking the ambient reading much closer to the case, seems like coretemp 0.94 does show temps below ambient. And much below the actual heatsink temp.
For my E6400 Revision B2, CoreTemp 0.94, CoreTemp 0.95 and SpeedFan v.4.32 all use a Tjunction of 85C which is what my processor is so it doesn't matter what program I use to report temps. All 3 of these report the exact same thing.
For some of the revision L2 processors that have a Tjunction of 100C then you need to use CoreTemp 0.95.
SpeedFan 4.32 will read approximately 15C lower which is wrong for these processors.
With all the crazy people out there running the hell out of their Core2Duos have you ever heard of even one of them frying their cpu? I haven't. These things are incredibly reliable and if they ever do fail, they have a 3 year warranty so no worries. After 3 years you'll probably be able to get an old C2D in a box of cereal or maybe it might cost you $20 in a used computer store. If it makes you feel any better my original Pentium III - 450 is still happily running at 600 MHz many years later. People used to freak out back then that a crazy overclock like that would "shorten the life of the processor." Call me in 10 years and I'll let you know if it only lasted 20 years instead of 30.
SuperKeijo: The testing you are doing now is at 3400 MHz and likely with a healthy amount of core voltage. Drop back down to those underclock settings I recommended and I'll see if I can make some sense of your data.
I wish I was there with your temp tool so I could learn a thing or two!
I think the results you came up with in post #67 on the previous page are accurate.
I wish I was there with your temp tool so I could learn a thing or two!
I think the results you came up with in post #67 on the previous page are accurate.
SuperKeijo
Gawd
- Joined
- Jan 18, 2007
- Messages
- 705
Yeah well sorry for not explaining the frequency thingy. I was really at 8 x 378 = 3024MHz using stock voltage 1,325. e4300 has a multiplier of 9 and when dropped to 8, programs think it is still at 9. So in this case 9 x 378 = 3402. Why it still reports 8 x 425 I really don't know, but fact is I can't even post at anything above 400, so...
I could do the same test with lower frequencies but I think the outcome would be pretty much the same.
I could do the same test with lower frequencies but I think the outcome would be pretty much the same.
SuperKeijo: I think part of what you're seeing is the temperature of your heatsink changes temperatures at a much slower rate, up and down, compared to the core of your processor. The core can jump up 10C degrees and go back down to its idle temp but the temperature of your big heatsink is not going to instantly change by that amount.
Try doing the low volt experiment again and give it some time for the temperature of everything to stabilize.
Was you case open or closed during your previous test?
Try doing the low volt experiment again and give it some time for the temperature of everything to stabilize.
Was you case open or closed during your previous test?
This is my first post here but I have been reading this thread for a bit because I was having problems getting the correct temp off of my cpu..it seems to me that the newer e4300 cpu's are 85c the older ones are 100.intel kind of messed that one up.I am guessing that's why they won't release that info to the public because it keeps changing.My date code is Q644 if that helps anyone.
idle
Coretemp 95=51c
coretemp 94=33c
speedfan=33c
load
coretemp 95=56c
coretemp 94=37c
speedfan=37c
the room temp is around 25c or so I am watercooled but it's not a perfect setup so I pretty sure coretemp95 is wrong atleast for me anyway.
idle
Coretemp 95=51c
coretemp 94=33c
speedfan=33c
load
coretemp 95=56c
coretemp 94=37c
speedfan=37c
the room temp is around 25c or so I am watercooled but it's not a perfect setup so I pretty sure coretemp95 is wrong atleast for me anyway.
Thanks for your numbers bill941.
That's what I found also. The later Q644 E4300 has an 85C Tjuntion and the early Q640 has a 100C Tjunction. I'm not sure if Intel made both varieties during the same week or switched over sometime between these two dates.
Without any official Intel documentation, all you can do is try to run a test and see if the temp numbers make any sense.
That's what I found also. The later Q644 E4300 has an 85C Tjuntion and the early Q640 has a 100C Tjunction. I'm not sure if Intel made both varieties during the same week or switched over sometime between these two dates.
Without any official Intel documentation, all you can do is try to run a test and see if the temp numbers make any sense.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
Sorry to dig up an old thread, but I'm working through the endeavor myself right now with my E4300 and am slowly coming to a conclusion.
I have a revision L2 E4300 with a date of Q702.
I haven't had time to do any extensive testing as my fiance is using the computer and I just came upon this thread...but is it a general consensus that E4300's later then Q644 have are 85C chips?
I know that I'm pushing things at 3.2GHz and 1.312v with stock cooling, but 73C+ load temps are a bit much to stomach, and this would at least clear things up.
I only ask because I haven't seen much mention of E4300's around mine with dates of Q702 or later, should it be a 85C "version" or is there any general agreement on that based on CPU dates?
EDIT: Sorry for posting in both threads, as I just noticed that you also replied in the thread I had posted...I'm fine if the discussion bridges between the two threads, as once I figure out once and for all what's really going on I'd like nothing less then for the information to be usefull for others as well.
I have a revision L2 E4300 with a date of Q702.
I haven't had time to do any extensive testing as my fiance is using the computer and I just came upon this thread...but is it a general consensus that E4300's later then Q644 have are 85C chips?
I know that I'm pushing things at 3.2GHz and 1.312v with stock cooling, but 73C+ load temps are a bit much to stomach, and this would at least clear things up.
I only ask because I haven't seen much mention of E4300's around mine with dates of Q702 or later, should it be a 85C "version" or is there any general agreement on that based on CPU dates?
EDIT: Sorry for posting in both threads, as I just noticed that you also replied in the thread I had posted...I'm fine if the discussion bridges between the two threads, as once I figure out once and for all what's really going on I'd like nothing less then for the information to be usefull for others as well.
I don't think there will ever be any consensus when it comes to the core temperature of an E4300.
The only ones that were proven to have a TjMax of 100C were the original ones that came out with a date code of Q640. CoreTemp 0.95 changed the TjMax to 100C for the Allendale processors and then it seems that Intel changed back to TjMax of 85C.
Run my test when you get a chance. You may not convince anyone but I'm pretty sure you'll be able to finally convince yourself.
The only ones that were proven to have a TjMax of 100C were the original ones that came out with a date code of Q640. CoreTemp 0.95 changed the TjMax to 100C for the Allendale processors and then it seems that Intel changed back to TjMax of 85C.
Run my test when you get a chance. You may not convince anyone but I'm pretty sure you'll be able to finally convince yourself.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
That's what's most important anyways...I'll let you know as soon as possible.
Any way to draw any kind of conclusion from temperatures at 3.2GHz and 1.350v in BIOS? I can doublecheck idle temps compared to room temps until I have time to do the voltage/clock testing.
If you run Orthos at 3200 MHz and post your core voltage and your room temperature then I can do the same with my computer and I can give you a temp number to compare to.
I'm still using the Intel heatsink and fan on my E6400 so we should have some similar numbers.
My main computer is busy at the moment so I'll check back tomorrow morning and let you know my temps.
I'm still using the Intel heatsink and fan on my E6400 so we should have some similar numbers.
My main computer is busy at the moment so I'll check back tomorrow morning and let you know my temps.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
Well...because I don't have time to do the full low clock/voltage testing that you guys were describing...but I closed F@H and my idle temps are 28/31 for the two cores using Core Temp 0.94...and considering my room temp right now is 20C on the dot, I'd say that 8-11C above ambient with less then stock voltage should be about right. Also, TAT and Core Temp 0.95 report about 46C idle at less then stock voltage with 20C rooms temps...and I'd be much more inclined to believe 31C then 46C on idle in my scenario.
What are some thoughts on this so far? I'll get back with more details on the extra testing tomorrow sometime.
What are some thoughts on this so far? I'll get back with more details on the extra testing tomorrow sometime.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
At 3.2GHz 1.312v under load Core Temp 0.94 reports 57C on both cores, and TAT reports exactly 15C higher at roughly 72C on each core while running Orthos Small FFT's.
Obviously 57C would be much more desirable and give me room to shoot for 3.3GHz on stock cooling, but let me know when you can, I'll try and push for 3.3GHz in the meantime anyways and see what happens.
TheRapture
Supreme [H]ardness
- Joined
- May 31, 2000
- Messages
- 6,885
At 3200mhz at a low vcore of 1.312v...and as long as your heatsink is properly attached......then now way in hell you are running at 72c.
In my warm ass room, at 3200mhz and a whopping 1.39v (true) under load....I never cross 70c under ORTHOS and right now since it is a cool evening, I am only at 65c. Even dual instances of Folding at home run at only ~55c.....which USED to be considered a heavy cpu load.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
And if that's with an E6400 monitored "properly" then I don't think I have anything to worry about.
No worries with the HSF, I've ditched the stock paste and re-seated the cooler and there's NO way I could push any of those 4 pins down any further.
I'll play a bit and see if I can get 3.3GHz stable while I wait for unclewebb to get back with comparative results tomorrow.
In a cool 19C room at 3200 MHz and 1.320 volts I'm steady at a core temperature of 63C today using the OEM heatsink and fan.
Cold temperature testing still seems to be the easiest and consistent way to prove or disprove a temperature monitoring program.
Cold temperature testing still seems to be the easiest and consistent way to prove or disprove a temperature monitoring program.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
Well, my room temps are now a little lower then 19C...but at the same clock/voltage, Core Temp 0.95 reports a full 10C higher then your results, and it would make sense that my Core Temp 0.94 temps may be a few C lower because of lower room temp, it's closer to 17C in here right now.
I'll do the low clock/voltage testing today sometime, but this is enough for now for me to rest a little easier.
Thanks again for all your effort thus far.
I'll do the low clock/voltage testing today sometime, but this is enough for now for me to rest a little easier.
Thanks again for all your effort thus far.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
Awesome news!!! I finished the testing to determine my tJunction temp.
My room temp is 18C at the moment and clocked at 1.2GHz /w 1.1v (actually a little higher) my idle temps are 19C and 23C on Core 1 and 2 respecitvely. There's no way at this low clock/voltage level I'm actually idling at 36C...so I'm glad.
Thank you so much unclewebb for all your help...I'm so relieved to finally have this figured out, and I'm spreading the word about the original thread/testing all over XS and wherever I can.
My room temp is 18C at the moment and clocked at 1.2GHz /w 1.1v (actually a little higher) my idle temps are 19C and 23C on Core 1 and 2 respecitvely. There's no way at this low clock/voltage level I'm actually idling at 36C...so I'm glad.
Thank you so much unclewebb for all your help...I'm so relieved to finally have this figured out, and I'm spreading the word about the original thread/testing all over XS and wherever I can.
cornelious0_0: Great to hear that. I find that simple test really works without stressing your processor whatsoever.
I was over at XS a while ago trying to convince some people to give it a try but I got a pretty cold reception over there. Tough crowd at XS!
I've seen a 101 links back to this thread so thanks for the promotion.
Until Intel comes clean and properly documents the TjMax of their processors, it's about all people can do to try and determine the real absolute temperature of their C2D. With the recent introduction of Conroe revision B2 E6600 cpus that are now showing up with aTjMax of 100C, there could be a few people giving this test a try.
I was over at XS a while ago trying to convince some people to give it a try but I got a pretty cold reception over there. Tough crowd at XS!
I've seen a 101 links back to this thread so thanks for the promotion.
Until Intel comes clean and properly documents the TjMax of their processors, it's about all people can do to try and determine the real absolute temperature of their C2D. With the recent introduction of Conroe revision B2 E6600 cpus that are now showing up with aTjMax of 100C, there could be a few people giving this test a try.
TheRapture
Supreme [H]ardness
- Joined
- May 31, 2000
- Messages
- 6,885
Damn right, this is a GREAT thread on determining correct temps and I have enlightened several other forums, directing to this thread. I have been playing around with voltages and am at stock volts, running at 2800mhz very cool, load temps on the order of 55c with ORTHOS. Going to continue to bump it up to see how high it will go without volt increases.
The more users we can get to contribute the better validation we have, although at this point it is pretty damn solid.
Maybe not sticky material but damn good none the less.
The more users we can get to contribute the better validation we have, although at this point it is pretty damn solid.
Maybe not sticky material but damn good none the less.
cornelious0_0
[H]F Junkie
- Joined
- Apr 6, 2003
- Messages
- 12,783
Technically it's not the voltage that'll "damage" the cpu...but extended exposure to high voltages and uncontrolled temperatures will. As long as your temps are reasonable I wouldn't worry about it. 1.55v or so on air should be fine, and beyond that is more or less "at your own risk".
unclewebb, I don't think I'm gonna bother actually doing any referral/writeup thread on XS, but whenever someone's talking about L2 revision temps I go into a fair bit of detail and link to thread, so hopefully some eye's will be opened.
unclewebb, I don't think I'm gonna bother actually doing any referral/writeup thread on XS, but whenever someone's talking about L2 revision temps I go into a fair bit of detail and link to thread, so hopefully some eye's will be opened.
What is the temp using TAT or SpeedFan? That seems low for that much voltage.