Sign in with
Sign up | Sign in
Your question

Worried that my 17 is running to hot

Last response: in Overclocking
Share
September 16, 2009 3:32:02 PM

Hi, I just recently got the i7 920 and installed it with the Xigatecj Dark Knight heatsink and fan. When running Prime 95 I am getting tempuratures of 70+ Degrees celcius at 3.6GHz. IS this too hot? I just installed it yesterday and read about how the artic silver 5 has a burn in period in which the temps will drop. Should I be worried? Are these temps high enough to require me to re-seat my heatsink? Also I have D0 stepping if that makes a difference.

Thank you for your time and help.

More about : worried running hot

September 16, 2009 3:46:12 PM

70+ IS Pretty hot, i get 57C w/ mo rosewill fort120 and thats at 3.8ghz maybe you should replace it but 70C won't blow, just hot

im sorry, by replace i meant reseat, check the thermal grease, spread it thin w. a spreader or credit card, you want a very thin layer, just enough to cover it all
m
0
l
September 16, 2009 5:21:58 PM

Ok thanks. So I can just put a bit on the processor itself and then spread it using a credit card instead of rotating the heatsink back and forth?
m
0
l
Related resources
September 16, 2009 6:53:07 PM

I'm trying to figure out whats up with your temps. I use a Dark Knight on my i7 and my temps NEVER exceed 52°C @ 3.66GHz on Prime 95. My temps idle in the high 20s and max at around 55°C (got that while playing Supreme Commander with a *** ton of units) My best guess is the following:

Poor airflow in your case. Make sure your airflow is equalized and your case isn't pressurized by too much flowing in and not out or vice versa. My CM Storm Scout has a front fan that is a pull fan, then a top fan wich is a push fan directly on the DK heatsink, then I have a side fan and rear exaust fan along with the DK fan as push fans moving air outside the case. Mess around with practical setups for whatever case you have, make sure that your case is pulling cool air inside, and pushing the hot air outside, this means having pull fans that are away from the heat.

The DK not being seated correctly with a terrible thermal compoud used. I'm guessing you used the stock thermal commpound on ur DK? Bad idea. You need to purchase a good TIM like IC Diamond 7. It outperforms Actic Silver 5 by a few degrees and if you're benching your system to find it's limits, then you need all the edge you can manage.

http://www.performance-pcs.com/catalog/index.php?main_p...

Also, read up on this guide to applying the TIM correctly, I didn't use any cards or anything to spread it. I used a smart application system, and the weight and pressure of the heatsink and i'm getting better than average temps from my DK. Good luck!
http://benchmarkreviews.com/index.php?option=com_conten...

Try using this method:

The Xigmatek crossbow bracket system is great. Make sure you secure the DK down well and there should be NO wiggle at all. NONE!
m
0
l
September 16, 2009 8:20:52 PM

I am using artic silver 5 thermal grease on it so I doubt it is that. Also I didn't use the actual instructions for the gease either, I use the one I found on here that is supposed to be better with the type of heatsink that the DK is. I don't think I can tighten it down any more. Also I am using the anted 300 case. I have two 120mm front fans pulling air in from the front and a 120 mm pushing it out the back as well as a 200mm pushing it out the top. Things are a tight fit so that might be hurting me as well. I think I just put too much thermal grease on it. When I get home I am going to try and reseat it gain (God I hate doing that and I have only done it once. XD) Thank you so much for your help.
m
0
l
September 16, 2009 8:36:57 PM

Absolutely, I forgot to say, acutally yes, too much thermal grease is also a bad thing lol. Yea it toatally does suck reseating that thing. Try the amount of paste used in the picture above. The purpose of TIM is supposed to be to fill the pockets and ridges between the bottom of the heatsink and the cpu in order to create a suface that is more conductive of heat. Good luck to you and let us know how this turns out.
m
0
l
September 16, 2009 11:24:13 PM

well I just reseated it using less paste in than before. I did it the same as the site you linked but it looks like there is little improvement. I was idleing at like 45 and at 100% load I am looking at high 60s to low 70s. It is better than before as long as it stays like this though. I am afraid that this time I used too little paste. I just hope it doesn't get worse. O_O Maybe my case isn't helping. Things are pretty tight in there and my PSU is under my mobo. Opening the side of the case seems to have dropped it a couple degrees. I can live with the 70's as long as that isn't going to shorten it lifespan.
m
0
l
a b à CPUs
a b K Overclocking
September 17, 2009 2:59:33 AM

People are outright lying to you when they say 20C core temps. They are either talking out of their A$$ or referring to CPU temp (tcase) and not core temp (tjunction).

Even still, Tcase can NEVER , NEVER be lower than ambient temps with air cooling. So when someone says he's 20C(68F) CPU Temp, means that his room temperature is around 65F or lower.
m
0
l
September 17, 2009 3:24:41 AM

I know that you can't get below room temp with air cooling. I am just worried that my CPU is getting too hot. I don't wanna shorten it life if I can help it. As far as I can tell I am messing up on the thermal grease, unless these are semi normal core temps, which I highly doubt.
m
0
l
a b à CPUs
a b K Overclocking
September 17, 2009 4:09:55 AM

saphirekosmos said:
I know that you can't get below room temp with air cooling. I am just worried that my CPU is getting too hot. I don't wanna shorten it life if I can help it. As far as I can tell I am messing up on the thermal grease, unless these are semi normal core temps, which I highly doubt.


You have nothing to worry about, If you read reliable sources. mid 70F ambient temps will get you around high 30sC on idle and mid to high 70'sC on load on Air cooling.

In essence your temps are normal for an air cooled i7 920.
m
0
l
a b à CPUs
a b K Overclocking
September 17, 2009 5:35:09 AM

saphirekosmos said:
Hi, I just recently got the i7 920 ... When running Prime 95 I am getting tempuratures of 70+ Degrees celcius ...

saphirekosmos,

Keep in mind that Prime95 Small FFT's is the standard for thermal testing, which is a steady-state 100% workload, unlike Large FFT's or Blend, which are cyclic worloads that yield variable temperatures.

Also, the Thermal Specification shown in Intel's Processor Spec Finder - http://processorfinder.intel.com/Default.aspx - is Tcase Max (CPU temperature) NOT Tjunction Max (Core temperature), which is an all too common misconception among most enthusiasts.

Since a 5c gradient exists between the CPU sensor and the Core sensors, which is shown in the following Intel document - http://arxiv.org/ftp/arxiv/papers/0709/0709.1861.pdf - simply add 5c to the value shown in the Spec Finder to get the Core temperature which corresponds to Intel's maximum CPU temperature.

The other specification, Tjunction Max, which is 100c, applies to overtemp conditions such as Throttle and Shutdown protection. As such, any Core temperatures on the i7 920, regardless of Stepping, which exceed 73c should be considered "overtemp".

All Core i7 9xx variants:

Vcore Max 1.375v
Tcase Max (CPU temp) 68c
Tjunction (Core temp) 73c

Comp :sol: 
m
0
l
September 17, 2009 5:43:54 AM

Thank you all for the replies. I think I can rest a bit easier knowing that it is unlikley I'm going to wake up with a fried 920. XD Thanks again for all the advice and held. I'll see if the temps go down a bit after the "burn in" period. Probably won't mess with the heatsink anymore unless i start getting low 80s on 100% load, which I highly doubt I'll ever reach outside of rime 95.
m
0
l
a b à CPUs
September 17, 2009 7:07:17 AM

CompuTronix: SmallFFT may be the "standard", but it is not representative of a maximum load case on an i7. LargeFFT is a better choice, since it also stresses the onboard memory controller, testing it for stability and resulting in an overall higher heat loading on an i7 based CPU.
m
0
l
September 17, 2009 7:34:11 AM

When using Arctic Silver5. it takes around 200 hours below it cures and temps stabalize. Your temps are within limit, just keep it under 90c when running Prime 95.
m
0
l
a b à CPUs
a b K Overclocking
September 17, 2009 10:00:16 AM

cjl said:
CompuTronix: SmallFFT may be the "standard", but it is not representative of a maximum load case on an i7. LargeFFT is a better choice, since it also stresses the onboard memory controller, testing it for stability and resulting in an overall higher heat loading on an i7 based CPU.
cjl,

You are absolutely correct, but the difference is almost negligible. As you might expect, I've extensively tested the thermal signatures of Prime95 Large FFT's. Small FFT's and Blend on i7's, many months ago. Large FFT's provides an overall higher heat load distibuted among the processor and memory components, which is also evidenced when observing Watts consumed at the power cord. Large FFT's maximum heat signature exceeds Small FFT's by just 1c on Core i variants, but does not exceed Small FFT's on Core 2 variants, where it's 2 to 3c lower during certain test segments.

Some latter-day utilities such as OCCT and Burn Test (reminiscent of TAT) use LinPack, which shows thermal signatures that resemble a bad day at the stock market, and cycle between light workloads, through test segments which spray all processor registers with all one's, (100% thermal load, which equates to 115% workload), and in turn can drive a high-end-air cooled, overclocked i7 at Vcore Max 1.375 with HT on, right through the ceiling to ring the Tjunction Max bell like a fire alarm! :o 

Since there are very few applications or games that will spike, let alone sustain processor workloads beyond 70% to 85%, utilities which load all registers with all one's are not representative of real-world computing. While these utilities are certainly very useful for stability testing, the author of Real Temp, Kevin Glynn, uses Prime95 Small FFT's unilaterally for thermal testing, and as the difference between Large and Small FFT's is only 1c on Core i's, it might not be worth throwing another variable at the n00bs.

Your point, however, is well taken, and is a very astute observation.

Comp :sol: 
m
0
l
September 17, 2009 3:07:11 PM

flyin15sec said:
People are outright lying to you when they say 20C core temps. They are either talking out of their A$$ or referring to CPU temp (tcase) and not core temp (tjunction).

Even still, Tcase can NEVER , NEVER be lower than ambient temps with air cooling. So when someone says he's 20C(68F) CPU Temp, means that his room temperature is around 65F or lower.



I'm guessing you were talking about me when I said my temps where in the 'UPPER 20s' That meas around 27-29°C and 80+°F for the ambient temps. Don't think that this isn't possible, i'm not lying if you wanna see screenshots i'll show them when I get home.

For me, i've got my Xigmatek Dark Knight with a Delta Ultra Xtreme High Speed fan attaced to it that produces 252 CFM @ 66db. Now of course that's loud but I listen to music and play games with my speakers loud so it might as well be 1db to me, I can't hear them. If you set up a smart case, there isn't any reason you can't get super low temps on air.
m
0
l
a b à CPUs
a b K Overclocking
September 17, 2009 5:49:39 PM

ChuvelxD said:
I'm guessing you were talking about me when I said my temps where in the 'UPPER 20s' That meas around 27-29°C and 80+°F for the ambient temps. Don't think that this isn't possible, i'm not lying if you wanna see screenshots i'll show them when I get home.

For me, i've got my Xigmatek Dark Knight with a Delta Ultra Xtreme High Speed fan attaced to it that produces 252 CFM @ 66db. Now of course that's loud but I listen to music and play games with my speakers loud so it might as well be 1db to me, I can't hear them. If you set up a smart case, there isn't any reason you can't get super low temps on air.


I apologize for my strong language.

I come across threads all the time, when someone ask if their temps are ok. Folks then respond with outrageous low temps. I know you edited your post, but when responding to a question, it's always best to be clear about what you are referring to.

As it is, you can only have high/mid 20C, when talking about CPU temp, not CORE temp, with an ambient of around 75F - 80F. It was obvious the OP was talking about CORE temp, yet you listed your CPU temp.

Many times newer folks start scratching their heads, as to why they can't get their temps to the same level. This just causes more confusion.

Because there is so many factors to temps, when it comes to i7 920 temps, always indicate CORE temp or CPU temp, Hyperthreading On or OFF, What ambient temps, Type of case used, CPU cooler /w fan, and CPU speed the reading was taken at.

The best answer is, an honest answer about your temps. It'll help those looking for answers.
m
0
l
September 17, 2009 6:24:32 PM

cpu's arent freezers thats all there is to it, when you get a cpu tahts labled as cooler than the room or ambient tem, theres something a little off.
m
0
l
September 17, 2009 6:48:12 PM

70's under load with HT enabled isn't bad. On mine with HT enabled I get to the upper 70's and I have two fairly strong fans on my TRUE being fed by a 180mm intake fan with a cool ambient of around 20°c (I keep it between 68 and 70°f). My idle is in the 30° mainly because I keep all power saving features enabled.

Here is a couple shots of idle and load temps I just took on my 24/7 settings (HT disabled) and one I did a while ago with HT enabled.

Idle


Load P95 small FFT's


Load HT enabled P95 blend and Linx
m
0
l
a b à CPUs
a b K Overclocking
September 17, 2009 7:52:46 PM

flyin15sec said:
... it's always best to be clear about what you are referring to.

As it is, you can only have high/mid 20C, when talking about CPU temp, not CORE temp, with an ambient of around 75F - 80F. It was obvious the OP was talking about CORE temp, yet you listed your CPU temp.

Many times newer folks start scratching their heads, as to why they can't get their temps to the same level. This just causes more confusion.

Because there is so many factors to temps, when it comes to i7 920 temps, always indicate CORE temp or CPU temp, Hyperthreading On or OFF, What ambient temps, Type of case used, CPU cooler /w fan, and CPU speed the reading was taken at.

The best answer is, an honest answer about your temps. It'll help those looking for answers.

flyin15sec,

Well said! :wahoo:  I can't over-emphasize the importance of this. You're my new best friend!

@ ChuvelxD,

The topic of temperatures, like most aspects of technology, is all about specifications. If we're not specific when discussing processor temperatures, then the topic becomes about as clear as apples and oranges thermal fruit salad in a blender. :non:  I wrote the Temperature Guide so everyone could make sense of the differences between CPU temperature and Core temperature, and I've made it my mission here at Tom's to help others understand Intel's over-complicated mess.

We can use all the help we can get to keep this confusing topic in perspective, but regardless, I applaud your achievement in lowering your processor temperatures. From the Guide:


Section 2: Specifications

Since temperatures can be confusing to decipher and compare, it is very important to be specific, so when listing Idle & Load test Results, it is also necessary to list the Variables as shown below:

Results

Tcase = Idle & Load
Tjunction = Idle & Load

Variables

Ambient = Room Temp
Chipset = Model
i7 / i5 / C2 = Model
CPU Cooler = Model
Frequency = CPU Clock
Load = Test Software
Motherboard = Model
Stepping = Revision
Vcore = CPU Voltage


Hope this helps,

Comp :sol: 
m
0
l
!