E8400 Temperature problems

Quick question about the temperature.

My e8400 at 4.05ghz (1.33v) and a Zerotherm Nirvana is running at 50C idle, 70C load according to CoreTemp. Does this sound reasonable? It is in an Antec 900 and room temp is about 25c.

Should I reapply the heatsink? Thanks.
 
Definitely a bit on the high side. What are you CPU (socket) temps?

I wouldn't expect those to be high, considering you have a 900, but I'd just check that. Otherwise though, the load temps in particular are somewhat high. Even given your voltage, I'd expect closer to 63-64C's, but not 70.
 
Quick question about the temperature.

My e8400 at 4.05ghz (1.33v) and a Zerotherm Nirvana is running at 50C idle, 70C load according to CoreTemp. Does this sound reasonable? It is in an Antec 900 and room temp is about 25c.

Should I reapply the heatsink? Thanks.

Ehh, mine runs at mid-50s idle and mid-70s load in Coretemp with an Ultra-120. Seemed high at first, but it's still 30C below Tjunction, so I'm not too worried about it anymore. Given the temp issues with the E8400s, as long as you aren't running insane volts I wouldn't worry about it.
 
My E8400 is wayyyy too hot:

My setup:
73F ambient
Thermalright SI-128 SE with Scythe S-Flex 120mm @ 1600rpm (63 CFM)
Arctic MX-2
Thermaltake VA-8003SWA with 5 case fans, including 250mm side intake fan
Gigabyte X38-DS4 BIOS F2

My CPU settings:
E8400 - Q746A518
4.005 Ghz (9x445) 1.3325 VCore in BIOS, 1.296 VCore CPU-Z
All voltages stock except VCore and VDIMM
Orthos Small FFTs stable for 12hrs

Reported by CoreTemp 0.96.1 in Vista (speedfan beta reports 5 degrees cooler b/c of lower tjmax)
Idle: 47C core0, 48C core1, 27 CPU temp according to EasyTune
Load: 74C core0, 73C core1, 61 CPU temp (mobo alarm sounds, set at 60C CPU temp)

At such a modest voltage, with excellent air cooling, case ventilation, paste, etc., how are my temps so high!

I can boot with 4.3Ghz @ 1.4 Vcore, but my mobo CPU temp alarm goes off after a couple min of Orthos. My chip's Voltage requirements are great, but the temps are very disappointing. I wonder if Intel will take an RMA for this. Even at stock everything I am over 60C core temp.
 
My E8400 is wayyyy too hot:

My setup:
73F ambient
Thermalright SI-128 SE with Scythe S-Flex 120mm @ 1600rpm (63 CFM)
Arctic MX-2
Thermaltake VA-8003SWA with 5 case fans, including 250cm side intake fan
Gigabyte X38-DS4 BIOS F2

My CPU settings:
E8400 - Q746A518
4.005 Ghz (9x445) 1.3325 VCore in BIOS, 1.296 VCore CPU-Z
All voltages stock except VCore and VDIMM
Orthos Small FFTs stable for 12hrs

Reported by CoreTemp 0.96.1 in Vista (speedfan is 5 degrees colder)
Idle: 47C core0, 48C core1, 27 CPU temp according to EasyTune
Load: 74C core0, 73C core1, 61 CPU temp (mobo alarm sounds, set at 60C CPU temp)

At such a modest voltage, with excellent air cooling, case ventilation, paste, etc., how are my temps so high!

I can boot with 4.3Ghz @ 1.4 Vcore, but my mobo CPU temp alarm goes off after a couple min of Orthos. My chip's Voltage requirements are great, but the temps are very disappointing. I wonder if Intel will take an RMA for this. Even at stock everything I am over 60C core temp.

Your Coretemp values look fine.
 
Well, after hitting the upper limits of the stock heatsink and all the fans i could cram into my case, ive hit a limit of about 3.6 ghz with the e8400. Anyone out there have a good recommendation for a good heatsink thats not over 50$? preferably something at newegg too
 
Zerotherm 120 or Xigmatek are both great heatsinks...tho I don't speak from personal experience. They're pretty close to the Thermalright Extreme on the charts, and I believe both come with a fan as well.
 
Even compared to other users, 74C Core @ 1.296V seems abnormally high with good air cooling. I guess my Cores are still 30C from TjMax. Still, if Intel's max CPU temp is 72.4 C, I am closing on 10 degrees from throttling which seems worrisome.

Some have suggested that degradation of these chips may not be due to voltage, but rather heat as many users have ignored high temps, assuming the DTS is wrong. I may be too risk-averse, but I am uneasy about running Orthos for 12 hours on a a CPU with <i> socket temp </i> above 60C. Anybody know if degradation has been reported at sub-1.35V yet?

I still can't get over that my E2140 at 3Ghz is cooler than my E8400 at 3Ghz---plus the
E2140 just has ACF7Pro in a poorly ventilated case.

Speaking of the Freezer 7 Pro, I'd say for CPU coolers it's the best bang for the buck at around $20 with fan; then again, as hot as these Wolfdales seem I'd go with something beefier for 45nm chips.
 
I'm using a Zalman CNPS9700 LED. With fan at 100%, I'm at 43 idle, second core fluctuating between 43 and 48 (seems normal judging by what I've read of others.) Load temps never went above 48/54 Core1/Core2 under Orthos/Prime95 torture test. I'm still fairly sure it's lower than that even. Heatsink is literally cold to the touch.

Using AS5 as well.

E8400 @ 3.6ghz/400fsb/1.25v actual.
 
Even compared to other users, 74C Core @ 1.296V seems abnormally high with good air cooling. I guess my Cores are still 30C from TjMax. Still, if Intel's max CPU temp is 72.4 C, I am closing on 10 degrees from throttling which seems worrisome.

Some have suggested that degradation of these chips may not be due to voltage, but rather heat as many users have ignored high temps, assuming the DTS is wrong. I may be too risk-averse, but I am uneasy about running Orthos for 12 hours on a a CPU with <i> socket temp </i> above 60C. Anybody know if degradation has been reported at sub-1.35V yet?

I still can't get over that my E2140 at 3Ghz is cooler than my E8400 at 3Ghz---plus the
E2140 just has ACF7Pro in a poorly ventilated case.

Speaking of the Freezer 7 Pro, I'd say for CPU coolers it's the best bang for the buck at around $20 with fan; then again, as hot as these Wolfdales seem I'd go with something beefier for 45nm chips.

You seem to have a very good chip and slightly worse cooling than me.
I cannot get the speeds you do at the low voltages you can use.
btw, Intels CPU max temp of 72C is the reported temp by the motherboard, not Coretemp value. Intel dont mention Coretemp values for recommended running temps.

I am using a Thermalright Ultra-120 Extreme cooler with 68CFM 120mm fan and MX2 paste.
At 4GHz 1.35V, my idle Coretemp values are 53C, load 68C (both cores are identical temps), at ambient (outside case) of less than 20C.

If you want to improve your temps further you need find the root of your slightly higher temps.
It could be down to your cooler not being quite so good, less airflow, badly fitting base of the cooler.
It could just be an erroneous reading that is a few C off.
They do seem about right to me though.

I'd start with checking the bow of your CPU cap and the base of your heatsink.
Check how flat they are with the flat edge of a ruler. If there is substantial warp or opposite warp (ie both bow out or both bow in) then they are not well suited to each other. Lapping is the cure if you dont want to replace anything.

You will know when your chip has degraded. you will no longer be able to get as high clocks with as low voltage.
 
I recently build my system this past weekend and have not overclocked anything. I have a Cooler Master 690 with a fan for each spot which brings the total to 7x120mm. Running a Gigabyte GA-P35-DS3L board. It seems like there is an adequate amount of air flow to keep everything nice and cool. I also have an Arctic Freezer 7 Pro as my Heatsink. In coretemp I'm hovering around Core0: 40-42 and Core1: 39-42 while idle. Does this seem normal or is this considered high? Not sure how to put it under stress to see the temps so any suggestions would help.
 
Personally I use TAT to stress the cores. It puts just as much stress on it as Orthos or any other program, and it's handier to use. As far as E8400 temperatures go, this chip just has flaky temp reporting period. It's all over the internet.

I've built several systems for people over the last month or so, and they all seem different. I've had 2 systems side by side with identical parts/airflow, and they have read 7-12c different. It's just not gonna have consistent temperature reporting no matter what anybody says until/if intel fixes it down the road.
 
Personally I use TAT to stress the cores. It puts just as much stress on it as Orthos or any other program, and it's handier to use. As far as E8400 temperatures go, this chip just has flaky temp reporting period. It's all over the internet.

I've built several systems for people over the last month or so, and they all seem different. I've had 2 systems side by side with identical parts/airflow, and they have read 7-12c different. It's just not gonna have consistent temperature reporting no matter what anybody says until/if intel fixes it down the road.

Have you built many X38 based systems?
If so, do they give seemingly more accurate temps with the E8x00 series or are they just as bad?
 
Have you built many X38 based systems?
If so, do they give seemingly more accurate temps with the E8x00 series or are they just as bad?


I haven't documented which board has which temps actually, and the only real test I did was with two identical gigabyte mb's with 8400's and all other parts completely identical. My next few builds I've considered documenting manufacturer data and temps etc, but I've proven to myself that it's just not consistent.

It really doesn't make much sense. My data, and comparing it to all the data on all the enthusiast sites has just simply been consistently inconsistent at best. Hopefully someone with more time on their hands than me will make some charts with all the different builds in a more controlled environment. I used a digi-therm on the intakes of the two identical cases, so that was enough to convince me.
 
It's just about done. This new program at least gives the E8000 series a chance of reporting correct core temperatures by using TjMax=95C like I discovered in recent testing.

It also gives you a way to test and correct for flaky sensors. Send me a PM if you want to do some beta testing.

 
It's just about done. This new program at least gives the E8000 series a chance of reporting correct core temperatures by using TjMax=95C like I discovered in recent testing.

It also gives you a way to test and correct for flaky sensors. Send me a PM if you want to do some beta testing.

Is this any different than setting Speedfan to -5C for each core? Thanks!
 
Wish I knew about all of this before, but I haven't been too involved with computers for quite some time. I built a computer for a friend, using the E8400 and Abit IP35Pro. Stock heatsink/fan from intel, and it was reading 12C all the time. Only doing updates and such on a fresh install cause he has dialup and the shop had broadband. Anyway, I thought it was strange to be so low, but his case has 4 fans. It never felt hot though and even the PSU fan was blowing cool air. I have no idea what revision the bios was, but the chip is the same as just about all of them I imagine (C0). I wrote it down and can get at it monday.

I noticed it was running at 1.14 volts too.
 
I just got my running, no programs installed yet. idles at 32 according to the bios, 33 after installing vista
 
Works under Vista 32bit just fine.

10*C lower than CoreTemp 0.96.1

:-\

Edit:
5*C lower than speedfan
 
cputempwatch gives me the same temps as speedfan 4.34 beta 40

Edit: Clarified which app I was talking about
 
speedfanrealtempwa6.png


Just to clarify things. SpeedFan 4.33 beta 40 uses TjMax = 100C. To get SpeedFan to read the same as my program you need to go into SpeedFan and use an offset of -5 on both cores to correct it to TjMax=95C that my program is using. That's for the E8400. On previous processors like the B2 - E6x00 my program should read the same as SpeedFan when they are both using TjMax=85C.
 
hmm, speedfan beta is giving me trouble in vista. I'm going to try real temp tonight.

Also, speedfan 4.33 seems to be reading correctly for me. 37-38 at idle, same as the bios
 
Ok..... Orthos for 2.5 hours now without problems.

coretemp1.jpg


Am I a lucky one, seems I'm getting close to correct temps, although there is a slight variation between cores.

Also, how do I adjust the tjmax? It seems to be locked for me
 
You can't adjust TjMax in CoreTemp but you can in SpeedFan. RealTemp works in Vista 32 so why not try that. :D
 
Works fine with vista32 for me, Xigmatek HDT-S1283 cooler/IP35/CM690 chassis with stock fans at low speed.

realtemp.jpg
 
Sorry to be slightly off topic but would anyone of you proud E8400 owners know the measurement of your CPU please ?

Pentium 4 = 35 mm x 35 mm
Pentium 1 = 50 mm x 50 mm


27-02-08_1446di9.jpg


Thank you :)
 
CPU ........................ Rel............. Frequency ........... Transistors .............. Process............... L2 Cache

8086 ....................... 1978 ........... 5 – 10 MHz .......... 29.000 .................... 3 µm
80286 ..................... 1982 ........... 6 – 25 MHz .......... 134.000 .................. 1.5 µm
80386 ..................... 1985 ........... 20 – 33 MHz ......... 275.000 – 855.000 .. 1 µm
80486 ..................... 1989 ........... 25 – 100 MHz ....... 1.6 Mio ................... 0.8 / 0.6 µm
Pentium 1 ............... 1993 ........... 60 – 233 MHz ....... 4.5 Mio ................... 0.35 µm
Pentium 2 ............... 1997 ........... 233 – 533 MHz ..... 7.5 – 19 Mio ............ 0.25 µm
Pentium 3 ............... 1999 ........... 450 – 1.4 GHz ...... 28.1 Mio ................. 180 / 130 nm
Pentium 4 ............... 2000 ........... 1.4 – 4 GHz ......... 42 – 55 Mio ............. 90 / 65 nm ............ 512 KB
Pentium D ............... 2005 ........... 2.8 GHz .............. 230 Mio .................. 65 nm ................... 2 MB
Core 2 Duo .............. 2006 ........... 1.86 – 3.16 .......... 291 Mio / 410 Mio ... 65 / 45 nm ............ up to 6 MB
Cor 2 Quad .............. 2006 ........... 2.4 – 3.2 GHz ...... 586 / 820 Mio ......... 65 nm / 45 nm ....... up to 12 MB

(Source: Wikipedia)

1 µm = 1000 nm
45 nm = 0.045 µm

Conroe is Inte's E6000 series
Kentsfield is Intel's Q6000 series 2 x Conroe on one die

Wolfdale is Intel's E8000 series
Yorkfield is Intel's Q9000 series 2 x Wolfdale on one die

Nehalem will use the 45 nm process and maybe due end of 2008 / beginning of 2009
Westmere shrinked version of Nehalem will use the 32 nm process and is due from 2009 onwards.
Sandy Bridge new generation comes first in 32 nm and later in 22 nm from 2010 / 11 ... if Intel stays on target with its roadmap.
 
I'm having similar temp problems and it appears one of my cores temp gauge is actually stuck. With this being the case how do you OC effectively without knowing your actual temps?

I"m running an 8400 with abit ip35-e board. Core Temp is saying idle 52c 42c ; while Real Temp is saying 42 c 31c but also claims my first core is stuck. I've got plenty of cooling using a Freezer 7 Pro and I've rechecked the arctic silver and seating of the heatsink. Do I just need stick under a certain voltage and stress test it? I'm just a little anxious not knowing my core temps.
 
If one of your cores is stuck then when you apply a Prime load, the stuck core should start to operate properly when things heat up. These Core processors are very well designed and will take care of themselves. If you are overclocking and Prime stable then you really don't have to give a second thought to temperatures. Always let stability be your guide and you can use my program to make sure that the processor is not throttling which will show up as a check mark in the PROCHOT section.
 
Asus MBs seem to also require the absolute latest beta BIOS to correctly display 8400 temperature.
I wondered why my Asus P5B rebooted so often-blaming Acronis TrueImage-wrongly!!!
If it don't blow up or feel hot, ignore with great caution!!! Apply thermal paste, seat heatsink well, proceed with caution!!! Fit more fans!!!
 
Asus MBs seem to also require the absolute latest beta BIOS to correctly display 8400 temperature.
I wondered why my Asus P5B rebooted so often-blaming Acronis TrueImage-wrongly!!!
If it don't blow up or feel hot, ignore with great caution!!! Apply thermal paste, seat heatsink well, proceed with caution!!! Fit more fans!!!

So when a new beta bios is released, does that stop the older BIOS showing correct temps?

;)
 
Back
Top