Solved

Using older CPU fan for my Core 2 Duo E8400

Posted on 2008-06-18
23
1,750 Views
Last Modified: 2013-12-09
Hi. I just purchased the Core 2 Duo E8400, before replacing the my old E6420 I was advised by the hardware vendor to keep the older fan and use it instead of the one that comes with my new processor. The reason for this is because the new processor is based on 45 nm technology which generate less heat hence comes with slower fan than my old E6420.

Can someone confirm if the fan that comes with the E6420 is indeed faster than the one that does with E8400 hence will provide better cooling. I want to make sure since this might cause an overheat issue if its the other way round

Thanks

a_anis3000
0
Comment
Question by:a_anis3000
  • 10
  • 6
  • 4
  • +2
23 Comments
 
LVL 6

Expert Comment

by:Sh4d1
ID: 21813516
Hmm i think they are same, Intel Boxed Cooler is without change from last Pentium 4. It is true that CPU produce less heat than older version, so dont vorry about it. But i suggest to use new one cooler, because it is without dust and etc...
0
 
LVL 19

Expert Comment

by:PeteJThomas
ID: 21813570
There's no reason to believe that it would be any different at all - If you REALLY want to know, you could try using the old one and just monitor the CPU temp to ensure it stays within reasonable levels. But I wouldn't bother, just install the new one, that's what it's there for! :)

Pete
0
 

Assisted Solution

by:sfletcher1959
sfletcher1959 earned 100 total points
ID: 21813623
Interesting question.  The E6420 fan has a copper heatsink, more surface area and greater fan speed (2500 to 4000 rpm).  The E8400 fan has a max fan speed of 2600 rpm and a slightly smaller heatsink.  The E8400 fan has an aluminum and copper heatsink.  Copper is a better conductor of heat, but Aluminum dispapates the heat faster.  So on first examination, it would appear that the E8400 might be the better choice because it can move the heat away from the processor quickly.  The E6420 compensates for the solid copper heatsink by moving more air, thus moving more heat.  
The key to the question in my mind would be the age of the E6420 fan.  I would expect that due to the volume of air that it might have an edge on the new E8400 fan.  But if the fan has any age to it at all, its mean time to failure would be much sooner than the newer fan.  So even if it is slightly better, a dead fan doesn't remove any heat.  The E8400 fan was designed for the processor and if it were me, I would stick with it.
0
 
LVL 70

Accepted Solution

by:
garycase earned 400 total points
ID: 21814076
First, the fans are absolutely going to be different [r.e. the comment "... no reason to believe that it would be any different ..."].   Intel provides different fans with different CPU's based on the thermal range for that CPU => even with Pentium IV's the fans varied quite a bit based on which specific CPU you ordered [I still have several new (unused) fans from P-IV's ... and those that came with, for example, a 2.8GHz Northwood are different than what came with a 3.2GHz Northwood, which is, in turn different than what came with a 3.2GHz Prescott, etc. => I have the fans simply because I used Zalman's instead of the stock fans when building those systems].

However, the difference in fans is based on the power and the thermal specifications of the CPU ... NOT the die size.   Since both the E6420 and the E8400 are 65 watt CPU's, they generate roughly the same amount of heat.   However, the thermal spec of an E6420 is significantly lower than an E8400 ... 60.1°C vs 72.4°C => so the fan that came with the E6420 is clearly designed to keep the CPU cooler.   It would clearly do the same thing with an E8400, so the answer to your question is Yes, the E6420 cooler will provide better cooling.

... although I'm also certain that the "stock" cooler shipped with an E8400 is just fine for that CPU.
0
 
LVL 19

Expert Comment

by:PeteJThomas
ID: 21814139
And thus, I'm made to feel small and insignificant... :) Well I was clearly wrong on that one, apologies if I mislead you... This site teaches me more than any job I've ever had!

Pete
0
 

Expert Comment

by:sfletcher1959
ID: 21814634
Again, the lifecycle or duty cycle of the older fan is what I would be concerned about.  As garycase indicated, both could clearly provide adequate cooling for the CPU design.  However, both are ball bearing fans and ball bearings wear out due to exposure to heat and contaminents.  I would stick with the newer fan.
0
 
LVL 70

Expert Comment

by:garycase
ID: 21816325
As I noted before, either fan will work fine; but the fan from the E6420 will keep the CPU cooler.   As for lifecycle of the fan ... yes, they can (and do) eventually fail; but it's rare, and I wouldn't be concerned about that in this case, since even the "old" fan is relatively new (an "old E6420" is still a relatively new CPU).   In addition, virtually all motherboards will shut down, or as a minimum provide a very noticeable alarm, if the CPU fan isn't working.
0
 

Author Comment

by:a_anis3000
ID: 21816711
>>If you REALLY want to know, you could try using the old one and just monitor the CPU temp to ensure it stays within reasonable levels.<<

Indeed I could have just done that, which was initially my first thought however I figured that I should try to find out in advance, if possible, since I want to minimize the number of removing and installation of the coolers while experimenting with each. Another reason is I would lose the thermal tape that comes with the E8400 cooler which is only good for one install. I would have to apply thermal paste (per the installation instruction) in case of reinstallation and thermal tapes that ships with the product do better job in conductivity

>>Again, the lifecycle or duty cycle of the older fan is what I would be concerned about.  As garycase indicated, both could clearly provide adequate cooling for the CPU design.  However, both are ball bearing fans and ball bearings wear out due to exposure to heat and contaminents.  I would stick with the newer fan<<

The E6420 processor was purchased on July 22, 2007 so the fan is roughly one year old so in terms of wear and tear I don't think it suffered much. I also clean it once every month or so which should minimize the accumulation of dust.
0
 
LVL 70

Expert Comment

by:garycase
ID: 21816871
As I noted before, your "... old E6420 ..." clearly isn't very old at all.   If you want the lowest possible temps for your E8400 (without buying yet-another heatsink), just use the E6420 heatsink/fan.   Be sure you clean off the old thermal compound [scrape what you can off with a credit card; then use a clean rag with isopropyl alcohol (rubbing alcohol).   Then apply a good thermal compound ... I'd suggest Artic Silver 5.
0
 

Author Comment

by:a_anis3000
ID: 21816993
Alright I will give the E6420 cooler a shot, I have Unreal 3 and Mass Effect installed atm should serve as good stress test particularly the later since rpg games are more CPU intensive, I will record and observe how the temprature goes. In case something goes wrong for E8400 what is the maximum allowed temperature that I should watch out for? for my older processor its 64 C presumably the threshold is larger for my new CPU since its faster.
0
 
LVL 70

Expert Comment

by:garycase
ID: 21817135
The thermal threshold isn't a function of how fast the CPU is ... it depends on the design.   Actually some of the highest thermal specs on Intel's CPU's were on their earlier, slower Pentiums !!

The thermal spec on your E6420 was (as I noted earlier)  60.1°C => not 64°C.   The thermal spec on the E8400 is 72.4°C.
0
Highfive Gives IT Their Time Back

Highfive is so simple that setting up every meeting room takes just minutes and every employee will be able to start or join a call from any room with ease. Never be called into a meeting just to get it started again. This is how video conferencing should work!

 

Author Comment

by:a_anis3000
ID: 21819933
Just installed the new cpu using the older fan, before booting into Windows I checked the bios which shows a temp reading of 56 C, after booting into Windows Vista I launched Everest and the temprature is reported as 53 C and roughly 45 and 46 C for each core separately. This idle temperature doesn't sound reassuring since according to the review below it should be around 33
http://www.xbitlabs.com/articles/cpu/display/intel-wolfdale_11.html#sect0

Since I live in a hot room (room temperature is 28.9 atm) and its summer time I understand that I should expect it to run at more heat, My old CPU ran would run at 44 C (with roughly 41 for each core) and 57 (53 & 55 for each core) during intensive gaming session.  Since my new processor is based on 45nm technology it should generate less heat and the faster fan should reduce the heat further. I was expecting it to run at a temperature less than my previous one at least

In terms of proper installation, I removed residual thermal paste and applied new one before installing the fan. I double checked for secure fitting of the cooler.

I am not sure if I should try to stress test the cpu given the idle temperature could already be too high

It might be helpful to mention that so far I left the PC running idle in Windows for around half an hour and the temperature seems to be settled at at 52 ( 45 for each core)
0
 
LVL 70

Expert Comment

by:garycase
ID: 21820103
In an air conditioned rom with a temperate of perhaps 22°C I'd expect idle temps in the mid-to-upper 30's (depending on the internal case airflow).   I get 35-36 on one system and 38-39 on another (the 2nd system's on a Micro-ATX board in a micro-tower with limited airflow).   But I'm not surprised yours are higher with an ambient temperature substantially above that --> the ambient temperature is about 7°C higher than a typical air conditioned environment; and your temperatures are about the same amount above what I'd usually anticipated ... which seems normal for your situation.

Just to be sure ... are you sure you cleaned all the old thermal compound off well ... and that you applied the new compound correctly?

As I noted before, however, your comment (repeated a couple of times) that "... Since my new processor is based on 45nm technology it should generate less heat .." is WRONG.  BOTH the E6420 and the E8400 are 65w CPU's ... so they consume the SAME amount of power.   In fact, the E8400 will actually generate more heat at the central core because it's concentrated in a smaller area (although the total heat dissipation is the same as an E6420).

I'd run a stress test on the CPU and watch the temps ... as long as the core temperatures don't exceed 60°C (which is higher than I'd normally expect ... but about what I'd think you'll see in a room that warm) you're fine => remember the CPU's thermal specification is 72.4°C, so even 60°C is well below that.

I'd also install CoreTemp to monitor the core temperatures ... it's much more accurate than most other monitoring utilities I've tried.   http://www.alcpu.com/CoreTemp/
0
 
LVL 19

Expert Comment

by:PeteJThomas
ID: 21820110
Well assuming Garycase is correct, then that cooler (and the cooler that came with it) should be ample to keep that CPU cool even under stress. Given that the new CPU has a higher thermal threshold as well, it really shouldn't matter...

Provided you're not overclocking the CPU of course! :) I personally always choose a 3rd party CPU cooler, as they tend to run quieter and keep things cooler... There are LOADS about too!

As for why the current CPU is running hotter when idling than the old, I couldn't say. I have seen many cases where one corner of the cooler isn't quite on properly, which can have this sort of affect, but you've already stated it's been double checked.

That's another good thing about 3rd party coolers, a lot use a mounting bracket that fits underneath the systemboard, and you get a much firmer nicer seat for the cooler.

Anyways, enough trying to sell you a new cooler, as you didn't ask! :) Maybe I should've just waited to see what Mr Case comes back with on this one! :)

Pete
0
 
LVL 19

Expert Comment

by:PeteJThomas
ID: 21820123
Ha, and had I waited another 2mins, I would've seen!
0
 
LVL 70

Expert Comment

by:garycase
ID: 21820240
Pete:  ... I also like 3rd party coolers :-)   ... as I noted before, the reason I have all those unused Intel fans sitting around (in the boxes for the CPU's they came with, so I can tell which ones were shipped with which CPU's) is that I used Zalman's when I built the systems.   But this question was about two specific fans -- the one that came with the E6420 and the one that came with an E8400 :-)
0
 

Author Comment

by:a_anis3000
ID: 21822156
Just performed benchmark/stress test using Devil May Cry 4 Demo, Unreal 3 Tournament and Sandra. The results are shown in the pictures below

Obviously Devil May Cry is not an ideal stress test since both cores are roughly 55% engaged, however I wanted to see how the temperature would behave on a moderate stress first in case the CPU would be prone to overheat

Temperature rise is more evident with UT3 with a maximum temperatures of 62, 58 and 53 for overall CPU, Core 0 and Core 1 respectively

It becomes quite a concern for Sandra. The CPU run at its peak usage all the time during the test, the maximum tempratures were 67, 60, 57  for overall CPU, Core 0 and Core 1 respectively

In this scenario I reckon the core temperatures (60 and 57) are acceptable however the 67 overall is pretty near the 72.

Note:I already tried CoreTemp it reads the same as Everest and RivaTuner, so I believe they all report accurately, however CoreTemp only report core temperature and lack the third temp reading (the overall CPU temperature presumably), and the log entries are fairly confusing so I switched to Everest & RivaTuner that offer easier graphical logs



Devil-May-Cry-4.jpg
UT3.JPG
Sandra.jpg
0
 
LVL 70

Expert Comment

by:garycase
ID: 21824472
I wouldn't be concerned about those temps --> in a room as hot as yours they're not unusual.   As I anticipated, the core temperatures are hitting 60°C under 100% load ... but will stay well under that under most other conditions (and this is still well below the thermal limit).   Don't worry about the

The stock cooler won't do any better (in fact, will likely be worse) ... but if you want to drop the temps a bit you could use a good 3rd party cooler (as discussed above ... a bit beyond what you asked; but always an option).   This would be an excellent choice if you want to do that:  http://www.newegg.com/Product/Product.aspx?Item=N82E16835118019

... but the best thing you could do to drop your temperatures is air condition your room !!
0
 
LVL 70

Expert Comment

by:garycase
ID: 21824487
... sorry for the Newegg link => clearly you can't use it, since you're in Europe.   But the Zalman cooler is still an excellent choice ... either the 9700 or the 9500 would do a very good job.
0
 

Author Comment

by:a_anis3000
ID: 21825809
Well my curiosity got over me so decided to swap the old fan for the new one to see the difference, after testing with Sandra the temperature results were slightly more hot (around 3-4C) other than that the difference is pretty subtle I would say. So in conclusion the E6420 is a bit cooler than the E8400, still the stock fan do a fine job like you said garycase

I also agree that ultimately I will have to look beyond the stock fans that come with processors If I want less temperature which I hoped to get from the hype stirred by the reviews (such as the one i mentioned above) about the 45nm in terms of less heat and power consumption which is apparently not quite true or biased

Plenty of thanks to Garycase and everyone's feedback

Regards
a_anis3000
0
 
LVL 70

Expert Comment

by:garycase
ID: 21826024
3-4°C isn't much, but as warm as your room is I'd use the E6420 cooler to get those extra few degrees of "headroom" for the CPU.   Also, be sure your case has good airflow ... and keep it reasonably dust-free (blow out the dust regularly) => these are always good ideas, but in a room as warm as yours they're even more important.

As for the various comments you've read about 45nm CPU's => there's a lot of "hype" about the 45nm fabrications ... they ARE better; and if the SAME circuits were simply replicated in a 45nm die they'd indeed consume less power.   But what they've done with the 45nm chips is add additional features; larger cache's; etc. ... and so you've got more circuitry that collectively uses the same amount of power as the earlier 65nm fabrications.   In addition, the CPU cores are more tightly packed, so the heat they generate is more concentrated.   Bottom line:  the important spec to look at is the total power consumption of the chip ... and for both the E6420 and the E8400 that is 65 watts.   Not bad, however, when you consider how much better the performance is for an E8400 => you're getting 65% more performance [Passmark CPU-Mark for the E8400 is 2004; for the E6420 it's 1216;  2004/1216 = 1.65] for the SAME amount of power consumption !!    Like replacing a 4-cylinder engine with a V8 and still getting the same gas mileage !! :-)
0
 
LVL 70

Expert Comment

by:garycase
ID: 21833472
a_anis3000 => You're most welcome.   Don't forget to close the question :-)
0
 

Author Comment

by:a_anis3000
ID: 21840149
sorry I forgot to close the question
0

Featured Post

How to run any project with ease

Manage projects of all sizes how you want. Great for personal to-do lists, project milestones, team priorities and launch plans.
- Combine task lists, docs, spreadsheets, and chat in one
- View and edit from mobile/offline
- Cut down on emails

Join & Write a Comment

Suggested Solutions

Hello It is a very old trick to use a ram disk in order to boost PC performances, in the past, when in home environments the fastest common devices were floppy disks a part of the very small ram memory available was used to create a virtual hard …
Great sound, comfort and fit, excellent build quality, versatility, compatibility. These are just some of the many reasons for choosing a headset from Sennheiser.
This video gives you a great overview about bandwidth monitoring with SNMP and WMI with our network monitoring solution PRTG Network Monitor (https://www.paessler.com/prtg). If you're looking for how to monitor bandwidth using netflow or packet s…
In this tutorial you'll learn about bandwidth monitoring with flows and packet sniffing with our network monitoring solution PRTG Network Monitor (https://www.paessler.com/prtg). If you're interested in additional methods for monitoring bandwidt…

746 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

11 Experts available now in Live!

Get 1:1 Help Now