Sign in with
Sign up | Sign in
Your question

New CPU cooler didn't affect temps?

Last response: in Overclocking
Share
August 24, 2008 4:46:28 AM

I just installed a new cooler for my e7200, an Arctic Cooling Freezer 7 Pro. My temps are:

New cooler/stock
Idle: 40C/42C
Load: 58C/58C

It seems like the cooler had 0 effect on temps? I was planning on overclocking a bit with the new cooler, but now I'm a bit confused...

*note the temps are high due to me using speedfan to monitor my temps, subtract ~10C for real temps I guess.
a b à CPUs
a b K Overclocking
August 24, 2008 5:06:33 AM

what case are you using? if the airflow in your case is inadequete there is nothing an aircooler will do.
August 24, 2008 5:14:23 AM

I'm using an antec 300 case, CPU on a gigabyte p35-ds3l motherboard.

I have to say I'm very disappointed. I saw the stock temps and hoped the load would be better... Apparently not :( .
Related resources
August 24, 2008 10:21:23 AM

you do know that the temp limit for OCer is 70C right?so you still have alot to go. but lapp your cooler and CPU to drop about 8-10C guarranteed!!!use as little thermal compound as possible.
August 24, 2008 3:51:49 PM

Well right now I'm still a little angry that I spent $30 on a cooler that had 0 effect on my temps...lapping is kind of out of the question, maybe in the future I'll do it but not right now. And MX-1 paste came pre-installed on the cooler...

I guess I'll try to move around the fans to change the airflow of my case, but I doubt it's gonna do much.
August 24, 2008 4:07:59 PM

Disable the fan control in BIOS and see if you can tolerate the noise.
August 24, 2008 4:37:01 PM

I've got a VERY quick question. I want to change the direction of the CPU fan, which requires the heatsink to be removed since it clips on only 1 way. Do I have to remove and reapply the thermal paste? I installed it yesterday and there's been 8 hours of use total.
August 24, 2008 4:46:08 PM

Yes.
August 24, 2008 8:08:44 PM

a few developments...

Installed the heatsink again, realtemp showed 1 sensor stuck and 1 sensor at 3-6 movement. Before they used to be 2 and 9 respectively. I assumed this was since they hadn't been under load yet. I ran 3dmark06, same scores as always, no problems.

After a few hours of running, Internet explored crashed (with error message). 5 minutes later it crashed with no error. I started to run crysis, it crashed to bsod!

On restart, my antivirus said it was disabled. I hoped it was a virus causing this, but some of it other programs did not work either...right now comp crashes to bsod every 5 minutes, aometimes on login. I'll probably reformat, hopefully this isn't a hardware problems but I just don't know. Had to type this message on my phone.
August 24, 2008 8:25:54 PM

erm....did you overclock your system??
August 24, 2008 8:35:31 PM

Nope. Didn't make any changes whatsoever after I reinstalled the heatsink, software or hardware. I've set up my old computer to check the forums for help, I'm reformatting using my original windows of copy this time (last time I used ryanVM's program to integrate SP2 into it since my 320gb HDD wasn't being read).

I'm guessing if this is a hardware problem, BSOD should start happening immediately correct? I ask this because the computer was fine for 3 days then randomly this happened.
August 24, 2008 8:40:13 PM

Run Memtest 86+ to check the RAM.
August 24, 2008 8:48:28 PM

^+1
memory could be faulty.
August 24, 2008 8:54:00 PM

You sure you got the thing install right. I had an AC7 and it worked great on my overclocked e4400

Never went above 48 degrees with fan speed 4.2 with a 700 mhz overclock. Idle was around 35 the bios read around 37 degrees idle.

If your case on its side has cooler temps then when it is upright that means it ain't installed correctly.

BTW I use AC5 thermal paste. I have an Aspire case don't remember the model but it has a 80 front fan a side 80 mm fan a 120 rear fan my psu has a 120 intake and an 80 mm exast.
August 24, 2008 8:55:05 PM

K a few more developments. During the windows install, I got a few "___.dll could not be installed correctly". I googled and a lot of people are saying this is a memory problem? Can someone confirm? I'm leaning towards returning the memory either way, but I HAD run memtest86 before, albeit only 1 pass, and it gave me no errors. Thanks again.

*edit: regarding the cooler thing, I already installed it again and it appeared to be a bit cooler during loads, but I didn't really get a time to fully test due to these other problems. I did do a pressure test with my finger though just to check, and that didn't affect much.
August 24, 2008 9:01:00 PM

sportsfan12321 said:
I'm reformatting using my original windows of copy this time (last time I used ryanVM's program to integrate SP2 into it since my 320gb HDD wasn't being read).



Original windows what version is it ?


I had made several windows installations using ryanVM's program and they all worked like a charm and worked Solid rock!

If u are having issues with ryanVM's program i would recommend that

1. Always apply the update pack on a clean xp Sp2 cd (i.e no updates and other addons are integrated)
2. integrate Sp2 manually i.e. do not use "nLite" or "ryanVM" program
3. After integrating sp2 download the last update pack and integrate the update pack ONLY using the ryanVM program. Do not integrate any other Addons. When i first came accross RyanVMs website i had also downloaded and integrated the Desktop themes from RyanVm site and when i had tried it out on MS Virtual Machine, it was sometimes getting stuck, then i had integrated the update pack only and it worked like a charm on MSVM. My 2 pcs are running xp with RyanVm update pack an never had issues. Both installations are 1 and half+ year old
August 24, 2008 9:06:28 PM

Well I was going to originally slipstream SP2 onto my SP1 version of the XP CD, but I read that ryanVM's method of integration was>>>manual slipstreaming. Anyway I'll give those points a shot as well on next install.

Some other updates: my reformatting encountered another one of those .dll errors that will not go away (the others worked on retry). I'll probably return the memory tomorrow just in case. Is there any chance this faulty memory caused permanent damage to my motherboard/other components? Thanks a bunch for your help btw I have a feeling the memory thing will work...
August 24, 2008 9:09:23 PM

sportsfan12321 said:
Is there any chance this faulty memory caused permanent damage to my motherboard/other components?

No.
August 24, 2008 9:10:09 PM

btw I forgot to mention that I'm running memtest 86 atm, I'll let it run for more passes this time.

*about 20 minutes into first pass I've encountered 6000 errors...now my concern is if the sticks are faulty or if the motherboard is faulty...6000 errors (most on test 6 if that matters) seems like the slots are faulty, but I'm not sure?
August 24, 2008 9:27:44 PM

the memory is bad quality thats the only reason since you havent done any overclocking on it.return them and get some decent ones.
August 24, 2008 9:31:14 PM

The memory is OCZ DDR2 PC2-6400 Platinum, I guess I'll return them for a better pair but I was under the impression that this was good quality memory.
August 24, 2008 9:34:58 PM

iluvgillgill said:
you do know that the temp limit for OCer is 70C right?so you still have alot to go. but lapp your cooler and CPU to drop about 8-10C guarranteed!!!use as little thermal compound as possible.



Does using less thermal compound make it better?
August 24, 2008 9:36:29 PM

Not enough is bad and too much is bad.
August 24, 2008 9:55:31 PM

just a skinny layer will be alright.
August 24, 2008 10:01:48 PM

Alright I memtested my sticks seperately in the same slot, stick 1 got hit with hundreds of errors on test 4, so I stopped and stick 2 has had 0 so far. I'll let it keep running but that kind of showed the slots are fine at the very least.

Now to find some better quality ram... this http://canadacomputers.com/index.php?do=ShowProdList&cm... is where I'll be returning the old for some new (probably PC2-6400 again to go with my e7200) so if anyone sees some good ram here please let me know. I'm guessing I somehow wasn't grounded and touched one of the sticks, since OCZ has a lot of good rep, so I don't know if I should replace it at all.

Thanks again (this topic certaily got out of hand!)
August 24, 2008 10:41:47 PM

Those certainly look good, I'm hearing some issues with OCZ platinum/gold and my EP35-DS3 motherboard though ugh. On the other hand, out of all OCZ the gold model is the only approved one on the qualified vendor's list, which is weird.
August 24, 2008 11:03:43 PM

I see that the Antec 300 does not come with a power supply,default = none,which begs the question,what kind of PSU did you buy and is it adequate for the job,,furthermore all OCZ ram is pretested before it is sold,they have to as they do not make it,would it be possible to bench your mobo outside of the case so as to isolate it away from any possible unknowns,i hope that when you installed your Arctic Cooler that you removed the mobo from the case,,otherwise there is no way you can be certain that the plastic thingy's are inserted properly,as for using AC5 you only need about a cigarette paper's thickness worth,after all it is just to fill in the small micro-gaps between the two surfaces,it is not an insulator..:>)
August 24, 2008 11:23:13 PM

I'm using an antec 500 earthwatts so that should be sufficient. On the other hand I didn't remove the mobo when installing the arctic pro, however there are white plastic tabs that rest on the motherboard that I used to determine flatness. A lot of times they weren't all the way in but in the end all of them were fully flat against it. After I sort out the ram and formatting I'll put it under load (which was crysis on high) and see how far the temp goes.
August 25, 2008 1:37:46 AM

that PSU is fine for what you got. the OCZ ram should be fine.i build several system with those 2 combination(although slightly different model/chipset) never experience a problem.
August 25, 2008 6:44:43 PM

Something I figured out was that this RAM is meant to run at 1.9-2.1volts. I didn't set this in BIOS, I left it at default 1.8v although latencies were higher (5-5-5-15 I believe).

Now my question is, should I buy the same RAM (platinum) again and change volts, or just go with the gold and leave it at 1.8v? Thanks
August 25, 2008 6:58:57 PM

just get the same for better capability and compatibilty
August 25, 2008 10:13:51 PM

Alright, got new mem, changed the settings and running memtest. My only Q atm is what is 'refresh to ACT delay' in the bios? My impression is this is something that will change when OCing but I'm not really sure.

For the record I put the timings at 4-4-4-12 and +0.3v. Nothing else was changed. Thanks again.
August 26, 2008 9:11:13 AM

leave that ACT setting at auto.i dont like fiddle with ram timing.
August 26, 2008 5:00:20 PM

I don't like to either, but after some research I found some posts on the OCZ official forums where employees told owners of my motherboard to set the timings and voltages manually. I really think this has to be written in big bold letters on the boxes...for a first-time builder like me and even those upgrading it's so easy to miss the sticker on the stick that says 4-4-4-@2.1v...
!