Heat Issues with Opty 165

D

Desdichado

Limp Gawd
Joined
Apr 22, 2006
Messages
331
I have an Opty 165 with an Arctic Cooling Freezer 64 on a DFI Lanparty DR-SLI expert. I was using the paste that came with the freezer 64, and I couldn't get it past 2.6 ghz (9x289) because of heat issues (core 1 was 60C under prime95), so I decided to move to AS5.

With it applied, my idle temps dropped from 37/33 (core1/core2) to 35/28. My load temps however, went up. I had to drop the OC down to 2.5 ghz, with my load temps hitting 58/50. My question is, why would this happen? and are the cores normally this different?
 
sometimes they can be out of wack.. but generally they should be within a few degrees of eachother but it all depends on the cores.. each ones different.. for example the c2q each pair of cores are similar but one pairs could be 10C higher/lower then the other pair.... but those temps arent all that bad.. unless your putting to much as5 on.. they seem about normal.. anything under 70C on the opty 165 at load is good.. the cpu's run hot.. nothing you can really do about it.. but it will last for ever between 55C-65C
 
re-adjust the heatsink, and make sure that it is sitting flat.
The efficiency of the cooler is more reliant on how flat the heatsink sits on the cpu, than the paste used.

K-TRON
 
AS5 also has a "burn-in" time that it needs in order to settle and give optimal results.

I would also check and make sure the HSF has been 100% installed correctly.
 
thats not a horrible load temp for that cooler and cpu at all, had the exact same cpu and cooler until i went with water and i did worse than that. my load temp was over 60C at best and i was only running 2.4Ghz undervolted...mind you my case ventilation sucks the big one - one of my reasons for going to water.

give the AS5 some time to settle (day or two) and see what happens. also you could consider lapping the base of the cooler (they're usually quite rough) and even the cpu top if you're feeling adventurous. if you really want something significantly better with that cpu look into watercooling.
 
Capt. Canada said:
AS5 also has a "burn-in" time that it needs in order to settle and give optimal results.

I would also check and make sure the HSF has been 100% installed correctly.
Click to expand...

Yes... the heatsink has been installed correctly, and the as5 has burned in... temps dropped about 1-2C since I first installed it about 2 weeks ago. I'm an old hat at overclocking, but this is my first time doing a dual core, so the core temps being up to 10C different at times kind of freaked me out.

I'm using an antec 300, so my case ventilation is pretty good. I was thinking that the max temp for these was 65C, so I was trying to stay below 50, but from what you guys are saying, I can go higher than that, so I might tweak stuff a little bit more.

Thanks all.
 
My shuttle has an Opteron 180 @ 2.8 and goes no higher than 55c and that's in a shuttle with the ICE heatsink and a quiet 80mm fan, maybe these were cooler than i expected.
 
I used to have a 165 and too thought the max temp was around 65. I used to run mine at 2.7Ghz 1.32v and never saw temps go into the mid 50's. Was also using the same SI-128 cooler I use now. Then I popped the lid and saw my temps go down to the mid 40's at load. :D

What stepping is the proc? The CCBBE, CCB1E were good clockers, but they ran hot as hell and needed tons of voltage to break 2.7Ghz. CCBWE couldn't get past 2.6. I had a golden LCB9E that clocked to 3Ghz w/ 1.45v on air.
 
HydroBudz said:
What stepping is the proc? The CCBBE, CCB1E were good clockers, but they ran hot as hell and needed tons of voltage to break 2.7Ghz. CCBWE couldn't get past 2.6. I had a golden LCB9E that clocked to 3Ghz w/ 1.45v on air.
Click to expand...

i must have an oddball CCBWE, its sitting at 2.8 with 1.41V...still some headroom i figure.
 
iirc, its a CCBWE.

The problem appears to be solved. I pulled the cpu out and looked at it, and the lid isn't anywhere close to flat, so I applied a little more as5 on the low parts. Apparently, that's where core 1 was, because now core 1 stays the same as core 2, and they're both around 50-52 under prime. Thanks for the help guys. I'm currently at 2.6 with 1.4V. I might try for 2.7, but this chip doesn't have much more left in it.
 
hey glad to see you got it worked out, in my opinion, if the IHS / HS mating isnt that good, a good lapping is due. and if your really up for it, or when you get ready to build a new system try taking the IHS off. From my experiences of Taking IHS's off of AMD chips its fairly easy if you take your time, so far iv done 2 opti 170's both of which now hit 3ghz, a newcastle 3500+ and a 146 opti. all of which still are functional and work great. iv also noticed no less than 5'c drops on all those proc's between IHS and no IHS. the last one i did was my opti 170' which was under water using the swiftech Apogee GT and a 120x3 rad , i used coolabratory LMpro literaly idle temp to load temp at stock was a difference of 1'c.
IMO if you feel confident Look up Opteron IHS removal and check it out deffinitly worth it, if youve got the Stones for it or a pocket full of money.
 
You must log in or register to reply here.
Back
Top