AMD Processors
Decrease font size
Increase font size
Topic Title: 8150= Hot, Hot, Hot
Topic Summary:
Created On: 05/28/2012 11:45 PM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
Answer This question was answered by jlindy, on Saturday, June 30, 2012 9:55 PM

Answer:
Okay, I've solved the problem! So what did I do? I picked up a Antec Kuhler H20 620 from Newegg.com on a "Shell Shocker" deal and now all is well. So if you're thinking about putting a 212evo on your 8150 DON'T, it can't handle the watts! this leaves the question as to why I was seeing such miserable performance out of the stock heatsink? Well I broke out the engineers square, and the bottom of that heat sink was warped. I guess that's something I should have checked "before" instead of "after". =|
 05/28/2012 11:45 PM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

?I'm having trouble trying to keep my new AMD 8150 machine within its temperature spec. (61° C). These are the results even after ditching the stock heatsink and fan for a Cooler Master Hyper 212EVO. When I run Prime95 at small FFTs with the default bios settings (x18 on the multiplier, and CPUvoltage at 1.2875v.) the temp. goes straight through 60° heading to 70° and from there to meltdown without pause in less than 20 seconds. When I run with the multiplier at x17.5, and the voltage above 1.1875v., but less than the normal setting the temperature will plateau at around 57°, and then slowly creep up to 61° in about 15 minutes. When the temperature exceeds 61° it rapidly shoots towards 70° where I end the test. With the multiplier set to x17, and the CPU voltage set at 1.1875 I've been able to run Prime95 for over 2 hours with the temperature never going over 59° , and by the looks of it, it would most likely survive a 24 hour run. I don't get any errors during the tests, just high temprature. Room temperature when I tested was about 22°C, and the CPU idle temperature was 31°. I'm running this on a Gigabyte 990XA-UD3 (most recent F10 bios) with 16GB Corsair Vengeance
DDR3 1600 memory. My Lian Li case is being ventilated with two fans in the top , one in the front, and one in the back. I'm also using Antec Formula 6 thermal compound. Other bios settings are as follows:

NB Freq. x11 2200Mhz
Core Boost Disabled
CPU Host Clock 200Mhz
HT Link Width 16bit
HT Link freq. x11 2200Mhz
Memory Clock x8.0 1600Mh
CPU PLL voltage normal 2.5v
DRAM voltage normal 1.5v
DDR VTT voltage normal .75v
NB voltage normal 1.1v
HT Link voltage normal 1.2v
NB/PCIe/PLL voltage normal 1.8
CPU NB VID normal

?Have I got a bum chip? I'd really like to be able to run it at it's rated settings. Would water cooling really make that big a difference?
 05/29/2012 07:28 AM
User is offline View Users Profile Print this message

Author Icon
TLH
Moderator

Posts: 3641
Joined: 11/19/2004

cpu voltage seems high. I would expect 1.35v stock or thereabouts.

-------------------------
The opinions expressed above do not represent those of Advanced Micro Devices or any of their affiliates.

1100t on M5A99x-EVO with 2x4gb and 6870 running Win7 on Force120 ssd in Corsair 600t cooled by H100. Folding for team 34106
 05/29/2012 10:52 AM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

Are you referring to the CPU PLL voltage? Umm, yeah I barley have a grasp on what that is, I believe PLL means Phase Locked Loop, and is used by the clock multiplier circuit in the CPU. The core voltage I have set at 1.1874 which is well below the stock setting. The PLL voltage is probably something better left to EE's who really understand how it works. I did check it however and the lowest it could be set is 2.02
 05/29/2012 11:51 PM
User is offline View Users Profile Print this message

Author Icon
Canis-X
The Frozen One

Posts: 4142
Joined: 01/19/2009

Do you have cool and quiet enabled....Turbo enabled....and.....what do you have your power option set to in the Control Panel (i.e. balanced, High Performance)? How did you apply the TIM (thermal grease)? I would run HWMonitor and have it run while you bench to see if your vcore is increasing, which to me sounds like what is happening.

Also, I would make sure that APM is set to disabled in your BIOS/UEFI.

I have been skimming through the link below, you might find more information on your specific motherboard in there...

http://www.overclock.net/t/113...dozer-owners-club/0_30

-------------------------
The opinions expressed above do not represent those of Advanced Micro Devices or any of their affiliates.
 05/31/2012 05:33 AM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

Cool & Quiet is disabled as are turbo, and APM. The power option is set to balanced, I did try it on high performance but the temps went straight to the Chernobyl level. I used CPUID HWMonitor to monitor the volts and AMD Overdrive to monitor the temps as the temps reported by HWMonitor were fubar. I ran a test with the multiplier at x18 and the CPU voltage set to 1.18v room temp was 20(C) idle temp was 32. HWMonitor reported the volts at 1.20, and when I started the test the volts went to 1.22, but held steady. The temps went to 56 held there for a bit then creeped up slowly over 43min. to 60. When the temps broke 60 the volts jumped to 1.30, and the temps skyrocketed. After giving the machine 15min. to cool down the volts were still reported at 1.30 so I rebooted and the volts went back down to 1.20. I made a second run with the multiplier at x18 and the volts at 1.16. HWMonitor reported the volts at 1.18. When I started the test the volts held steady, and the temps went to 56 for a bit longer than the first run but slipped up slowly over 1hr 47min., and when the temps broke 60... Chernobyl!

BTW, I applied the TIM with the pea sized dot method.
 06/01/2012 12:14 AM
User is offline View Users Profile Print this message

Author Icon
Canis-X
The Frozen One

Posts: 4142
Joined: 01/19/2009

hmmmm, I'm stumped. I did however post up your issue on the thread that I linked to previously in hopes that someone over there would be able to give some more advice. I would definitely follow that thread as there are a lot more people on that forum that check in frequently.

Good Luck!

-------------------------
The opinions expressed above do not represent those of Advanced Micro Devices or any of their affiliates.
 06/01/2012 09:34 AM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

Thanks Canis-X. This one is definitely a head scratcher.
 06/02/2012 07:00 PM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

Okay, quick update... I flashed the BIO's to the most recent version (F11 dated 5/12), and no I didn't do it over the net. I think the 'net can be just a bit too flaky to trust to something as critical as BIO's flashing. I removed the heatsink and cleaned the mating surfaces with 91% isopropyl, and reapplied the TIM using the dot method again. Antrec Formula 6 is just too thick to spread properly. I remounted the heatsink. I ran the TIM in by putting the CPU under load for 20min., idling for 20min., and off for 20min. then repeating the cycle for 6hrs.
Now for the bad news, at the stock settings (multiplier x18, cpu volts 1.28) the temps skyrocket instantly. But there is good news... The temps are stable at 57 deg. (C) with the multiplier at x18 and the volts dropped to 1.16, so the BIO's update has proved to be a limited success.
 06/06/2012 09:02 PM
User is offline View Users Profile Print this message

Author Icon
elite_commando
Overclocker

Posts: 843
Joined: 01/11/2006

try using a different TIM and reapply it , im thinking your method of applying the TIM is faulty.

-------------------------
I7-975 at 4.2ghz//noctua ND-H14//Gigabye X58-UD5// Gskill 6gb ddr3 at 2ghz// intel x25 80gb// 1 1TB seagate hdd// antec tpq 1200 watt psu// CM stacker 832 case/ ECS GTX 460 1GB black edition



Opteron 165 at 2.2ghz 1.4Vcore//Asus formula HD4770//2gb OCZ platinum 2-3-2-5//Asus A8VE-SE//Seagate 2x 250gb SATA2 RAID0// thermalrock dragon gaming case 5 90mm fans// XP pro SP2//
 06/08/2012 08:29 PM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

I have tried the TIM that came with my heatsink (Cooler Master) as well as some Prolimatech PK-1 that I just aquired, and the results have been the same... Runaway temps. when CPU volts are set over 1.21v. My initial thought as well was that I wasn't getting good coverage of the TIM, but the contact patch(es) say otherwise. The conclusion I'm coming to is that the heatsink isn't up to the task despite what Cooler Master may say, or (god forbid) I have a bum chip.
 06/30/2012 09:55 PM
User is offline View Users Profile Print this message

Author Icon
jlindy
Lurker

Posts: 8
Joined: 05/28/2012

Answer Answer
Okay, I've solved the problem! So what did I do? I picked up a Antec Kuhler H20 620 from Newegg.com on a "Shell Shocker" deal and now all is well. So if you're thinking about putting a 212evo on your 8150 DON'T, it can't handle the watts! this leaves the question as to why I was seeing such miserable performance out of the stock heatsink? Well I broke out the engineers square, and the bottom of that heat sink was warped. I guess that's something I should have checked "before" instead of "after". =|
Statistics
112018 users are registered to the AMD Processors forum.
There are currently 0 users logged in.

FuseTalk Hosting Executive Plan v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.



Contact AMD Terms and Conditions ©2007 Advanced Micro Devices, Inc. Privacy Trademark information