• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

SOLVED Troubles overclocking AMD 980BE

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

scrambler

Member
Joined
May 9, 2012
Hello, brand new here. I recently purchased an AMD 980 CPU. I took to overclocking it the other day and noticed it was throttling itself at 50 degrees Celsius. At any overclock. I'll include some pictures here, one at 4.1ghz and one at 3.9.

This is a pic I snapped after about a 20 minute prime run. Using both HWinfo 64 and Core temp to look at CPU temps. I stopped the run when I noticed the cpu was going from 4.1 back 800 every 10-20 seconds. Which, I then noticed happened when the temps reached ~50 degrees. CPU voltage is incorrectly displayed in HWinfo as 1.350. It is actually 1.42 volts verified in the BIOS and in CPU-Z.

This one I snapped again after a 20 minute prime run when I noticed it jumping in between the new 3.9 ghz clock @ 1.37 volts and 800mhz. Same problem as before, 50 degrees seems to be the limit to which the CPU won't go over.

Tested in game as well, same problem. After about 10 minutes, it throttles and my frames dip way down low. Stock appears to be fine. My mainboard is a very cheap Asus board at the moment. The RAM is the G-skill Ripjaw series 8 GB (4 x 2) @ 1600MHz. PSU is a 700 (or 600, I can't remember anything about it) watt unit. Some help would be very appreciative. Am I missing something obvious?
 
Look at the CPU support list on the Asus web site for that board. Glance down through the list of CPUs that are compatible with this board and take note of their wattage rating. What is the model of the board, anyway?
 
Your board is rated for handling 125W CPUs and the X4 980 is on the CPU list. But the 980 is 125W at stock frequencies and voltages, not overclocked. You are overloading your motherboard and it's protesting to protect itself.
 
Your board is rated for handling 125W CPUs and the X4 980 is on the CPU list. But the 980 is 125W at stock frequencies and voltages, not overclocked. You are overloading your motherboard and it's protesting to protect itself.

It's more likely that 50C is set as the max allowed temperature under the BIOS, thereby initiating throttling. Or, the core temp is much lower than the CPU temp which is over 62C. My old Asus board would do the same thing described when my core temps went over 62.
 
Cool N Quiet is inactive under load. And yes, thermal throttling is definitely happening here and you can increase the threshold temp for that but it is still true the board will be overtaxed by overclocking that CPU. You don't want to build a skyscraper on top of a chicken coupe foundation. There are no sinks on the mosfets.
 
Cool N Quiet is inactive under load. And yes, thermal throttling is definitely happening here and you can increase the threshold temp for that but it is still true the board will be overtaxed by overclocking that CPU. You don't want to build a skyscraper on top of a chicken coupe foundation. There are no sinks on the mosfets.

Sad news. I am back to stock, have been for today, looks like I'll keep it there until I buy a new mobo.
 
Make sure your next board has a heat sink on the mosfets. They are the most fragile electronic component on the motherboard. See the area I have circled in red. Boards like this will be in the $90-$110 range.
 

Attachments

  • mosfet.JPG
    mosfet.JPG
    42.8 KB · Views: 142
Last edited:
That looks like a worthy board. Just read the reviews on NewEgg from customers and be aware of what the potential problems are. All boards have at least a minor quirk or two. Just make sure its not serious and something you can live with. And don't believe every review, either. Some of those reviewers are technological idiots. Look for patterns. If the same issue is being reported by a lot of people it probably is an issue.
 
That looks like a worthy board. Just read the reviews on NewEgg from customers and be aware of what the potential problems are. All boards have at least a minor quirk or two. Just make sure its not serious and something you can live with. And don't believe every review, either. Some of those reviewers are technological idiots. Look for patterns. If the same issue is being reported by a lot of people it probably is an issue.

It's the same board a friend of mine had about a month ago. He ruined it when he attempted to update the BIOS in hopes of preventing the BSOD he kept getting when he tried to play Fear 2. It... didn't work and the board never worked again. It got power, fans turned on, HDDs spun, leds turned on, but no post. I think he just flashed to the wrong BIOS version and eliminated support for his 8150.
 
A lot of people ruin their boards by carrying out the flashing process incorrectly. First, all bios settings should be returned to stock. Second, if you are doing it through Windows, all open programs need to be closed. Third, you should follow the post up prompt to enter bios before entering Windows. The safest way to flash is not in Windows, however, but through a self booting floppy disk or usb flash drive. But it requires using DOS commands and not many people still know how to do that. When you flash from DOS you are usually given the option to backup the old bios to disk so if the new bios is bogus you can reverse it.
 
A lot of people ruin their boards by carrying out the flashing process incorrectly. First, all bios settings should be returned to stock. Second, if you are doing it through Windows, all open programs need to be closed. Third, you should follow the post up prompt to enter bios before entering Windows. The safest way to flash is not in Windows, however, but through a self booting floppy disk or usb flash drive. But it requires using DOS commands and not many people still know how to do that. When you flash from DOS you are usually given the option to backup the old bios to disk so if the new bios is bogus you can reverse it.

He used the built in flash system in the bios (Ez flash?), however he did not back up anything.
 
Just an update. I got a 4.4 Ghz overclock going. In a prime run lasting 30 minutes, I reach temps of 61 degrees but no errors. However, while gaming, I never break the sacred 55 degree limit. CPU voltage is at 1.48. Scale back or all good?
 
Just an update. I got a 4.4 Ghz overclock going. In a prime run lasting 30 minutes, I reach temps of 61 degrees but no errors. However, while gaming, I never break the sacred 55 degree limit. CPU voltage is at 1.48. Scale back or all good?

Long term instability . After about 5 hours I got a BSOD. I am thinking of dropping down to 4.2 @ 1.45/6.

EDIT: Went down to 4.2 at 1.47 and again did a 30 minute Prime run, no errors popped up. Temps peaked at 59C.
 
Last edited:
I'm guessing TMPIN2 in HWMonitor is your CPU (socket) temp. With good aftermarket cooling the CPU temp is usually about 10c higher than core temps. At 70c it is at the cusp of being dangerously high. AMD recommends not exceeding about 70c (they are coy about whether its core temp or socket temp and tell me "either"). I would back off on the vcore a little more to shave off four or five degrees and settle for whatever overclock that will give you.

Are you running the new motherboard now? You don't really say.

What make and model case are you using? How many case fans are you employing and how is each one oriented with regard to intake/exhaust? Good case ventilation is as important as a good CPU cooler in keeping temps down.

As a courtesy to those trying to help you, you should put your system info in your "Sig". To do this, go to the top of the page and click on Quick Links and Edit Signature. You have made enough posts now to have access to this feature. You may refer to mine if you wish for a pattern. It's the stuff below the actual post.
 
Last edited:
Back