M5A99X EVO R2.0 and AMD 8300 Runs Hot

Toggle sidebar Toggle sidebar
R

RRTRACKS

Honorable
Aug 25, 2013
11
0
10,510
My CPU is running very hot and I am at a loss as to why. Any comments as to what the problem is, would be appreciated.

Motherboard Asus M5A99X EVO R2.0
CPU AMD 8350
Cooler Corsair H80
Memory Corsair 4 x 8GB 1600 MHz
Power Supply Antec Green 500 Watt

One thing that seems a bit wacky right away is the case temperature. Upon starting the computer it is reported to be 30 degrees C, while the room temperature is 23.8 C. The CPU temp checked as quickly as I can get to it starts out at 39 degrees C. This also seems a bit high.

If I run Prime95 with no overclock and the stock voltage (1.36v) the CPU in less than 2 minutes goes to 70 degrees C and is still climbing before I stop the test. The cooler seems to be working since as soon as I stop the test the temperatures drop fairly fast, about as fast as they go up. I can also feel warm air coming off the radiator while the test is being run. I tried re-seating the cooler on the CPU twice using Artic Silver Thermal paste one time and Dow Corning Thermal paste (I have a 5 ounce tube of this stuff) the other, with no effect.

The only way I can get the CPU to run Prime95 without over heating is to reduce the voltage down to 1.23 volts and the temps still get to 62 degrees C before stabilizing. I have run the test for 15 minutes with this voltage and did not have any of the cores fail (stop calculating). After running for about 10 minutes the idle CPU temp is 43 to 44 degrees C. The case temperature goes up to about 33 degrees F. I have a top mounted case fan running at low and would rather not crank it up because of noise. I also tried putting two high output fans in push pull on the radiator. This helped a little bit.

One other item that seems wrong is the memory only seems to be running in single rather than dual mode as reported. When I run SuperPi to 1M I get times of 22 seconds. This is the same time I was getting with 800 MHz memory with an Intel Q6600. I have tried this with 16GB pairs twice using different memory modules each time with the same result.

 
Solution
RazerZ
Sort by date Sort by votes


I should have said, "Prime95 was stopped when the temperature reached 70 degrees C with no sign the increase in temperature was going to stop. I have never run the CPU for more than a minute about 65 degrees C and never above 70 degrees C.

The cooler is brand new and the tubes are not kinked in any way. It feels as if liquid is flowing through the tubes, plus as mentioned the CPU cools at a rate suggesting the cooling is at least partially working. Also when the temperature gets above 50 or so degrees C you can feel warm air coming off the radiator.

Can you damage the CPU by under volting it as I am doing? As mentioned all 8 cores pass 15 minutes of running Prime95 at 1.23 v indicating it is stable..
 
Upvote 0 Downvote


If its stable, nothing to worry about.

 
Upvote 0 Downvote
Solution


Thanks to all for the comments and your patience with me. Having rebooted my computer and played with the voltages there is a little more information to report. All the test run below were done at the stock CPU Speed.

I can get the Core Voltage down to 1.93750v and a stable test with Prime95 for 15 minutes. At 1.87500v one core fails on Prime95 after about 6 minutes, 2 minutes at 1.1825v. Testing is stopped once a single core fails.

Unfortunately there was not much improvement on the temperatures. After 10 minutes the CPU temps were about 59C, but rose to 61C and appeared to be stable by the end of 15 minutes.

WarSam71, thank you for the link to the AMD Overdrive Tool. The CPU core temperatures at idle were 23C to 24C. At 1.93750v all the CPU cores’ temperature stabilized around 41.5C. If you could tell me what that means it would be appreciated. TmP1 and TmP2 had the same readings as Asus’s CPU and Motherboard temperature readings, respectively. In all the testing I have done the motherboard temperature has never gotten above 36C and was usually around 34C. Air temperature in my room has been between 75 and 78F by air conditioning. This is August in Inland Southern California for me.

In determining if the cooler is the problem, it would seem useful to have some data on how quickly the CPU cools down once the test load has been taken off it. To give you what I found might be useful so here goes. In my latest test 15 minutes duration, in the first 30 seconds the CPU went from 61 to 52C while the CPU Cores went from 41.5 to 30C. It is worth noting both the CPU and cores had roughly the same final temperature increase during the test +/-19C. After 2 minutes CPU 48C and cores 27C. Maybe I should have timed how long it took to get to the idle temperatures.

Does anyone think there appears to be a problem? I will run a couple of test at stock voltage to see how long it takes the CPU to hit 65C. As I recall it was far less than 10 minutes and climbing quickly. I will probably also run a 5 percent overclock to see what happens to the stability and temperatures at the lower voltages. Any other suggestions for other testing would be greatly appreciated.

Thank You,
Rich

Update August 28, 2013

After a little more testing I have a little more data.

Putting the core voltage up to 1.325v, Idle CPU 48C, Core 33C, after 135 seconds of running Prime95 CPU 65C Core 47C

30 Seconds after Prime95 was stopped – CPU 56C, Core 32C, 60 Seconds CPU 52C Core 30C

Attempting a five percent CPU overclock with a core 1.200v resulted in a BSD the moment Prime95 started. On the second try with the same settings the computer just went directly to the BIOS start up.

Running it again at the stock CPU frequency of 4.0 GHz with core 1.200 volts produced a stable Prime95 test for 15 minutes with CPU 61C and Core of 42.2C. Paying a little more attention to the temperatures after Prime95 stopped, I noticed the Core temp went from 42.2C to 30.0C in 10 seconds, in 30 seconds the Core went to 52C, similar to the previous tests.

What was not mentioned before: CPU fan (pump in this case) 4300 to 4400 RPM, Radiator push, pull fans +/-1975 Rpm and top of case fan +/= 1050 Rpm. I am running the push pull fans a higher noise level than I would prefer. Since the case does not have a cooling problem the top fan has been left on a low speed. The primary voltages, 12, 5 and 3.3 drop less than 0.1v before or during Prime95 testing. Beyond that I do not know what to look at.

What I do not like is my system appears to running successfully with little room for error, with a CPU voltage way below spec. Is this normal? If there is something wrong with either the motherboard or CPU I would rather have it fixed while it is under warranty or before it does something like wipe out a hard disc’s data. The power supply seems to be doing what it is supposed to be doing, and the cooling system would not appear to have the effect on voltages seen but could be part of the problem with heat.

If there is not much data on the rate of temperature drops after a load has been removed from the CPU, someone might want to start a thread on this. The rates of temperature drops I am getting seem OK, but it would be nice to have something to compare them to.

Any comments, opinions, guesses, or even wild thoughts about all this would be appreciated.

The next step for me at this point is to summarize data here and submit it to Asus and AMD and see what they have to say. What I hear back will be presented here as it might be helpful.

Thank You,
Rich
 
Upvote 0 Downvote


A summary of the problem provided to AMD follows. Asus/a board were off lime at the time my attempt to post was made.

Problem: The CPU will not run at the stock voltage and frequency without overheating.

With a core voltage of 1.325v, after 135 seconds of running Prime95 the following temperatures were recorded CPU 65C Core 47C.

Unless the core voltage is less than 1.250v or less running Prime95 causes the CPU temperatures to exceed 70C. At a core voltage of 1.93750v Prime95 will run all 8 cores stable for 15 minutes, but at a temperature of 62C.

At a core voltage of 1.200v and five percent over clock of the CPU speed, starting Prime95 immediately causes either a BSD or a BIOS reboot.

After running Prime95 for 15 minutes, generating a CPU temperature of 62C, and core temperatures of 47C upon stopping Prime95 the CPU cooling system dropped the CPU temperature by 9C in 30 seconds and the core temperature by 12C in 10 seconds.

According to the Asus motherboard and its monitoring software the voltages for 12, 5 and 3.3 were never exceeded and did not show more than 0.1v drop during any of the tests running Prime95.

To me this appears to be a problem with either the CPU or motherboard and as such this communication has been addressed to both.

It would be appreciated if you could provide an answer and a direction I should take with this problem.

Thanks,
Rich

 
Upvote 0 Downvote
Finally I got a technical support submittal to Asus. Neither the links in my user guide nor a search by Google worked. From a previous service ticket I found the following link that worked.

http://support.asus.com/servicehome.aspx?SLanguage=en

Once this problem is resolved to make it easier for people to figure out what is going on, unless someone objects, all my posts on this thread leading up to the summary of the problem will either be deleted or edited to the minimum allowed.
 
Upvote 0 Downvote
Both AMD and Asus provided responses.

Asus Response

Thank you for contacting ASUS Customer Service.
My name is Carter and it is my pleasure to help you with your problem.

For this problem, I suggest you to enabled the Cool'n'quiet and the CPU C1E under the CPU Configuration in the bios for a try.
If still, please enable the CPU Q-Fan Control and set the CPU Fan Profile to Turbo under the Monitor menu in the bios for a test.

Welcome to refer Troubleshooting & FAQ for ASUS products in ASUS website:
http://support.asus.com/servicehome.aspx?SLanguage=en

If you continue to experience issues in the future, please do not hesitate to contact us.


AMD's Response

Response and Service Request History:
I understand that your CPU suffers overheating when it is running Prime95.The maximum temperature for the FX-8350 CPU is 61.1C, I would definitely be concerned about those types of temperatures. However, certain temperature monitoring programs can actually report incorrect temperature values, as the 61.1C maximum is for the CPU internally, not the CPU socket (where the CPU plugs into the motherboard). The motherboard will report the socket temperature, which can be much hotter than the CPU itself, and most temperature monitoring programs will use the socket temperature, or variations on those readings, to report the CPU temperature.
If you download the AMD OverDrive software at http://sitesprod.amd.com/us/game/downloads/amd-overdrive/Pages/overview.aspx , it will provide an accurate temperature reading from inside the CPU. If you still see those much higher-than-normal temperatures in OverDrive, I would look into software updates first (FX patch from Microsoft for Win7, Chipset/BIOS updates for the motherboard) to see if that addresses the issue. In addition, it is advisable to put your CPU into another motherboard to detect its temperature because it can help to verify whether the problem lies in your CPU or your present motherboard. If the problem lies in your CPU, you may wish to replace the CPU under warranty, which you can do at www.amd.com/warranty
If you have any other questions or concerns, please do not hesitate to reply to this e-mail directly and I will try to provide any additional information that you may require. Thank you for contacting AMD!

So far I have taken the recommendations provided by Asus. Unfortunately, there was no improvement. I have a AMD FX-8150 chip I can use with the board to see if there is any difference in voltages and temperatures. Maybe what I have is normal and the methods of my testing is flawed. Once I do a bit more testing I will report back what is found. I was pleased at both the quickness and answers provided by Asus and AMD. They can only do so much with an email.

Alll the Best,

Rich
 
Upvote 0 Downvote
Hi to All,

Before beginning, I would like to once again acknowledge both ASUS and AMD customer service for providing both a timely, helpful and professional response to the inquiries made. Neither ASUS nor AMD were told their responses were going to be posted on any type of forum.

It turns out the CPU I have for exchange testing is a AMD FX-8120 rather than the FX-8150 mentioned earlier. A brief summary of the equipment test and results follows:

Same setup as before except for the items noted below:

CPU: AMD FX-8120

Core Voltage: 1.35

Test: Prime95

Results after 10 Minutes:

CPU socket temperature: 58 C


CPU core: 37.5 C

These do not seem to be very good results but are much better than what the FX-8350 did with the same voltage. From what I have heard the FX-8350 is supposed to be better at handling the voltage / work load better than the FX-8120.

This would seem to suggest the difference in CPU performance in heat with the CPU I have versus the other test results is with the CPU rather than the motherboard.

Given I am able to get the CPU I have to work properly by under volting it with the apps I use, should I give any concern to this being a possible factor for failure down the line?

One of the reasons I am posting here is because I am lost and need help. When someone attempts to aid me I act like an expert and refuse advice. I listen with both ears closed. If you can get beyond that and make some meaningful noise, it would be appreciated.

All the Best,
Rich
 
Upvote 0 Downvote


Hi Rich,

don't be surprise of the temp reading on your M5a99x, as AMD claims on their reply to you, Socket Temp and Core Temp will vary, i have the same motherboard as yours and i'm running an Old Phenom X6 1055T 90W version, i have the same problem as yours where the variation between CPU Socket temp and CPU Core Temp will be 20 - 22 C, this is normal for this motherboard, it is the Sensor type that Asus use in this motherboard that does that, the sensors do varies across motherboard from same manufacturer itself, i had the same CPU tested on MSI motherboard, the Core Temp was exactly same as Asus counterpart but the socket temp difference was just 10 degree C on MSI, so don't worry as long as you keep Core temp under 55C nothing will go wrong and please take note AMD's Heat management protection is really good, the moment it notice overheating on the core it will throttle the core speed, u will notice a drop in the frequency at least by half and fyi my fren had a customer that ran an AMD Athlon X2 with a heatsink that slipped off from the contact between CPU, it somehow dented about 45 Degree and never touch the CPU, the CPU Ran without Heatsink for 2 years and it was just fine, all this was possible because of AMD frequency throttle management for overheating
 
Upvote 0 Downvote
Oxide007,

Thank you for your time and comments. My thoughts on your observations follow.

Since Asus is a fairly reputable company it would be expected their quality control would have picked up and corrected the 20 C temperature difference between the core and CPU temperature in 99 percent of the cases. The lack of other people bringing this up in forums such as this would indicate this problem is limited.

After all the tweaking, testing, posting and communicating with Asus and AMD, I experienced burn out with this issue. My computer works even though it makes a lot of noise because of the high volume fans and the extra money spent on CPU cooling, might have just corrected what would have otherwise been a problem. The slightly over clocked Intel Q6600 based system with slower memory is faster than the subject of this topic (The AMD 8350 system) with some of the applications I use.

My biggest problem as I see it is I did research on my system after rather than before the system was purchased. This was probably compounded with a quality control issue. Although this is my first experience with an AMD processor and a bad one, I will still give AMD consideration before purchasing another computer.

All the Best,
Rich
 
Upvote 0 Downvote


Hi Rich,

another bigest culprit that contributes for overheating that i just discovered is PSU issue, when PSU is maximum pushed even-thou it is pure power, the Voltage it supplies on the CPU 12 V rail will be fluctuating, at a time it will be low and at a time it will be high, my Asus 990X just had the USB 3.0 Fried out and temp of my Phenom was quite high, i OC'd my Phenom ii x6 1055T from 2.8 GHZ to 3.5 GHZ, my CPU Core temp was about 64 C and CPU socket temp was 84 C, and to get there i have to tweak my asus 990X CPu vcore to higher to achive it, the most stable clocking i can get without voltage increase on vcore was at 3.4 GHZ, where 100% stress test was revealing about 60 C core temp and 80 C CPU Socket temp, after a while My USB 3.0 Fried out as well as 1 of the Sata port and once i sent the motherboard for warranty, i used a spare motherboard which my fren gave me, i hocked it on my same Gigabyte 550 Watts existing PSU and guess wat, the motherboard fried, worst than my 990X, this time got smoke came out from my frens Asrock motherboard, and again 1 more motherboard was fried by this PSU and after successfully Frying 3 motherboard that PSU went dead, before my Asus 99X came from warranty i have purchased a Corsair CX750M 750 Wats Power supplier and kept it, than i received the Asus 990X motherboard from warranty, exact same unit, they have repaired the USB 3.0 Connectivity and the Sata connectivity, i install everything with the new Corsair PSU, guess wat, i was amazed, the CPU was able to clock till 3.5 GHZ without a Voltage tweak on the Vcore and it runs way cooler, the CPU core temp was never high than 46 C and most of the time it will be 45 C and my CPU Socket temp was at 59C - 60C, all this happen just a week ago, so my personal experience, PSU can cause serious heat issue and stability problem, if you search online, allot people wont agree, they will say if voltage fluctuate most of the time the system will froze and the temp will not get affected, but 1 thing they never understand , what if the CPU 12V rail was pumped in higher voltage spike at times ?, say 13V at a second and another second was 10V and so ?
 
Upvote 0 Downvote
Hello. I have been running the ASUS 99x EVO r2.0 for about half of a year now and have finally got around to messing around with settings for OC. As many have discovered, the socket on this board runs really hot with the FX 8350 or I am sure any 125watt CPU. I have tried multiple solutions including:
1. Blowing case fan directly on Northbridge Heatsink and surrounding area. Made no noticeable change and still receive up to 85 degrees celcius.
2. Replace thermal pad for entire heatsink on board. Made absolutely no difference at all.
3. I have enough room in my CASE for a fan blowing directly on the back of the motherboard where the Northbridge is located. (THIS HAD DROPPED MY SOCKET TEMP SENSOR DOWN 10+ degrees.

Now, while stressing Prime95, I still get high socket temps but at least now it is around 72 degrees celcius and that is only while stress testing. While hardcore gaming (far cry 4) I do not recieve anything over 49-50 degrees celcius for socket and 54 degrees celcius for the physical cpu.

I have tried flashing to the most updated bios, lower voltage, change digi power settings and nothing had worked. Unfortunately my only work around(not fix) is to have a big enough case to have a fan blowing directly on the back of the mobo. This will slightly raise your ambient case temps so please be advised that proper case cooling is just as essential.

As for a fix to the issue. I would suggest considering a purchase of another board model after checking forums and doing research. Most people having issues with this board are trying to overclock and in all honesty, if overclocking your hardware is really that important to you, invest the money into a proper OC board, and spend the time you would trying to solve this issue, making money toward the newer board. Thank you.
 
Upvote 0 Downvote
On a side note, if you are looking for a compatible board for any FX type CPU, this board has many great features. If you do consider buying this board for a future pc, the first thing you will want to do is jump into your bios, set it to advanced mode.
1. Head to the AI Tweaker tab and scroll down to the voltage section.
2. Select manual voltage and manually set yor CPU/NB voltage for mine was set stock at 1.4 when it should be around 1.2(Doing this will drastically lower your CPU temp and prevent any future damage from being done to the board.
 
Upvote 0 Downvote



Make sure you use a good PSU, i had the same problem, i changed my PSU to corsair CXM 750Wats, thing changed my CPU socket temp droped from 85c to 60c Max for Phenom X6 1055T at 3.5ghz OC and my current new CPU FX8320 runs allot cooler at 4GHZ OC - not reaching anything pass 60c for the socket and 45c for the core.., hope this helps

 
Upvote 0 Downvote


check my answer the PSu should be the issue, i change PSU my socket temp drop from 85c to 60c max for the oc i did...
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom