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

MSI 880GM-E35

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

Electric Flux

Registered
‍
Joined
Apr 4, 2013
I have a AMD Phenom 2 955 BE and the board in the title. Currently I can overclock it to 3.8 and remain stable but when i go for 4.0 and run Prime95 the tmpin1 in HWMonitor shoots up, past 90 degrees C, and the PC will just shut off as if it were overheating, no BSOD. My CPU is liqiud cooled btw, that doesnt even hit 45.
So, I sent an email off to MSI and got this reply.
"Dear User, It's an invalid temp. it's not temp. sensor diode. thank you"

If thats the case what is causing it to shut off, there are no errors in P95.
 
Capture HWMonitor and post it. You and about the other 10 users are the only ones that I count as having an AMD MSI mobo and posting in the AMD sections is about 2 years or close thereto. So no idea of what the actual readings may be pointing toward. Maybe the capture of just HWMonitor and not the whole screen will give some idea.
RGone...
 
Also need to be able to see the CoreTemp or in the newer version of Free HWMonitor they call it a "package" temp. Need a larger capture.
RGone...
 
RGone said:
Also need to be able to see the CoreTemp or in the newer version of Free HWMonitor they call it a "package" temp. Need a larger capture.
RGone...
Click to expand...
+1
If tmpin is the socket temp and it's not under load then your idle temps are already high. As Rgone said we need to see the full HWmonitor and idle temps don't help us. We need to see what is going on under load. With that said when overclocking, you put your motherboard under greater stress. The section that provides the cpu with power gets hotter then normal. Motherboards that can overclock and ones that are recommended are two completely different things. Motherboards that are recommended have heatsinks to help cool the VRM section which provides the power to the cpu. I'm posting a picture of your board with the VRM section highlighted. Your's doesn't have any heatsinks on it, and if you are water cooled it only increases the issue, due to lack of airflow over this section. On top of that your board only has 3+1 a power phase count in that section, which could also be causing the issue you are having. The cpu may not be getting enough clean power from it, especially if it's overheating.
 

Attachments

  • Capture.PNG
    Capture.PNG
    204.5 KB · Views: 2,653
Last edited:
Electric Flux you should put a list of the specs of your pc in your signature. What speed were you running during this test? You still have about 10-15c of head room on the temps. How are you overclocking and do you have the Cpu V on auto?
 
I do not like Giga boards for the reason we are seeing with your temp readouts. Now this MSI thing is acting just as stupid.

On most giga boards the TMPIN0 is the board temp.
On most giga boards the TMPIN1 is the North Bridge.
On most giga boards the TMPIN2 is the Socket Temp.
Of course the old Core Temp is now "Package" temp so your HWMonitor is an Old version.

It took using 3 different Temp reading programs just to figure out what I showed above on the silly acting giga board. Had to cross reference each temp from knowing what the three separate programs called Socket Temp. That was the key.

Not a single one of your TMPINx makes any d*mn sense at all. In one of your uploaded images I would "think" that TMPIN2 is the mobo temp. I would think that TMPIN0 is the North bridge temp and then the next image down the page shows some sequence of temps that throws the whole dang thing in the garbage.

But one thing is for "sure" that TMPIN1 at 95c is not good. It could be the Socket temp or it could be the north bridge temp. If it happened to be the VRM temp >> then that is why the computer just shuts down like you turned the power off when you push it.
RGone...
 
Considering there are no heatsinks on the VRM section I guess its a possibility.
 
I'm 99% sure TMPIN1 is VRM related. On the cheapo MSI board in my lanbox, that section dropped a good 10C or so in HWMonitor after tossing a couple copper heatsinks on the MOSFETs.

Edit: It also seems to be the only temp that seems to climb steadily & rapidly under CPU load. TMPIN0 is climbing, but at a much slower rate. TMPIN2 is steady, so that may be northbridge/chipset temp.
 
Last edited:
Specs
AMD Phenom 2 955 BE, OC'd to 4.0 GHz
MSI 880GM-E35 Mobo
2 Corsair 4 gig sticks of ram running 666.7MHz
Corsair 800 watt PSU
EVGA 560ti GPU
2 SSD's and an HDD
I control the voltages.

ram.png
cpu.png
 
Also I leave the side of my case off because i have my radiator mounted outside my case it has two fans on each side of it. As of now I have no fans in my case, waiting for my PWM splitter and im going to be putting a 120mm fan in
 
Electric Flux, I believe the issue lies more with your motherboard then it does with cooling your CPU. A this point you should add your system specs to your signature. It is a courtesy to us so we only have to look at your last post to see whats in the rig. You might as well post a spd tab in cpu-z as well so we can see if your timings are correct. Why are you running your HT link so high? There is little benefit, if at all running it that high, it could also be causing issues The biggest issue though is finding out a way to cool the VRM section of that motherboard. If that can be accomplished you "may" have better luck overclocking it.
 
The RAM timing are on auto, I will post a pic in a few
The HT link being so high, no reason, its at the same speed as my NB and other than the cooling part of things my pc was "stable" when i did that. I have no problem kicking it back down, how low can i go tell I start to throttle?
 
AMD Phenom 2 955 BE = IF you study and g00gle recommendattions concerning that cpu and not the later FX or even Thuban core cpus you will; within a 99% rate, find the admonition to run the CPU_NB speed faster than the HT Frequency.
RGone...
 
The only part that I recall dolk talking about the Ht Link in that guide is where it's said that it cannot be at a higher multiplier then the CPu NB Freq and that you don't have to do much with the HT Link, if anything you may need to lower it?
 
Mandrake, here is the quote from Dolk's guide the thread starter *may* be referring to:

This is in the THUBAN BE section:
On a side note, I did mention that the HTT will be used a lot more with Thuban. For some reason, Thuban loves to have the HTT just as high as the CPU-NB. It creates a much more stable environment if you do this. But, on another hand, it may not work at all. There is a bit of mystery surrounding this phenomenon, of which I hope to figure out soon. As for now, just keep that bit of information with you.

And as already mentioned, the general practice is to keep the HT around stock and OC the NB within reliability limits (~2600-3000mhz) so yes the guide may be throwing the newbies off....
 
freakdiablo said:
I'm 99% sure TMPIN1 is VRM related. On the cheapo MSI board in my lanbox, that section dropped a good 10C or so in HWMonitor after tossing a couple copper heatsinks on the MOSFETs.

Edit: It also seems to be the only temp that seems to climb steadily & rapidly under CPU load. TMPIN0 is climbing, but at a much slower rate. TMPIN2 is steady, so that may be northbridge/chipset temp.
Click to expand...

Another thing you can do is physically check the VRM area to see if it is infact superheated...
 
You must log in or register to reply here.

Similar threads

K
MSI K9A2-Platunum power problem
Replies
7
Views
566
KenM2k
K
Wolfrequiem
Prime95: Passing Small FFT's, Failing Blends (Mild OC-3.6 Ghz-Phenom II 955 BE)
Replies
9
Views
1K
DNRDustin
DNRDustin
M
Overclocking Phenom II x4 955 and G.Skill 1600
Replies
7
Views
905
caddi daddi
caddi daddi
D
High voltage on AMD Phenom II X4 965 BE?
Replies
19
Views
11K
RGone
RGone
Wulfuhrier
MSI 880GM-E43 - Tweeking/OC Help
Replies
2
Views
2K
Wulfuhrier
Wulfuhrier
Back