R3 Temps Higher After 1 week

Hey guys,

I repasted and had my CPU never hit above 75 degrees celsius on a flat table in a room about the same temperature. My GPU never hit above 75 either. This is after running prime95 and furmark for 20 minutes and playing hours of games.

Now today I play a game and notice my GPU hits 78 C. So I run prime95 and furmark for 10 minutes and my CPU hit 83 and GPU hit 80. The temperature of the room is about the same, it may be 1-2 degrees warmer today. However, I idle at the same area, around 50-52 CPU and 54-60 GPU.

You guys think I need a repaste again? If so how much did you put on your GPU/CPU?

Thanks

 

I don't know much about pasting but I recommend buying a notebook cooling pad.. Lowered my idle temp's by over 10 degrees C

__________
M17xR3 Nebula Red | i7-2630QM | Nvidia GTX460m 1.5GB | 8GB DDR3 1333 | 320GB 7200RPM | CM Storm inferno Gaming Mouse | CM Notepal Infinite Black Cooling pad | Nvidia 3D Vision

 

sounds like a bad paste job...air gets in and drys it out, higher temps as a result imo.

 

I'm pretty sure I did a good job. Very thin layer on the heatsink and a dot in the middle of the die the size of small rice.

Then I put the heatsink on and tightened via the numbers order.

Anybody else can chime in? This thread should be full of posts.

BTW on furmark I do xtreme burn in mode at 1920x1080 and prime 95 I run the blend test at the same time.

 

I repasted again, this time spreading it evenly by ripping off a plastic bag and stretching it on my finger so it is smooth. I put some on the heatsink and some of the die and my idle temps are about 1-2 C cooler on the CPU and 3-5 C cooler on the GPU. My GPU hit 80C fans kicked in full blast, staying at 77C now.

I would like all of you to try furmark in xtreme burn in mode at 1920xs1080 on a flat surface for 10 minutes and tell me your GPU temps. I can guarantee they won't be lower than 75 C. If they are I will rep.

 

I think this might cause a problem. Spreading it out may trap bubbles in the pasted surface. The pea method is a safer way to ensure good results as it will not trap bubbles.

 

The chances of having air bubbles are pretty low.

Before, I would spread a very light coat on the heatsink then put a dot in the middle of the die.

I did this twice.

Now I tried this method and I am running cooler than before. I believe that air bubbles are a myth, either way whether you spread it or put it on as a dot, the heatsink goes on the same way every time.

 

I have tried all sorts of method on the 6970m using rice dot,pea,line,finger spread,etc,using a variety of thermal paste such as MX-4,MX-3 & Cooler Master Thermal Fusion 400,& with new thermal pads,each & every time the idle temps will loiter around 57-61 degrees.I know i am doing it correctly coz i did repasting for my previous M15x & the 5850m idle temps always hover around 45-49 degrees.Pretty much given up repasting my 6970m since max temps do not exceed 78 degrees with 780/1000 overclock.I believe its the nature of the card

 

I believe it has to do with the fan and heatsink. The previous M17x R3 had a more snug fitting fan than mine does. Plus there was a bit less of a gap. And you know what, it is still at my house but all boxed up. I am somewhat tempted to exchange the heatsinks because I remember it idling at like 47-49 for the CPU. Ah, whatever

I still want people running furmark on a flat table without elevation using the 1920x1080 xtreme burn in config.

 

my temps did not change a bit when i repasted with ic diamond 24, which has me a bit alarmed since my temps have been climbing over the past week as well. and i know for sure its not the compound or the way i do it because it significantly lowered temps on my qx9100 gx720 laptop via same method and compound.

 

Why would anyone do this? If the system is working fine and has idles in the 50s-60s with load up to the 80s, I don't see any reason to purposely stress your GPU.

Can you play games, does the card down-clock when idle, do the fans kick in when they're supposed to? If the answer is yes, why would I bother running a stress test when I don't have any symptoms?

Cheers,

 

Well to compare to temps across R3s.

Though, me being me, I switched heatsinks and my GPU is 3-5C cooler on load...

So many factors here, I mean come on, the same heatsink but from a different R3 makes that happen??????????????

And technos I know how you feel. The cooling system on the R3 is great but I still think temps are a tad bit high. On my x305 I repasted my GPU and noticed differences of about 10-15C (very bad factory job). Perhaps they need better heatsink design/quality.

 

I'm a big fan of "if it broke don't fix it". I've had my Macs re-pasted and it can make a big difference, especially with the low profiles and 6000rpm fans in a MBP, however, I using don't benchmark anything.

On the G72 I used to have it started throttling while playing games so I checked the temps and they were over 100C while playing and 75-80C at idle. Took it to authorized service center and had it repasted.

I was just wondering why you would ask everyone else to try something that will needlessly stress the GPU. I now understand you are curious.

Cheers,

 

Yeah those temps are pretty high. Higher temps means a warmer room, unable to have it on your lap comfortably, lower life of the component, and less stability.

Well stressing the GPU is just to ensure that it is operating under proper thermal conditions. I had a scare today where I didn't put the fan in right I guess and my card hit 95 C!

So I redid the job and bam, cooler than before. The CPU is still a bit hot but I am not interesting in taking this thing apart for at least a week

 

when i repasted my M17x i did notice something funny. My heatsink had scratches everywhere where it make contact with the CPU die. I've NEVER seen that before on any laptop heatsink, and I cleaned it off using a cotton ball, so there is no way i caused that damage. maybe the lack of contact where all those scratches are is the cause for higher temps?

 

I had a lot of scratches too. To remedy that I spread paste with my finger on the heatsink.

And using that rice sized amount, well it is dumb because the die is not square shaped and it did not spread evenly and that contributed to my 95 C temp.