At first I was a bit septic regarding the gain I could get if I changed my standard AS5 replacement paste for some IC Diamond 7, and was expecting 2 to 3°c max gain !
...but after carefully repasting and also changing a couple of pads, I just noticed a 5 to 9°c gain depending on the benchmark used, on both cpu and gpu![]()
![]()
On idle the gain might not be really noticeable - hardly 1 to 2°c as fans are not kicking anyway - but on Furmark I'm now at 72°c stabilized without even having to lift the back of the beast, this to be compared to the original 81°c with back lifted ( no ac here, ambient temperatures are around 22/24°c !!).
I guess the guys with gtx580m could really give it a try as it would probably allow them to avoid being in the throttle gap crap![]()
Now I can start to really enjoy this beast.
Cya
ps : be careful while changing the pads, I screwed up once when I first changed the Dell paste for the AS5 plus one pad at the end of the graphic card as it was not really making contact with the card components![]()
I screwed up somehow and had temperatures going by +8°c on gpu Temp #3 sensor in gpuz...![]()
-
IC Diamond 24 is even better. I get 66C max on Furmark Extreme Burn-In stock clocks with my 580m.
-
Did you guys apply this on your own?
-
Yep, it is fairly easy. I've done this for every computer I own.
-
I bet it was about applying the paste better than it had been done the last time around which provided the gains.
-
-
Alienware M17X R3 Re-pasting Guide and 6970 re-pasting guide - Newegg Promo Code - YouTube
Edit: Except I wouldn't use a card to spread the thermal paste. I'd just apply the paste and let the heatsink do the spreading. -
I stopped counting the repasting I do a while ago
The thermal pad problem is another story as I probably didn't pay enough attention and used a thicker one than the oem.
Btw I own no share in their company
I was only a bit more cautious to make sure everything was alright from 1st time as I read IC Diamond paste is a bit of a pain to remove. So I heated the syringe a bit before and applied the same amount of paste as usual.
@widezu69 : I thought they were the same, only the size of the syringe was different ! Anyway it is Diamond 24 that I used (4.8g syringe).
By lifting the back of the beast I get 3°c less in Furmark
I was about to return my rig - again - and go back to Clevo's P170HM (a bit more performance but much more noisy, no switchable igp, poor audio and no backlit keyboard, all these apart from poor design) but now I have a beast that really fits my needs and eyes
-
I'm wondering how changing pads affects GPU temp since they are cooling RAM not the GPU itself. My understanding is that GPU temp reading is coming from the core itself and there are no pads involved there. -
On ATi cards, more than one section of the GPU can be monitored using GPUz. Thus the effect of changing the thermal pads will be shown. We can't see this on our nvidia cards.
-
-
I'm just trying to understand this as I have not tried to replace thermal pads on my GPU, mainly becasue I don't see the benefits of doing so. As far as I can tell, thermal pads are there to cool down the memory and any other circuitry on the dauther board, that's it. -
-
I'm just trying to share my experience, not selling anything here !
I'm the first one impressed as until now my favorite paste was AS5.
Both Furmark tests, before with AS5 and after with IC Diamond 7, were done with the following : stability test, windowed 1280x1024, 15mns duration.
The gain was 9°c but should be counted 12°c as the 1st test was with rig's back lifted while it was not in the second ( -3°c in addition in my case when I lifted the back during short 3rd test).
To confirm my experience I just repasted my old desktop rig that was already with AS5 : [email protected] on Rampage Formula and HD6970rev1@950/1400 with musashi rad.
For the cpu it's a 5 to 6°c gain on all cores (max is now 69°c on core 0).
For the gpu, 4°c gain in idle and 6°c gain full load(max is now 84°c, to be compared to 115°c with +10% power control ...but that was with oem rad and paste
).
For a few bucks and a bit of time I think it's worth giving the try. -
does repasting void warranty ? also is there any risk of damage while repasting ?
-
TostitoBandito Notebook Evangelist
Yes there is a risk of damage if you aren't careful, but it isn't that hard. Follow the repasting guide in this forum and you should be good. -
Repasting does not void warranty but changing pad does I think.
It's always a good idea to keep the old pads in a safe place, just in case.
As I travel a bit I always have my original ram and hdd in my bag ...just in case -
TostitoBandito Notebook Evangelist
The "pads" they have in there are little more than silly putty and do very little to conduct heat. It would be irresponsible not to replace them when repasting.
Like you said, you should definitely keep the original hardware for anything you replace (HDD, RAM, etc...) so you can swap it out in case you need to send it in for replacement. -
SlickDude80 Notebook Prophet
Did you keep the stock thermal pads or did you put new ones in? -
TostitoBandito Notebook Evangelist
I strongly recommend this thread:
http://forum.notebookreview.com/alienware-m17x/561599-m17x-r3-cpu-gpu-re-pasting-guide-w-pics.html
This is what I followed. It uses some different techniques than that video, has more explanation, and does full repadding. -
soon as i get my R2 back from repair i will be repasting myself with IC diamond 7 paste
Now I'm happy - IC Diamond 7's help
Discussion in 'Alienware 17 and M17x' started by MetaBaron, Sep 8, 2011.