How is the health of me NP9370? [Afterburner dump inside]

I re-pasted my notebook (both GPUs + CPU) some months ago with AS5 and would like to know if it's doing well.

- MSI Report: HardwareMonitoring.hml (long 11.4MB MSI Afterburner .hml report, that's why I simple didn't take a screenshot of it)

This report was captured while I played Rise of the Tomb Raider (up-to-date) with the following settings:




Only the game + Kaspersky (not scanning, just running) + Dropbox (not syncing, just running) + ROCCAT Tyon mouse software were running, OS running on SSD, game running on Momentus XT HDD. I started the game at ~17:52. It crashed out of the blue at ~20:46 (I wasn't even in an action/busy scene), I re-started it and played until ~21:17.

I appreciate any comments, thanks!

 

Anyone?

 

Posting up a nice screenshot of the graph is a lot quicker to read

 

The issue is that it's gigantic, one SS for example:



As you can see, I have enormous horizontal and vertical scrolls and this is just 30min of gameplay in the SS.

Is there a way to export it to images easily?

 

From that Screenshot alone I can tell you that you should re-paste the GPU. GPU1 reached 94°C, and GPU2 91, definitely undesirably high temperatures. Look up what the max CPU temps were, MSI Afterburner can also log it. They should stay below 90°C too, otherwise you are best advised to repaste everything. When repasting, check whether the thermal paste has been spread evenly and covers the entire die, and you should also see whether you used too much or too little paste, and adjust your new application accordingly.

 

Well, just checked, GPU1/2 temp limit, with only 2 minutes of game it goes to the 1, also I got GPU1 94c max temp and GPU2 91c max temp. Rise of The Tomb Raider is pretty demanding and I'm playing with high/very high settings, but this shouldn't happen, right? Repaste would fix that? On the other hand, on normal use I get 45c/45c max, is this normal?

PS: Running everything stock.

 

Repaste with some IC Diamond or Gelid GC Extreme.. I wouldn't run my GPUs at those temps, they will die really quickly at that rate!!

 

Well, too bad where I live I can't really get my hands on those pastes, I only have Arctic Silver 5, wouldn't that work?

 

.
AS5 isn't as good at high temp scenarios (more at lower 50s-60s which is why it was favourable with desktops). I suggest Gelid GCE as well. If only because it's cheap and easy to apply. ICD is a more permanent solution (at the least you don't want to waste it if you have to remove the heatsinks again) and better once you've eliminated any other failures.

The most worrying thing I can see in the graph is while those temps are basically on the thermal throttle, your GPU1 usage is not 100% and STILL generating huge temps. The 680Ms are pretty damn hot though so don't expect much change, but perhaps better clocks (you're undoubtedly throttling atm).

That makes me wonder if your heatsinks are actually contacting correctly or if your pasting is not up to snuff (too much or too little). Furthermore, it's probably a good idea to give the heatsinks a good blow out. If you haven't, chances are they're chock full of dust.

 

I'll do my best to find this Gelid GCE here... Well I cleaned the PC not long ago (1mo or so), I believe it was a messy repaste I did last time with AS5, I'm not very good at doing this

 

Nobody gets it right the first few goes.

If you can't find GCE, have a look for Thermal Grizzly Kryonaut. It's a newer non-conductive paste (like GCE) and performs very well for high-temp usage (better than GCE in many tests I've read).

 

Just import some IC Diamond 24 and be done.

You can find some on Ebay.