07-26-2025 07:25 AM
I bought my laptop in July 2023, it was fine for a an year but then it started to randomly crash while gaming.
Some instances are
1. Playing dota 2: it crashes either loading into the match, or towards the end of the match.
2. Marvel Rivals: crashes in queue, or character selection screen or during the character selection during the second round.
3. Ghost of Tsushima: crashes after a cutscene ends.
I have cleaned my laptop fans, done all the thermal pasting, used DDU, downgraded and upgraded drivers. NOTHING seems to fix it, it just stopped for a couple months in between, but the issue started again. I have no idea whats going on.
I dont even get a BSOD, just laptop crash straight away, it loses power. Then I have to reboot it right away(and during that reboot sometimes some kind of windows update thing starts). I used HWinfo to monitor temperature and noticed that even while idle charging, laptop starts thermal throttling (temperature starts to reach 100C in some cores)
I tried undervolting, it doesnt crash during undervolting but apparently it seems like for 6-7 minutes in between it starts to act like it is on battery and not on power to run the game. In other words, the performance suddenly drops(as if I unplugged my charger, but the laptop is actually charging).
I am so tired of this issue, please help. Even view doesnt show anything significant. I also got BSOD on battery a couple time 'internal power error'. Recently I got a BSOD after a windows update 'error nvlddmkm.sys', and it didnt detect my dGPU for a while.
IDK what is going on with this laptop.
07-27-2025 04:10 AM - edited 07-27-2025 04:11 AM
Hello,
I dont even get a BSOD, just laptop crash straight away, it loses power. Then I have to reboot it right away(and during that reboot sometimes some kind of windows update thing starts). I used HWinfo to monitor temperature and noticed that even while idle charging, laptop starts thermal throttling (temperature starts to reach 100C in some cores)
This is a problem and needs attention before it leads to permanent damage.
When you changed the thermal paste last time what exactly did you do and use?
CPU = Liquid metal
CPU VRM = Thermal putty (usually blue or pink)
GPU = Liquid metal or Thermal paste [depends on model]
GPU VRM = Thermal putty (usually blue or pink)
And how did it look back then?
Because Liquid metal over time will always eventually shift off the CPU / GPU due to heat cycles. As parts expand, think of the heatsink expanding, its gets larger in all directions, in doing so it will spread the liquid metal or thermal paste over the edge of the CPU/GPU die. This cycle repeats each time the heatsink expands and contracts due to heat so over the course of many months its going to thin the liquid metal to the point it will no longer be on the CPU / GPU and needs to be redone. It's actually high maintenance because the more runny the thermal paste, the easier it is to push off the CPU / GPU.
Like this, it has thinned out over the P-cores and run down the sides.
So you need to check this again as what you are seeing is that the CPU is definitely running too hot and it's now in a state of thermal throttling to the point its running extremely slow. Same goes for the GPU, if it gets to hot it will switch off when it hits a critical temperature limit.
Stop using it and check / redo all the thermal paste again including the VRM (power circuitry for CPU and GPU) and GPU memory chips.
Just be extremely careful with liquid metal, it must not get on the board.
07-28-2025 01:48 AM
I understand. I actually asked a repair guy to do it for me. He used that blue putty and a grey colored thermal paste in an injection. However, I wasnt physically present there to see how he applied it, but it has only been 2 weeks, yet the issue persists from months.
07-28-2025 04:38 AM - edited 07-28-2025 04:42 AM
With those temps it wasn't done correctly and that sounds like regular thermal paste used on the CPU / GPU. Plus if the liquid metal wasn't cleaned correctly it can impact the temperatures.
To try and temporarily help can you run with Turbo boost disabled to see how hot it idles?
Open the registry (Start >Run, then type regedit)
Find this location
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\54533251-82be-4824-96c1-47b60b740d00\be337238-0d82-4146-a960-4f3749d470c
Change the attribute option to 2
Then go to your power settings and set the Processor performance boost mode to disabled on battery and power.
This will help until your resolve the problem but if the thermal paste is completely terrible your CPU will run hot no matter what. (performance is limited in this mode as the CPU P-cores will only run 2.6GHz)
Check temperatures like this to see how bad it is.
But what do you want to do next? It's needs to be taken apart and redone so are you able to get it done properly or do it yourself?