cookieoreo18 Posted October 6 Share Posted October 6 Recently the laptop started completely freezing or just shutting down when playing 7 days to die or running 3DMark time spy. Most of the time it wouldn't crash immediately. The game would suddenly start running at 2 fps while the sound is stuttering and then after 2 seconds the computer would COMPLETELY freeze or it would shut down by itself. In 3D mark it would crash halfway through most of the time, almost never at the start. In 7 days to die it could take over an hour. Now sometimes the driver could manage to recover and the laptop wouldn't crash, but only the game would crash. In the event viewer i would get this error about nvlddmkm: The description for Event ID 14 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer. If the event originated on another computer, the display information had to be saved with the event. The following information was included with the event: \Device\Video7 CMDre 00000002 00000ffc ffffffff 00000007 00ffffff After getting this error, the GPU would sometimes disappear from the device manager and i would have to restart the laptop multiple times or go into the bios and load the default settings. The reason i said in the title that it crashes even under medium load is because sometimes in 7 days to die where i was CPU bound, the GPU would be at 70% and still crash. I never encountered this issue in DIRT 3. I haven't played any other games to check. The maximum temperature i reached was 81 Celsius so the temps are not the issue. Im also getting a DCOM error related to game bar in the event viewer. I dont know if it has something to do with this. Here is what i tried to fix this issue but none worked: Tried older GPU drivers. Tried the versions 388 (stock windows driver), 462, 511, 551, 555, 556 (I had to modify them with NVClean because my laptop never came with this GPU and thus i have to mod the drivers to support it) I did a 130Mhz underclock on the core and 500Mhz on the memory. Gave to my user full permissions for the file nvlddmkm.sys in C:\Windows\System32 Created a DWORD (32BIT) in the registry named TdrDelay and set the decimal value to 10 Flashed the VBIOS with one for a clevo 980M. The clevo bios is known to work because many others have used it for this upgrade on this laptop. Did a fresh windows installation and installed only the GPU drivers and 3DMark. Ran memtest86 and the ram was totally fine. Removed the GPU and cleaned the MXM contacts with alcohol. Modified the GPU VBIOS to undervolt/underclock it at -200Mhz -100mV core core, -500Mhz memory and get it as cool as possible. It still crashed at only 66C Celsius. Noticed that the card is somewhat bent in one side. The guy who did the upgrade did not properly trim the backplate and one side of it was covering one of the screw holes. So in one side the card was screwed in, but in the other the backplate was hitting the screw hole on the motherboard thus pushing the GPU upwards. I am guessing this caused the GPU to get bent over time and some balls under the core may be slightly broken. So i reflowed the core, which did nothing. Maybe needs a reball? So i guess the card is dying. Is this a common issue with these cards? Is it the silicone of the core dying or could something like the vcore mosfets be dying? Link to comment Share on other sites More sharing options...
sliderfra Posted October 7 Share Posted October 7 980m is super fragile as it gets really hot, plus power rail is underrated (not enough MOSFET for a 100W TDP+ card) so usually the card dies quite quickly. 1060s are much better (run cooler, faster, gen 10...). MSI 1060 mxm can be grabbed for "cheap" on aliexpress (~150e), they are much better than the gecube version. Gecube will also do fine as long as you don't mess with TDP. Zotac mxm version should be avoided as they miss a temperature resistor so fan will either not work (alienware, dell, HP) or run at 100% (msi, clevo). On alienware you can force fan with hwinfo64 so it's still ok. Link to comment Share on other sites More sharing options...
cookieoreo18 Posted October 7 Author Share Posted October 7 might as well go to a 1070 if im going to change the GPU. Will try to replace all mosfets and reball the core on the 980m anyways and see how it goes 1 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now