MSI GT70 motherboard bad?

Hi All, just picked up another GT70 off ebay. Seller stated it was a no POST after replacing the RAM. I got it and took out the RAM they had in there and replaced it with some known good RAM and it POSTed fine. Threw the SSD from my other GT70 into it and it booted into Windows fine. Ran for maybe 5 minutes before the screen went fuzzy and it shut off. Figured it was just a driver issue so I rebooted it to safe mode and ran DDU. After it came back up I went to install the drivers and it did it again. Figured the GPU might be bad and I had a spare 880m to throw in there. Put that in, as well as I changed the CPU as well for a 4930MX. Rebooted it and it went to windows again, as I was going through to see what everything was doing it crashed again. It won't POST again for about an hour or so, it almost seems heat related? I'm guessing the motherboard must be cooked or something, however when I inspected it I didn't see any obvious signs.

Reference image of what the screen does:

 

try external monitor
GPU "or only cable or display"

when the boot succeeds .,,.coolerboost !!
and HWINFO in windows --determine heat problem

 

Yeah I did try these already. I turned on the cooler boost and went into HWInfo but the crashed before I could get to the GPU temps. I would like to rule it out as a GPU specific problem as two different 880m's have caused this exact same behavior. I did try an external monitor as well but when it crashes it crashes, doesn't matter if it's the internal display or an external display.

 

but sometimes works ??? VR PADS !!!

 

It sometimes boots into Windows for like 5 - 10 minutes, and all the thermal pads are in place. Are you suggesting I look at the thermal pads on the voltage regulators for the GPU or for the CPU?

 

all,.,. you must see the PRINTED hw on all pads


or try cpu lock / gpu lock "minimal state + cooler boost"
check stability ,.,.,but VR cooling is slow "some models --passive " impossible use fans for cooling


,,best basic inspection ?? ,.,.,.new pads / thermal paste
PADS - TG8 "thermal grizzly is good" soft and working or here ARCTIC 6w/mk -low cost / soft

you have to evaluate the problem yourself., from a distance it is difficult

//////////////
REPASTE / new pads --basic step "when have random problem" this look like temperatures problem"
cpu / gpu look OK ?? ,,.,,.VR or others

check BD-PROCHOT status ?? when is possitible or all for CPU
Throttle stop is best for this

,.,.some status is hidden after EC chip "impossible find real data"

translation

edit ,.,.hwinfo screen or log ?? for gpu / cpu zone ???

 

https://forum-en.msi.com/index.php?...clocks-itself.362762/#lg=post-2058517&slide=0
gpu zone

https://forum-en.msi.com/index.php?threads/less-processor-speed.362105/#lg=post-2054824&slide=0
cpu zone

+ check all others ??

 

https://www.myfixguide.com/manual/wp-content/uploads/2014/08/MSI-GT70-Disassembly-12.jpg

??
here is to many PADS where is possitible problem

 

Okay I mean, I haven't tried NEW thermal pads yet as the pads I had didn't match the same thickness as the ones that came on it stock, however I'll see if I can swap some out. Also, I did already repaste everything with Arctic silver already. I could try setting the CPU and GPU to minimum states however it usually doesn't stay on long enough to really get anything meaningful done. Also, I have the GPU drivers uninstalled so I would think it would mostly just be using the iGPU... Although like the GPU, I have tried swapping the CPU for a known good one and got the same results.

 

when is problem igpu --CPU zone + VR

this is old machine ,.,.what contains bios ??
disable speed step ?? -- and have BASE CLOCK lock for cpu

GPU ? no drivers - LOCK "lower p-state"
similar problem ,.,,or here is NVIDIA INSPECTOR ?? for GPU OC / setup "try lock p-state"
unfortunately I don't remember

best way ,.,.try correct pads on VR !

 

another message for PADS ,.,HINT

translation

 

Alright just as a follow up in case anyone else finds this thread when trying to diagnose a similar issue.

I decided to see if it was somehow a Windows related issue so I booted it off a flash drive into Ubuntu. Seemed to run fine, ran some 4k60 youtube videos to stress the CPU, everything checked out. Couldn't get it to crash again. So, since I had been kind of lazy and just swapped a drive from another computer previously without reinstalling Windows I decided to go through and reinstall Windows. Everything worked fine, got it installed, updated, and even flashed the MSI vbios to the 880m that was in there. No crashes or anything. I go back and run some 4k60 youtube, no crashes. I run Heaven Benchmark for a while, let the GPU and CPU get nice and warm, nothing.

Right when I was about to call it fixed for the second time I go to shut it off and it crashes AGAIN. Then does the whole no boot thing. I figure that surely, there must be some bad hardware. So, I took out a stick of RAM that I hadn't confirmed to have been working in a GT70, and left the other two sticks in. Sure enough, I have not been able to get it to crash again since. After I removed that stick it booted right up, no more crashes. I tried over and over to see if I could get it to crash.

I guess for now the answer is that my "known good" RAM wasn't as good as I believe it to be. I know that it was working in the laptop I pulled it out of, however maybe I somehow zapped it or something on my way to putting it in this GT70?

Anyways, I will be giving it a good 8-9 hour stress test tomorrow, if there are no other updates from me about this, that means it has stayed fixed!

 

gaming machine --gaming stress

MSI KOMBUSTOR
gpu render + cpu burner

or full VRAM + cpu burner
or + MEM test --extreme stress

no need 10h or similar .,,.,.need point where temperatures go down

you have to be sure it was just ram and not temperature !
or you only risk HWdefect !

 

Small update on this issue.

Been running fine for 8-10 hours per day, no crashes or anything. Added 2 more sticks of ram to fill up all available slots, and I noticed shortly after that the crashing started again. Seems that it's possible the RAM I had wasn't bad at all, rather the slot has some sort of issue. Seems whenever I have one specific slot populated it will crash randomly and not reboot. Guessing that was the real issue after all.