[SOLVED] Games Hard crash/freeze with solid display color and audio loop, can only shutdown holding power button

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
Hey guys, thank you in advance🙏

Laptop:
Acer Predator 17x,
Specs:
i7-7820hq,
Gtx 1080, dedicated card no integrated
32gb Ram,
512gb ssd raid, 1Tb hdd
4k display 17" inch
Windows 10 shipped

My problem is in the title.

Description is:

Complete effed up display all artefacting and weird, sometimes flickering white lines sometimes all solid white or some other color. This happens when playing any resource intensive game obviously 🙄 all games are resource intensive. When it happens is completely random but now happens early like within 5 minutes. So the display will be as described but the audio of the game still works and seems to be on a loop. No button no ctrl-alt-delete no nothing will work and only solution is to power down the laptop by holding down its main button.

Chronology:
The problem surface around November 2019 and I scoured every forum and applied nearly everything previously worked for people but nothing worked, not even a complete windows reinstall and all drivers rolled back or not. Only happens during gaming and not anything else, not during all kinds of stress tests/benchmarking and not videos and browsing and neither daz studio.

And then my corsair keyboard k95 and Asus monitor pg279q arrived which i had ordered before the problem surfaced. The weirdest thing for me in troubleshooting this particular problem is that it Stopped when i connected the laptop via display port to the monitor. Henceforth, no issues have arisen and none shall arise i know when I connect my laptop to this setup. It was a magical fix I almost forgot any problem existed in the first place. No matter how much temps skyrocketed never an issue.

Ok so a couple days ago I had to go to my house in another city but couldn't bring my keyboard monitor with me.

Now I am here trying to game on my laptop alone and this problem has resurfaced.

I think it has something to do with some circuit shorted in the screen of my laptop because what else can it be, if the graphics card was faulty there would be no way it would work butter smooth through display port. Ok, so with this in mind I connected my laptop to the tv in this house through Hdmi. But nope it white screen of deathed still.

A couple solutions i have implemented so far are disabling Nvidia audio driver, uninstalling rivatuner, updating all sortsa drivers with Iobits Driver Booster, (Bios was already updated to latest). So far nothing has worked.

I am truly perplexed as to what the hell the core issue might be, why it fixes itself with display port to monitor but Not with the hdmi to the TV.

And most importantly why it happens in the first place :(

Maybe you guys can figure something out.
 
Last edited:
Solution
Here is how to disssembly it :
View: https://www.youtube.com/watch?v=S2KgD1KuoAY

Just do not use metallic things like knifes etc in the disassembly, ony plastic (except the screwdriver)
You need:
-screwdriver -thermal grizzly kryonaut paste -90% isoprophyl alcohol -q-tips or cloth wipes
- card (soft card is best but credit card etc willl do)

1. Disassembly like he does until 09min05 seconds, after you remove the back plate. First thing is to DISCONNECT the battery connector from the motherboard which he does at 10min00sec, this is very important. Do not start taking connectors off after removing the back plate like he does, you are only removing 2 fans and the heatsink covering the cpu and gpu chips...
Any warranty left?

Go to power options, advanced settings and from there put "sleep with battery and power cable off" also "hybrid sleep off for both", then put "hibenation off for both" too. This is just to test if this is causing the issue.

Also are you using original charger and what wattage model it is?
 
Last edited:

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
Any warranty left?

Go to power options, advanced settings and from there put "sleep with battery and power cable off" also "hybrid sleep off for both", then put "hibenation off for both" too. This is just to test if this is causing the issue.

Also are you using original charger and what wattage model it is?

No warranty plus Im in south asia and there is no official repair centre here.
I did the power settings as u described but it did not fix the issue.
I am using original charger but I don't know the wattage or can figure out. Here is a photo of my charger

https://ibb.co/0thgmvz
 

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
its 330watts. It is the right model. It is not LCD related if you got problems with hdmi aswell. Is your operating system installed in the ssd or hdd (512gb ssd raid, 1Tb hdd )?
In ssd, also i dont have a multimeter but charging is fine nothing extraordinary

Update: I managed to catch this in "view reliability history", this is most recent and while other times other applications failed at time of crash, this time windows failed:

description: ''The computer has rebooted from a bugcheck. The bugcheck was: 0x00000133 (0x0000000000000001, 0x0000000000001e00, 0xfffff8065ab73358, 0x0000000000000000). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 00000000-0000-0000-0000-000000000000.''

Unfortunately, trying to access memory dump with who crash says it is corrupted and inaccessible, manually going to the minidump reveals there is no file there.
 
Last edited:

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
Actually just update the firmware drivers with this intel software, when you do the upgrade before that close all apps and disable antivirus completely for the update:

https://downloadcenter.intel.com/download/29205?v=t

Download the intel firmware software update tool. Restart after update if it asks.
Done .

Update: All individual cpu and gpu stress tests used to pass as well as memtests. Then after you mentioned the possibility of voltage related issues I searched for a Psu stresstest.

So i ran the OCCT power stress test and lo and behold i crashed similarly. Now that I have replicated the problem what does it mean? That I should get it repaired? A temporary solution was to downclock both cpu and gpu and undervolt the cpu. It helps crash later and gives me some time but it crashes nevertheless because no matter how much i try the gpu temperature albeit slowly but readily climbs to 86C at which this happens
 
So you have multiple problems at the same time, the error code you had with bsod translates to : DPC_WATCHDOG_VIOLATION which is one of more "rare" ones and its mostly caused by SSD firmware or more uncommonly other drivers. Where you able to fix this bsod issue with the firmware update or do you still get bsod with the code 0x00000133? This is not temperature or hardware related.
 
OCCT power stress test can cause stability issues with even desktop pc:s that is the only one i would not test your desktop with. For gpu you can run something like 3d-mark, heaven/valley/superposition benchmark. For CPU stress testing use Intel XTU or aida64.
Just please dont run OCCT power stress test, i have had desktop pc crash due to it with one of best power supplies, and similar issue was only happening with this particular test.
 
Was the OCCT bsod the only crash you had after installing latest SSD firmware?
Do you get any BSOD error codes than : 0x00000133 ?? (you can see error code when you get bluescreen)

Unplug the power cable and try to run on only battery, do you get shuttdown right away or unable to even boot? (issue with battery or battery sensor)

Think closely, does this ONLY happen when your using your laptop only on battery instead of power cable + battery?
  1. What is your bios version, check the msinfo method : https://www.howtogeek.com/196916/how-to-check-your-bios-version-and-update-it/
  2. If you get shutddown without any blue screen (dont count OCCT power test crash) then no need to try this but if you always get bsod blue screen reinstall your nvidia drivers with DDU (display driver uninstaller)
- download ddu, download latest drivers for your nvidia card , put both desktop
-boot into safe mode (google how), then disable your internet cable and wifi and open ddu, select GPU and then nvidia, press clean and restart recommended, after reboot in normal windows "clean install" your nvidia drivers from desktop, reboot after. Plugin internet.
 
Last edited:

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
@lga1156_fta

Hey mate thankyou for your support and ideas.Crashes never trigger a bsod. Also they were never really crashes but hard freezes with artefacts on screen and i had to manually shut down using power button. And after firware update it still happened.

I have found the defect.

Its my Graphics Card. I have very closely observed through countless testing and have finally found that the Graphics Card throttles when it reaches a certain temperature which is caused by it sucking alottt of tdp, upwards of 180.

I have successfully corrected the problem at hand by using 1. Throttlestop: for cpu, 2. Afterburner: for gpu, 3. Hwinfo64: for detailed info.

Basically, i undervolted my gpu and cpu to keep them from drawing power out of their ranges and leading to boiling temps. I was able to fix the clock speed of my gpu to a certain tdp level and after much tweaks, the performance inpact is minimal while cpu temps have decreased from 96 to early 70s or late 60s.

Similarly my gpu temperature limit was 87 and by constantly drawing more power it exceeded that limit while increasing clockspeed to 1900 mhz whereas my gpu was originally manufactured to effectively run between 1556-1733 mhz.
I now fix my gpu at 1620mhz at 0.8 v which gives me max temps at max loads in 70s and normally 60s which tdp max hits 110.

So all in all, my laptop is performing better while running way cooler and most importantly it fixed the hard freeze artifacting.
 
Last edited:
I can give you More precise guiding if you really attempt To do it.

Too high temperature wont make you To freeze like that and get artifacts. Your right about it probably been power related issue, you never answered about does this issue happen when your running it with power cable only and battery disconnected? And like i said running it other way around without power cable and only battery connected can give issues like this If battery Is faulty.
 
Last edited:

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
I can give you More precise guiding if you really attempt To do it.

Too high temperature wont make you To freeze like that and get artifacts. Your right about it probably been power related issue, you never answered about does this issue happen when your running it with power cable only and battery disconnected? And like i said running it other way around without power cable and only battery connected can give issues like this If battery Is faulty.
Yeah it happens only with the power cable connected and not on battery. Thanks mate but thermal paste is kinda hard to get where im from but i will check online if i can get it.
 
Last edited:
Here is how to disssembly it :
View: https://www.youtube.com/watch?v=S2KgD1KuoAY

Just do not use metallic things like knifes etc in the disassembly, ony plastic (except the screwdriver)
You need:
-screwdriver -thermal grizzly kryonaut paste -90% isoprophyl alcohol -q-tips or cloth wipes
- card (soft card is best but credit card etc willl do)

1. Disassembly like he does until 09min05 seconds, after you remove the back plate. First thing is to DISCONNECT the battery connector from the motherboard which he does at 10min00sec, this is very important. Do not start taking connectors off after removing the back plate like he does, you are only removing 2 fans and the heatsink covering the cpu and gpu chips

2. Also you should clear the cmos when you have disassemby this far. You can see round coin battery on the left side on the video, this is the cmos battery, remove it like this :
View: https://www.youtube.com/watch?v=UaFzyGOcrtM

This will clear cmos, just put it back when youve done changing the thermal paste.

3. Skip until 11min 25sec, you only need to remove these 2 fan screws after removing the back plate, both fans are secured by 2 screws. Remember to remove the fan connector from the motherboard, then take fans off. You can clean them with clean soft brush and compressed air, clean all dust you can. then again skip the video:

4. Skip until 19min20sec, and remove heatsink screws, if you see numbers on the heatsink, remove the screws in opposite order, like first screw 8 then 7..6..5.. etc. then doublecheck all screws are out and remove it carefully same time from both side.

5. Heatsink has also thermal pads which connects to vram, vrm and other chips. Make sure those are in same place, sometimes thermal pads can fall off. Clean heatsink off ALL dust, expecially those exhaustion parts those are usually full of it.

6. First clean off thermal paste from the heatsink with isoprohyl alcohol wipes, until its very clean. Then to the cpu and gpu. First use cloth wipes dipped to alcohol and take most of it off, if it is hardened use something like a credit card to carefully remove the hardened stuff and then continue with isoprophyl alcohol q-tip or wipes. Check this video 4mins onwards :
View: https://www.youtube.com/watch?v=XTi4LGu9SOs

Like he say do not touch those small capacitors and resistors near the CPU / GPU die, Make the die super clean and leave alcohol on it, let it dry and apply thermal paste. You can use slighty more than in the video, it is not conductive (thermal grizzly kryonaut)

7. apply heatsink carefully and tighten the screws by number order on the heatsink 1...2...3.. etc after you tightened all. Dont use excess force but make it tight.
Put cmos battery back and, put your fans back , connect both of the fan connectors to the motherboard. Tighten the fan screws. Last thing before applying the backplate connect the battery connector to the motherboard and assembly it together.
 
Solution

Only a noob

Commendable
Apr 19, 2020
12
0
1,510
Thankyou bro that was mighty helpful. I ordered the kryonaut online and will do exactly as you say.

Hey man i am planning to build my first pc complete with hard tube watercooling. I am just waiting for 10900k and the 3080ti to hit the market. I will definitely ask your help when i run into problems :p I definitely expect issues.

Anyway thanks again mate. Stay safe.
:)