P65 Creator dGPU not working

[BCN1000]

I have a P65 Creator with an Nvidia GTX 1050Ti GPU.

The 1050 doesn't work to work at all. Connecting the HDMI just causes hanging and constant switching between the integrated Intel GPU and back to the 1050.

I have had a similar problem recently with my Ge75 but this was caused by old battery + underpowered charger. However I the original 19.5V 180w charger for this laptop and have tried with a 20v 260 watt charger for the GE75 as well and it doesn't remedy it (I'm presuming that tryign with a 20v charger instead of 19.5 volt wouldn't itself casuse a problem).

Here are the screen shots and monitoring hml from afterburner and GPU-Z.

Have done a bunch of stuff (reflash original vBIOS, update BIOS and EC firmware, reinstall windows, change driver, change RAM, underclock, replace thermal paste for CPU and GPU but left existing thermal pads on other components).

Is this GPU dead? I gather that the 'reballing' service offered by some stores nearby is a non-fix, as if it's a dead GPU the problem just resurfaces soon after.

Thanks.

 

[BCN1000]

?

 

[BCN1000]

Can anyone suggest why this is not receiving any replies?

 

[BCN1000]

Not all at once people.

 

[StefanM]

Can anyone suggest why this is not receiving any replies?

Probably because of the obscure title, add sth. like MSI laptop...

Anyway, "OpenGL 1.1" indicates that the driver has not been properly installed.
Check device manager for errors

Error codes in Device Manager in Windows - Microsoft Support

Lists the error codes that may be reported by Device Manager and the possible resolutions in Windows.

support.microsoft.com

 

[Nike_486DX]

I have a P65 Creator with an Nvidia GTX 1050Ti GPU.

The 1050 doesn't work to work at all. Connecting the HDMI just causes hanging and constant switching between the integrated Intel GPU and back to the 1050.

I have had a similar problem recently with my Ge75 but this was caused by old battery + underpowered charger. However I the original 19.5V 180w charger for this laptop and have tried with a 20v 260 watt charger for the GE75 as well and it doesn't remedy it (I'm presuming that tryign with a 20v charger instead of 19.5 volt wouldn't itself casuse a problem).

Here are the screen shots and monitoring hml from afterburner and GPU-Z.

Have done a bunch of stuff (reflash original vBIOS, update BIOS and EC firmware, reinstall windows, change driver, change RAM, underclock, replace thermal paste for CPU and GPU but left existing thermal pads on other components).

Is this GPU dead? I gather that the 'reballing' service offered by some stores nearby is a non-fix, as if it's a dead GPU the problem just resurfaces soon after.

Thanks.

It could be that you damaged the dgpu while reassembling the laptop (for maintenance maybe, its a common issue i usually see when someone brings in an asus tuf/ hp omen/ lenovo legion or other notebook for repairs. Chipped gpu die is what i usually see. Please omit all of this if you had never touched the internals of your laptop and its still got the factory seals. What looks weird to me, at least according to the graphs, is the massive difference in temperature readings, on core 4 you have 64C and on core 3 you have 91C, thats 27C difference, to me that sounds like a bad mount. And because there is one heatsink assembly for both cpu + gpu, there could be something else wrong on the inside.

btw keep in mind thats a 6c 12t cpu, so there are 12 duplicated readings, because they are 2 threads on each core.

 

[BCN1000]

It could be that you damaged the dgpu while reassembling the laptop (for maintenance maybe, its a common issue i usually see when someone brings in an asus tuf/ hp omen/ lenovo legion or other notebook for repairs. Chipped gpu die is what i usually see. Please omit all of this if you had never touched the internals of your laptop and its still got the factory seals. What looks weird to me, at least according to the graphs, is the massive difference in temperature readings, on core 4 you have 64C and on core 3 you have 91C, thats 27C difference, to me that sounds like a bad mount. And because there is one heatsink assembly for both cpu + gpu, there could be something else wrong on the inside.

btw keep in mind thats a 6c 12t cpu, so there are 12 duplicated readings, because they are 2 threads on each core.

By bad mount do you mean a chip which is badly soldered onto the board? Yes I have disassembled to add RAM previously but didn't take off the heatsink.

Probably because of the obscure title, add sth. like MSI laptop...

Anyway, "OpenGL 1.1" indicates that the driver has not been properly installed.
Check device manager for errors

I can confirm there are no errors in device manager under this adapter.
I also notice that it no longer says OpenGL 1.1 and after I made the various above repair attempts it now reads OpenGL 4.6, but there is no change in the symptoms. Could you explain why OpenGL 1.1 was relevant? Does an outdated version there indicate that the driver wasn't installed properly?
I can't edit the title it seems.

 

[StefanM]

Could you explain why OpenGL 1.1 was relevant?

That's the capability of the "Microsoft Basic Display Adapter", which takes over in case of codes 43 and 52.

 

[BCN1000]

That's the capability of the "Microsoft Basic Display Adapter", which takes over in case of codes 43 and 52.

So does this mean the card is throwing one of those errors necessarily? Perhaps there is an error log file somewhere I could upload here?

 

[StefanM]

So does this mean the card is throwing one of those errors necessarily? Perhaps there is an error log file somewhere I could upload here?

The log can be found at device manager -> properties of GPU -> events -> show all events

In case of successful installation it should read status=0

 

[BCN1000]

In case of successful installation it should read status=0

Status = 0x0
Problem=0x0

In administrative events system log, when running the GPU-Z test, an event with ID 4109 is generated saying GPU-Z was blocked from accessing hardware.

Installation and uninstall logs are OK (device was removed; device was configured Message ID 400; device was initiated, etc.)

 

[BCN1000]

Anyone seen a problem like this before?

 

[chaoshusky]

By bad mount, they mean the cooler isn't mating properly with the surface of the dies.. However, this is an MSI laptop, i've seen many turn faulty including one that was DOA, had to 'warm up' for a couple of minutes, then hard power it off and back on and it would POST/Boot.. In theory a re-ball/reflow may actually fix it for a while if it isn't the VRM circuitry or other problems causing the fault, or you may have to manually force a driver to install. Possible for a bad cooler mounting to cause this as well, lets just hope nothing broke away from the mounting mechanism.. Have you checked, re-pasted and re-seated the cooling heat-sink?

Sadly the main problem here though is its MSI, don't mean to sound like a jerk but...avoid! We also aren't aware of your ability level so if you're worried you might fudge something up, don't attempt to remove the heatsink!

 

[BCN1000]

...or you may have to manually force a driver to install.

Which driver? As stated in the OP, I have tried different drivers and as stated above, the system event log indicates no driver installation error.

By bad mount, they mean the cooler isn't mating properly with the surface of the dies...

If this were the cause, would there not be high temperatures on the chip temperature readings?

 

[BCN1000]

This thread is a month old and the only instructions I've received so far are to review the GPU driver events log. Can anyone offer further instructions to diagnose the problem please? I have basic tools + multimeters, soldering/hot-air rework station and could buy further tools if necessary.