ASUS ROG G703GX - likely about to blow up

[cucubits]

I think we're close! I've started to copy all my files to the storage as I type this but it's not looking good.

 

It started doing something very worrisome all of a sudden. For about 2 days now, every time it comes out of standby, as soon as the display comes on, it shows the whole screen full of random lines (like old computer monitors used to do with bad cables for example). It does this every time and it lasts for a few seconds, seemingly until I move the mouse. I don't know if it would recover by itself every time but so far I instinctively started moving the mouse when I noticed this. 

 

For now I didn't want to experiment with it until I save everything but I'm worried. I have to say, I'm not surprised, but I am worried. Asus build quality seems to have gone downhill a lot over the past few years. This was the first ROG laptop I actually had to mess with out of the box just to get it to a usable state (changed thermal paste and lowered clocks just to keep things from throttling under regular gaming load).

 

I'll try to get a video/picture of this happening after I save everything.

 

Btw, nothing changed when it started doing this (no driver updates, it wasn't moved from my desk at all, nothing new installed). I did however get it pretty toasty over the past few weeks daily playing Star Citizen. It's very resource intensive and gets the CPU easily to over 90C all the time.

 

 

 

Idle temps are ok... not great but ok:

 

[jaybee83]

34 minutes ago, cucubits said:

I think we're close! I've started to copy all my files to the storage as I type this but it's not looking good.

 

It started doing something very worrisome all of a sudden. For about 2 days now, every time it comes out of standby, as soon as the display comes on, it shows the whole screen full of random lines (like old computer monitors used to do with bad cables for example). It does this every time and it lasts for a few seconds, seemingly until I move the mouse. I don't know if it would recover by itself every time but so far I instinctively started moving the mouse when I noticed this. 

 

For now I didn't want to experiment with it until I save everything but I'm worried. I have to say, I'm not surprised, but I am worried. Asus build quality seems to have gone downhill a lot over the past few years. This was the first ROG laptop I actually had to mess with out of the box just to get it to a usable state (changed thermal paste and lowered clocks just to keep things from throttling under regular gaming load).

 

I'll try to get a video/picture of this happening after I save everything.

 

Btw, nothing changed when it started doing this (no driver updates, it wasn't moved from my desk at all, nothing new installed). I did however get it pretty toasty over the past few weeks daily playing Star Citizen. It's very resource intensive and gets the CPU easily to over 90C all the time.

 

 

 

Idle temps are ok... not great but ok:

 

likely gpu related, could be defective vRAM. 

[cucubits]

14 minutes ago, jaybee83 said:

likely gpu related, could be defective vRAM. 

 

Any specific tests I could try to narrow it down?

[Tenoroon]

25 minutes ago, cucubits said:

 

Any specific tests I could try to narrow it down?

Could either be damaged display cable or defective vRAM/memory controller as Jaybee said. 
 

MATS is a good way to test for defective vRAM. Watch this video and it will help you diagnose your problem. Also, use the exact same version of RUFUS that this guy uses. Newer versions tend to not work right when partitioning the flashdrive in the desired way for some reason.
 

 

[cucubits]

Thank you. I'll do that.

 

As soon as all the data is transferred, first I'll let it go to standby to see if the problem can indeed be replicated and then I'll try to diagnose.

[cucubits]

 

[cucubits]

For some reason I just can't get it to boot from a usb drive running mats. I set up a couple drives exactly as the dude says in the video above but it just won't boot from them. The drive shows up in the boot menu in the bios, I set it first priority but it still just loads windows...

[1610ftw]

You may want to connect it to an external monitor to see if it is anything display related but from your error description I doubt it.

 

I hope you get this sorted out somehow as I do not even want to know what Asus would charge for a motherboard with CPU AND GPU!

[cucubits]

12 minutes ago, 1610ftw said:

You may want to connect it to an external monitor to see if it is anything display related but from your error description I doubt it.

 

I hope you get this sorted out somehow as I do not even want to know what Asus would charge for a motherboard with CPU AND GPU!

 

Yeah, if it gets to that, I'm not even considering replacing the motherboard.

 

Not really sure what to try next... still no luck booting from the USB drive with mats but I think I know why. I can't find a way to disable EFI boot from the bios. There is no CSM/legacy boot setting. I did find "Secure Boot" but it still doesn't work if I set that to disabled.

 

It's pretty strange it does this only when coming out of standby/powering on. Once the artefacting goes away after a few seconds, it all works, doesn't do it again.

Also,  it does this only after it's been off for at least a few mins. If I just hit sleep/power off/on, it will work fine. If I wait a few mins before powering it back on, it does it again.

 

I really don't want to take it apart to replace the thermal paste but it may come to that...

[1610ftw]

Might be an idea to contact Asus about it. If it was temperature related then I would expect more of an unpredicatable on/off behaviour. Maybe really some driver / windows related stuff or windows corruption? In any case this is so specific that there is a chance somebody at Asus may know.

 

Would you be able to go back to a backup of the system before all that happened by any chance?

 

[cucubits]

No backup, I usually disable system restore - please don't ask why 🙂

 

I did finally update the nvidia driver, it's been quite a few months since I did that and this morning when it came out of standby, it seemed fine. More testing to follow but it's possible it was just something corrupt.

[cucubits]

11 hours ago, cucubits said:

No backup, I usually disable system restore - please don't ask why 🙂

 

I did finally update the nvidia driver, it's been quite a few months since I did that and this morning when it came out of standby, it seemed fine. More testing to follow but it's possible it was just something corrupt.

 

No change. Besides this morning at the first startup which was clean, then throughout the day, every time it came out of standby, it did the thing again. Every time the same way, a few seconds after coming out of standby. Unfortunately I don't have time to deal with this for the next week or so. Please do keep suggestions coming and we'll figure this out eventually.

[Tenoroon]

23 hours ago, cucubits said:

 

Yeah, if it gets to that, I'm not even considering replacing the motherboard.

 

Not really sure what to try next... still no luck booting from the USB drive with mats but I think I know why. I can't find a way to disable EFI boot from the bios. There is no CSM/legacy boot setting. I did find "Secure Boot" but it still doesn't work if I set that to disabled.

 

It's pretty strange it does this only when coming out of standby/powering on. Once the artefacting goes away after a few seconds, it all works, doesn't do it again.

Also,  it does this only after it's been off for at least a few mins. If I just hit sleep/power off/on, it will work fine. If I wait a few mins before powering it back on, it does it again.

 

I really don't want to take it apart to replace the thermal paste but it may come to that...

The Legacy boot thing could be the problem. Is there a way in ASUS BIOSes to unlock settings like MSI laptops? Maybe you can figure out how to turn on legacy boot. 

If not its no big deal, I could see if I can re-create the problem on another ASUS I know of and see if I can get a fix.

Just keep exhausting all options if you can, while I don't care for ASUS's laptops all too much, I hate seeing good hardware being thrown away/destroyed because it has prematurely failed. Good luck!

[cucubits]

Sadly no updates on this. I have been away for a while so the only new thing I tried was a clean GPU driver reinstall (ran DDU before) but it still happens. Same as before, only when waking up from sleep, only for a few seconds and then it's all fine.

 

 

[cucubits]

It's been quite a while since the last update on this story.

 

Sadly it looks like this problem is not one of those that goes away if left alone for long enough. I tried. 😆

 

The laptop was off for over a month during our move and I just unpacked it a couple days ago. Slightly different symptoms now. It does the usual screen goes crazy routine when waking up but it lasts much longer than a few seconds. Last night it would do it constantly for over a few minutes and it did not stop. I powered it off. When powering it back on, it worked for a while, I even started a game but after about 10 mins of playing, it slowly started showing worse and worse lines/artefacts to the point the whole screen filled up. I had to power it off again.

 

Strangely, during a reboot the screen was showing these lines right from the start - this is also new - and I would assume this rules out software. Windows wasn't even loaded at that point.

 

My plan for now is to take it apart, clean it and re-apply thermal paste. The problem here is that I can't find it. Haven't unpacked all the boxes and I have no idea where I stashed my paste, lol.

[cucubits]

Found the paste, started taking it apart. 

 

Maaan, liquid metal after about 3 years looks nasty. Completely dry and it left almost like a rough crystalized surface on the heat sink. It's late now, I'll sleep on it and figure out a way to clean it off tomorrow... likely dremel or sand it off while trying to keep the surface straight. I might apply something else when I put it together... I doubt the surfaces are going to be good enough for LM this time.

 

 

[Tenoroon]

5 minutes ago, cucubits said:

Found the paste, started taking it apart. 

 

Maaan, liquid metal after about 3 years looks nasty. Completely dry and it left almost like a rough crystalized surface on the heat sink. It's late now, I'll sleep on it and figure out a way to clean it off tomorrow... likely dremel or sand it off while trying to keep the surface straight. I might apply something else when I put it together... I doubt the surfaces are going to be good enough for LM this time.

 

 

I think that's your problem; having LM on the GPU...

LM on GPUs is almost never a good idea unless it is applied well and has very good barriers. Because a GPU die is so big, small clumps of LM like to sometimes move around (especially on a laptop where mounting pressure is low) and sometimes fall of the die. When LM escapes the die and substrate, thats where everything goes wrong as cleaning it is risky and difficult. If one little drop gets under the substrate, the GPU is dead as it is basically impossible to get out.

[cucubits]

I was open to the possibility that LM might have had something to do with this but no, that wasn't it.

 

If LM would've been the culprit, we'd have seen the laptop start acting up a long time ago. I applied it almost 3 years ago and I knew what I was doing. I didn't use too much and I protected the sections around the CPU/GPU very well. Started with healthy layers of conformal coating. The chances of something shorting out components around there were extremely low. I also knew this would be used as a desktop ...laptop only. It never moved from my desk.

 

Having said that, I cleaned up the heatsink and dies really well and applied LM again. I still had some of the original Conductonaut left. I should've done this a long time ago. With the fans also cleaned up, fresh LM, it's like a whole new machine. Way more quiet and so far I couldn't push CPU/GPU temps above 75C but I did only fire up FarCry6. It never ran at these low temps in-game before.

 

 

 

Now for the fun part. We did figure out the problem, and by we I mean you guys and me being a knucklehead for having ignored one specific advice.

 

Yes, I finally connected an external monitor and it works perfectly fine. No issues, no artifacts on wake, nothing. As soon as I switch to its display, it goes crazy again. Easy to replicate, after every sleep/wake cycle like before. Now that we narrowed it to the display, I started poking around it and pressing/touching the lower section of the laptop screen does actually make a difference on how the artifacts show up. The bezel on the bottom of the screen is also pretty toasty but I never touched that area before so I don't know if it's normal. From some pics online it looks like it has a 40pin connector on the back of the screen. I did not disassemble it, the bezel is full of "don't touch" metallic tape under the plastic cover. I didn't want to mess with it but I may as well try, maybe I just need to re-seat the connector, if it's a connector and not soldered on. 

 

I could also just buy a replacement display, but, to be honest, I always wanted to move to an external monitor and 24'' vs 17'' is quite a compelling argument towards just buying a nice 144hz monitor. Right now I have my 24'' work monitor hooked up and while it's working fine, it is only 60hz which is quite noticeable in games.

 

So yeah, that's where we are now 🙂

 

 

On 6/27/2022 at 11:22 AM, Tenoroon said:

Could either be damaged display cable or ....

 

On 6/27/2022 at 11:10 PM, 1610ftw said:

You may want to connect it to an external monitor to see if it is anything display related ....

 

[1610ftw]

Glad you got it repasted and that it is "only" an issue with the internal display. Your decision of course what to do next but you might want to at least keep it in some kind of shape that allows you to take it traveling as after all that is what a laptop is for 🙂