Jump to content

my galax rtx 3090 artifacting

dezan

i just broke my rtx 3090, it because I'm changing the thermal pads to the cheap one, i think all thermal pads are all the same, once i turn on my pc the temps immediately rise to 87 degree on start menu, and i immediately turn it off, now I've changed the thermal pads to thermal odyssey, but now it's heavy artifacting😭 it is heavy artifacting now, I can't even shut down my pc through windows, i need to manually pressed, and i void the warranty too, is there any hope for me? thanks

Screenshot_20210725-231648.jpg

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, dezan said:

i just broke my rtx 3090, it because I'm changing the thermal pads to the cheap one, i think all thermal pads are all the same, once i turn on my pc the temps immediately rise to 87 degree on start menu, and i immediately turn it off, now I've changed the thermal pads to thermal odyssey, but now it's heavy artifacting😭 it is heavy artifacting now, I can't even shut down my pc through windows, i need to manually pressed, and i void the warranty too, is there any hope for me? thanks

Screenshot_20210725-231648.jpg

Why did you change the original thermal pads?

Link to comment
Share on other sites

Link to post
Share on other sites

If the GPU temp (not memory or VRM) went to 87 degrees, you messed up cooling the die as well - that needs to be done by a thermal paste (although there are other options as well, they aren't that widely used)

If you are lucky, you might be able to fix this by installing proper thermal pads. If not, some components got permanently damaged.

HAL9000: AMD Ryzen 9 3900x | Noctua NH-D15 chromax.black | 32 GB Corsair Vengeance LPX DDR4 3200 MHz | Asus X570 Prime Pro | ASUS TUF 3080 Ti | 1 TB Samsung 970 Evo Plus + 1 TB Crucial MX500 + 6 TB WD RED | Corsair HX1000 | be quiet Pure Base 500DX | LG 34UM95 34" 3440x1440

Hydrogen server: Intel i3-10100 | Cryorig M9i | 64 GB Crucial Ballistix 3200MHz DDR4 | Gigabyte B560M-DS3H | 33 TB of storage | Fractal Design Define R5 | unRAID 6.9.2

Carbon server: Fujitsu PRIMERGY RX100 S7p | Xeon E3-1230 v2 | 16 GB DDR3 ECC | 60 GB Corsair SSD & 250 GB Samsung 850 Pro | Intel i340-T4 | ESXi 6.5.1

Big Mac cluster: 2x Raspberry Pi 2 Model B | 1x Raspberry Pi 3 Model B | 2x Raspberry Pi 3 Model B+

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, dezan said:

i just broke my rtx 3090, it because I'm changing the thermal pads to the cheap one, i think all thermal pads are all the same, once i turn on my pc the temps immediately rise to 87 degree on start menu, and i immediately turn it off, now I've changed the thermal pads to thermal odyssey, but now it's heavy artifacting😭 it is heavy artifacting now, I can't even shut down my pc through windows, i need to manually pressed, and i void the warranty too, is there any hope for me? thanks

Screenshot_20210725-231648.jpg

 

You either static shocked the card or you ripped off a cap or some component.

Also temps of 87C on the start menu dont happen because of bad or even missing thermal pads.  They happen because of bad or incorrect thermal paste or mounting pressure on the core. 

Were you working on carpet ?
Is this your first time taking apart hardware like this?

Link to comment
Share on other sites

Link to post
Share on other sites

54 minutes ago, Falkentyne said:

 

You either static shocked the card or you ripped off a cap or some component.

Also temps of 87C on the start menu dont happen because of bad or even missing thermal pads.  They happen because of bad or incorrect thermal paste or mounting pressure on the core. 

Were you working on carpet ?
Is this your first time taking apart hardware like this?

yeah I'm working on carpet, I'm also touching the chip without gloves, it is my first time doing this

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, jj9987 said:

If the GPU temp (not memory or VRM) went to 87 degrees, you messed up cooling the die as well - that needs to be done by a thermal paste (although there are other options as well, they aren't that widely used)

If you are lucky, you might be able to fix this by installing proper thermal pads. If not, some components got permanently damaged.

i already reapplying new thermal paste and paste, it keeps getting worst by the day

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, Mel0nMan said:

Yeah why did you change the thermal pads? You definitely broke the card by doing so.

this galax rtx 3090 are too hot, like 80 degree while gaming

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, Murasaki said:

Memory modules could be fried. F

it is looking so bad, can't even start any games

Screenshot_20210725-231636.jpg

Link to comment
Share on other sites

Link to post
Share on other sites

58 minutes ago, dezan said:

yeah I'm working on carpet, I'm also touching the chip without gloves, it is my first time doing this

Don't work on carpet next time.  Take your card apart in the kitchen on a countertop.

It also helps to take off your socks and shirt as well and go barefoot.

And even before touching the card or PCB, touch your stove or refrigerator first.  Then you won't destroy your next card.

I've taken multiple cards and laptops apart for thermal paste/pads and have never destroyed one.

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, dezan said:

it is looking so bad, can't even start any games

At this point, I'd take it to a PC shop and ask them to double check it for me.

Link to comment
Share on other sites

Link to post
Share on other sites

7 hours ago, Falkentyne said:

Don't work on carpet next time.  Take your card apart in the kitchen on a countertop.

It also helps to take off your socks and shirt as well and go barefoot.

And even before touching the card or PCB, touch your stove or refrigerator first.  Then you won't destroy your next card.

I've taken multiple cards and laptops apart for thermal paste/pads and have never destroyed one.

i just found out something weird about this gpu, i mined a bit on it, one i started mining, gpu is going normal without any problem whatsoever, i can even game normally on it, but once i stopped mining, the gpu is artifacting and can't play games, it is weird

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, dezan said:

i just found out something weird about this gpu, i mined a bit on it, one i started mining, gpu is going normal without any problem whatsoever, i can even game normally on it, but once i stopped mining, the gpu is artifacting and can't play games, it is weird


Post a screenshot of your HWINFO64 GPU Temp, Hotspot, Memory junction and *ALL* power and *ALL* voltage GPU rails expanded in hwinfo64.  Don't forget to expand the values so all of them are shown.  There are a lot of values.

 

Also if you have 91% or better isopropyl alcohol, take apart the card, completely drown it (Submerge) in 91% alcohol, let it sit there for 10 minutes, then clean the card fully and completely with a lint free coffee liner or microfiber cloth.  99% isopropyl alcohol is MUCH better to use for this though, but it's not cheap.  

 

Make sure the card is fully and completely dried.  Then you will need new thermal paste and thermal pads.  Make sure the pads are the correct thickness and the spacing.

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Falkentyne said:


Post a screenshot of your HWINFO64 GPU Temp, Hotspot, Memory junction and *ALL* power and *ALL* voltage GPU rails expanded in hwinfo64.  Don't forget to expand the values so all of them are shown.  There are a lot of values.

 

Also if you have 91% or better isopropyl alcohol, take apart the card, completely drown it (Submerge) in 91% alcohol, let it sit there for 10 minutes, then clean the card fully and completely with a lint free coffee liner or microfiber cloth.  99% isopropyl alcohol is MUCH better to use for this though, but it's not cheap.  

 

Make sure the card is fully and completely dried.  Then you will need new thermal paste and thermal pads.  Make sure the pads are the correct thickness and the spacing.

 

ss.png

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, dezan said:

 

ss.png

i realize this gpu now cannot be overclocked, can't go pass 300w now

Link to comment
Share on other sites

Link to post
Share on other sites

18 minutes ago, dezan said:

i realize this gpu now cannot be overclocked, can't go pass 300w now

You didn't do what I told you to do, even though I said it *TWICE*.

I'm sorry if English isn't your first language but you really need to listen better.

 

Please EXPAND the POWER (WATTS) and VOLTAGE collapsed menus.  You have them collapsed.  Look at the arrows on the left side....

Link to comment
Share on other sites

Link to post
Share on other sites

8 minutes ago, Falkentyne said:

You didn't do what I told you to do, even though I said it *TWICE*.

I'm sorry if English isn't your first language but you really need to listen better.

 

Please EXPAND the POWER (WATTS) and VOLTAGE collapsed menus.  You have them collapsed.  Look at the arrows on the left side....

is it like this? i don't really understand this, sorry

ss.png

Link to comment
Share on other sites

Link to post
Share on other sites

6 hours ago, dezan said:

is it like this? i don't really understand this, sorry

ss.png

Yes, great.  But I'm in bed now.

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, Falkentyne said:

Yes, great.  But I'm in bed now.

ok thanks bro, see you soon

Link to comment
Share on other sites

Link to post
Share on other sites

12 hours ago, dezan said:

is it like this? i don't really understand this, sorry

ss.png

 

Your memory power draw is 200-207W.

Memory cannot draw this much power.

You have a problem with your card.  Probably something is shorted on the PCB or damaged.

You should clean the card PCB fully and completely (submerge it 15min then clean) with 99% isopropyl alcohol.

 

Link to comment
Share on other sites

Link to post
Share on other sites

Guys, maybe he just mounted everything wrong and its just overheating… we'd need pics of the pcb, power delivery ,vrms, etc… should be visible if something broke.

2 hours ago, Falkentyne said:

Memory cannot draw this much power.

I admit that sounds hella much but how do you know this? That info is not on google… the best guess i found is "probably something like 1 watt!"  🤣

 

Does nvidia disclose the vram power draw or?

The direction tells you... the direction

-Scott Manley, 2021

 

Softwares used:

Corsair Link (Anime Edition) 

MSI Afterburner 

OpenRGB

Lively Wallpaper 

OBS Studio

Shutter Encoder

Avidemux

FSResizer

Audacity 

VLC

WMP

GIMP

HWiNFO64

Paint

3D Paint

GitHub Desktop 

Superposition 

Prime95

Aida64

GPUZ

CPUZ

Generic Logviewer

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, Mark Kaine said:

Guys, maybe he just mounted everything wrong and its just overheating… we'd need pics of the pcb, power delivery ,vrms, etc… should be visible if something broke.

I admit that sounds hella much but how do you know this? That info is not on google… the best guess i found is "probably something like 1 watt!"  🤣

 

Does nvidia disclose the vram power draw or?

Because many people have tested this before.

Memory drawing 200W only happens because of a 1) bad VRM/trace. 2) faulty shunt mod.   Period.

That's a false reading.  If the memory GDDR6X were actually drawing 200W, it would be uncoolable without an active backplate mod.

 

There is also a bios editor where various bioses power values were dumped.

The memory power limit varied between 82W to 125W.  Memory at higher than 125W causes a "TDP Normalized%" power limit.

 

BTW I did a quick mining benchmark earlier today, on my 3090 FE, at 400W power limit (114%), and +150 core, +600 memory.

My memory power was about 126W.  
So his card is defective somehow or broken.

 

Here is a person who didnt mod his PCIE slot shunt when shunt modding.

 

https://www.overclock.net/threads/rtx-3090-founders-edition-working-shunt-mod.1773780/page-4#post-28656402

 

Here is what happened when he modded all the shunts.

 

https://www.overclock.net/threads/rtx-3090-founders-edition-working-shunt-mod.1773780/page-4#post-28656411

Link to comment
Share on other sites

Link to post
Share on other sites

13 hours ago, Falkentyne said:

 

Your memory power draw is 200-207W.

Memory cannot draw this much power.

You have a problem with your card.  Probably something is shorted on the PCB or damaged.

You should clean the card PCB fully and completely (submerge it 15min then clean) with 99% isopropyl alcohol.

 

if i submerged it in 99% alcohol, and dried it completely, is there any chance that this card will die? because now i still can used it.

Link to comment
Share on other sites

Link to post
Share on other sites

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 account

Sign in

Already have an account? Sign in here.

Sign In Now

×