Jump to content

Bad GPU

Stefan Kasev
By Stefan Kasev
in Graphics Cards
 Share
Posted

Hello guys! I have a little problem. I was using my  MSI RTX 2060 GAMING Z gpu for mining with the settings of 65% power limit,+700 mhz memory overclock, and core clock-50. The card was working fine for the first month while the problems started to apper. When i boot my PC to windows logo the screen goes black and says no signal. I managed to boot several times but i get artifacts and the screen goes black and restarts. Sometimes i get BSOD too. I tried the card in other pc and when i start to install the drivers the same thing happens, bsod, black screen, no signal etc. So i am going to RMA the card because it still has warranty. Any suggestions? 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/
Share on other sites
Link to post
Share on other sites
Posted
  • Author
Just now, emosun said:

yep that cards toast , buy another one and this time don't do that

But i read that this memory clock was normal and i didnt have any issues with testing ans gaming with it too. In the options for mining and forums there are people with way higher memory clocks? I dont think that is the problem. It may be but i didnt apply an insane clock. 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673158
Share on other sites
Link to post
Share on other sites
Posted
1 minute ago, Stefan Kasev said:

i didnt have any issues

well how about now? is there an issue now?

Most people here aren't miners but I can tell you that overclocking a component that has no temperature sensor on it and isn't designed to be run at that clock is why components die early. More than likely , your gpu did not do enough mining to pay for itself in that time period so you can see the downsides of pushing hardware past what it's designed to do to trickle in a few dollars. So yes I'd recommend not doing that in the future but I'm sure the mining community with 20 of these gpu's would say otherwise.

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673178
Share on other sites
Link to post
Share on other sites
Posted

An overclock applied for long periods of time and run at 100% (like mining) will deteriorate a GPU over time.

Hell, just running the GPU at 100% for long periods of time will degrade it.

 I have mined, folded, farmed and plotted my way through a few broken GPUs over the years and I never even OC them while running them like that.  

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673185
Share on other sites
Link to post
Share on other sites
Posted
13 minutes ago, Stefan Kasev said:

Hello guys! I have a little problem. I was using my  MSI RTX 2060 GAMING Z gpu for mining with the settings of 65% power limit,+700 mhz memory overclock, and core clock-50. The card was working fine for the first month while the problems started to apper. When i boot my PC to windows logo the screen goes black and says no signal. I managed to boot several times but i get artifacts and the screen goes black and restarts. Sometimes i get BSOD too. I tried the card in other pc and when i start to install the drivers the same thing happens, bsod, black screen, no signal etc. So i am going to RMA the card because it still has warranty. Any suggestions? 

Isn't +700mhz is too high for vram?

I think the vram got toasted, since in perfectly normal case scenarios Those ram chip will produce less heat than the gpu die and possibly overheat and burned

01110100 01101000 01100001 01110100 00100000 01110111 01100001 01110011 00100000 00110111 00110000 00100000 01101001 01101110 01100011 01101000 00100000 01110000 01101100 01100001 01110011 01101101 01100001 00100000 01110011 01100011 01110010 01100101 01100101 01101110 00100000 01110100 01110110

 

 

 

 

 

 

 

 

 

 

Audio Interface I/O LIST v2

 

 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673200
Share on other sites
Link to post
Share on other sites
Posted

You can try to RMA the gpu, (idk exactly if the rma is gonna get accepted because of an overclock fail) but here's my opinion:

When mining with expensive rigs, you want your GPUs to survive as long as possible before breaking, and when they do eventually, then moving on to buying new ones. That keeps replacing costs down. Overclocking only shortens the lifespan of a GPU and is only recommended for gamers/overclockers who don't run their card 24/7 under 100% load, so it doesn't affect them too much. Believe me, keeping a card running any amount of time more, rather than trying to increase mining performance with OC but not being sure how much longer it will survive is better.
Additionally, you applied a really high overclock that isn't recommended to run on 24/7 under 100% load

 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673211
Share on other sites
Link to post
Share on other sites
Posted
  • Author

So better run them without any oc and things, just stock? Because from what i read everyone wants better profit  but with higher profit the higher risk goes. I have MSI RTX 2080 Ti Ventus again running at 60% power limit - 200 core clock, +1100 memory clock. Maybe i  should turn them off. The card is running at these settings for second month. 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673232
Share on other sites
Link to post
Share on other sites
Posted
4 minutes ago, Stefan Kasev said:

Maybe thats what happend. I dont want to toast my gpus, thats why i ask and read about all of these things. And there were people with way higher clocks and setttings running fine. But guess its just my luck. 

No it's not your luck. Yes people can run with those oc's BUT you do not know the temp of your vram as it doesn't have a temp sensor. Hence where silicon lottery comes in once more or simply different vram. You don't know the temps so it could be that they were running at or over their thermal limit. Hence why the card died so soon. The amount of times i haven't see a dead card due to a agressive memory oc, a watercooling block that didn't cover the memory, too short thermal pads,... is too many too count.

 

Simply put if you want your components to last a long time leave them be at stock if you cannot know the temperature of all the components on the board. If you do know the temps keep them a good bit under max and chances are you can run the oc to the end of time basically.

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673244
Share on other sites
Link to post
Share on other sites
Posted
Just now, Stefan Kasev said:

So better run them without any oc and things, just stock? Because from what i read everyone wants better profit  but with higher profit the higher risk goes. I have MSI RTX 2080 Ti Ventus again running at 60% power limit - 200 core clock, +1100 memory clock. Maybe i  should turn them off. The card is running at these settings for second month. 

In the long run, yes, that's preferred. But it's really your preference whether or not you want to apply a very light OC or just run stock for longer gpu lifespan. In the end, it's your decision.

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673248
Share on other sites
Link to post
Share on other sites
Posted
2 minutes ago, Stefan Kasev said:

So better run them without any oc and things, just stock? Because from what i read everyone wants better profit  but with higher profit the higher risk goes. I have MSI RTX 2080 Ti Ventus again running at 60% power limit - 200 core clock, +1100 memory clock. Maybe i  should turn them off. The card is running at these settings for second month. 

Dude that 2080ti is going to die. 1100 on mem clock WITHOUT knowing the temps of the vram is basically BEGGING for a dead card ESPECIALLY when mining. Stay at stock or at best an underclock and leave them be. It has nividia turboboost so it will go as high as it can where it was tested.

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673251
Share on other sites
Link to post
Share on other sites
Posted
  • Author
2 minutes ago, jaslion said:

Dude that 2080ti is going to die. 1100 on mem clock WITHOUT knowing the temps of the vram is basically BEGGING for a dead card ESPECIALLY when mining. Stay at stock or at best an underclock and leave them be. It has nividia turboboost so it will go as high as it can where it was tested.

Is there any way to know the temp or vram? 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673273
Share on other sites
Link to post
Share on other sites
Posted
Just now, Stefan Kasev said:

Hope i get lucky with the RMA and get replacement with other. Is it possible for them to refuse it? 

I'm not completely sure, but there are lots of places where rmas/repairs get rejected due to strict electronics fault policies about not having insurance. We're talking about GPUs though and a certain company that's selling them. They decide. 

Link to comment
https://linustechtips.com/topic/1330095-bad-gpu/#findComment-14673354
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
 Share
Go to topic listing Graphics Cards

×