Dead GPU warranty void

[IgorNew]

Ok so i was a little bit stupid I decided i will use my GPU RX 470 Sapphire Nitro+ 8Gb to mine ETH when i' m not using my PC. Any way I used GPU-Z to create safe copy, Polaris Bios Editor to edit Bios and Atiflash to flash it on my GPU. I did Get my core clock down to 1175 MHz with 0.965 V and 2175 MHz on memory with 1750 MHz timings. Last week i started playing Mount and Blade and found out i can't mine and play it in same time so i aborted mining for that period. Any way after about a month of mining today when i came back from work i found out my PC was at black screen with no signs of life. I took out GPU and tried to use igpu but it was mess in the end i figured out and now i' m on igpu.

I have no idea what is with my GPU but i' m positive it is not bricked as it is dual Bios card (i did not play with other bios) and other bios does not work as well. Any way monitor shows no card detected msg "can't recall word to word message but you know what I mean" Bios does not see it fans on it don't spin and led sapphire text don't turn up when i turn on PC. Normally my
Asus Z170-a don't see igpu when there is gpu installed now it see igpu and don't see gpu.

I guess I voided my warranty but question is: Is it even worthy trying to get repair / new card from my warranty will they notice bios change?

I mean looking from that point of view high temps did not kill it (during night to lower the noise i did lower fan speed to keep gpu in 82-85 range and during day rise it to have my gpu in 75is range.

Also if my "bad" bios is problem then i would still be able to use other one no?

So what do you thing will they fix it and send it (or new card) or i' m <Removed> up now?

Point of the sorry kids don't do drugs and stay away from grey warranty area.

System if it means anything
i5 6600K stock speeds most of the time.
Asus Z170-a
Antec HCG 620W
Kingston HyperX 2x4 GB DDR4 2666 MHz
Seagate and WD hdd and Kingston SSDnow 120GB
Sapphire Nitro+ RX 470 8GB (past tense)

 

[bignastyid]

Please watch the language profanity is not allowed here.

 

[IgorNew]

Ok will look on it for now on

 

[johnnyq1233]

You probably burnt your card... and yes they'll notice the bios change.
Mining generates a lot of heat and probably should have had fans at 100%.
I personally don't mine and like my GPU ready to do what it was designed to do...

 

[IgorNew]

It is designed to run at higher temps that mid 80's it's target is 75 and max 90 it auto shut down at 94 and my tems are rarely above 82... if it is burned between target " fans don't turn on below 50 and below 70 they are naturaly below 30% its max rpm" and max safe temp then i call that bad product. Also I had higher temps playing Witcher 3 and GTA 5 than mining so... Any way do you think they can notice bios change before they fix it if it is broken as you say burned i assume you mean core is dead without taking it complete apart and testing every bit of it as they can't visual observe that one bios is changed considering it is just unlocked noting else no shady 3rd party etc. Also on their website does not mention software changes at all.
https://www.sapphiretech.com/support.asp?lang=eng

 

[johnnyq1233]

The thing is you said that it was mining while you were at work... how do you know what temp it was at? Also, the bios you flashed onto the card might disable that safety shutdown...
All you can do is send it in and hope they will repair or replace it!

 

[TMTOWTSAC]

If I had to guess, I'd say you probably fried your VRAM. They get cooled as a byproduct of the GPU and VRM temp sensors telling the fans to spin up, but don't usually have sensors on their own. By having an undervolt on the GPU and an OC on the RAM, you skewed the ratio. RAM runs hotter than usual, GPU and VRM's run cooler than usual, and it doesn't spin the fans fast enough to keep the RAM from overheating.

 

[IgorNew]

GPU Z is tracking temps all the time. You can see average max and min if for month it was average 82 max 84 (1450 rpm) during night and 75 is during day (and i place it to 2200 rpm when i woke up for job) i don't believe anything unusual would happen that would rise temp for more than few C also Pc would simply turn off if temps would go crazy anyway.

But if sending card for warranty is pointless now, is there anything i can do to try to fix it?

 

[Niruvius]

For a very long time vram had very small 1cmx1cm heatsink or no heatsink at all in a video jays2cents removed heatsink to instal water cooling and was concern about vram not being cooled then result lower temps as source of vram heat was heated heatsink they share with gpu. With that said im very confident vram did not die at least not from heat

 

[TMTOWTSAC]

For a very long time vram had very small 1cmx1cm heatsink or no heatsink at all in a video jays2cents removed heatsink to instal water cooling and was concern about vram not being cooled then result lower temps as source of vram heat was heated heatsink they share with gpu. With that said im very confident vram did not die at least not from heat

I'm not saying it definitively or ruling out other causes. Especially not with a custom flashed BIOS of indeterminate providence. But, with the GPU downclocked and undervolted, the only components under abnormal strain would have been the OC'd RAM.