Common manufacturing defects or faults in GPU's.

[up667664]

My GPU (gtx 1060 6gb) recently failed just outside of the 1 year warranty the supplier provided. I went and had a diagnostics test done at a local repair shop and it stated that the card was faulty and "This would appear to be a typical GPU failure most likely from a faulty component or manufacturing defect."

What and how could this faulty component or manufacturing defect only become apparent after a year of use and being well looked after?

 

[Phazoner]

Maybe you just had bad luck, maybe your PSU or motherboard weren't correctly managing electricity or maybe your GPU was getting older very, very fast because of overheating or overvolting.

 

[up667664]

That's not answering my question, I asked what faulty components or manufacturing defects are common in GPU's that might present after a period of time, not "What did I do to break my GPU" I've already had a professional state that I was not at fault.

 

[Dunlop0078]

Well I would not say there is any "common" reason a 1060 would fail so soon. What 1060 is it specifically?

Common causes of GPU failures I have seen are typically relegated to the VRM, a cap or some other component fails.

This professional did not tell you what failed? Just called it a typical GPU failure and said it wasn't your fault? Well like I said a 1060 failing that early is not typical at all.

 

[up667664]

The card is a "Gigabyte GeForce GTX 1060 Mini Gaming 6GB GDDR5 VR Ready Graphics Card"

 

[spdragoo]

Not sure what the repair shop meant by "typical" defects. If there were "typical" defects that were that common in GPUs, you would either a) see more news stories identifying them, or b) you'd see more stories about "Avoid company X because their GPUs tend to have defective parts".

 

[Karadjgne]

Common faults with any gpu are the exact same as with any electronic device. Capacitor failure, resistor failure, diode failure, silicon grade in ic's etc. A gpu is made up of 1000 tiny components in different circuits and under normal circumstances generally do not fail. The causes of failure, ultimately, lie in outside influences, as mentioned irregular voltages or current, overheating, user abuse through OC or fan curve settings, environmental abuse from lack of airflow, can even be as stupid as a wire or dirt preventing the fan from spinning and supplying the heatsink.

So actual, spontaneous failure is rare, but failure by outside influence is far more common.

 

[ss202sl]

The word typical was added to make it seem like they knew more than you. if you take the word typical out of his statement, then you just have a failed GPU.

What I fon't know is how they were able to tell that the failure was caused by a faulty component or defect without actually identifying the specific component or defect.

 

[Peter Martin]

you down-voted one of the good answers. yes, indeed, a failing PSU can and will destroy hard drives and GPU cards... even memory and cpu, etc...if your card burned up in just over a year, something is wrong with either, their parts on the card or yours in the PC, and you have to figure it out.

 

[up667664]

I'm not after specifics towards my Card. I'm just curious as to what issues could be present at time of purchase and not become apparent until further down the line.

 

[Karadjgne]

Nothing. Any issues known at time of purchase are known, it's that simple. Anything that develops later is an unknown, unforeseeable happening.

 

[smorizio]

an old engineer and F/A tech here. first to help you out you have to think of your gpu as a car. standard car 36,000 warranty. your the owner and change the oil and add gas and follow the vendor recommended servicing. then one day your driving down the road a a rod or cyclender fails in the engine. car has just over the 36,000 warranty. if you did change the oil and was not beating the car up what you had was a defect that was in the part or when the car was made. f/a tech like me look for trends and high failure rates in parts. a few years ago a lot of cheap caps from china got into everything from hard drives to cars. there were a lot of recalls and class action from those bad caps. on your part there may have bee na bad batch of parts or a bad solider joint. or someone esd shock a roll of parts. anyone even a board repair guy now or the vendor can only tell that part a failed and may have taken out b and c. why a failed most times is never found unless there a trend and it found that part z that powers the gpu card was under sized and part a was the first part to be killed from bad power. as your a little over the warranty...start with calling the vendor ask for customer relation dept. let them know the card failed one or two days out of warranty and your looking to replace the card under warranty you may have to ask for a maanger or higher. most time if these people are trained right they just do the return for you. if the vendor says no and you used a credit card some extend the warranty on things you buy if not try doing a charge back. if you got the part on amazon try contacting amazon there realy helpfull to there customers.

 

[uppercut4u]

A number of things could determine the outcome and or life of the GPU. 1- Did you buy new, used, refurb? 2- Component failure such as bad caps, resistors, GPU, vram. 3- GPU cooling, airflow, ambient temps. 4- Mini cards typically are a budget PCB smaller in size doesnt mean its more prone to failure but it typically has 1 fan for cooling as opposed to 2 which extends the life of the GPU. Sometimes you can appeal to the manufacturer to consider your case even if its outside of the warranty period. Give it a try and see what they say, IF you meet all the requirements of their RMA policy of course.

 

[Peter Martin]

crapola happens too. components give up as mentioned above. it all depends on all those components themselves being sound and manufactured to high standards, etc.... who knows and frankly, who cares? All you need to know is you probably need a new Video card