No it's not but the same basic cause of failure about the hardware can be determined. An example is when the starter in your auto goes bad and burns up, requiring a new one.
It can be caused by several things but in the end it's always one of two things - Wear and tear (age) or it got too hot and cooked, these two things being root causes of failure which can be determined. Wear and tear is a given, if it's worn out it's worn out and actually expected since it has a certain "Life Span" anyway but premature failure is another thing.
Let's run down the list of what can cause it in this example off the top of my head:
Weak/bad battery, battery getting weak because the alternator wasn't charging it, bad battery cables, weak/loose battery wiring connections, the engine's ignition timing being too far advanced making it difficult to turn it over, the vehicle always being driven short distances so the battery never recover's it charge from being started (Slowly dying as a result - Weak battery), improperly maintained battery (Dirty posts or dry cells) or the battery going bad by being too close to a heat source causing it to cook and swell up along with drying out internally....
There are many things that can be a cause of failure - True, but the failure based on the hardware is almost always the same if it wasn't engineered correctly and that can be determined in-house if they really took time to investigate it like Igor did.
A small lab solved what a big corp with all it's resources coudn't during R&D time - There is no denying it. |
To be fair, there is no way to take into account every possible scenario but at least it's entirely possible to cover most things so you won't have failures on such a large scale by the numbers.
I don't see how any of this is in any way in conflict with what I've been saying. At all. Lab testing to provoke failures allow you to identify many possible failure modes, as well as (hopefully) identify which of these are the most likely. This is a general type of troubleshooting, identifying
potential risks of various severity. Investigating actual, real-world failures allows you to identify
specific failure modes that appear in the real world. These two modes of investigation complement each other, and neither is sufficient on their own. All I've been arguing here is that there is
absolutely nothing suspicious about Nvidia asking for these adapters to be sent to them,
because specific investigation of real-world failures is a crucial step of fixing this.
Depends on the photo and camera taking the shot.
No. A photo of a failed part is a photo of a failed part - its quality and informational value is variable and can be debated, but at the very least it serves as documentation of an actual failure of some sort occurring. It might not be sufficient evidence for Nvidia being at fault, but in a civil suit, such a photo + no compelling evidence for user error + a mysteriously disappeared burnt adapter in Nvidia's custody? That'd be an easy win for the customer. Not in a criminal case, but in a civil case? Either that, or it'd see the case handed over to the relevant authorities for a criminal investigation.
Not to mention for the sake of a lawsuit photos can be "Doctored" to try and prove it as real if the one filing suit was trying to sue for the sake of getting paid - It's been done before and will happen again at some point.
Which would put the burden of proof for this on Nvidia to show that the photos were doctored if this was the case.
Of course a pic can go a long ways to that end but still, there is nothing like having the actual article physically onhand to present as evidence.
And if this was necessary, any lawsuit would require Nvidia to provide ample documentation of any investigation of the returned adapter, if not access to the adapter itself, as a part of the discovery process. Alongside all RMA and shipping logs. Witholding relevant, requested evidence from discovery is illegal, and will be sanctioned by the court, and might even result in a default judgement against Nvidia.
If there is no fried card to present, there is nothing to cover up in the first place in the legal sense of it.
See above. If a photo of a fried adapter exists, as well as a document chain confirming its submission for RMA and subsequent return to Nvidia, no court in any sensible country would accept Nvidia saying "whoops, you know, we might have misplaced that" and not follow up on that part of the suit.
You're not well versed in legal stuff or so it seems to me - I don't get why you don't understand the simple fact about evidence to support a claim.
Hearing about it is one thing, physically seeing it up close and personal is way better and stronger too than word alone.
Photos are routinely used as evidence in both civil and criminal cases thousands of times every single day, and if sued, Nvidia would need to prove beyond a reasonable doubt that any such photos were doctored - which they wouldn't be able to. Faking a low quality phone photo convincingly is
incredibly difficult. They could no doubt find some "expert" to testifythat it "might" be doctored, but then the person/group suing them would find their own expert to counter this, rendering the argument moot, word-v-word. And, of course, the lack of a photo but existence of RMA logs and shipping logs of a burnt adapter wouldn't help Nvidia anyhow.
As for the burger, yes it's still relevant in how or what the company could try to cover their asses to avoid a payout.
... except for the fact that a burger would spoil within days, and the evidence would disappear by itself unless work is done to preserve it - unlike this case, where there'd need to be actual effort put into destroying it, and a document chain showing what has happened to it. You're conflating two different ends of a vast spectrum of malfeasance here, pretending that they're one and the same. Whisking away a burger with whatever crap in it a minute or two after a customer has a disgusted outburst after finding that thing in their food is
vastly different from Nvidia asking their AIB partners to return power adapters submitted for RMA to them. If you don't see the difference between these two, you really need to take a look in the mirror regarding that "not well versed in legal stuff" claim.
I don't get why this concept eludes you, it's not about THE burger or THE card, it's about a claim being factually proven and justification of a ruling made - The end result (Ruling) they are trying to dodge is a judgement resulting in payout to the one(s) filing suit.
... that doesn't elude me whatsoever. I'm arguing that there's a vastly different basis for making evidence-backed claims in these different cases, making them essentially uncomparable. The nature of the basis of the claims, and the things involved in causing the claims, are vastly different and operate differently in material, temporal and causal ways. A GPU power adapter does not spoil or go bad if left unattended. A burger does. A melted GPU adapter literally cannot be taken from the customer by an employee before they have a chance to document the incident - which is what gives companies the opportunity to hide and destroy evidence. And, crucially, failure to uphold hygiene standards and failure to design a safe power adapter are quite different causes of failure, attributable to entirely different people and/or entities.
Seriously - Twisting the meaning of what I posted will get you exactly nowhere.
I'm not twisting it whatsoever - I'm demonstrating the logical leaps you're making in your reasoning. They're there for all to see - if you don't, then that's on you. That's why I'm pointing them out.
You're trying to make it all based on the physical rather than the concept, the concept which is my point about it.
... except that such a lawsuit, if it were to happen, would be based on
actual events, not conceptual ones. I don't care whatsoever about any abstracted conceptual argument you're making - I'm arguing that
in the real world, anyone interested in fixing a flaw with a product will collect as many samples as feasible of the failed product in order to identify and fix the flaw for future revisions.
Concept always being someone has the opinion they were slighted, screwed over, hustled, taken from unfairly which is why suits are filed but at the same time, if you know you're been screwed over then it's more than just simple opinion - You still have to have the evidence to prove it as fact to more folks than yourself alone.
Uh ... so ... what? Sorry, I
really don't follow here. You're arguing for Nvidia collecting these being suspect because
conceptually they might do so to avoid lawsuits? Is that what you're saying? 'Cause if that's the case, then
conceptually it's more likely that they're doing this to better identify the specific modes of failure found in the real world. These are not different ontological levels.
And, as I've said above: "disappearing" the adapters
wouldn't help. There would be photographs. There would be RMA logs - most likely with extensive photos attached from AIB partner RMA techs. There would be shipping logs from board partners of adapters being sent to Nvidia, which would be traceable back to specific RMA numbers. There would be tracking numbers and receipt signatures. All of which would be relevant evidence for
either the fault at hand,
or Nvidia actively working to destroy evidence of a fault. The latter of which would be rather interesting to the relevant authorities, and would land Nvidia in a lot more hot water than any civil suit. You generally don't risk criminal liability to avoid civil liability - that's pretty damn stupid.
No - It proves they can be sued exactly as said - The reason(s) why, real or imagined, faked or truthful can vary to about any degree you can name.
And how is this relevant? Is "frivolous or unfounded lawsuits exist" an argument for or against anything at all in specific? No. The entire gist of your argument seems to be that if Nvidia can make the physical adapters go away, this would all go away, which just shows a
massive blindness to how obvious this would be to any court.
Seriously, the logic underpinning this whole "Nvidia is doing this to hide evidence" spiel is woefully naive and out of touch with the actual workings of both the corporate world and legal systems. Unless they'd be able to
also delete all relevant RMA logs, including emails sent to the customer about this, it wouldn't help them at all - it would in fact harm them, as it would show clear signs of
trying to hide evidence. Which is, in fact, illegal! Whoops!