NVIDIA GeForce RTX 4090 Power Adapter Failures Can Be Blamed on User Error and Foreign Debris, according to New Investigation

Grimlakin said:
I can say for a fact I would build out my power adaptor and seat it into the card BEFORE I put the card in the case.
Click to expand...
That doesn't mean much it can still wiggle its way out if it is put under pressure. Like for me who has very limited space in my case, and even with my current card the PCIE connector is literally dragging on the fan grill when I remove or install my video card.
 
LazyGamer said:
I'm going to need help getting milliseconds out of the (s) label on the graph...
Click to expand...
I could zoom it in for you, but the major intervals are 10sec, and there are many, many more samples than 10 in between each interval marking on the graphs.

So it may not be a resolution of 1 millisecond exactly, but it's much, much better than 1 second. Hence - milliseconds. Probably somewhere around 20 msec just eyeballing it, but I don't have first hand knowledge and I didn't zoom in to count.
 
What requirements must be met for failures to no longer be blamed on user error? Serious question.

It seems like it's going to be very easy for a cable or adapter to exceed its useful service life just through "normal" use, or somehow become compromised without any reliable way for the user to diagnose a marginal connection. Parts get swapped around for troubleshooting as a matter of course, and transferred to other systems on occasion. Maybe it's a user error to buy one of these cards. ;)
 
Brian_B said:
I could zoom it in for you, but the major intervals are 10sec, and there are many, many more samples than 10 in between each interval marking on the graphs.
Click to expand...
I've got it zoomed in; I see that you could maybe make out 150mm to 200ms, but the question is what the power draw is looking like at sub-millisecond time slices.
 
Nvidia released a statement basically any melted 12vhp adaptors/cards/cables supplied in the box with your nvidia card will be honored for RMA.
 
LazyGamer said:
I've got it zoomed in; I see that you could maybe make out 150mm to 200ms, but the question is what the power draw is looking like at sub-millisecond time slices.
Click to expand...
I don’t think anything sub-millisecond is damaging cables. That’s … really really short
 
Brian_B said:
Sounds like we agree then.
Click to expand...
Yeah, the power draw isn't an issue - yet.

We may see power transient problems with future higher-powered Ada GPUs, (4090 Ti, hypothetically), but that's not going to be a source of cable melting unless someone ships some seriously substandard cables.
 
Tempest said:
What requirements must be met for failures to no longer be blamed on user error? Serious question.
Click to expand...
Well Asus put a part on backwards on certain of their Z690 Hero motherboards which caused them to puff some magic smole, something like that.
 
You must log in or register to reply here.
Become a Patron!
Back
Top