-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Check Your 4090's 12vhpwr Connectors To Make Sure They are in Pristine Condition
- Thread starter EastCoastHandle
- Start date
feel sorry for the people who think it's an issue effecting more than a handful of people. been using mine since the 12th or whenever they came out had 0 issues, but then I know how to fully insert a connector.
Gamers nexus have been workingon a vide for 3 days trying to get one to melt, seems it didn't happen yet.
Strange no youtubers or reviews got a melt so far when it's meant to be so common
Last edited:
Permabanned
The fact it happens at all is BS and you know it, stop choking on jacket mans sausage please.
could be user error...
one one of them the guy made a photo of how his lead was inserted... then when people pointed out it wasn't even fully clicked in place he backtracked and said thats as far as it goes in after the melt LOL....
One of the others was clearly only 2 leads from the PSU going so must have been daisy chained.
A few of them are clearly with the cable so bent when everyone knew not to bend it, but then jay2cents bent on in a video and it did nothing. it's probably a combination of not being fully clicked in securely and also being bent so one side has a barely connection pin and almost no contact area where the electrons go.
people can go on about the soldering all they want but that's not the area which is melting is it
Last edited:
Permabanned
Considering Nvidia are taking no precautions shows they barely tested it or they just sent it out of the door knowing it could fail, which you picking?
There is no guide saying "Don't bend" From Nvidia..
Plus the factor of it being installed on the side of the card.
There is zero excuse.
Last edited:
Nvidia are obviously going to take it seriously...
Not from nvidia but from SIG or whomever it was, didn't all the reviews and previews mention it.
It's not even a n nvidia proprietary connection so they could easily push the blame at intel
Last edited:
Permabanned
I never knew everyone was watching all reviews and all people who bought one are on Youtube 24/7 or any forums. Your world view is very narrow.
And this being fixed after is exactly WHAT WE DON'T WANT.
It's like me sending out below temperature food as a chef and making someone ill, oh it's ok we will tell them afterwards, no GET RID.
Last edited:
Yep, Steve says he's tried a bunch of different theories out but no melting yet
It's kind of a problem that no one can seem to replicate the issue, it means it's more complicated than first thought. But on the positive side, it means it's also not very common
Last edited:
yea people are spending 1600-2300 on a gpu blindly now apparently, no research at all, never watched a single video about them.
copium
Last edited:
he must be doing it wrong according to the people on these forums that hate nvidia for pricing them out of a gpu, maybe they could get on twitter and give him some tips
Last edited:
Permabanned
I want to punch your head in.
Nvidia THEMSELVES HAVE NOTHING, Why is this the case?
Nvidia is selling products to customers with zero warning.
Relying on Youtubers?
My god you will make any excuse won't you?
Sit down and take it deep from jacket man.
Last edited:
Soldato
so you think people blindly buy a 4090? you know how hard it is to even find one in stock since the 12th right? did they just happen to stumble upon one instock somewhere and think I'll take it @ £1600-2300 without knowing anything? what are you smoking pal? the fumes from a burning connecting pin or what?
Enthusiasts just take a random punt on anything now apparently.
How lucky these people are, they didn't even need to rely on an app to tell them about drops
Last edited:
Permabanned
You think people don't buy them completely blindly?
You also think people watch every single review?
You also think people should show a distrust in the manufacturer when all their prior GPU's worked ok?
You seem to think only enthusiasts buy it lol, you are so tied to status and what you own you absolute joke.
Is your head on straight? I would love to fix it.
No one by your mind is capable of making a mistake by being blinded by the flair and marketing by Nvidia.
I say no more.
Last edited:
Imagine threatening violence over a difference of opninion on the internet twice now in as many posts.
This guy must have some real issues with trying to find a 4090 in stock somewhere.
you think these people would be totally oblivious to the internet too? they have 0 friends who would tell them about the "problems"
- Joined
- 8 Jun 2018
- Posts
- 2,827
Not the same thing in this context. AMD/Intel didn't provide an essential accessory to make their CPUs work correctly.
Nv mandated that you use "their" adapter which has turned out to be faulty. And threaten to deny any warranty claims if they find out that you used any 3rd party solution. I could care less of what you think the warranty means. This makes the whole situation a lot worst optically. Which is why they need to address it. Sooner rather then later. For starters, they should pivot on their website the wording of the warranty to include some sort of alternate solution since it's been proven that "their" adapter was not designed correctly.
Exactly.
Last edited:
So I just visually checked my adapter for any sign of damage from melting etc, nothing to be seen, looks like new
Associate
there seems to be a pattern to this with no FEs being reported yet, though lets wait for the sample to reach 30-50
I also kind of see the pattern that "inferior" 600W boards are affected more than others.
Its mainly TUF and Gigabyte, the only inferior boards allowing 600W PL. No Strix, no FE.
Of course, Suprim X cases contradict this, but some pattern is there.
Also, it seems strange that all partner boards have worse power efficency at idle and worse transistent spikes than FE, even if they seemingly use the same power delivery with more phases(Strix/Suprim use the same voltage controller with more power phases).
This and the fact that no FE has melted seems to point to the fact that FE is still somehow using superior board layout than the rest.
Last edited:
Associate
i think the defect is on the onboard connector, given the pattern of failures, like how the exposed ends are showing burn marks and not the cable side
i also had a look at the stripped down schematics of the adapter yesterday, what they have done is soldered the wires on a single plate like contraption
so lets for a moment assume that the solder's an issue so we have 2 scenarios
a) a clean separation at the solder: the current flowing through joined wires increases, while the current going through the plate/pins should still be constant.. so there should be burn marks on connected cables
b) an unclean separation that causes voltage drop near the solder for the loose cable: again we should have burn marks cable side but this time it would be visible on the loose cable not the ones that are properly connected
i also doubt if power spikes could do that because a large intermittent shock (like a dirac delta function) wont be able to raise temps for such a thing to happen
GN is doing some lab tests, hopefully we should know something before nvidia announces their findings
i also had a look at the stripped down schematics of the adapter yesterday, what they have done is soldered the wires on a single plate like contraption
so lets for a moment assume that the solder's an issue so we have 2 scenarios
a) a clean separation at the solder: the current flowing through joined wires increases, while the current going through the plate/pins should still be constant.. so there should be burn marks on connected cables
b) an unclean separation that causes voltage drop near the solder for the loose cable: again we should have burn marks cable side but this time it would be visible on the loose cable not the ones that are properly connected
i also doubt if power spikes could do that because a large intermittent shock (like a dirac delta function) wont be able to raise temps for such a thing to happen
GN is doing some lab tests, hopefully we should know something before nvidia announces their findings
Last edited: