News AMD Exec Burns Nvidia Over Melting Connectors

Page 4 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

eazen

BANNED
Nov 17, 2022
25
4
35
Given the same user doing this exact same procedure dozens or hundreds of times in the past on earlier components, and now, with a somewhat different connection and more power running through the connection...and now we have "problems to the point of Flame On"....to me, that would point to a design issue.

As in....why is it so easy for seemingly experienced users to get it wrong?
What is now different?
Well as Jonnyguru and GN a bit later pointed out, only a very bad connection can cause this problem. Anything else they tried didn’t reproduce the burning issue. So I suspect that the cable (as Steve also said), is kinda hard to attach and not very well visible as well. So if you’re not that careful it can absolutely happen to not attach it properly, as in, it doesn’t “click” and you just think it’s attached. And then maybe you do some cable routing as well, and tadaa, the cable is even looser after that, leading to the burning problem of the cable or connector. The issue is that it’s a very small connector but it has a lot of smaller PINs, so it’s not as easy to attach than the 8 Pin connector.
 

USAFRet

Titan
Moderator
...kinda hard to attach and not very well visible as well. So if you’re not that careful it can absolutely happen to not attach it properly, as in, it doesn’t “click” and you just think it’s attached.
Which, to me, says poor design.

That is what testing and design iterations are for....to design out 'user error'.
 
Fact is, I don’t know why you’re attacking me here, are you one of them, or why so personal?

All problems so far had burnt connectors which all tests have proven to be by faulty connection, and can only be caused by the user. Enough said. Inform yourself better if you don’t understand what’s going on here.

lol, you should really go see GN video a few more times. Steve said faulty connection is only 1 of the reason (in fact, the only one he could reproduce). Thats far away from been the only reason.

Since you are new to the forum, I already said and wrote this many, many times, of course there will be those who maybe didn't plug the connector correctly, but I found very hard to believe all of them were so dumb and uncareful to make the same mistake.
And let say you are right, then Why is nvidia taking so long to said something?. If theres nothing else going on but user error, Why letting this go on and on for weeks?, Why did the CEO traveled from far away to get to know whats going on if this was only user error?

Enough indeed, take care.
 

eazen

BANNED
Nov 17, 2022
25
4
35
Which, to me, says poor design.

That is what testing and design iterations are for....to design out 'user error'.
Like I said in my first post, I don’t disagree. The connector could’ve been better designed or at least in a way, that the GPU is not usable unless you properly connect it first. But this doesn’t change the fact that the cable isn’t at fault, as only a few people had this problem so far. Even if I’m generous and count in unpublicized cases, I doubt it would be more than 100 people. I think and I expect that most people, and especially most enthusiasts, are rather careful with cable connections to high powered GPUs.