News Corsair’s PSU Expert Jonny Guru Weighs in on Nvidia Connector-Gate

[-Fran-]

Here.

Regards.

 

[Makaveli]

I agree that blaming the customer is not the answer, but he still may be correct about the root cause of the problem. The plausibility and/or correctness of his explanation, as far as I'm concerned, is the standard for judging his credibility. Clearly his expertise is in power supply issues, not customer relations.

His response is one I would expect from technical person. If one wants the PC answer that usually comes from marketing. And I know this because i'm a technical person myself and that is the same answer I would give. EVGA knew this was going to be a problem and they were smart for making an exit and not have NV make them look stupid.

And the blame for this is 100% on NV, AMD was also smart for avoiding this because they saw it coming. So NV knew didn't give a damn and sold it as is because they know they have a fanatical fan base that will pitch a tent infront of best buy and sleep there for nights to get first dibs at anything they release.

If there fans would have said no we are not buying this you can be sure it would have been recalled and resigned already.

 

[hotaru251]

, but he still may be correct about the root cause of the problem.

maybe, but doubtful.

the standard as a whole has had warning issues since back in Sept.
example below

Melting 12VHPWR Connectors - The Real Story - Cultists Network

WCCFTECH stated that PCI-SIG sent an email warning against using non-ATX 3.0 PSUs and 8-pin to 12VHPWR adapters. Except, this was made up...

cultists.network

then again it could just be multiple issues that build up t the major issue together.

 

[Amdlova]

That connector only made for money... now I want really bad people sues nvidia.

 

[Inthrutheoutdoor]

You know where I'd like to stick a greased up connector Jonny Guru? Ya...
Everyone is plugging in 8 pin connectors right but all of a sudden we can't plug in a new one? Please.

Nope, he doesn't get da grease... at least not the 1st (or 3rd, 7th, or 13th) time, hehehe....

To quote Dr. Smith from the original '60's Lost in Space....

"Oh the pain, the pain"

 

[logainofhades]

So he was the Corsair employee, that Corsair reached out to GN, to apologize over.

 

[thisisaname]

Maybe, instead of calling the adapter 'octopus' we should call it 'hydra' 🤣

Hail hydra? 🤣🤣😱

 

[truerock]

Wouldn’t it be fairly straightforward to replicate the problem to prove this theory?

Yes

 

[truerock]

yeah...no.
if you spend 1500$ on a gpu it shouldnt require you buying grease to PLUG something in.

if your plug is issue then the plug needs revised.
the seller's responsible for making sure its idiot-proof.

it isnt users fault the selelr wanted to make a new adapater to save space on pcb...they are massive already wouldnt of changed anything making it fit 3 or 4 8pins.

I avoid idiot proof products

I've designed information systems for decades. The users were/are generally around IQ and technical competence top 10% to top 5%. For example, a financial analyst with an MBA from UCLA. So, the system designs were for that type of individual.

I had a rare situation where I was on a team designing a public terminal for general users... a passport validation system. One member of the design team created a user registration system that had a simple IQ test, technical competence test built into the user registration system. Based on the results, the user would be setup for the idiot's system or the regular user system.

 

[ManDaddio]

Anytime you have a connector plug you should always make sure it is fastened in tightly all the way in. The fact that people can't do that or don't know that, if that's part of the problem, says something about our society today. We are not evolving.

The fact that this is still being talked about is crazy. How many reports have been made of burnt connectors? More people have probably tested the problem without replicating than have actually complained.

I am not saying there should not be a warning issued now for users/dyi-ers since an issue has been raised. But I doubt Nvidia is at fault since the cord is built to spec as far as they know. If not it's the cord manufacturer who is at fault.

That is just my thoughts.

 

[InvalidError]

Claiming the error is in insertion is BS.

Since nobody has been able to reproduce the problem on-demand with any combination of suspected material and manufacturing defects along with "normal bending abuse" yet, user error seems like a perfectly probable and likely option.

 

[Gunkk0]

Blame the customer is ALWAYS the BEST solution.

 

[cfbcfb]

You don't need dielectric grease. Just touch your finger to some vaseline or the end of a chapstick and rub it on the connectors, and wipe off any residual as it will liquify when it gets warm. Just enough to barely stick to the plug. Push it in, take it back out, wipe the connector a second time to remove any squished up extra and plug it back in.

And no, it doesn't damage the plastic at all. I've done this with 24 pin mobo connectors and other large/tight connectors for decades. Chap stick is what I usually use, because if you're in a room with 3 people, one of them has some.

Of course, "don't put too much vaseline on it before you stick it in" is advice that will serve you well on a universal basis?

Perfectly reasonable explanation right here, I've even had the 8 pin connectors get difficult on me, especially when they're really close together with other 8 pins.

 

[Papusan]

As in the article....

Nvidia Remains Quiet
It has been more than two weeks since the the first 16-pin 12VHPWR melting adapter problem reared its smoldering head. While various third parties and tech experts have come forth with reasonable theories about the underlying issue(s), Nvidia and its partners have been conspicuously quiet.

Yep, but not the buyers of the new gaming flagship from Nvidia. Some of them have already taken action

Ahdoot Wolfson is investigating reports that Nvidia’s GeForce RTX 4090 graphics cards contain a defect that causes them to melt and burn. "If you have purchased a Nvidia GeForce RTX 4090 graphics card or other Nvidia 4000 series graphics card and have experienced melting or damage at the power supply, please fill out the form below and one of Ahdoot Wolfson’s lawyers promptly will contact you."

Nvidia GeForce RTX 4090 Graphics Cards Overheating – Class Action Investigation... https://www.ahdootwolfson.com/blog/...cards-overheating-class-action-investigation/

If Nvidia hope this will blow over of itself, they should think twice.

 

[JarredWaltonGPU]

How is it BS if this could very well be the issue. What qualifications do you have to say otherwise? Again as I have posted before, he blamed nobody. He emphasized greatly that the plug was truly difficult to fully insert to the extent that someone could believe that it is fully inserted when in fact it is not. This is in no way blaming the user, if anything its implying a fault with the design of the connector itself.

The interesting thing is that 12-pin cables tend to insert easier than 16-pin, from personal experience. Given tens of thousands of RTX 4090 cards potentially in the hands of users, is it possible some of them won't push the connector all the way in? I'd say absolutely. Compared to an 8-pin or 12-pin, I probably have to push twice as hard to get the click with 16-pin. So from that perspective, Jonny Guru may be absolutely correct.

I know that every time I test a different 4090 card, I'm double and triple checking that the connection is all the way in. Because for now, that's all I can do short of just not testing the cards, and I'm not ready to do that. Also, if one of the cards I'm reviewing then has a meltdown, I can be sure that it wasn't from an improperly inserted cable.

The thing is that no one will come forward and publicly admit to having "inserted it wrong." All you get is a melted connector and a Reddit post with a picture or two after the melting has occurred. Does that mean for sure that every melted connector is from user error? No. But the lack of complaints from the past two years of 12-pin cables and adapters does make the whole situation rather suspect. It's either manufacturing errors that only affect a few adapters (including direct to PSU cables), or user error on insertion, or both.

 

[Keng Yuan]

Always thought the internal power cable connector designs are stupid as heck since like 30 years ago when I was putting together computers back then,

 

[helper800]

The interesting thing is that 12-pin cables tend to insert easier than 16-pin, from personal experience. Given tens of thousands of RTX 4090 cards potentially in the hands of users, is it possible some of them won't push the connector all the way in? I'd say absolutely. Compared to an 8-pin or 12-pin, I probably have to push twice as hard to get the click with 16-pin. So from that perspective, Jonny Guru may be absolutely correct.

I know that every time I test a different 4090 card, I'm double and triple checking that the connection is all the way in. Because for now, that's all I can do short of just not testing the cards, and I'm not ready to do that. Also, if one of the cards I'm reviewing then has a meltdown, I can be sure that it wasn't from an improperly inserted cable.

The thing is that no one will come forward and publicly admit to having "inserted it wrong." All you get is a melted connector and a Reddit post with a picture or two after the melting has occurred. Does that mean for sure that every melted connector is from user error? No. But the lack of complaints from the past two years of 12-pin cables and adapters does make the whole situation rather suspect. It's either manufacturing errors that only affect a few adapters (including direct to PSU cables), or user error on insertion, or both.

To include another for the list of potentials, the port itself on the card could be the issue in one of the cases. I used to work in quality control at an aerospace company and even with very tight manufacturing tolerances and high quality materials used to make a product you still always get 1-2% of products getting rejected from being sold and get recycled due to defects. I cant imagine the manufacturers of the ports and cables are near 100% quality in their manufacturing.

 

[InvalidError]

even with very tight manufacturing tolerances and high quality materials used to make a product you still always get 1-2% of products getting rejected from being sold

1-2% for finished, fully-assembled products that managed to slip through QA. The connectors are only a single component of such a product and if you count everything down to SMD resistors, then you need a ~0.003% individual component failure rate to hit that 1% premature failure rate on finished products.

 

[Svarra42]

Ya, no.... every plug I've used on hundreds of builds doesn't have that problem. I've not plugged in my connector many times all the way and never once have I had a issue with melting. If your paying that much for a card it shouldn't melt period.

The fact is and will remain those plugs are faulty.

 

[watzupken]

I believe there may be a subset of people that may not have connected the power connector firmly. But his explanation is unlikely to cover all the scenarios because I am pretty sure we have seen experienced reviewer/ testers running into the same issue. Some blame the cable quality, some connector quality, some says you cannot bend (not even a little bit), here we see poor connection, etc. All these just shows there is no conclusion as to the root cause. Whatever the root cause, I do feel that this connector in its current form is a fail and don't go well with high power draw, even though it is created for this purpose. The reason why I came to this conclusion is because when you have these many exceptions/ potential points of failures to use a cable, and some in which you have no control over, it just don't sound right nor safe to use. This incident reminded me of the infamous Apple antenna gate where Steve infamously said, "you are holding it wrong".

 

[InvalidError]

But his explanation is unlikely to cover all the scenarios because I am pretty sure we have seen experienced reviewer/ testers running into the same issue.

Many "experienced reviewers / testers" have COMMENTED on the issue and shared their thoughts on what they think the root cause might be but I am not aware of any of them having personally witnessed or reproduced it themselves yet.

 

[aberkae]

Jonny Guru just lost all credibility for me. User error and dialectic grease - what nonsense.

last month when I raised my concerns of the issue he downplayed the whole thing on techpowerup just an FYI. He was sure to promote the corsair adapters though!

 

[Co BIY]

A better plug would have some haptic fee back and maybe a little latch to let you know when it was seated properly for a good connection. Think about an RJ45 - or any electrical connector in a car. You always know when they are properly seated.

Why the hate for JonnyGuru - If I had a 4090 I would definitely be double checking for full insertion. If I found it difficult then I would appreciate the di-electric grease suggestion (right lube for purpose). Although I would recommend paying only 99 cents at an auto parts counter.

 

[KTB84]

Source of this revelation (that it was him)? Because I'm pretty sure you just made it yourself to justify baseless accusations.

See -Fran-'s post at the very top of the 2nd page of this thread for a screenshot. In the initial GN video the name was blurred out, but some folks were able to screen capture the original posts, and the comments were definitely made by JonnyGURU. Additionally, I made NO accusations, other than the mention of the GN / Guru3D controversy itself. You're very rude.

 

[KTB84]

Source?

See -Fran-'s post at the very top of page 2.