Nvidia has spent several weeks examining melted RTX 4090 adapter cables and has finally come to a conclusion. It turns out people weren’t plugging them in properly, as was reported by several outlets previously. Though the solution to the mystery is rather simple, and we imagine quite embarrassing for those affected, it still raises questions about Nvidia’s adapter design.
Team Green posted the results of its findings to its support site. If you were expecting an elaborate breakdown of its investigation, you will be disappointed. Nvidia’s post is surprisingly light on specifics, but Gamers Nexus was able to confirm some additional details. The gist is that Nvidia is aware of 50 instances of cables melting, which works out to a 0.04 percent failure rate based on discussions Gamers Nexus has had with Nvidia’s partners. According to Nvidia, the adapter can melt when it’s not fully plugged into the GPU. This is apparently a “common issue,” according to the update; users may think it’s fully seated, and the card is operational despite it not being fully inserted. Nvidia provided a clear illustration of what this can look like. It says all the melted cables it examined showed a clear “wear line” that indicated it was not fully seated:
In other words, it’s down to user error, but the design of the connector makes it difficult to fully insert, as it seems to require more force than people are used to using. Also, given the 4090’s $1,599 price tag, it’s understandable people are being careful. To remedy the situation, Nvidia advises people to plug the adapter into the GPU first. This allows you to examine the connection from all angles so you can ensure it’s connected properly. Once you know it’s secure, then plug the GPU into the PCIe slot.
Of course, this is cold comfort for those with melted GPUs. After all, it wasn’t like they were trying to melt their brand new video card. For those affected, Nvidia says it will prioritize RMAs for anyone dealing with the issue. That will apply to any card and cable combination as well, which is a laudable development. Nvidia says it’s also looking at ways to make sure the connector is secured before the card receives power.
Now, this raises a few questions about Nvidia’s adapter. First, why is the card even allowed to operate when the connector isn’t fully seated? We’re not engineers, but it sounds like a change to the cable that would prevent it from receiving power if a certain connection isn’t made would be a sound remedy. Gamers Nexus said this could be accomplished by shortening the length of the sense pins on the belly of the 16-pin cable. This would prevent it from powering on unless it was fully inserted.
Finally, why was this literally never an issue on any GPU prior to this generation? It seems like the design of the adapter is the issue, as previously it was reported that Corsair’s PSU chief described it as being very difficult to insert fully. Therefore, we’re left to conclude that it can be both user error as well as a miscalculation on Nvidia’s part in the adapter design.
For now, we can put this controversy to bed it seems. The likely cause has been unearthed, and Nvidia says it’ll take care of all affected customers. All’s well that melts well, we guess. The only remaining unknown is what changes — if any — Nvidia will make it its 40-series adapter cable. Since the problem is minuscule in scale, we can see Nvidia just putting a label on future adapters warning of the risks of incomplete insertion. That seems less expensive than redesigning the darn thing.
Now Read:
- Class Action Suit Against Nvidia Filed Over Melted RTX 4090s
- Redditor Enables DLSS 3 on Nvidia Turing GPU With Simple Config File
- Nvidia RTX 4090 Sets New Records in Password Cracking Benchmarks
from ExtremeTechExtremeTech https://ift.tt/eMOtZNU
ليست هناك تعليقات:
إرسال تعليق