Testing Burning NVIDIA 12VHPWR Adapter Cable Theories (RTX 4090)
The plot thickens but this is what feels like the most plausible.
The cable Igor got was rated for 150 volts vs the 5 cables GN got and tested with are rated for 300 volts. According to AIBs, the cable that everyone with a 4090 supposed to get is the one for 300 volts. There's also a number of big differences between the way Igors cable was soldered vs how GNs was. GNs cable was soldered in such a way where bending it even very aggressively wouldn't cause it to fail where as with Igors cable, it would. He also mentions that if ATX 3.0 PSU cables are made the same way as the 150 volt cable, those could also fail.
GN intentionally damaged their cable and put it through 8 hour 100% workload and it didn't fail. So it sounds like it could be a bad batch issue which is still pretty bad.
Imo:
What a clusterfuck of a mess. Nvidia needed to do stringent QC on this and not cheap out, especially for 4090s. There shouldn't be a scenario where some users are getting cables with different voltages and more so, significantly different soldering quality. This may be why we aren't seeing any FEs fail or higher end cards like the Strix fail because Nvidia and Asus could be ensuring they are shipping with the correct voltage cables. But in reality, all cables should have the correct voltage cable.
Last edited by Jizz_Beard_thePirate - on 30 October 2022 
PC Specs: CPU: 7800X3D || GPU: Strix 4090 || RAM: 32GB DDR5 6000 || Main SSD: WD 2TB SN850
