I’ve said this when the original spec was released, it’s a bad design and a bad spec. 600W of total power on tiny, fragile, endpoints is not a good idea.
I do like the idea of getting rid of the double 8 pin connector, but the new one could have had 30%< 50% more area, been way more secure, and STILL more compact than the old one
NVIDIA enjoys apple like religious fervor from its users. Anyone with two neurons to rub together could see this connector was a bad idea… Nobody from the consortium has implemented it other than NVIDIA.
I’ve said this when the original spec was released, it’s a bad design and a bad spec. 600W of total power on tiny, fragile, endpoints is not a good idea.
That’s what annoys me.
I do like the idea of getting rid of the double 8 pin connector, but the new one could have had 30%< 50% more area, been way more secure, and STILL more compact than the old one
NVIDIA enjoys apple like religious fervor from its users. Anyone with two neurons to rub together could see this connector was a bad idea… Nobody from the consortium has implemented it other than NVIDIA.