It doesn't mean that Nvidia will use sk hynix instead of micron. Considering that they partnered to endeavor to alleviate the memory supply pressure, why would they then simultaneously create a single point reliance and choose to limit their supply? I'm other words, micron is still required in its full capacity.