Samsung has just announced that it is increasing the production of its 8GB HBM2 chips, in order for the company to "address rapidly growing market demand". Remember, that Samsung isn't providing HBM2 to AMD for their new Vega GPU architecture, but NVIDIA is tapping Samsung's own HBM2 chips.
The increased HBM2 production is to ensure that important and emerging markets like AI, HPC, advanced graphics, network systems, and enterprise servers are flooded with 8GB HBM2 chips.
Jaesoo Han, Executive VP of Memory Sales and Marketing for Samsung Electronics said: "By increasing production of the industry's only 8GB HBM2 solution now available, we are aiming to ensure that global IT system manufacturers have sufficient supply for timely development of new and upgraded systems. We will continue to deliver more advanced HBM2 line-ups, while closely cooperating with our global IT customers".
Now, before you get all excited and think this means that there will be a million Radeon RX Vega graphics cards released in the coming months because Samsung is ramping up 8GB HBM2 production... remember that Samsung is the main provider of HBM2 to NVIDIA.
NVIDIA uses HBM2 on their high-end Volta GPU architecture, including the new Tesla V100 graphics card. I'm sure NVIDIA will release a new HBM2-based consumer GeForce graphics card next year alongside GDDR6-based cards... but this is something to look forward to in 2018, and beyond.