In a round of questions and answers, NVIDIA CEO Jen-Hsun Huang explained why they chose to continue using GDDR6 memory rather than the HBM memory that AMD people continue to use in their graphics cards.
NVIDIA prefers GDDR6 for cost over HBM
During a recent Q&A session, NVIDIA responded to the possibility of using HBM memory in the near future, for which Jen-Hsun Huang was decisive, suggesting that if everything is fine, the price of memory is HBM’s disadvantage over GDDR6.
“Huang: It’s much more expensive for consumer goods. But storage prices will fall. Everything is fine with HBM. I love HBM. But I love GDDR6 (graphics memory) so much more.”
The GDDR6 memory is used by the new series of GeForce RTX graphics cards based on the Turing architecture. GDDR6 exponentially increases bandwidth over GDDR5, as does HBM memory, but is cheaper to make. Nevertheless, they say it’s 70% more expensive than GDDR5, which would explain the prices we see for the RTX 10 series.
One of the main complaints about AMD’s new product in this segment, the Radeon VII, is the price of $699. It is said that this price is mainly due to the 16GB of HBM memory being used.
Looking at it from this point of view, from a cost point of view, it seems that NVIDIA is very reasonable to rely on this type of memory for its consumer graphics cards.