Jensen didn't want to comment on when the Fermi GT300 will launch, but we anticipate that it should happen within a few months. We don't know if this particularly means late Q4 or early Q1 2010 though. However, one thing that he has confirmed is that Fermi architecture is very scalable and that it will end up in many GPUs.

Nvidia will implement a top-to-bottom release strategy from high end to entry level. While he didn't talk about it during the keynote presentation, this release strategy also includes a high end dual-GPU configuration that should ship around the same time as the high end single-GPU model.

It is important to note that a chip with 64 clusters can easily be divided in 32, 16 or any similar configuration of clusters.