[Manuel Calavera]

Speaking with PCPer’s Ryan Shrout*after the event, Raja revealed a bunch of interesting stuff about Polaris that wasn’t touched upon at the event. Chief among which is performance per dollar and market positioning.*To that effect*Raja said that he’d like to bring 14nm technology and all its goodies to as many people as possible.*Furthermore, Raja said to*expect Polaris based graphics cards cross the entire performance stack. From the entry level to the very top-end. He also*stressed that everyone will be “very pleased” and “surprised” at what*he*and his team are going to do in terms of positioning.

...

Raja also spoke about the pivotal change in the economics of GPU manufacturing that’s taking place. Where larger*GPUs are becoming*exponentially more difficult to produce and smaller dies yielding much better in comparison than ever before. It’s an infliction point in moore’s law that’s going to have a long-term effect extending into 2017, 2018, 2019 and beyond. The economics of smaller dies have always been better than larger dies. This has been true forever, but as process technologies grow even more complex and expensive it*gives all the more*reason*to chip makers to favor smaller dies now more than ever before.

...

Raja continued to outline his belief that multi-GPU solutions will be much more prominent in the future, extending up and down the stack rather than be relegated to the high-end enthusiast segment.*

...

Finally, Raja confirmed that indeed HBM2 will be coming next year with Vega but not before then. A decision that was made because HBM2 is simply not ready for 2016 from an economic standpoint Raja explained. It wasn’t going to allow him and his team to make a compelling performance per dollar product in 2016 due to the high costs associated with it in its early ramp. Thankfully however, we’re not going to have to wait until 2017 for Polaris.

Kao što sam pisao na pascal temi - prerano je za hbm2 grafe