Recent content by el etro

2019 but what quarter is a mystery.

With triple the transistor density and double the power efficiency, real challenge will be to keep power consumption on the line. Don't expect Nvidia don't haves a proper answer to this.

I am thinking at 300% of GTX1080ti DX10/11 performance, for the biggest Navi.

Two reasons: GF node instead of TSMC's and underutilization of theorectical FP power. RX480 haves around the same power consumption for the around the same FP power as a GTX 1070. Radeon begs for a real new uarch.

12n may play a part on this power reduction result.

It seems that the 5k shaders are not properly fed in most scenarios.

So Vega was fabbed at the wrong fab and else lacks its most important features? Raja needed a leader to backup his work and to administrate RTG, both things at once was proven he can't do it. Lack of potential of him is highly doubtful otherwise Intel would not hire him as the Chief Engineer/SVP...

So, technically is hard to extract maximum performamce of an Epyc-style GPU.

For the GPU uarch experts: how hard is to write drivers for a chiplet-based Navi? Especially in the case the MC is shared across the four dies of the chiplets.

15W i5.... I don't believe in unicorns. As the AMD chip boosts to out-of-spec power limit to raise its benchmark scores, is naive and partial to believe that KBLr-U doesn't do the same. Intel lost this one, deal with it.

NBC review bashed Mobile Ryzen heavily. Look at the Laptopmag review(R5U versus I5U): https://www.laptopmag.com/articles/amd-ryzen-mobile-vs-intel-8th-gen-core

RX560D exists here in Brazil.

My indignation rises up very much. Why not stay with the 560D naming?

For sure. In practice we have a six-year old architecture that begs for a real revamp. But i discord on power measurements of Nvidia vs. AMD cards. I really don't think that GTX 1080 consumes just ~150W.

Mobile Ryzen review: https://hothardware.com/reviews/ryzen-mobile-benchmarks-and-performance-analysis?page=1