AMD introduces sub-$1,000 Radeon Pro WX 8200 at SIGGRAPH

92 0

AMD has announced the latest high-performance addition to its Radeon Pro workstation graphics lineup with the AMD Radeon Pro WX 8200 graphics card, packing heavy workstation graphics punch for under $1,000.

Based on its Vega GPU architecture, the Radeon Pro WX 8200 graphics card should give the performance required to drive increasingly large and complex models through the entire design visualisation pipeline, with planned certifications for many of today’s 3D CAD packages, as well as workloads including real-time visualiaation, physically-based rendering and VR.

The GPU features 8GB of HBM2 Memory, 56 compute units, and 8K display support, while its GPU ‘multitasking’ capabilities are claimed by AMD to be 13 times more powerful than Nvidia’s P5000 card.
AMD states that its High Bandwidth Cache Controller (HBCC) on the Radeon Pro WX 8200 graphics card memory system removes traditional GPU memory capacity limitations, letting engineers and designers work with much larger, more detailed models and assets in real time.

The Vega GPU architecture’s enhanced pixel engine also looks to reduce GPU limitations, increasing efficiency by batching related work into the GPU’s local cache to process them simultaneously.

Also built-in to the card is Error Correcting Code Memory3 – billed as helping guarantee the accuracy of computations by correcting any single or double-bit error resulting from naturally occurring background radiation.

“Professionals can fully unleash their creativity with the ‘Vega’ architecture2 at the heart of the Radeon Pro WX 8200 graphics card,” said AMD Radeon Pro general manager Ogi Brkic at the launch.

Advertisement

“This powerful new workstation graphics card empowers creators to improve collaboration among remote teams with VR, create exciting new cinematic experiences and visualise their creations with ease, all at an incredible price point.”

AMD also unveiled major updates to Radeon ProRender, including a plug-in for PTC Creo users.