AMD today revealed a new tier of workstation graphics cards. The company announced the upcoming Radeon Pro WX 8200, which is expected to ship to graphics professionals next month. The new card, which is based on AMD’s existing Vega GPU architecture, is a slightly cut down version the flagship Radeon Pro WX 9100, which debuted last summer, for less than half the price of the top-tier card.
AMD’s Radeon Pro WX 9100 was a big jump in performance compared to the company's previous products, but it also carried a $2200 price tag, which of course limits potential customers. The company’s new Radeon Pro WX 8200 looks to offer most of the performance of the WX 9100 for under $1000.
AMD managed to trim the price by a huge margin by cutting the memory capacity and lowering the CU count of the GPU. The Radeon Pro WX 8200 features a Vega 10 GPU with a 1500MHz boost clock, like the WX 9100. However, the GPUs in the new cards have eight of their 64 compute units disabled. The WX 8200 also offers half the memory capacity, but with slightly higher memory bandwidth.
Despite the trimmed specifications, AMD said the new card should perform admirably in advanced workloads such as real-time visualization, virtual reality, and photorealistic rendering. The WX 82200 produces 10.8 TFLOPS of single precision and 21.5 TFLOPS of half precision floating point procedures, compared to the flagship WX 9100’s 12.3 TFLOPS and 24.6 TFLOPS, respectively.
The Radeon Pro WX 8200 also supports the same features and technologies as the higher-end Radeon Pro WX 9100, such as a High Bandwidth Cache Controller (HBCC), Enhanced Pixel Engine, Error Correcting Code (ECC) Memory, and AMD's Secure Processor. AMD said it also plans to have the WX 8200 certified for “many of today’s most popular applications – including Adobe CC, Dassault Systemes SOLIDWORKS, Autodesk 3ds Max, Revit, among others.”
AMD says the Radeon Pro WX 8200 will be available for pre-order for $999 at Newegg on August 13. The company expects to begin shipping cards in early September.
AMD Workstation Card Specification Comparison
|Radeon ProWX 9100||Radeon ProWX 8200||Radeon Pro WX 7100||FirePro W8100|
|Memory Clock||1.89Gbps HBM2||2.0Gbps HBM2||7Gbps GDDR5||5Gbps GDDR5|
|Memory Bus Width||2048-bit||2048-bit||256-bit||512-bit|
|Single Precision||12.3 TFLOPS||10.8 TFLOPS||5.7 TFLOPS||4.2 TFLOPS|
|Half Precision||24.6 TFLOPS||21.5 TFLOPS||5.7 TFLOPS||8.4 TFLOPS|
|ECC||Yes (DRAM)||Yes (DRAM)||No||Yes (DRAM)|
|GPU||Vega 10||Vega 10||Polaris 10||Hawaii|
|Architecture||Vega(GCN 5)||Vega(GCN 5)||Polaris(GCN 4)||GCN 2|
|Manufacturing Process||GloFo 14nm||GloFo 14nm||GloFo 14nm||TSMC 28nm|
|Launch Price (MSRP)||$2199||$999||$799||$2499|
Table credit: Anandtech