Unparalleled Compute Performance. Unparalleled Performance per Watt. Do the Math.
Up to 5.07 TFLOPS of peak single-precision and 2.53 TFLOPS peak double-precision compute performance, at 235 Watts maximum power consumption. Even the most compute-intensive workloads are no challenge for the world's most powerful server GPU
28nm Graphic Core Next (GCN) Architecture
The new 28nm GCN Architecture with more efficient process technology puts more transistors in less space, enabling a dramatic increase in processing power and graphics performance.
- Each compute unit can execute instructions from multiple kernels at once
- Design for high-utilization, high-throughput, and multi-tasking
- Dual geometry Engines for faster triangle ("primitive") processing
- First GPU family to fully support OpenCL 1.2 and DirectX 11
Half-Rate Double Precision
Unlike competing server cards that are not optimized for double precision, AMD FirePro new server GPUs offer the most double-precision performance in its class, completing compute-intensive tasks faster than ever before.
AMD STREAM Technology
Powers the ecosystem that enables AMD FirePro S-Series server cards to be used for compute-intensive workflows leveraging the massively parallel processing power of AMD GPUs, and accelerate many applications beyond just graphics.
Full 30-Bit Display Pipeline
With a full 30-bit display pipeline producing more than one billion colors (10-bit per RGB component). The AMD FirePro™ is ideally suited for applications that benefit from accurate color reproduction and superior visual quality.
This state of the art power management technology provides direct control over GPU power usage. Applications enjoy ultimate performance with dynamic clock optimization, while minimizing workstation energy usage.
Quad HD Display Support (4K Support)
4K (4096 x 2160-pixel) support for ultra-high-resolution display devices and advanced configurations through DisplayPort 1.2 and AMD Catalyst Pro drivers ensures future-forward compatibility for advanced workflows.
Error Correcting Code (ECC) Memory
Correct single-bit memory errors in hardware
- ECC memory maintains a memory system effectively free from single-bit errors: the data read from each word is always the same as the data that had been written to it, even if a single bit actually stored, or more in some cases, has been flipped to the wrong state. Some non-ECC memory with parity support allows errors to be detected, but not corrected; otherwise errors that may occur are not detected.
- Electrical or magnetic interference inside a computer system can cause a single bit of dynamic random-access memory (DRAM) to spontaneously flip to the opposite state. It was initially thought that this was mainly due to alpha particles emitted by contaminants in chip packaging material, but research has shown that the majority of one-off ("soft") errors in DRAM chips occur as a result of background radiation, chiefly neutrons from cosmic ray secondaries, which may change the contents of one or more memory cells or interfere with the circuitry used to read/write them.
Server-friendly, low-profile design enables maximum device density and carefree maintenance. Cooled solely by internal server airflow, FirePro GPUs using this approach can effectively reduce complexity and maximize reliability for IT organizations requiring server-based visual computing.