Orders placed after 4PM on weekdays will not ship until the next business day. Orders placed after 11AM Fridays will not ship until the following Monday.
Faster shipping methods may be available; just upgrade during checkout.
*Some exclusions apply.
Enjoy Extended Returns thru Feb 1, 2017!
You have until Feb 1, 2017 to return or exchange items purchased between Nov 1, 2016 and Jan 1, 2017. All other standard return policy conditions apply.
Enter new zip code to refresh estimated delivery time.
The FirePro W9000 Workstation Graphics Card from Sapphire is designed for use in applications that demand high-performance dedicated graphics processing power. Although this can card can drive up to six displays at DCI 4K (4096 x 2160) resolution, it is capable of much more than just rendering display graphics. The parallel processing power locked up in the GPU can be leveraged for a wide range of computationally intensive tasks, including transcoding video, rendering 3D objects and environments for CAD, analysis of complex data sets, and more. The card features a 975 MHz GPU clock speed, and has 6GB of GDDR5 RAM over a 384-bit memory interface - providing a memory bandwidth of up to 264 GB/s.
Six mini DisplayPort outputs allow you to connect the card to a variety of display types, including computer monitors, HDTV TVs, digital signage displays, projectors, and many 4K and 3D devices. Two mini DisplayPort to DVI adapters are included. Additionally, the outputs can be adapted to standard DisplayPort and HDMI using separately available adapters or cables. Using DisplayPort 1.2-compliant displays the card supports a maximum resolution of 4096 x 2160 and with DisplayPort 1.1 displays, 2560 x 1600. There is a stereo connector to trigger left/right eye alterations in select active 3D systems.
Popular API Support
DirectX 11, OpenCL 1.2, OpenGL 4.2 and Shader Model 5.0 support allows GPU-aware applications to benefit from the acceleration provided by a dedicated graphics card, taking much of the workload off the server's CPU
4.0 TFLOPS Performance
This card is capable of processing up to 4.0 trillion single-precision floating point operations per second as well as 1.0 trillion double-precision floating point operations per second
AMD Eyefinity Technology
This card supports up to six displays connected concurrently. Eyefinity helps you manage multi-display setups, providing flexible configuration options beyond what the operating system offers for applications as diverse as video walls to immersive viewing
Allows the GPU to process geometry data at a rate of twice per clock cycle resulting in a doubling in the rate of primitive and vertex processing. Triangle rates increase two-fold relative to a GPU that does not possess GeometryBoost
Partially Resident Textures (PRT)
PRT can utilize texture files up to 32 terabytes large typically with minimal performance impact. PRT accomplishes this by streaming small bits of these massive textures into the GPU as needed, giving compatible applications a virtually endless supply of unique texture data
AMD Power Tune
Dynamically boosts the core clock speed up to 30% wherever possible to allow applications to maximize the processing power they have available while reducing it when not required to conserve energy
A unified driver across the entire FirePro line simplifies IT maintenance and administration
AMD Graphics Core Next (GCN) Architecture
Features a 28nm core architecture designed to increase efficiency by putting more transistors in less space. In addition, the architecture is engineered to be capable of performing many of the processing tasks that previously could only be handled by a computer's CPU
Enables low latency peer to peer data
transfers between devices on the bus and AMD
FirePro GPUs. Devices supporting DirectGMA can
write directly into the local memory of the GPU and
vice versa - the GPU can directly access the memory
of a peer device
By enabling application access to the frame
buffer before content reaches the display engine,
ISVs are granted the ability to create new GPU-based solutions
As a user moves between applications, or opens
new ones, the graphics driver settings are
automatically configured for maximum performance
Full 30-Bit Display Pipeline
When connected to 30-bit displays, enables four times more color values than 24-bit
devices for more accurate color reproduction and
superior visual fidelity
Use HydraVision to manage desktop displays and
workspaces in multi-display environments:
Explicitly set the behavior of pop-up windows
and dialogs in different applications though
Provide multiple layers of the desktop workspace
through Multi Desktop
Restrict application window placement across
multiple desktops through HydraGrid
Give the card the ability to power down when any attached displays are off - also known as the "long idle state"
Genlock facilitates synchronization to external sources, such as a common house clock in a broadcast studio. Meanwhile Framelock synchronizes 3D rendering across multiple GPUs installed in different machines
Video Codec Engine (VCE)
A multi-stream hardware H.264 HD encoder, for significantly faster H.264 encoding than is generally possible when relying on the CPU
ECC Memory Support
Designed to help ensure the accuracy of computations by correcting single or double bit errors caused by naturally occurring background radiation
PCI Express 3.0 x16 Support
Provides twice the bandwidth of PCI Express 2.0; for example, an x8 PCI Express 3.0 slot has the same bandwidth as a x16 PCI 2.0 slot. In addition, the card remains backward-compatible with motherboards using earlier generations of PCI Express
1 x Free PCI-E x16 expansion slot - use PCI-E 3.0 x16 for best performance
2 x PCI-E slots physical
1 x 8-pin PCI-E power connector
1 x 6-pin PCI-E power connector
2 GB or more system memory
Internet connection for driver installation
Windows (32- and 64-bit): 8, 7, Vista, XP
Linux (32- and 64-bit): Support varies depending on distribution