Earlier this week, Nvidia unveiled the GeForce GTX Titan. It weilds 2,688 CUDA cores for 4.5 teraflops of processing power. It uses a 384-bit interface for memory at 6 Gigabits for 288 Gigabytes per second memory bandwidth.
The card gets its name from the Titan supercomputer housed at the Oak Ridge National Laboratory in Tennessee. It was mostly funded by the US Department of Energy. The supercomputer has ten-times the processing power of its predecessor and utilizes AMD Opteron CPUs and Nvidia Tesla GPUs to provide a (theoretical) peak of 27 petaFLOPS and was benchmarked to 17.59 petaFLOPS.
The GeForce GTX Titan card uses internal water cooling to improve heat transfer and a design to keep it very quiet. It also utilizes GPU Boost 2.0 to keep the GPU running at its peak for the highest frame rate. It supports 3-way SLI for maximum power.
- NVIDIA® TXAA™ technology
- NVIDIA GPU Boost 2.0
- NVIDIA PhysX® technology
- NVIDIA FXAA™ technology
- NVIDIA Adaptive Vertical Sync
- NVIDIA Surround™
- Support for four concurrent displays including:
- Two dual-link DVI
- DisplayPort 1.2
- Microsoft DirectX 11.1 API (feature level 11_0)
- NVIDIA Project SHIELD™-Ready
- NVIDIA 3D Vision®-Ready2
- NVIDIA SLI®-Ready
- NVIDIA CUDA® technology
- PCI Express 3.0 support
- OpenGL 4.3 support
- OpenCL™ support
The card is expected to be available March 21st. ASUS has a GTXTITAN-6GD5 card available for pre-order through NewEgg for the MSRP $999.99 and EVGA has the 06G-P4-2790-KR card listed at the same price.