Delivering a Generation Ahead at 28nm
Xilinx is delivering a generation ahead in 3 areas:
- Portfolio: All Programmable FPGAs, SoCs, and 3D ICs available today
- Product: An extra node of performance, power and integration
- Productivity: Unmatched time to integration and implementation
Portfolio: All Programmable 28nm FPGAs, SoCs, and 3D ICs Today
The Industry’s Only Optimized Performance/Watt All Programmable FPGA Family
Xilinx provides an extra generation of performance/watt by combining innovations in process technology (with TSMC) and numerous design optimizations for maximum performance, maximum I/O and SerDes bandwidth, and minimal static and dynamic power.
The Industry’s Only All Programmable SoC
Xilinx SoCs represent the first hardware, software, and I/O programmable devices in the world. Through the integration of CPU, DSP, logic, and mainstream Analog Mixed Signal (AMS), Xilinx provides an extra generation of system performance, integration, and lower power.
The Industry’s Only All Programmable Homogeneous and Heterogeneous 3D ICs
Xilinx All Programmable 3D ICs more than double the logic capacity, system level performance, and level of systems integration beyond what is possible with a monolithic solution; providing an extra generation of value.
Product: An Extra Node of Performance, Power and Integration
Across every critical element, Xilinx provides an extra 1.2x performance, integration, and power advantage; which is effectively a generation ahead in system value.
Productivity: Delivering Unmatched Time to Integration & Productivity
Xilinx developed from the ground up a next generation design environment and tool suite. This development took over 4 years of time and 500 man-years. Without this design suite, Xilinx’s 3D ICs could not be effectively leveraged. For FPGAs and SoCs, the new design suite further improves the quality of results (QoR) of designs by up to 3 speed grades, cuts dynamic power by up to 50%, improves routability and resource utilization by >20%, and speeds time to integration and implementation up to 4x.