XC5VSX95T-1FFG1136C

Manufacturer: Xilinx
Logic Cells: 98,304
Logic Slices: 15,360
Embedded RAM (eRAM): 3,072 Kb
Package: FFG1136 (Flip-Chip BGA)
Operating Temperature: Commercial (0°C to +85°C)

SEND US A MESSAGE

    Specifications

    MODEL P/N SERIES NUMBER OF LABS/CLBS SPEED GRADE NUMBER OF LOGIC ELEMENTS / CELLS TOTAL RAM BITS NUMBER OF I/O VOLTAGE – SUPPLY MOUNTING TYPE OPERATING TEMPERATURE PACKAGE / CASE SUPPLIER DEVICE PACKAGE
    XC5VSX95T-1FFG1136C Virtex-5 SX 14,72 -1,00 94208 8580000 640 ~1.0 V Surface Mount 0 °C ~ +85 °C (C) 1136-BBGA / FCBGA 1136-FCBGA

    The XC5VSX95T-1FFG1136C is a capable mid-to-high density FPGA from the Virtex-5 SXT family (Xilinx, now AMD), crafted on the reliable 65nm process and tailored for applications heavy on DSP and high-bandwidth serial I/O.

    Engineers usually jump to these key points:

    • ~94,208 logic cells (with 14,720 adaptive logic modules / slices) — decent headroom for multi-channel DSP filters, video processing paths, FFT cores, or complex signal-handling pipelines
    • 16 RocketIO GTX transceivers — these provide solid multi-gigabit serial performance, typically supporting up to ~6.5 Gbps in deployed systems (excellent for 10Gb Ethernet, PCIe Gen1/2, Serial RapidIO, or custom high-speed links)
    • 640 user I/Os — plenty of flexible pins for connecting to external DDR2/3 memory, fast ADCs/DACs, or system interfaces
    • Embedded block RAM around 8.8 Mbit (plus ~1.5 Mbit distributed RAM for quick buffers and small FIFOs)
    • Strong DSP48E slice count (around 488) — second-gen 25×18 multipliers with 48-bit accumulators, great for high-throughput fixed- or floating-point arithmetic, filters, and math-intensive tasks
    • Package: 1136-pin FFG (flip-chip BGA, ~35 × 35 mm), commercial temperature range (0°C to +85°C junction)
    • -1 speed grade — offers balanced timing performance and closure headroom while keeping power draw reasonable for many DSP/SerDes-focused designs

    The SXT series is all about DSP acceleration plus fast transceivers — it prioritizes those DSP48E blocks and GTX ports over embedded processors (no PowerPC here; save that for FXT) or maximum plain logic density (that’s more LX/LXT territory).