Technology|dsp_central

XtremeDSP 48 Slice

High performance XtremeDSP™ DSP48 slices allow designers to implement multiple slower operations using time-multiplexing methods. They provide:

  • Improved flexibility and utilization
  • Improved application efficiency
  • Reduced overall power consumption
  • Increased maximum frequency
  • Reduced set-up plus clock-to-out time
  • Support for many independent functions, including multiply, multiply accumulate (MACC), multiply add, three-input add, barrel shift, wide-bus multiplexing, magnitude comparator, bit-wise logic functions, pattern detect, and wide counter
  • Support for cascading multiple DSP48s slices to form wide math functions, DSP filters, and complex arithmetic without the use of general FPGA fabric
XtremeDSP DSP48s Slice Comparison
Function Spartan®-3A DSP DSP48A Virtex®-4 FPGA
DSP48
Virtex-5 FPGA DSP48E
Multiplier 18 x 18 18 x 18 25 x 18
Pre-adder Yes No No
Cascade inputs One One Two
Cascade output Yes Yes Yes
Dedicated C input Yes No Yes
Adder 2 input 48-bit 3 input 48-bit 3 input 48-bit
ALU logic functions No No Yes
Pattern detect No No Yes
SIMD ALU support No No Yes
Carry signals Carry in Carry in Carry in and out
RTL support Main functions + pre-add Main functions Main functions

XtremeDSP DSP48E Slices for the Virtex-5 Family

DSP48E slices, available in all Virtex-5 devices, accelerate algorithms and enable higher levels of DSP integration and lower power consumption than previous-generation Virtex devices. They:

  • Support over 40 dynamically controlled operating modes including: multiplier, multiplier-accumulator, multiplier-adder/subtractor, three input adder, barrel shifter, wide bus multiplexers, wide counters, and comparators.
  • Enable efficient adder-chain architectures for implementing high-performance filters and complex math efficiently.
  • Draw only 1.38 mW/100 MHz at a toggle rate of 38% - a 40% reduction from previous-generation slices.

t

Virtex-5 FPGA DSP48E Feature Benefits
Feature Benefit
25-bit by 18-bit, two's complement multiplier with full precision 48-bit result Enable higher precision for greater dynamic range, single-precision floating-point math, and wide filters with fewer slices.
Enhanced second stage

Enable three input, flexible 48-bit adder/subtracter with optional registered accumulation feedback.

Implement pattern detector for convergent rounding, underflow/overflow detection for saturation arithmetic, and auto-resetting counters/accumulators.

Support SIMD operations.
Over 40 dynamic user-controller operating modes Adapt DSPE slice functions from clock cycle to clock cycle.
18-bit B cascade routing Support input sample propagation.
New 30-bit A cascade routing Enable advanced filter implementations and reduce power.
Independent, 48-bit C input Multiply, add, use large three-operand addition, or flexible rounding mode. Increase usability by eliminating sharing of C input across slices to simplify design and increase performance.
Cascading, 48-bit P bus Support output propagation of partial results.

XtremeDSP DSP48 Slices for the Virtex-4 Family

DSP48 slices are available in all Virtex-4 family members to accelerate algorithms and solve complex DSP challenges. They provide:

  • 500 MHz performance independently or when combined within a column to implement DSP functions.
  • 2.3 mW/100 MHz power consumption per slice, at a typical toggle rate of 38%.
  • Support for over 40 dynamically controlled operating modes including; multiplier, multiplier-accumulator, multiplier-adder/subtracter, three input adder, barrel shifter, wide bus multiplexers, or wide counters.
  • DSP48 slice cascading without using device fabric or routing resources to perform wide math functions, DSP filters, and complex arithmetic.

t

XtremeDSP DSP48 Slice Highlights
  • 18-bit by 18-bit, two's complement multiplier with full precision 36-bit result, sign extended to 48 bits.
  • Three input, flexible 48-bit adder/subtracter with optional registered accumulation feedback.
  • Over 40 dynamic user-controller operating modes to adapt XtremeDSP Slice functions from clock cycle to clock cycle.
  • Cascading, 18-bit B bus, supporting input sample propagation.
  • Cascading, 48-bit P bus, supporting output propagation of partial results.
  • Multi-precision multiplier and arithmetic support with 17-bit operand right shift to align wide multiplier partial products (parallel or sequential multiplication).
  • Symmetric intelligent rounding support for greater computational accuracy.
  • Performance-enhancing pipeline options for control and data signals are selectable by configuration bits.
  • Input port "C" typically used for multiply, add, large three-operand addition or flexible rounding mode.
  • Separate reset and clock enable for control and data registers.

XtremeDSP DSP48A Slices for the Spartan-3A DSP Family

The 250 MHz DSP48A Slice provides an 18-bit x 18-bit multiplier, 18-bit pre-adder, 48-bit post-adder/accumulator, and cascade capabilities for various DSP applications.

dsp48 Slice

DSP48A Slice Highlights
  • 18-bit by 18-bit, two's complement multiplier with full precision 36-bit result, sign extended to 48 bits.
  • Pre-adder saves 9 logic slices per DSP48A used.
  • Two input, flexible 48-bit adder/subtracter with optional registered accumulation feedback.
  • Cascading, 18-bit B bus, supporting input sample propagation.
  • Cascading, 48-bit P bus, supporting output propagation of partial results.
  • Multi-precision multiplier and arithmetic support with 17-bit operand right shift to align wide multiplier partial products (parallel or sequential multiplication).
  • Symmetric intelligent rounding support for greater computational accuracy.
  • Performance-enhancing pipeline options for control and data signals are selectable by configuration bits.
  • Input port "C" typically used for multiply, add, large three-operand addition or flexible rounding mode.
  • Separate reset and clock enable for control and data registers.
/csi/footer.htm